The Time is Now How Can Uzbekistan Leverage Urbanization as a Driver of Sustainable Development? 2022 © 2022 International Bank for Reconstruction and Development / The World Bank 1818 H Street NW, Washington DC 20433 Telephone: 202-473-1000; Internet: www.worldbank.org Some rights reserved. This work is a product of the staff of The World Bank with external contributions. The findings, interpretations, and conclusions expressed in this work do not necessarily reflect the views of The World Bank, its Board of Executive Directors, or the governments they represent. The World Bank does not guarantee the accuracy of the data included in this work. The boundaries, colors, denominations, and other information shown on any map in this work do not imply any judgment on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. Nothing herein shall constitute or be considered to be a limitation upon or waiver of the privileges and immunities of The World Bank, all of which are specifically reserved. Rights and Permissions This work is available under the Creative Commons Attribution 3.0 IGO license (CC BY 3.0 IGO) http://creativecommons. org/ licenses/by/3.0/igo. Under the Creative Commons Attribution license, you are free to copy, distribute, transmit, and adapt this work, including for commercial purposes, under the following conditions: Attribution—Please cite the work as follows: Sivaev, D., Kamilov I., Rossitti G., Orlova, N., Vaggione 2022, P. The Time is Now: How Can Uzbekistan Leverage Urbanization as a Driver of Sustainable Development? Washington, DC: World Bank. License: Creative Commons Attribution CC BY 3.0 IGO. Translations—If you create a translation of this work, please add the following disclaimer along with the attribution: This translation was not created by The World Bank and should not be considered an official World Bank translation. The World Bank shall not be liable for any content or error in this translation. Adaptations—If you create an adaptation of this work, please add the following disclaimer along with the attribution: This is an adaptation of an original work by The World Bank. Views and opinions expressed in the adaptation are the sole responsibility of the author or authors of the adaptation and are not endorsed by The World Bank. Third-party content—The World Bank does not necessarily own each component of the content contained within the work. The World Bank therefore does not warrant that the use of any third-party-owned individual component or part contained in the work will not infringe on the rights of those third parties. The risk of claims resulting from such infringement rests solely with you. If you wish to re-use a component of the work, it is your responsibility to determine whether permission is needed for that re-use and to obtain permission from the copyright owner. Examples of components can include, but are not limited to, tables, figures, or images. All queries on rights and licenses should be addressed to World Bank Publications, The World Bank Group, 1818 H Street NW, Washington, DC 20433, USA; e-mail: pubrights@worldbank.org. Contents ACKNOWLEDGEMENTS..................................................................................................................................................... 9 EXECUTIVE SUMMARY....................................................................................................................................................... 11 ACRONYMS..........................................................................................................................................................................14 CHAPTER 1. INTRODUCTION...........................................................................................................................................17 CHAPTER 2. MEASURING URBANIZATION IN UZBEKISTAN AND ITS DEVELOPMENT OUTCOMES................23 2.1. Cities in Uzbekistan are yet to contribute as drivers of economic growth and structural transformation .......................................................................................................................................................23 2.2. Understanding urbanization in Uzbekistan..............................................................................................26 CHAPTER 3. KEY URBANIZATION PATTERNS THAT EXPLAIN UNDERWHELMING OUTCOMES ......................43 3.1 Distortions to population mobility limit cities’ contribution to the national economy and signal the inefficiency of factor markets.................................................................................................43 3.2 The way cities in Uzbekistan are spatially organized is affecting urban productivity, livability, and sustainability.................................................................................................................................45 CHAPTER 4. KEY FACTORS THAT DISTORT URBANIZATION IN UZBEKISTAN.....................................................57 4.1 Critical structural constraints to the development potential of urbanization.................................57 4.2. Policies and other conditions leading to inefficient urban growth...................................................62 4.2.1. What conditions are preventing densification? .......................................................................63 4.2.2 What conditions promote sprawl?................................................................................................ 67 4.2.3 Underdeveloped land markets are impeding the efficiency of land use in cities............. 72 4.2.4 Urban planning system is inadequate for the needs of cities............................................... 75 4.3 Local government shortcomings that impede potential .....................................................................80 4.3.1. Local government finance challenges.......................................................................................... 81 4.3.2 Local government capacity constraints...................................................................................... 86 THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT? 3 CHAPTER 5. POLICY PRIORITIES FOR SUSTAINABLE AND PRODUCTIVE URBANIZATION.............................95 5.1 National urban and territorial development strategy ...........................................................................95 5.2 Empowering subnational governments ....................................................................................................98 5.3 Critical structural reforms ............................................................................................................................103 5.3.1 Developing land markets ................................................................................................................103 5.3.2 Reforming urban land use planning.............................................................................................109 CHAPTER 6. CONCLUSION.............................................................................................................................................121 REFERENCES.......................................................................................................................................................................127 APPENDICES Appendix A . Urban sprawl in Uzbekistan, additional analysis .............................................................................137 Appendix B. Using night-time lights to measure city productivity......................................................................141 Appendix C. Analysis of productivity elasticity of Uzbek cities............................................................................143 Appendix D. Effect of adopting a different definition of cities and urban areas on perceptions of Uzbekistan’s urban system...........................................................................................................145 Appendix E. Tashkent relative to its comparators ...................................................................................................149 Appendix F. Analysis of the relationship of population growth to access to markets in the cities of Uzbekistan...............................................................................................................................................151 Appendix G. Analysis of historic locations of human settlements in Uzbekistan..............................................157 Appendix H. Analysis of the link between urban shape and productivity.........................................................163 Appendix I. Urban expansion forecast methodology ............................................................................................165 Appendix J. Analysis of specialization and productivity of cities in Uzbekistan..............................................167 Appendix K. Draft outline of a comprehensive land reform roadmap for Uzbekistan....................................171 FIGURES Figure 1.1 Correlation between economic growth and urbanization, 1960–2020............................................17 Figure 1.2 Urbanization of Central Asian countries (and China) ...........................................................................18 Figure 1.3 Growth of urban and rural populations in Uzbekistan, 2010–2019....................................................19 Figure 2.1 Urbanization as the driver of the transition to high- and upper-middle income in Brazil, Malaysia, and Turkey.......................................................................................................................................24 Figure 2.2 Role of larger cities in driving economic growth in Brazil, Malaysia, and Turkey...........................24 Figure 2.3 Contribution of major industries to GDP in Uzbekistan vs ECA comparators................................25 Figure 2.4 Structural transformation in countries of the world by level of urbanization.................................25 4 THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT? Figure 2.5 Share of the service sector in economies of the largest global cities.............................................26 Figure 2.6 Population growth in Uzbekistan’s urban areas, 2010–19 (%).............................................................27 Figure 2.7 Agglomeration around Samarkand ..........................................................................................................28 Figure 2.8 Agglomeration around Andijan..................................................................................................................29 Figure 2.9 Urban population’s share using administrative vs. gridded population data in Europe and Central Asia.................................................................................................................................................29 Figure B2.1.1 Difference between administrative, urban agglomeration, and urban center definitions of cities and urban areas in Uzbekistan ................................................................................................31 Figure 2.10 GDP per capita growth in urban centers in Europe and Central Asia ..........................................31 Figure 2.11 Distribution of cities’ productivity in five Central Asian nations.......................................................32 Figure 2.12 Productivity of cities in Uzbekistan..........................................................................................................33 Figure 2.13 Relative size of cities in Central Asian countries..................................................................................34 Figure 2.14 Benchmarking Uzbekistan using various measures of urban primacy...........................................35 Figure 2.15 Service accessibility in the core vs peripheries of urban clusters....................................................36 Figure B2.2.1 Night-time lights in urban centers and agglomerations, January 2020–March 2021 compared to the same months in 2019....................................................................37 Figure B2.2.2. Night-time lights in urban centers, January 2020–March 2021 compared to the same months in 2019............................................................................................................................................39 Figure 3.1 Five-year internal migration rates in selected countries (rough estimates)....................................44 Figure 3.2 Urban shape of cities in Uzbekistan and other low- and middle-income countries in Europe and Central Asia.............................................................................................................................................47 Figure 3.3 Urban land use pattern metrics of cities in Uzbekistan and other low- and middle-income countries in Europe and Central Asia..........................................................................48 Figure 3.4 Urban structure metrics of cities in Uzbekistan and other low- and middle-income countries in Europe and Central Asia............................................................................................48 Figure 3.5 Fullness index and productivity of urban areas in Uzbekistan compared to the rest of the region...................................................................................................................................................49 Figure 3.6 Comparison of different models of urban growth in Atlanta and Barcelona................................50 Figure 3.7 Built-up expansion forecast for selected cities in Uzbekistan by 2050...........................................51 Figure 3.8 Costs of urban infrastructure provision under different models of accommodating population growth...........................................................................................................................52 Figure B3.1.1 Population growth forecasts for the 10 selected agglomerations...............................................53 Figure B3.1.2 Three scenarios of accommodating population growth in the largest agglomerations of Uzbekistan......................................................................................................................................53 Figure B3.1.3 Costs of infrastructure provision costs in the largest agglomerations of Uzbekistan across different growth scenarios......................................................................................................54 THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT? 5 Figure 4.1 The effect of propiska restrictions on population growth across the country................................ 59 Figure B4.1.1 Share of people living in the largest city (urban primacy) of former Soviet countries............ 61 Figure B4.1.2 Actual and predicted size of the largest city..................................................................................... 62 Figure B4.2.1 Map of built-up area expansion in and around Karshi.................................................................... 65 Figure B4.3.1 Map of built-up area expansion in and around Namangan.......................................................... 70 Figure 4.2 The hierarchy of land use planning documents in Uzbekistan........................................................... 76 Figure B4.6.1 Map of expansion in Kasansay............................................................................................................... 79 Figure 4.3 Revenues and spending of subnational governments in 2020 (UBZ billion).................................... 84 Figure 4.4 Location and economic performance of Piskent and Akhangaran.................................................. 87 Figure B4.8.1 Share of Akhangaran city in the industrial production of Tashkent region, 2020 (% of total).................................................................................................................................................................. 89 Figure B4.9.1 Share of Piskent district in the industrial production of Tashkent region, .................................... 91 2020 (% of total)................................................................................................................................................................... 91 Figure 5.1 National Urban Policy development process........................................................................................... 96 Figure B5.2.1 Land privatization in Georgia (ha per year) ......................................................................................105 Figure 5.2 Global perspective of a modern land management system.............................................................106 Figure B5.3.1 The structure of the urban planning system....................................................................................... 111 Figure B5.3.2 The historic evolution of the urban planning system in Korea ...................................................... 112 Figure A.1 Population growth and urban sprawl (built-up area expansion) in cities of Uzbekistan and the ECA region (non-high income) ......................................................................................................................137 Figure A.2 Change of density in cities of Uzbekistan and Central Asia..............................................................138 Figure A.3 Map of extensive urban area forming Karshi ........................................................................................139 Figure A.4 Merging urban agglomerations: Number of urban clusters in 2000 and 2015..............................139 Figure B.1 Night-time lights per capita and GDP per capita in cities.................................................................141 Figure B.2 Comparing different night-time lights-based measures of productivity.......................................142 Figure C.1 City-level agglomeration elasticities in Uzbekistan............................................................................143 Figure C.2 Population growth and agglomeration elasticity at the city level..................................................144 Figure D.1 Fergana Valley’s administrative boundaries of cities and built-up areas......................................145 Figure D.2 Fergana Valley’s urban agglomerations.................................................................................................145 Figure D.3 Uzbekistan’s functional urban areas........................................................................................................146 Figure D.4 Uzbekistan’s urban regions.........................................................................................................................146 Figure D.5 Zipf’s curve representing Uzbekistan’s urban system ..........................................................................147 Figure E.1 Key indicators of Tashkent’s growth relative to its closest comparators, 1990–2015....................149 Figure G.1 Map of the historic tract of the Silk Road, and current night-time lights emissions in Uzbekistan.................................................................................................................................................... 161 6 THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT? Figure H.1 Results of regression analysis of effects of various urban shape and structure indicators on productivity or urban areas in Uzbekistan and the ECA region......................................................................163 Figure J.1 Extractive specialization and output per capita................................................................................... 167 Figure J.2 Extractive specialization and productivity of local economies.........................................................168 Figure J.3 Districts’ specialization and productivity................................................................................................ 169 Figure J.4 Specialization and productivity by district type................................................................................... 169 BOXES Box 2.1 The definitions of urban areas used in this report....................................................................................... 30 Box 2.2 COVID-19 had a relatively large impact on cities, but is unlikely to diminish their importance..... 37 Box 3.1 Estimated costs of urban expansion .............................................................................................................. 53 Box 4.1 What the size of major cities in post-Soviet countries tells us about the impact of the propiska regime........................................................................................................................................ 61 Box 4.2 Karshi—a large city where expansion is driven by poor infrastructure and long-term failures of urban planning ........................................................................................................................... 65 Box 4.3 Namangan—a large city stuck in an inefficient growth pattern.............................................................. 69 Box 4.4 Role of land markets in urban transformation............................................................................................. 72 Box 4.5 Summary of Law No. 728 on Privatization of Non-Agricultural Land Parcels........................................ 74 Box 4.6 Kasansay’s ongoing expansion into agricultural land .............................................................................. 78 Box 4.7 Local budget revenues in Akhangaran city and Piskent district ........................................................... 85 Box 4.8 Akhangaran’s need for greater infrastructure provision........................................................................... 88 Box 4.9 Piskent’s struggle to overcome path dependency ..................................................................................... 91 Box 5.1 International experience of competitions (Russia) and asymmetric devolution of powers (United Kingdom) as drivers of local capacity building ......................................................................101 Box 5.2 Privatization of nonagricultural (urban) land in Georgia..........................................................................104 Box 5.3 Korea’s spatial development process: Providing a vision while addressing evolving qualities of a place .......................................................................................................................................... 111 Box 5.4 Seoul’s urban master plan: Adjusting a spatial strategy to reach urbanization goals.................... 112 Box 5.5 Land readjustment in Japan allows cities to expand in a planned manner and to retrofit areas developed with insufficient infrastructure....................................................................................115 Box 5.6 A general plan is not enough: Negative consequences of lack of enforcement in Vietnam...................................................................................................................................................116 Box F.1 Measuring market access using different measures..................................................................................153 Box F.2 Overcoming data limitations: Market access using night-time lights..................................................155 Box G.1 The impact of the Silk Road on the location of cities in Uzbekistan....................................................160 THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT? 7 TABLES Table B4.3.1 Key characteristics of Namangan, and plans..................................................................................... 72 Table B4.6.1 Key parameters of previous and current general plans of Kasansay...........................................80 Table 4.1 Proposed distribution of tax and nontax incomes among budgets of different levels of government in 2021...........................................................................................................................................82 Table 4.2 Incomes, expenditures, and equalizing interbudgetary transfers planned for 2021 (combining regional and local governments in each region; in UBZ billion).......................................84 Table B4.7.1 Distribution of budget incomes in Akhangaran city and Piskent district, 2021 Q1 (%)...............86 Table A.1 Population elasticity of density...................................................................................................................140 Table F.1 Market access and population growth......................................................................................................151 Table F.2 Market access and population growth decomposed by type of urban places.............................152 Table F.3 Alternative measures of market access.....................................................................................................153 Table F.4 Nearby population changes and population growth in Uzbekistan’s UTSs....................................154 Table F.5 Market access and population growth using night-time lights..........................................................155 Table G.1 Adjusted R-squared results for relationship between log night-time lights and base, agriculture, and trade fundamentals...........................................................................................................................158 Table G.2 Variables’ contribution (Shapley values)..................................................................................................159 Table G.3 Natural determinants of economic activity............................................................................................160 8 THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT? Acknowledgements This report has been prepared by the core study team lead by Dmitry Sivaev (Urban Specialist – Task Team Leader) and including Ildus Kamilov (Senior Urban Specialist), Nadege Orlova (Senior Land Admin- istration Specialist), Giuseppe Rossitti (Urban Economist), Thomas Gertin (Geographer), Pablo Vaggione (Senior Urban Planner), Manjusha Rai (Urban Planner), David Egiashvili (Senior Land Policy Consultant), Bakhrom Khodjaev (Urban Planner). The team is grateful for the important contributions and guidance from Ellen Hamilton (Lead Urban Spe- cialist), Roland White (Lead Urban Specialist), Thomas Farole (Lead Economist), Olga Kaganova (Lead Land Specialist), Luis Quintero (Urban Economist), Rosanna Nitti (Senior Urban Specialist). The material in the study was enriched significantly thanks to peer review input provided by Nancy Lozano (Lead Urban Economist), Ayah Mahgoub (Senior Urban Development Specialist), David Mason (Urban Specialist). This Report benefited immensely from close collaboration with representatives from the Government of Uzbekistan, including the Ministry of Economic Development and Poverty Reduction (MoEDPR), Min- istry of Foreign Investment and Trade (MIFT), Ministry of Construction, Cadastral Agency, Hokimyats of Namangan region, Kashkadarya region, Namangan, Karshi, Kasansay, Piskent, Akhangaran. The team would like to express its gratitude especially to Jamshid Kuchkarov, Deputy Prime Minister and Minister of Economic Development and Poverty Reduction, Ilkhom Norkulov, First Deputy Minister of Econom- ic Development and Poverty Reduction, Adiz Boboev, Deputy Minister of economic development and Poverty Reduction for their close involvement and support, Bakhadir Sadikov from the Ministry of Con- struction for his invaluable contribution and experience sharing and Kamiljon Khanarov, Ravashanbek Rustam-o'g'li, and Umidjon Oktyabrinov for close collaboration and overall coordination of joint work with the Ministry of Economic Development and Poverty Reduction. This report was prepared with support from the Sustainable Urban and Regional Development program (SURGE) https://www.worldbank.org/surge, a World Bank Umbrella Multi-Donor Trust Fund financed by the Swiss State Secretariat for Economic Affairs (SECO). THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT? 9 Executive summary Urbanization presents a huge development opportunity for Uzbekistan. As Uzbekistan urbanizes, and its cities grow, the country can reap substantial development benefits, as many other countries have in the past. With a large share of population still dispersed across small towns and rural areas, it is more than likely that further concentration in bigger cities will create opportunities for advancing productivity and well-being. But this process requires favorable conditions that allow market forces to bring resources to their most productive uses, boosting economies of agglomeration. The first step toward creating such conditions is to gain a deeper understanding of the urbanization process. The dynamics of urbanization in Uzbekistan are not yet well understood. Today, official statistics suggest that only about half of the population lives in urban areas, and their share is declining. Yet, there is good reason to believe that these data do not reflect the current reality. A visit to the country quickly confirms that the cities of Uzbekistan are growing and contain a lot of energy. Thus, even if the estimate offered in this report, of 85 percent urbanization (based on global gridded population data and an international definition of urban areas), is an exaggeration, it points to the fact that urbanization in Uzbekistan is prob- ably more advanced and more dynamic than official statistics suggest. This discrepancy highlights the overall inadequate understanding of key urbanization trends and challenges in the country to date. While in several countries in the world urbanization has been a key driver of growth, it is not a guar- antee, and Uzbekistan has yet to leverage its potential fully. From historic cases of the United Kingdom and the United States to more recent cases of the Republic of Korea, China, and Thailand, countries have proven that urbanizing can help them reach greater levels of productivity and prosperity. However, the experiences of other countries show that urbanization does not always guarantee economic growth and development. Unfortunately, in the recent past Uzbekistan’s economic growth has been decoupled from urbanization, and structural transformation of the economy (associated with urbanization) has been lagging. A lot of this has to do with the performance of cities. In countries that successfully leveraged ur- banization to advance to upper-middle-income status (Brazil, Malaysia, Turkey), large cities led the way in output per capita growth. Uzbek cities need to do the same, but today they are unproductive and unable to generate agglomeration benefits and drive structural transformation. They also struggle to provide ur- ban services, thus failing to attract people with the promise of a better quality of life. Limited labor mobility along with unplanned and poorly managed urban growth are two key symptoms of the underperformance of Uzbekistan’s cities.   Functioning factor markets are critical for unlocking urbanization opportunities. The popula- tion’s mobility is required to make sure the economic potential of various locations is maximized. However, in Uzbekistan limited internal migration leads to a potentially inefficient distribution of population across the urban system, resulting in lost opportunities for individuals and in losses for the economy. Similar inefficiencies plague capital and land markets, creating a barrier for structural transformation and limiting the potential of cities.   Well-managed and properly planned urban growth makes cities attractive and sustainable and mitigates the negative effects of density. Well-planned and managed cities provide conditions that make them attractive to people and businesses and build a foundation for long-term resilient and sustainable development. But in Uzbekistan, cities have been growing through sprawl, making services and jobs harder to access, productivity benefits of agglomeration more muted, service and infrastructure provision extremely expensive and logistically difficult, and thereby, sustainable growth and resilience to climate change harder to achieve. THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT? 11 While structural reforms are making important advances, remaining structural issues impede the mo- bility of factors of production in Uzbekistan. Currently discussed and implemented reforms to liberalize and develop financial markets, improve the business climate, and reduce the role of the state in the econ- omy provide a baseline for the successful development of urban economies. It is critical that their imple- mentation continues. The relaxation of the propiska regime, which is underway, will remove one of the key structural barriers that has been holding large cities back from realizing their potential. But other barriers to domestic migration, related to unaffordable housing and inadequate service provision (among others), still need to be addressed. As barriers are lifted, it is likely that the population will flow into cities; govern- ments should see this as an opportunity, rather than a threat. It is critical to prepare cities to accommo- date growth with appropriate services, infrastructure, and efficient institutions and systems. Getting there requires understanding the root causes of the inefficient urban growth of recent years. The sprawling model of urban growth is a result of a system intended to prevent rather than encourage growth, epitomized by inefficient land markets and an inadequate urban land use planning system. Sticking to the current model of urban sprawl would cost an extra US$1.2 billion in basic infrastructure in- vestments (compared to growth through modest densification) over the next 30 years in just 10 major cities of Uzbekistan. To shift to a more compact model of growth, Uzbekistan needs to reform key systems that shape cities. Land reforms—including urban land privatization—are ongoing, and their completion is critical to ensure that urban land can be reallocated for its most productive and efficient use without barriers. Urban planning is failing to provide solutions to challenges of urban growth as it stays rooted in the Soviet central planning approach, is not responsive to private sector needs for varying and changing land uses, lacks transparency, stringent enforcement, and remains unsupported by adequate staff capacity. The potential of cities is further hindered due to weaknesses of subnational governments that lack the authority, resources, incentives, and technical capacity to become local development champions. The ongoing administrative reform intends to transfer powers and revenues to local governments. But for now, intergovernmental budgetary mechanisms keep evolving and struggle to provide sufficient and predicta- ble revenue streams. Local governments are also facing confusing incentives, and are incapable of initiat- ing substantial local development programs, aside from national initiatives like industrial zones, that tend to underdeliver on their promises. While setting an urbanization target of 60 percent, as the president of Uzbekistan has done, is an im- portant signal of the importance of cities, further reforms are needed to create conditions for market signals to drive urbanization, and the growth of cities, which will help maximize their development po- tential. Three priority areas of reform are further discussed in this report:   Adoption of a national urbanization strategy, rooted in analysis of unbiased data and the real situation on the ground, and offering a vision of the qualities that cities should possess, is needed to set clear policy priorities and coordinate different branches and levels of the government in ad- vancing urbanization-related reforms.   Empowerment of local governments through coordinated delegation of powers and resources (in- cluding ensuring provision of financial capacity for long-term capital investment in infrastructure and services), and through building capacity and managing incentives. This is not a problem that can be solved overnight, but by utilizing international best practices and adopting an iterative adaptation mindset Uzbekistan can ensure that local governments become the leaders of a sus- tainable urbanization process.   Continuation and completion of critical structural reforms will set the foundation on which sus- tainable and productive cities can be developed. Success in further removal of migration barri- ers (including propiska system relaxation and housing reform), financial market development, and 12 THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT? business environment improvements will be paramount for the whole economy, cities included. Two structural reforms in particular can be highlighted as having the most direct relevance to the chal- lenges cities face today:   Development of land markets, through full implementation of initiated reforms and devel- opment of a land management system.   Modernization of urban planning, through adjusting the system to market conditions. Urbanization can still be a boon for development in Uzbekistan, but for that to come true, the Govern- ment of Uzbekistan must act now. As population growth continues, cities have a chance to become bus- tling centers of innovation and ladders to people’s greater well-being. They can provide better access to jobs, education, health care, and other services. Cities can also help Uzbekistan develop in a sustainable and climate-friendly way. This is possible. But it would require allowing people and investments to come to cities where they see opportunities and ensuring that cities create conditions for compact and well-man- aged growth. Global experience shows that this is possible, and the tools have been developed and test- ed. But decisive and coordinated action needs to start now, otherwise Uzbekistan is at risk of missing the historic development opportunity that urbanization provides. THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT? 13 Acronyms APZ Architectural And Planning Task (In Russian) CA Cadastre Agency CNTP Comprehensive National Territorial Plan CSC Chamber Of State Cadastres ECA Europe And Central Asia EU European Union FAR Floor Area Ratio FEZ Free Economic Zone GDP Gross Domestic Product GHS Global Human Settlement GIS Geographic Information System IISRPRC Integrated Information System For Cadastre And Real Estate Registration km2 Square Kilometer LMS Land Management System LR Land Readjustment m2 Square Meter MoEDPR Ministry Of Economic Development And Poverty Reduction NTL Night-Time Lights SIZ Small Industrial Zone SOE State-Owned Enterprise UN United Nations UTS Urban-Type Settlement 14 THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT? Chapter 1 Introduction THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT? 15 Chapter 1 Introduction Urbanization and growth of cities are the defining development trends in Uzbekistan, and the country has a unique opportunity to use them as drivers of development. Uzbekistan, unlike most countries in Europe and Central Asia, is still experiencing rapid natural population growth, of 20 percent between 2010 and 2020. This rapid demographic expansion fuels the growth of cities. Additionally, the official lev- el of urbanization is close to 50 percent, which is relatively low compared to countries with similar levels of development. These two factors suggest that there is likely still substantial scope for further urban growth and the concentration of the population. This is a unique once-in-history opportunity. Urbani- zation and growth of cities can become the drivers of economic development, poverty alleviation, and better well-being as cities offer better access to jobs, infrastructure, and services. But this result is not guaranteed. To make the most of the urbanization opportunity, Uzbekistan needs to learn from global experiences, understand the challenges that growing cities face, and act quickly to create favorable conditions for urbanization. Cities are the economic engines of the world. While just over a half of the world population is urban (UN 2018a), 80 percent of global gross domestic product (GDP) is concentrated in cities (McKinsey & Co 2013). The 750 largest cities account for 60 percent of new private sector jobs in their countries, while accounting for only 25 percent of the total population (World Bank 2015). In many countries of the world, urbanization is closely associated with progress. Generally, as the share of people working in agriculture declines due to growing labor productivity, workers transition to other more productive sectors, and this structural transi- tion drives economic growth. Historically, this was the phenomenon observed during the Industrial Revolu- tion in England. Later, the growth of cities was the engine behind the United States’ rise. More recently, the economic acceleration seen in the Republic of Korea, China, and Thailand has been closely associated with rural-to-urban migration and the rise of cities. And even if the global COVID-19 pandemic adjusts this trajectory (see Box 2.2 in chapter 2 for further discussion), there is little reason to believe that it will diminish the role of cities in the long run. Figure 1.1 Correlation between economic growth and urbanization, 1960–2020 10000 9000 GDP per capita USD$ China 8000 (constant 2010) 7000 Korea 6000 5000 Kyrgyz Republic 4000 Nigeria 3000 2000 Thailand 1000 0 Uzbekistan 10 20 30 40 50 60 70 80 Rate of Urbanization (%) SOURCE: WDI. NOTE: GDP = gross domestic product. NGA = Nigeria; THA = Thailand; UZB = Uzbekistan. THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT? 17 Not all countries have managed to leverage cities’ potential to achieve economic growth and a bet- ter quality of life for the population. Historical data prove that for some countries (e.g., Korea, China, and Thailand), urbanization was closely associated with economic progress and achieving higher GDP per capita. However, this does not hold true everywhere. Nigeria and Kenya saw much less economic progress as they urbanized in recent decades (Figure 1.1). Similarly, while in some countries urbaniza- tion paved a path to prosperity for people, in others (like India or Bangladesh) poorly managed urban growth trapped people in slums with poor access to services or jobs and high levels of deprivation. Un- fortunately, Uzbekistan’s trajectory (particularly up until 2009) resembles the trajectory of countries that experienced urbanization without growth—in other words, a higher level of urbanization did not lead to higher levels of prosperity (Figure 1.1). As Uzbekistan continues to urbanize, what does the country need to do to ensure urbanization brings development benefits? What conditions are getting in the way of leveraging urbanization for growth? And how can government policy improve them? The current urbanization trends in Uzbekistan might not be understood correctly. According to official statistics, after urbanizing faster than its neighbors for decades, in recent years, Uzbekistan has seen its share of urban population decline (Figure 1.2). While the cities were still growing, rural population grew faster than the urban population in 2010–19 (Figure 1.3). But it would be wrong to understand this trend as a reversal of urbanization. First, cities are still growing, and they are full of vibrant activity, as can be seen on the streets of Tashkent, the neighborhoods of Namangan, or along the road from Shahrezabs to Kitab. But the other reasonable question is whether the measurement of urbanization is accurate, and if the city boundaries that it uses captures the urban reality of the country. Rapid peri-urban growth and the expan- sion of urban build-up that can be seen across the Fergana Valley, around Samarkand and Bukhara, and even in smaller towns, suggests that the urban landscapes of Uzbekistan might be more expansive and more complex than official estimates suggest. Figure 1.2 Urbanization of Central Asian countries (and China) 60% Uzbekistan 50% Tadjikistan 40% China Kazakhstan 30% Kyrgyzstan 20% 10% 2000 2004 2008 2020 1980 1960 2016 2012 1984 1964 1988 1968 1996 1992 1976 1972 SOURCE: World Development Index (WDI) 18 THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT? Figure 1.3 Growth of urban and rural populations in Uzbekistan, 2010–19 122,0 2.900.000 121,0 2.800.000 Population in 2019 as % of 120,9 2.700.000 population in 2010 120,0 2.600.000 119,0 2.500.000 118,0 2.400.000 117,0 2.300.000 116,0 116,7 2.200.000 115,0 2.100.000 114,0 2.000.000 Urban Population, Rural Population, Total Total Population increase, 2010-2019 (right scale) Cumulative population growth (%), 2010-2019 SOURCE: Uzbekistan State Committee for Statistics. While reforms directly affecting urbanization are already underway in Uzbekistan, they can be more effective if they are based on a better understanding of the ongoing urbanization process and its chal- lenges. Government of Uzbekistan initiatives such as the propiska reform, land privatization, and adminis- trative reforms (among others) will have a direct impact on how cities develop, and whether they become drivers of social and economic progress. These policy reforms may be the key to unlocking the path to greater prosperity for Uzbekistan through leveraging urbanization. But to do so, they need to be grounded in an understanding of the issues that have been limiting the success of urbanization. Today there appears to be a lack of detailed understanding of the way urbanization is unfolding, the forces that shape it, and its impact on the country’s development. This report offers an overview of key characteristics of the urbanization process in Uzbekistan and issues associated with it, and thus sets a baseline for detailed work on advancing policy reforms related to urbanization and urban development. The second chapter of the report offers a review of the key out- comes of urbanization in Uzbekistan thus far and opens discussion about the two critical conditions that could be holding Uzbekistan back: factor market distortion and poor urban management. The third chap- ter presents more detailed analysis of key urbanization trends in Uzbekistan today and demonstrates the channels through which inefficient factor markets and deficient urban management limit the development benefits of urbanization. Chapter 4 digs deeper into these factors that might be causing the inefficient and unsustainable urbanization trends and attempts to identify policy-relevant conditions that might be driv- ing these trends. Finally, chapter 5 discusses key directions for policy reform that are pivotal for effective and sustainable urbanization and urban growth in Uzbekistan. THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT? 19 Chapter 2 Measuring urbanization in Uzbekistan and its development outcomes THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT? 21 Chapter 2 Measuring urbanization in Uzbekistan and its development outcomes What are the current urbanization trends and what are the development outcomes of the urbanization process in Uzbekistan? This chapter shows that urbanization in Uzbekistan is not understood properly and most likely substantially undervalued—which is, to a large extent, a result of the (hidden) urbanization of peri-urban areas. Furthermore, while cities will need to lead the way in delivering development outcomes, today they struggle to grow productively, drive structural transformation of the economy, or provide ade- quate access to services. The chapter closes with a discussion of possible conditions that might limit the contribution of urbanization to development: (1) the distortion of factor markets, and specifically barriers to internal migration, and (2) inefficient urban management. 2.1. Cities in Uzbekistan are yet to contribute as drivers of economic growth and structural transformation Urbanization contributes to economic growth through the success of cities, and in Uzbek cities oppor- tunities are still to be unlocked. In countries where urbanization was associated with economic growth, cities, specifically larger cities, were growth champions. Figure 2.1 compares the examples of Brazil, Turkey, and Malaysia to Uzbekistan. In all three comparator countries, large cities (with populations of more than 100,000) were the drivers of growth, as their GDP per capita growth was highly correlated with the national economic progress measured by the same indicator (Figure 2.2). While the results of this analysis should not be interpreted as a demonstration of a causal relationship between economic development and the evolution of productivity in large cities, it suggests that the growth of city economies has been a major part of economic development in urbanizing countries. For Uzbekistan, this means that opportunities for unlocking further growth should be sought in cities. The contribution of cities to national economic development can be assessed through the lens of the structural transformation of the economy. The development trajectory of every high-income country in the world today has depended on the reallocation of land, labor, and capital—the key production fac- tors—to their most productive uses in the economy. This process of structural transformation starts with the agricultural sector dominating total employment, relatively low skill levels among most workers, low productivity, and low average wages. But as agricultural productivity begins rising with better technology, demand for manual labor in rural areas falls (even as output often increases). Higher incomes in agriculture create demand for goods and services that, due to agglomeration effects, are more efficiently produced in cities. Growth in services and manufacturing thus starts pulling more workers away from agricultural employment and into urban areas. This process is the driver of labor productivity growth as it puts the talents and energy of the people to more productive uses and inevitably leads to greater economic com- plexity, rising productivity, and increasing incomes. This process is clearly closely associated with urbani- zation (Satterthwaite 2007, 28–31; Michaels, Rauch, and Redding 2012, 535–86), and for it to succeed, cities need to create conditions for the development of nonagricultural industries that will attract investors and workers alike. THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT? 23 Figure 2.1 Urbanization as the driver of the Figure 2.2 Role of larger cities in driving transition to high- and upper-middle income economic growth in Brazil, Malaysia, and in Brazil, Malaysia, and Turkey Turkey Brazil Malaysia Turkey Uzbekistan Brazil Malaysia Turkey Uzbekistan 10.5 15000 2019 of urban centers >100k (GHS-POP) GDP per capita (2010 USD) Log sum of GDP per capita 2015 10 2019 2019 10000 2000 2015 9.5 2015 2000 2000 9 5000 1960 8.5 1960 2019 2015 1960 1987 0 8 20 40 60 80 7.5 8 8.5 9 9.5 10 Share of urban population Log country GDP per capita SOURCE: WDI. SOURCE: WDI and GHS-UCDB 2015. NOTE: GDP = gross domestic product. NOTE: Comparator countries have a level of log GDP per capita that is within a standard deviation of Uzbekistan’s in 2015 in the initial time period and above that in the final time period. Uzbekistan’s log GDP per capita in 2015 is included for reference. GDP = gross domestic product. The stalled process of structural transformation in Uzbekistan is a cause for concern about the role of urban economies. In high-income countries (all of which are highly urbanized), agricultural employment accounted for only about 3 percent of the workforce on average in 2020, while services accounted for more than 64 percent. Uzbekistan is far below this threshold, and it appears that the continued weakness of non-agricultural sectors has slowed its structural transformation. Agriculture in Uzbekistan accounted for a greater share of GDP in 2019 than it did in 1996 (World Bank, forthcoming). In 2019, the share of ag- riculture was five times larger in the economy of Uzbekistan than in other non-high-income Europe and Central Asia countries on average, while services were contributing almost half as much to the national GDP compared to the same regional counterparts (Figure 2.3). Compared to other countries with similar levels of urbanization, Uzbekistan has one of the largest shares of agriculture in its GDP, and one of the smallest shares of services (Figure 2.4). These observations suggest that cities in Uzbekistan might not be driving structural transformation and economic growth. This is confirmed by comparative analysis of the sectoral structure of Tashkent’s economy, presented in Figure 2.5. The national capital has one of the smallest shares of the economy dedicated to services among cities at a similar level of development and is lagging behind even neighboring country capitals like Bishkek and Dushanbe. What conditions might be constraining the ability of cities to lead economic transformation and development? 24 THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT? Figure 2.3 Contribution of major industries to GDP in Uzbekistan vs ECA comparators 60 50 40 30 20 10 0 Agriculture, forestry, and fishing, Services, Manufacturing, Total natural resources rents value added (% of GDP) value added (% of GDP) value added (%of GDP) (%of GDP) Europe & Central Asia (excluding high income) Uzbekistan SOURCE: World Bank (2021) WDI. NOTE: GDP = gross domestic product. Figure 2.4 Structural transformation in countries of the world by level of urbanization Agriculture Services 60 90 Agriculture value added 50 80 Services value added as a % of GDP as % of GDP 40 70 30 60 20 50 10 40 0 30 0 50 100 0 20 40 60 80 100 % of urban population % of urban population SOURCE: World Bank (2021) WDI. NOTE: GDP = gross domestic product. THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT? 25 The development of productive and economically dynamic cities is an essential part of the road to prosperity for Uzbekistan. While stylized facts presented in this opening section do not allow for any conclusive judgement about the causal relationship between national economic growth and well-be- ing and the economic performance of cities, they indicate clearly that cities are an important part of economic development. At the same time, data snapshots suggest that, so far, urbanization in Uzbeki- stan has not necessarily gone hand in hand with economic growth and structural transformation. Figure 2.5 Share of the service sector in economies of the largest global cities 1 City services share of total GVA in 2019 Tashkent ECA (non high-income) .8 Central Asia Rest of the world .6 Tashkent Tashkent .4 .2 -1 1 3 5 Log GDP per capita USD in 2019 SOURCE: Based on Oxford Economics (2021), Global Cities Database. NOTE: ECA = Europe and Central Asia; GDP = gross domestic product; GVA = Gross Value added; USD = US dollar. 2.2. Understanding urbanization in Uzbekistan The first step toward making Uzbek cities development champions is making sure that the ongoing urbanization process is understood properly. The opening section of this chapter outlined that the road to prosperity usually requires economic growth in cities, which is also a part of the structural transforma- tion process that assures more productive use of a country’s resources. The rest of this chapter will aim to look deeper into the current state of urbanization and the performance of cities in Uzbekistan. First, it is worth considering the level of urbanization in Uzbekistan. The most recent population census took place in Uzbekistan in 1990, before the turmoil and readjustment that happened after the country became inde- pendent. Thus, without new census data (expected in 2022) the full picture of urbanization is unknown. Of- ficial statistics suggest that urbanization in Uzbekistan has been stalling over the past decade. However, the reality on the ground is that cities are growing and expanding rapidly, suggesting that administrative population data may fail to capture the extent of urbanization in Uzbekistan. Thus, a deeper discussion of ways to make urbanization a development driver starts with questioning the top-line urbanization figures that the official statistics offer. 26 THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT? It is likely that the official statistics do not fully capture the extent of urban growth, and thus some of the ongoing urbanization remains “hidden”. Official population growth numbers suggest that in Uzbekistan the population of rural areas is growing faster than the population of urban areas, and that the population of small towns is growing faster than the population of large cities1 (Figure 2.6). Most small towns have the official status of urban-type settlements (UTSs, a Soviet legacy name for small towns of around 10,000– 20,000 inhabitants). UTSs tend to concentrate near cities. 2 In fact, 79 percent of them are located within 50 kilometers (km) of a city with a population larger than 100,000 people. 3 This leads to the conclusion that while population growth is the highest outside cities themselves, a lot of the fastest-growing settlements are just outside official city boundaries, and functionally within urban agglomerations. This phenomenon is closely associated with urban sprawl, which has become the main form of urban growth in Uzbekistan. As a result of population growth in peri-urban territories, their density increases and reaches urban levels. This leads to the formation of extensive continuous urbanized territories or agglomerations, which cross administrative boundaries of cities, and include multiple settlements in one spatial superstructure (Figure 2.7 and Figure 2.8 show examples of agglomerations around Samarkand and Andijan). This observation suggests that the administrative boundaries of cities, which are used for population statistics, do not capture the real extent of urbanized territories. This process is not unique to Uzbekistan and is most prom- inently observed in the countries of South Asia and referred to as “hidden urbanization” (Ellis and Roberts 2016) (appendix A further demonstrates that cities in Uzbekistan tend to accommodate growth through sprawl rather than densification, and that that this pattern of growth leads to the formation of large ag- glomerations more than in the rest of the Europe and Central Asia region). Figure 2.6 Population growth in Uzbekistan’s urban areas, 2010–19 (%) 25,00% 20,00% 15,00% 10,00% 5,00% Cities larger than Cities smaller than Rural areas 50,000 50,000 and UTSs SOURCE: State Committee on Statistics of Uzbekistan. NOTE: UTSs = urban-type settlements. 1  The reliability of official population data is limited overall since the last population census was held in 1990. With the next census planned for 2022, better data will be available soon. However, the observations presented here with regard to administrative boundaries remain relevant. 2  The spatial distribution of cities, towns, and villages in Uzbekistan reflects the fact that only a relatively small part of the country offers favorable conditions for agriculture. See appendix D for a detailed discussion. 3  Calculation based on analysis of the Uzbekistan State Committee for Statistics data on population, and geolocation of UTSs derived from Yandex maps, 952 out of 1,099 UTSs were successfully located accounting for 92 percent of the total UTS population. THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT? 27 Applying a different definition of cities reveals a very different picture of urbanization in Uzbekistan. Since administrative definitions of cities may not be capturing the full extent of urbanization in Uzbekistan, it is worth trying to redefine urban areas using an improved methodology. This report uses the Global Hu- man Settlement (GHS) database, one of the most widely used gridded population datasets, 4 and follows the Degree of Urbanization methodology for defining urban areas endorsed by the United Nations (UN) (Box 2.1). This approach provides a very different picture of urbanization in the country, with the share of population in built-up areas reaching an urban level of density exceeding 85 percent, far more than the 50 percent of urban population reported by official population statistics. While this is likely an overesti- mate, as GHS data tend to overconcentrate estimated populations in built-up areas (Leyk 2019), it sug- gests that the available administrative data provide only a partial picture of urban growth in the country. This is further highlighted by the fact that Uzbekistan, along with other Central Asian countries, is a clear outlier in terms of the difference between the official level of urbanization and estimates using a universal definition and global population data (Figure 2.9). The failure of administrative boundaries to capture the extent of urbanization is a common issue in developing countries where peri-urban areas accommodate a substantial part of population growth (Vishwanath et al. 2013). Figure 2.7 Agglomeration around Samarkand SOURCE: World Bank analysis using WorldPop, Open Street Maps data, and Yandex Maps. 4  Gridded population datasets use official population statistics as inputs and, accounting for the density of built-up (as suggested by satellite images of all territories of the earth), assign population to specific grids of territory. Such data allow a much more detailed picture of the distribution of population. 28 THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT? Figure 2.8 Agglomeration around Andijan SOURCE: World Bank analysis using WorldPop, Open Street Maps data, and Yandex Maps. Figure 2.9 Urban population’s share using administrative vs. gridded population data in Europe and Central Asia 100 Uzbekistan 90 Urban population (% of total) Central Asia from administrative data 80 Other ECA 70 60 50 40 45 degree line 30 30 40 50 60 70 80 90 100 Urban population (% of total) from GHS-POP 2015 SOURCE: WDI and GHS-POP 2015. NOTE: GHSL = Global Human Settlement Layer. THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT? 29 Box 2.1 The definitions of urban areas used in this report Demographic data defined by administrative boundaries fail to capture the extent of the urban- ization phenomenon in many countries across the world. Each country uses its own definitions of cities and urban areas, rendering meaningful international comparisons virtually impossible. To address this issue, international practitioners have attempted to develop global definitions of cities. One such effort is the European Commission’s Degree of Urbanization methodology (endorsed by the United Nations [UN 2018b; Dijkstra et al. 2020a]), which proposes two types of urban definitions:   Urban clusters — contiguous grid cells of 1 square kilometer (km2) with a density of at least 300 inhabitants/km2 and a minimum total population of 5,000 (low-density threshold); and   Urban centers — contiguous grid cells of 1 km2 with a density of at least 1,500 inhabit- ants/km2 and a minimum total population of 50,000 (high-density threshold) (Dijkstra et al. 2020b). To paint a more complete picture of urbanization patterns in the country, and benchmark Uzbek- istan against its regional neighbors and relevant comparator countries using a consistent defini- tion of cities, this work applies the Degree of Urbanization “urban cluster” definition to GHS-POP data (Schiavina, Freire, and MacManus 2019) to define urban agglomerations (i.e., areas that are defined using the “urban cluster” technical definition above). The choice of focusing on the lower-density definition was influenced by the consideration that most cities in this part of the world are small (indeed, the median city in the resulting dataset has a population of 11,508) and have low density. Nonetheless, some selected parts of the analysis will also use urban centers as the definition specifically to make use of the Global Human Settlement’s Urban Centre Da- tabase, which uses this definition and includes a number of additional characteristics of urban areas (Florczyk et al. 2019). Finally, this work also utilizes administrative boundaries downloaded from Yandex Maps5 and administrative population data provided by Uzbekistan’s Ministry of In- vestment and Foreign Trade and produced by the State Committee on Statistics of Uzbekistan to analyze population changes over time (global population datasets offer much more limited opportunities for timeseries analysis) (Figure B2.1.1). Thus, throughout this report, cities will be defined in different ways depending on the objectives of the analysis. 5  https://yandex.com/maps. 30 THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT? Figure B2.1.1 Difference between administrative, urban agglomeration, and urban center Box 2.1 definitions of cities and urban areas in Uzbekistan cities Chirchik cities city administrative boundaries city administrative boundaries urban centres Keles urban centres urban agglomerations urban agglomerations Chartak Tashkent city Turakurgon Namangan Yangiyul SOURCE: Original elaboration using EU (2020) and GHS-POP (2015) Dataset. Figure 2.10 GDP per capita growth in urban centers in Europe and Central Asia 10 Log average city GDP per capita ECA High Income 9.5 Eastern Europe (unweighted) Central Asia 9 Uzbekistan 8.5 8 7.5 1990 Year 2000 2015 SOURCE: GHS-UCDB 2015 NOTE: ECA = Europe and Central Asia; GDP = gross domestic product. After applying comparable urban definitions, cities in Uzbekistan are less productive than cities in comparator countries and fail to capitalize on agglomeration. Figure 2.10 compares the average GDP per capita in Uzbek cities to regional neighbors and other countries in Europe and Central Asia.6 The graph shows clearly that Uzbek cities lag behind average levels of economic development of cities in other countries of the region. The estimated baseline level of GDP per capita was also the lowest before the collapse of the Soviet Union and, as in Central Asian and post-communist Eastern European countries, it falls in the decade 1990–2000 and then rises until 2015. As countries’ GDP converge over time (Sala-i- Martin 2006), the same should happen in cities where most of a country’s production occurs. However, 6  This analysis uses GHS urban centers’ database (Florczyk et al. 2019) as it includes GDP estimates that use the GHS-based definition as well. GDP estimates in constant 2011 US dollars. See Kummu, Taka, and Guillaume (2018). THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT? 31 while the average city in Uzbekistan in the 2000–15 period grew its economy faster than ECA counterparts, they underperformed compared to their neighbors in Central Asia despite starting from a lower level of GDP per capita. The productivity of Uzbek cities7 is below comparators, including those in Central Asia. As shown in Figure 2.11, the graph showing the distribution of productivity of city economies in Uzbekistan peaks at a lower level than all other Central Asian countries but Tajikistan. Indeed, when looking at individual cities’ performance in Figure 2.12, it is apparent that all of the largest cities in Uzbekistan perform below the Central Asian median and well below the European median. Among those, Tashkent is one of the best performers, but it still lies below the regional peers’ median (meaning that at least half of cities in Central Asia are more productive than the Uzbek capital). While analysis presented here refers to the time be- fore the COVID-19 pandemic, Box 2.2 presents a look at the economic performance of cities during the pandemic that demonstrates that large cities were the most affected by the restrictions related to the pandemic (analysis also shows strong recovery of Uzbekistan’s urban economies). Uzbek cities are not converting density into productivity, raising questions about factors that diminish benefits of agglomeration. One of the reasons why it is widely accepted that urbanization is a driver of economic growth is due to the role of cities as centers of high productivity, which rises in denser agglom- erations as they generate different types of productivity benefits due to the proximity of workers and firms (Glaeser 2011). This process, however, does not seem to be present in Uzbekistan, hinting at some obstacles that prevent Uzbek cities from capitalizing on agglomeration economies. Using administrative population data and relevant boundaries, 8 it is possible to estimate the elasticity of productivity to population density at approximately -0.67.9 This means that as the density of cities in Uzbekistan increases by 1 percent, their productivity goes down by 67 percent. Among international comparators, meanwhile, an average 1 per- cent increase in density would lead to a 0.47 increase in productivity (Graham and Gibbons 2019) (further analysis of the relationship between productivity and urban density is presented in appendix C). Figure 2.11 Distribution of cities’ productivity in five Central Asian nations .6 Uzbekistan Kernel density Kazakhstan .4 Kyrgyz Republic Tajikistan .2 Turkmenistan 0 -4 -2 0 2 4 Productivity SOURCE: Original elaboration using EOG VIIRS night-time lights 2013–19, GHS-POP 2015 using the European Commission’s urban clusters definition. 7  Productivity is measured by night-time lights; see appendix A for details. 8  The administrative boundaries are downloaded from Yandex Maps (https://yandex.com/maps). For a discussion of the different definitions of cities, see Box 2.1. 9  The estimate is obtained using year and city fixed effects. 32 THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT? Figure 2.12 Productivity of cities in Uzbekistan 1 0 Productivity -1 -2 -3 10 50 100 500 1000 Population ‘000s (log scale) Uzbekistan Central Asia All ECA median ECA cities median (excluding >100k median Uzbekistan) SOURCE: Original elaboration using EOG VIIRS night-time lights 2013–19, GHS-POP 2015 using the European Commission urban clusters definition. NOTES: Vertical axis represents the productivity of the urban agglomeration measured using night-time lights data; horizontal axis represents population size of urban agglomerations; larger cities are located further to the right. Dotted lines represent regional median values for comparison (as indicated in the legend). ECA = Europe and Central Asia. Applying a different definition of cities changes the picture of the urban system. Following the definition introduced in Box 2.1, this report identifies and studies 349 urban agglomerations with a population of at least 5,000 including 7 agglomerations with a population of over 1 million (Tashkent, Andijan, Kokand, Sa- markand, Karasu, Denau, and Namangan).10 Application of this definition across the Central Asian coun- tries shows that Uzbekistan has a larger number of midsized and large agglomerations than its neighbors (Figure 2.13). Another conclusion is that the agglomeration of Tashkent, the capital, appears to account for a smaller share of population than the largest cities in neighboring countries (the point furthest to the right is located lower on the graph than in the comparator countries). Thus, the urban population in Uzbekistan appears to be less concentrated in the main agglomerations, than it is in other Central Asian countries,11 but the second tier of agglomerations appears to play a larger role than in other countries of the region (see appendix D for a more detailed discussion). 10  The structure of the urban system depends a lot on how urban areas are defined (for a detailed explanation, please see appendix E). 11  Kazakhstan, where the largest city is not the capital, is an exception here. THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT? 33 Figure 2.13 Relative size of cities in Central Asian countries Uzbekistan 20 (% of total country population) Kazakhstan Kyrgyz Republic 15 City population Tajikistan Turkmenistan 10 5 0 2 4 6 8 Log population SOURCE: Original elaboration based on GHS-POP 2015 using the European Commission urban clusters definition. The urban population in Uzbekistan is less concentrated in the primary city than in comparators and is spread across several large agglomerations. Figure 2.14 presents various measures of urban primacy (the relative size of the largest city—also called the primary city). Compared to other Central Asian and Eastern European countries, Uzbekistan displays low urban primacy on most indicators, confirming a low level of concentration of the urban population. One measure of primacy in which Uzbekistan consider- ably lags behind comparators is the ratio between the size of the largest (primary) city, Tashkent, and that of the second and third largest in the country. In comparator countries, the largest agglomeration tends to be much larger than the second and third agglomeration, but in Uzbekistan the difference is less stark. For example, in the Europe and Central Asia region (excluding high-income countries) the median relative size of the second-largest urban agglomeration is about 70 percent smaller than the largest, while in Uzbekistan, Andijan’s and Kokand’s agglomerations are respectively 36 percent and 47 percent smaller than Tashkent’s. Furthermore, these second-tier agglomerations play a very large role in Uzbekistan’s urban system overall. The share of population in agglomerations of over 1 million people in Uzbekistan is more than twice as large as in either Central Asian or Eastern European countries. This clearly shows that while Tashkent plays less of a dominant role in the urban system of Uzbekistan than the largest cities in other regional counterparts tend to, Uzbekistan has a number of large agglomer- ations that collectively form the core of the urban system. This should be interpreted as an opportuni- ty. With multiple large agglomerations that can generate substantial productivity benefits, Uzbekistan has a chance to build several dynamic economic hubs, enabling a more diversified model of economic growth. However, this will depend on these secondary agglomerations creating the right conditions to harness agglomeration benefits and drive growth. Tashkent is relatively more productive than other cities in the country, yet relatively smaller than comparable cities internationally, which signals unutilized opportunity. This work introduces a novel measure of primacy that is based on cities’ productivity, measured using night-time lights12. (Figure 2.14) 12  See appendix A for a more detailed discussion of productivity measures using night-time lights. This measure uses urban agglomerations defined as in box 2.2. 34 THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT? This measure compares the productivity of the largest city to that of the average (weighted) city in the country. Tashkent is relatively much more productive than the other cities in Uzbekistan, revealing a larg- er productivity gap than can be observed in the rest of Europe and Central Asia. This suggests that it should be offering much better economic opportunities and higher wages (than other cities) and thus be a magnet for migration. But relatively low population growth of Tashkent and low share of population liv- ing in the city (or low primacy as demonstrated on Figure 2.14) indicate that there are obstacles prevent- ing city growth. This could be a result of low fertility rates in the city, but potentially suggests that barriers to migration might be obstructing the growth of the capital. (These barriers will be discussed further in chapter 4.) Following this logic, it is reasonable to suggest that if these constraints are addressed and people can move to places of opportunity (one of which appears to be Tashkent), the national economy will benefit from the reallocation of its labor resources to more proactive activities. (appendix E offers a visual representation of Tashkent’s underperformance relative to international comparators). Figure 2.14 Benchmarking Uzbekistan using various measures of urban primacy 0.5 Uzbekistan Central Asia 0.4 Eastern Europe 0.3 0.2 0.1 0 -0.1 Largest / Total Largest / Urban Largest (2nd +3rd) Share in agglom. > 1M / Largest prod. (GHSL) (GHSL) rescaled Urban (GHSL) / Avg. prod. SOURCE: GHSL and WDI. NOTE: Definition of measurements presented in the graph: Largest/Total represents the population of the largest city over the total population of the country, while Largest/Urban represents the ratio between the largest city’s population and the country’s urban population. WDI and GHSL represent the data source used for the largest city and urban population (the latter is measured as the sum of the population of all urban agglomerations in the GHSL case). Total country population is always from the WDI. All remaining measures rely uniquely on GHSL data. “Largest / (2nd + 3rd)” measures the relative size of the largest urban agglomeration vis-à-vis the second and third largest; while the “Share in agglom. >1M” is the share in urban agglomeration of over a million people compared to the total population in urban agglomerations. In the latter, averages for Central Asia and Eastern Europe are lower than other measures of primacy as some countries have no agglomerations over 1 million and an overall value of 0 in this metric. The last measure represents the ratio of the productivity of the largest urban agglomeration over the average productivity (weighted by population) of all the other urban agglomerations in the country. Productivity is measured as the residual of regressing the logarithm of the sum of night-light emissions per capita on the logarithm of population. All values are group averages. In the latter measure, the value for Eastern Europe is negative and close to zero, indicating that the largest urban agglomeration narrowly underperforms the average urban agglomeration in the country. GHSL = Global Human Settlement Layer; WDI = World Development Indicators. THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT? 35 Despite low economic productivity (relative to regional peers, Figure 2.12), urban agglomerations in Uz- bekistan have lower poverty rates than rural areas. The inverse correlation between city size and poverty levels has been well established in international economic studies (Ferré, Ferreira, and Lanjouw 2012). Anal- ysis of small area poverty estimates for Uzbekistan confirms that this rule holds for Uzbekistan, and thus even though cities do not appear to make the most of productivity benefits of agglomeration, they help residents achieve higher levels of income and consumption than in rural areas (Seitz 2020). This points once again at the critical importance of the urbanization process for development and people’s well-being and suggests that free movement to places of opportunity is not only important for economic growth, but is also critical for providing a pathway out of poverty. Despite having lower poverty rates, cities in Uzbekistan still fail to maximize their contribution to well-be- ing due to gaps in service provision. Disparities in access to services is an important factor of nonmonetary well-being, and often a major driver of rural-urban migration as well as migration to bigger cities. Larger cities usually have an advantage because the cost of providing public services tends to be lower at higher densities, and the variety of private services is often defined by the size of the market. Figure 2.15 shows the number of education, health, leisure, and retail venues per 1,000 inhabitants, calculated using points of inter- est (which include privately run facilities) from Open Street Maps.13 The capital, Tashkent, offers much higher service accessibility than the rest of the country, as expected. However, the most interesting comparison is that between other urban areas. It is particularly striking how the other agglomerations with populations above 500,000 offer a lower level of education and health services per capita, as well as of leisure and retail venues, in both cases below the overall country average.14 Additionally, the analysis confirms that the pe- ripheries of agglomerations, where most growth takes place, are underserved by public and private services, which is not an uncommon phenomenon, yet it is worth mentioning due to a vast gap in service access be- tween centers and peripheries.15 Figure 2.15 Service accessibility in the core vs peripheries of urban clusters Core E&H All Core L&R Periphery E&H <100K Periphery L&R >100K >500K Tashkent 0 0.5 1 1.5 2 2.5 3 Venues per 1000 people (Education & Health or Leisure & Retail) NOTE: E&H = education and health; L&R = leisure and retail. 13  https://www.openstreetmap.org. 14  Open Street Maps data were used, because the official data on service provision are highly fragmented and not available in Geographic Information System (GIS) format. Partial validation using data on location of hospitals in Tashkent regions, showed that coverage of Open Street Maps exceeds 70 percent. 15  International comparisons of these indicators were outside of the scope of this report. 36 THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT? This section demonstrated that cities in Uzbekistan are underperforming compared to their potential and there is scope to improve their development results through harnessing the forces of agglomera- tion for productivity growth and maximizing their potential to improve people’s well-being. This obser- vation suggest that higher urban incomes (closely related to productivity), and better urban quality of life (related to access to services), may not be working as major incentives for migrants, as in other countries (see a deeper discussion of migration barriers in chapter 4). The end of this chapter will turn to discussing the factors that might be affecting the underwhelming performance of cities before they are analyzed in more depth in the following chapters. Box 2.2 COVID-19 had a relatively large impact on cities, but is unlikely to diminish their importance The COVID-19 pandemic has disproportionately affected cities and the largest urban centers. The ongoing global pandemic caused by the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-​ 2), commonly referred to as COVID-19, has upended the lives of nearly all the global population. It is impossible for any urbanization study to completely ignore the pandemic as it has affected the way cities allow individuals and businesses to interact with one another, and generate the positive spillovers that turn urban places into the engines of economic growth. Restrictions aimed at curbing the virus muted some of the advantages of cities (consumption availability, job proximity) while maintaining the disadvantages (high rental prices, limited hous- ing space, crowding) and potentially increasing the risk of catching the disease due to popu- lation density. Figure B2.2.1 Night-time lights in urban centers and agglomerations, January 2020–March 2021 compared to the same months in 2019 100 Stringency index (monthly average) 1.8 Ratio of monthly NTL to 2019 value Centers Agglomerations* 80 1.6 5 km Buffer 60 1.4 10 km Buffer Oxford University 40 1.2 Covid-19 restrictions' stringency 20 1 .8 0 Jan 2020 Feb 2020 Mar 2020 Apr 2020 May 2020 Jun 2020 Jul 2020 Aug 2020 Sep 2020 Oct 2020 Nov 2020 Dec 2020 Jan 2021 Feb 2021 Mar 2021 SOURCE: Original elaboration based on GHS-POP 2015, EOG VIIRS night-time lights (monthly) 2020–21, and the Oxford COVID-19 government response tracker (OxCGRT). NOTE: *Excluding urban centers. Rural areas are defined as either a 5 or 10 km buffer around urban agglomerations. km = kilometer; NTL = night-time light. THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT? 37 This Box aims at providing a snapshot of how the pandemic has affected cities in Uzbekistan. Box 2.2 Figure B2.2.1 shows the ratio between night-time lights (NTL) in 2020 and the first three months of 2021 and the same month in 2019,16 overlayed with a measure of stringency of government-im- posed restrictions in the country from the Oxford COVID-19 government response tracker.17 The graph looks at the different monthly responses of NTL emissions to COVID-19 restrictions in urban centers, urban agglomerations, and rural areas relative to a normal year.18 The comparison with 2019 shows that NTL emissions dropped across the country during the first wave of the pandemic with the introduction of the strictest restrictions toward the end of March 2020. Urban centers in particular display a level of economic activity, proxied by NTL, that is lower in April and May 2020 than in the same months of the previous year. The summer months clearly show how the most populous city centers were the most affected. While restrictions started to be gradually relaxed, rural areas and low-density agglomerations rebounded, but activity stayed depressed in cities relative to the previous year. Only later, cities recovered strongly, when the severity of restrictions reached the lowest level since the beginning of the pandemic. Focusing on urban centers (Figure B2.2.2), it is immediately clear how Tashkent, which is by far the largest and most populous city, has borne the heaviest burden, as its economic activity has constantly been below 2019 levels and the city witnessed the slowest recovery in early 2021. Among the other cities with populations above 500,000, Karshi experienced the fastest drop between March and April 2020, albeit starting from the highest level of activity relative to 2019 pre-pandemic. The faster recovery in January and February 2021 is witnessed in the smallest ur- ban centers (with populations under 100,000). Nonetheless, the graph shows how cities’ recovery has been strong overall, emphasizing the resilience of Uzbekistan’s urban economies. It is fair to admit that some of the adjustments made to cope with the new reality (such as working from home) could survive the end of the pandemic and create a fundamental shift in how cities and their economies evolve. However, the strong recovery of the largest cities after the first wave of the pandemic suggests that even if the “new normal” might be affecting them, for now there is little reason to believe that the effects of the pandemic will deem their productivity advantages irrelevant, which is also the argument made by prominent economic thinkers using global evi- dence (Glaeser and Cutler 2021). 16  The monthly comparison is necessary due to seasonality in NTL emissions, which are affected by the length of day at different times of the year (Elvidge et al. 2019). 17  The stringency index is preferred to an indication of the presence of a “lockdown” as the severity of restrictions varies across countries and across time within the same country. The Oxford University tracker aggregates different types of restrictions on a daily basis to compute the stringency index. The graph uses the monthly average of the index for consistency with the NTL data. See Hale et al. (2021a; 2021b). 18  Monthly NTL emissions data allow to track population movements and economic activity within the country. 38 THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT? Figure B2.2.2. Night-time lights in urban centers, January 2020–March 2021 compared to the Box 2.2 same months in 2019 100 3 Stringency index (monthly average) Ratio of monthly NTL to 2019 value Tashkent Namangan 2.5 80 Samarkand Karshi 60 2 Pop 100-500K Pop <100k 1.5 40 Oxford University Covid-19 restrictions' stringency 20 1 .5 0 Jan 2020 Feb 2020 Mar 2020 Apr 2020 May 2020 Jun 2020 Jul 2020 Aug 2020 Sep 2020 Oct 2020 Nov 2020 Dec 2020 Jan 2021 Feb 2021 Mar 2021 SOURCE: Original elaboration based on GHS-POP 2015, EOG VIIRS night-time lights (monthly) 2020–21, and the Oxford COVID-19 government response tracker (OxCGRT).; NTL = night-time light. 2.3. Preconditions for effective and sustainable urbanization While cities can pave a path toward economic progress and development, their growth does not guar- antee the desired development results. Economic research confirms (as suggested in the introduction) that not every country manages to convert urbanization into economic growth. The academic consensus these days suggests that it is not urbanization itself, but the form it takes, that defines whether urbaniza- tion enables productivity growth and development. Two conditions that go hand in hand appear to have a major role to play: (1) the efficiency of factor markets—the mobility of people and capital and ease of reallocation of land, and (2) the quality of city management that is needed to provide benefits of agglom- eration while mitigating the congestion costs (Turok and McGranahan 2013). Both of these conditions are to a significant extent shaped by government policy. Distortions that limit the mobility of labor and capital, and complicate changing the use of land, prevent factors of production from being allocated to the most productive uses, thus limiting cities’ economic potential. Early studies of urbanization without economic growth, that focused on African countries, asso- ciated unproductive urbanization with location incentive distortions that lead to the spatial misallocation of factors of production and the dysfunctionality of cities (Fay and Opal 2000). In other words, if it is hard for people to move to places where they can find a better job (because of a high cost of moving or be- cause of administrative constraints or other factors), or there are barriers for investment to flow to places or industries of highest potential, or for land to change use toward a more productive one—cities are un- likely to have resources needed to make the most of their competitive advantages. This means they will struggle to achieve productivity growth and contribute to economic development and restructuring away from agriculture. Thus, it falls to the government to create conditions under which markets drive resources to their most productive uses. THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT? 39 Additionally, to unlock urban productivity growth, public policy and investments at the local level should enable agglomeration while mitigating congestion costs. Cities are known to generate produc- tivity benefits for firms through three key mechanisms: matching firms to workers and suppliers, sharing expensive services and infrastructure, and learning through knowledge spillovers enabled by density. (Duranton and Puga 2004; Storper 2010). While matching, sharing, and learning are products of density, their benefits can be offset by growing congestion costs: overcrowding, overburdened infrastructure, traffic, pressure on ecosystems (water and air quality), and the higher costs of living and housing in cit- ies. As Turok and McGranahan (2013) point out: “These negative externalities tend to increase as cities expand, especially if urban development is haphazard and there is insufficient public investment to maintain and expand essential infrastructure. Dysfunctional systems, gridlock, power cuts and insecure water supplies increase business costs, reduce productivity and deter private investment.” This points at the fact that urban growth does not guarantee productivity growth as such, rather its effect on produc- tivity to a large extent depends on whether the right supporting and enabling conditions are created by means of local and national policy and investment (Henderson and Vernon 2010). An array of other conditions also affect the economic performance of cities and well-being of residents and should not be ignored. While the efficiency of factor markets and quality of urban management are identified throughout this report as the key factors required for sustainable and productive urbanization, several other conditions play a critical role and should not be left out of policy discussions. Examples in- clude education and skills, business environment and innovation potential, social support, and safety nets. While these and many other relevant issues are outside the scope of this report, it is important to empha- size that cities do not exist separately from the country, and thus all major structural conditions affect their performance. Policy has a role to play in maximizing development potential of urbanization, but to understand pol- icy priorities it is critically important to understand the causes of urban underperformance. The rest of this report will largely focus on two groups of issues related to urbanization in Uzbekistan: first, whether the market-driven allocation of factors of production is distorted, and second, whether urban growth and congestion costs associated with it are managed in a way that creates conditions conducive to improved productivity and livability. The next chapter will aim to advance along both axes by presenting further analysis of critical characteristics of the urbanization process in Uzbekistan. 40 THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT? Chapter 3 Key urbanization patterns that explain underwhelming outcomes THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT? 41 Chapter 3 Key urbanization patterns that explain underwhelming outcomes Urbanization in Uzbekistan is falling short in terms of development outcomes because it is distorted by the limited mobility of production factors and inefficient management of urban areas and their growth. The distribution of population growth across Uzbek cities suggests the presence of barriers to mobility. Urban form characteristics and growth patterns, visible in cities of Uzbekistan, suggest that inefficient urban management might be one of the causes of urban underperformance. 3.1 Distortions to population mobility limit cities’ contribution to the national economy and signal the inefficiency of factor markets Population distribution and mobility in Uzbekistan suggest that factor markets are distorted, which limits the potential of cities. Both people and capital tend to flow to places of highest potential. Barriers to the reallocation of land or capital likely result in the suboptimal utilization of local potential, and thus far less economic opportunities to attract migrants. At the same time even if economic opportunities exist, it is hard for migrants to take them because moving is restricted or unaffordable—and so the city economy will not grow. Thus, trends of population growth in different places can be used as a proxy for the effectiveness of factor markets. The second chapter of this report showed that the urban system in Uzbekistan differs from comparators. If people could move freely to the most productive locations (like Tashkent), the national economy would benefit. This is closely linked to the fact that in Uzbekistan the estimated domestic migration rate is among the lowest globally (Figure 3.1, according to calculations presented in Seitz [2020])19. This is particularly unexpected, given the stark gap in poverty levels and dis- posable incomes observed in rural and urban areas of Uzbekistan, a dynamic that should create incen- tives for moving to cities. 20 The canonical relationship between city size and proximity to other large cities and population growth is broken in Uzbekistan. Cities that are located close to other big population centers usually grow faster and larger. This relationship between market access (defined as the population living within or in easy reach of the city) and population growth is found in many countries around the world. 21 Usually the plac- es with larger populations (internal markets), or transport access to other places with larger populations (external markets) grow faster (Henderson and Wang 2007). However, in Uzbekistan, this relationship is not observed. This fundamental observation deserves further analysis. 19  Seitz uses various data sources and statistical methods to calculate several estimates of internal migration, which translate into a 1.24 percent crude migration rate over a five-year period (which is methodologically comparable to international estimates). The methodology behind this has limitations, however. Seitz uses data from the State Committee of Uzbekistan on Statistics, based on a partial population survey that uses sampling frames from the 1990 population census, as well as data from household registration. See Seitz (2019a; 2020).  20  Despite equalization of wages and insufficient service provision, analysis by Seitz clearly shows that more urbanized areas have lower poverty rates, and higher disposable incomes (based on Seitz [2019b]). 21  This analysis is conducted using administrative data for 119 cities and 920 UTSs provided by the State Committee on Statistics of Uzbekistan with population records for 2010, 2015, and 2019. The final dataset utilized excludes the UTSs that have no population data at any of the three data points (reclassified to rural or fully absorbed by a larger settlement) and exclaves in the Kyrgyz Republic such as the Sokh District. THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT? 43 Figure 3.1 Five-year internal migration rates in selected countries (rough estimates) 20 15 10 5 0 Indonesia 7 Uzbekistan 160 Indonesia 26 Philippines 16 Nicaragua 17 Portugal 22 Philippines 83 Canada 11 Argentina 24 Indonesia 280 Colombia 33 Honduras 18 Philippines 1610 Australia 8 Venezuela 24 Mexico 2438 Greece 54 Ecuador 22 Ghana 110 Colombia 10 Barbados 11 Domincan Republic 225 Vietnam 1203 USA 9 Portugal 308 Argentina 511 Vanuatu 6 Uruguay 19 Senegal 34 Mongolia 21 Peru 25 Chile 44 Brazil 1520 Costa Rica 81 New Zealand 16 Peru 1833 South Africa 52 Portugal 4000 Australia 69 Chile 178 New Zealand 73 SOURCE: Seitz 2019a. NOTE: Migration rates for the circa 2000 round of censuses (originally from Bell and Charles-Edwards [2013]); estimates for Uzbekistan are based on Uzbekistan State Statistics Committee 2018 permanent migration flows data, using the 2018 L2CU baseline survey. Numeric values in country labels indicate the number of territorial units. Proximity to large markets (or, simply put, a large population living within or near a city) is usually a strong predictor of growth, because it is associated with cities that are attractive places to live. The more populous a city and the larger the cities it is easily connected to via transport, the larger is the market available to it. Such cities enjoy more agglomeration economies, which make them more produc- tive, pushing salaries of residents up and raising the cities’ overall appeal. Larger markets also offer more consumption options, making them relatively more livable (Glaeser, Kolko, and Saiz 2001). For example, in cities of Northern, Western, and Central Europe, a positive relationship between market access on one side and growth in GDP, population, and employment on the other side is well established (Adler 2020). These relationships can be associated with an environment that allows market forces to bring people to places where they maximize their incomes and well-being by choosing locations that offer the best consumption options associated with overall quality of life. If these forces are disrupted, suboptimal outcomes for indi- viduals and for the society are to be expected, as people would be stuck in places where they cannot put their talent to best use, thus making a lower income and contributing less to the economy. Greater access to internal or external markets does not result in higher urban population growth in Uz- bekistan. The analysis presented in detail in appendix F shows that while proximity to population centers (external market access) has no impact on the population growth of Uzbek cities and towns, larger popu- lation of the city or town itself has a negative impact on its growth. It is hard to conclusively say why (one possible explanation is poor-quality data). On one hand, these results may mostly be capturing fertility rates, which are higher in rural areas. And while internal migration is very low, the attractiveness of cities does not compensate for lower natural urban population growth. On the other hand, they may also reflect 44 THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT? that cities do not appear as attractive as they could for people of Uzbekistan. A high cost of living coupled with a limited availability of services in large cities outside Tashkent, and the fact that greater size does not necessarily mean higher productivity and wages (demonstrated in chapter 2), may limit the attrac- tiveness of cities. They may also reflect the presence of administrative barriers to migration. Whatever the cause of these disruptions, they have a direct impact on the economic contribution of Uzbek cities. Low levels of internal migration, and a missing relationship between market access and urban popu- lation, are among the multiple distortions of factor markets. When a population is not flowing to places of highest potential, the labor market functions inefficiently. Distortions exist in other factor markets as well. The capital markets in Uzbekistan are dominated by state-owned banks whose rationale for allo- cating loans may be affected by policy priorities and thus is not consistent with market signals (World Bank, forthcoming). In the land market most nonresidential urban land remains under state ownership, and its change of use and development rights are hard to obtain. The roots and policy implications of these inefficiencies will be further discussed in the next chapter. The rest of this chapter will investigate whether the dominant patterns of urban growth in Uzbekistan are consistent with effective and sustain- able urban management. 3.2 The way cities in Uzbekistan are spatially organized is affecting urban productivity, livability, and sustainability Uzbek cities accommodate population growth through sprawl and formation of vast agglomerations, rather than through densification. Chapter 2 showed that Uzbek cities tend to form sprawling agglom- erations that cross administrative boundaries. Appendix A demonstrates further analysis, confirming that this is a product of the model of urban growth that is different from how cities tend to grow in the Europe and Central Asia region. In fact, cities in Uzbekistan are defined by more rapid built-up expansion per unit of population growth than the comparators, and by more rapid mergers of separate urban settlements into agglomerations occurring due to built-up area expansion. What drives this process, and can it help explain the underwhelming urbanization outcomes? The phenomenon of rapid urban sprawl in Uzbek cities is closely connected to low internal migration. At first glance, this might appear counterintuitive. Usually, sprawl is a result of people migrating to cit- ies and settling on their edges. But things appear to differ in Uzbekistan. Because of weak migration in combination with rapid natural population growth, smaller towns and villages grow faster than cities. This is a phenomenon that is commonly known in global practice as urbanization in-situ (or in place), when population growth leads to densification and conversion of rural settlements into urban ones over time. What appears to be different in Uzbekistan is that, in a country where a large part of the territory (mountains, deserts) was historically not suitable for settlement (leading to a concentration of population in areas with fertile soils, as shown in appendix G), a lot of these towns and villages are clustered close to cities. Thus, their growth, and expansion results in different settlements merging and the formation of large, urbanized territories, or agglomerations. So, counterintuitively, in Uzbekistan the lack of migration, rather than the rapid inflow of rural migrants (in combination of the nascent state of land markets, limited infrastructure availability, urban planning and urban governance issues that will be discussed later) is the explanation for urban sprawl. Furthermore, this form of urban growth produces unsustainable and unproductive urban agglomerations. Urban sprawl and the merger of cities into vast agglomerations result in the formation of urban areas that have complex shapes, inconsistent land use, and a lack of well-structured road networks. These THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT? 45 characteristics, all of them typical consequences of unplanned growth, often impede urban develop- ment. In a city that is sprawling and lacks density and good road networks, it typically takes longer to get to work, school, hospitals, places of interest, or destinations for the delivery of goods. Because of the increased transport costs, it might be more expensive to run a business and harder to reach consumers. The lack of density may reduce the benefits of knowledge spillovers that result from frequent interac- tions between people. Low density sprawl may affect all three key drivers of urban productivity—sharing, matching, and learning—by reducing the size of the accessible market and frequency of interactions (Duranton and Puga 2004). This pattern of spatial growth may also impede access to public facilities (schools, hospitals, and parks) and private services (shops, restaurants, theaters), thus affecting overall quality of life. Analysis in various countries has linked the characteristics of urban form to economic per- formance (Ciccone and Hall 1996), sustainability (Roo and Miller 2000), quality of life (Squires 2002), and knowledge spillovers through human interactions (Glaeser 1998). And it is not just the density of urban population that defines the quality of urban spatial form, but a number of more specific characteristics of cities’ spatial organization. Urban shape, the internal structure of a city, and land use patterns are the three characteristics that define the efficiency of a city’s spatial form.   Urban shape: The more a city looks like a circle, the better its urban shape. This is because a circle is the most compact geometric feature, and within it, distance between points would be shorter than in a city shaped like an octopus. Rounder cities often also have lower per capita cost of infra- structure provision due to their compactness (Litman 2015).   Efficient land use: Cities can be defined by maximum utilization of available land or by large stretches of underused land amid patches of build-up. These patterns naturally result in lower den- sities and diminish agglomeration benefits. Furthermore, research shows that cities that fail to use the land consistently are inefficient in providing infrastructure and public services because the per capita costs of providing these increase with sprawl (Fallah, Partridge, and Olfert 2011).   Internal structure: This refers to the extent to which a city’s road network simplifies connectivity between places. If the road network is dense and has a lot of intersections, commuters can more easily avoid indirect routes and find alternative paths to destinations, thus making it faster and cheaper to get around the city. This facilitates better access to services and enhances agglomer- ation effects (Giacomin and Levinson 2015). Recent studies have confirmed the role of these characteristics in cities in different parts of the world. A study of Indian cities has shown that better geometry is associated with higher population growth and higher housing rents, which points to the fact that people enjoy living in such cities and are willing to pay for it (Harari 2020). Such compact cities are also defined by faster economic growth (Tewari, Alder, and Roberts 2016). A recent World Bank study of Latin American cities has shown that all three types of characteristics of urban space matter for the economic productivity of cities (Ferreyra and Roberts 2017). The shape of Uzbek cities is less efficient than that of comparators. Figure 3.2 demonstrates that on metrics of urban shape, Uzbek cities rank way below comparator countries in the Europe and Central Asia region. Agglomerations in Uzbekistan have shapes that are less round, and distances between places within them are on average larger (with lower scores on the proximity and roundness index). What is spe- cifically striking is that larger and more populous agglomerations in Uzbekistan get a substantially lower score than small cities (the trend lines on the graph are negatively sloped), which does not hold true for 46 THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT? comparator countries. This appears to be consistent with the earlier discussion of rapid expansion result- ing in a merger of urban areas, thus producing bigger agglomerations with very complex and noncircular shapes. Land use patterns in Uzbek cities also show substantial inefficiencies. Figure 3.3 demonstrates that rela- tive to other cities in Europe and Central Asia, Uzbek urban agglomerations, on the one hand, have a lower share of land that is occupied by structures or improvements (lower fullness index values), and on the other hand, a greater share of total territory with a lower-than-average population density (the sprawl index is higher). This observation is consistent with the previous finding that population growth in Uzbek cities is accommodated through low-density sprawl. Road networks in Uzbek cities do not help address the inefficiency of their shape. A well-developed road network can help cities compensate for some of the inefficiencies of their shape and land use. Better road networks help people get to places faster even if the distances are long. However, Figure 3.4 demonstrates that this is hardly the case in Uzbekistan. Not only is the density of streets and inter- sections in Uzbek cities lower than in the comparator countries, it also does not increase as cities get larger—which is the case in the other countries. This once again points to the prevalence of unplanned and chaotic growth, which allows cities to expand and merge without adequate planning and develop- ment of the road network. Larger urban agglomerations are not necessarily defined by a larger density of population and build-up in Uzbekistan (which in most cases would imply a denser road network); in- stead, a lot of large urban areas are just more expansive, while large portions of their territory retain a low peri-urban density level. Figure 3.2 Urban shape of cities in Uzbekistan and other low- and middle-income countries in Europe and Central Asia Proximity Index Roundness Index 1 0,9 0,95 0,85 Roundness Index Value 0,9 Proximity Index Value 0,8 0,85 0,75 0,8 0,75 0,7 0,7 0,65 0,65 0,6 0,6 0,55 0,55 0,5 0,5 0 12,5 13,5 14,5 15,5 0 12,5 13,5 14,5 15,5 Log population Log population ECA (non hign income) Linear ECA (non hign income) ECA (non hign income) Linear ECA (non hign income) Uzbekistan Linear Uzbekistan Uzbekistan Linear Uzbekistan SOURCE: Original calculations based on GHS-BUILT. NOTE: ECA = Europe and Central Asia. Ln = linear. The proximity index is the ratio of the average distance from all points in the equal-area circle to its center and the average distance to the proximate center from all points in the shape. The roundness index is the share of the total area of the urban extent that is inside the equal-area circle about its center of gravity. THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT? 47 Figure 3.3 Urban land use pattern metrics of cities in Uzbekistan and other low- and middle-income countries in Europe and Central Asia Fullness Index Sprawl index 0,6 0,8 0,7 Fullness Index Values 0,5 Sprawl Index Values 0,6 0,4 0,5 0,3 0,4 0,2 0,3 0,1 0,2 11,5 12,5 13,5 14,5 15,5 11,5 12,5 13,5 14,5 15,5 Log population Log population ECA (non hign income) Linear ECA (non hign income) ECA (non hign income) Linear ECA (non hign income) Uzbekistan Linear Uzbekistan Uzbekistan Linear Uzbekistan NOTE: ECA = Europe and Central Asia; Ln = natural logarithm. Figure 3.4 Urban structure metrics of cities in Uzbekistan and other low- and middle-income countries in Europe and Central Asia Street density Intersection density 20000 300 Intersection density (per km sqr.) 18000 Street Density (m/km.sqr.) 250 16000 14000 200 12000 10000 150 8000 100 6000 4000 50 2000 0 0 11,5 12,5 13,5 14,5 15,5 11,5 12,5 13,5 14,5 15,5 Log population Log population ECA Uzbekistan ECA Uzbekistan Linear (ECA) Linear (Uzbekistan) Linear (ECA) Linear (Uzbekistan) SOURCE: Original calculations based on GHS-BUILT. NOTE: ECA = Europe and Central Asia; Ln = natural logarithm. The fullness index indicates the fraction of the total area of the urban extent that is built up. The sprawl Index measures the share of the total area of the urban extent that is of lower-than-average density. 48 THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT? Before diving deeper into the negative impact of the urban forms that are prevalent in Uzbekistan, it is important to acknowledge that urban sprawl is not always a negative aspect of growth. While sprawl is associated with a lot of negative outcomes, it can be seen through a very different lens. In particular, sprawl is seen by some experts as a process that allows people more choice regarding the urban environments and conditions that they want to live in. As prosperity grows, it is natural for people to desire more choices; thus sprawl in some countries can be a product of economic growth and wealth accumulation (Bruegman 2005). If households pay for the cost of the infrastructure they use (which is the case in many urban and suburban areas in the United States, for example), and if settling in suburban areas is a choice that house- holds make while other options also exist—then sprawl may not be that detrimental to the societies (even though the negative impact on the environment and climate would still apply). These considerations lead to two conclusions. First, it is important to understand the conditions and incentives that lead to sprawl: do households have other options, and does sprawl offer a better quality of life to suburbanites? Second, while there is no doubt that sprawl puts extra pressure on urban infrastructure and services, are these ser- vices provided in sufficient quantities, and do the suburbanites cover the costs? These questions will be discussed in detail in chapter 4. Globally, better urban forms and structure are associated with higher productivity. Economic litera- ture establishes a link between the quality of urban form and the productivity of an urban area. Spe- cifically, the innovative work of Harari (2020) introduces the notion of urban form as an amenity and shows that compact cities are more attractive and command a greater willingness to pay as they offer a superior quality of life mainly due to greater accessibility and lower transport costs. Figure 3.5 shows the positive correlation between productivity and the most common measure of city compactness: the fullness index. Uzbek cities are on average less productive and less full than their counterparts in Europe and Central Asia. This correlation should not be interpreted as causality, but it suggests that better land use might help improve urban performance. Beyond simple correlations, modeling that replicates approaches from Harari (2020) shows that the characteristics of urban form are significant predictors of urban performance in Uzbekistan (a technical analysis of the link between productivity and urban form in Uzbekistan is presented in appendix H). Figure 3.5 Fullness index and productivity of urban areas in Uzbekistan compared to the rest of the region 5 Uzbekistan ECA Fitted line Productivity 0 -5 0 .2 .4 .6 .8 Fullness index SOURCE: Original elaboration based on GHS-POP 2000 and 2015 data using the European Commission’s definition of urban clusters. NOTE: ECA = Europe and Central Asia. THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT? 49 Beyond productivity losses, global experience demonstrates that various models of urban growth offer very different outcomes in terms of the costs of service provision, residents’ well-being, and sus- tainability. For example, consider the cities of Barcelona (Spain) and Atlanta (United States). They are comparable in terms of population, but Atlanta’s urban footprint is almost 27 times larger. Its much lower urban density results in transport-based carbon dioxide emissions per hectare that are 10 times higher than Barcelona’s (Figure 3.6) (Bertaud and Richardson 2004). But costs of low-density growth are not limited to greenhouse gas emissions. Provision, operation, and maintenance of water, wastewater, road, heating, and power infrastructure in low-density areas is far more expensive for the public sector on a per capita basis. This is a result of having to install and operate larger networks (more pipes, roads, and wires per person), which usually implies greater transmission losses (of water and power) and mainte- nance costs per capita. A study commissioned by the World Bank in the Kyrgyz Republic demonstrated that service provision in high-density residential areas is at least 50 percent lower than in low-density areas (Amerasekera et al. 2015). Also, the extra costs—both in terms of time and money—of traveling larger distances to access jobs and services (including more time in traffic) are borne by households. At low densities, dependence on private vehicles usually grows as public transportation becomes less via- ble, leading to increased greenhouse gas emissions. Lower-density developments also tend to produce negative societal impacts like social polarization and segregation, and even increased crime. Finally, in the Uzbek context, significant additional costs of expansion may be associated with greater exposure to the risk of disasters, such as floods and earthquakes (Amerasekera et al. 2015). Figure 3.6 Comparison of different models of urban growth in Atlanta and Barcelona Atlanta Barcelona Built-up area Built-up area Population Population 5.25 5.33 million million Urban area Urban area 4.280 162 km2 km2 Transport carbon Transport carbon emissions emissions 7.5 0.7 tonnes CO2/ha/an tonnes CO2/ha/an (public + private transport) (public + private transport) SOURCE: Bertaud and Richardson 2004,293–310. NOTE: CO2/ha/an =CO2 emissions per hectare per year in tons; km2 = square kilometer. Expansion forecasts for the cities of Uzbekistan show that continuing current growth trends will lead to highly inefficient urban forms and major loss of arable land. Using a spatial modeling technique (see ap- pendix I for a description of the methodology), we can demonstrate the predicted growth of the built-up areas in the largest agglomerations in Uzbekistan in the next 30 years (Figure 3.7). These expansion projec- tions assume that urban growth will be following the same model, and expansion will be happening at the 50 THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT? same rate as it did in 2010–15 (years for which the latest data are available). And it clearly demonstrates that retaining the current model of growth and expansion will result in a further increase of sprawl and the greater complexity of urban structures across multiple urban areas. Forecasts of the expansion of the 10 largest urban agglomerations in Uzbekistan show that following the current mode of urban expansion will result in the loss of 272 square kilometers (27,200 hectares) of arable land by 2050, which according to recent agricultural yields in Uzbekistan would produce a loss of 1,252 tons of cereals harvested per year, or 12,216 tons of vegetable harvested per year. 22 Figure 3.7 Built-up expansion forecast for selected cities in Uzbekistan by 2050 Legend 1985 Andijan Angren 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 Kokand Samarkand 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2050 SOURCE: Original elaboration using GHS-BUILT data. Accommodating population growth through an increase in density, rather than the current model of sprawl, would lead to savings exceeding US$1.2 billion by 2050—and this is only for infrastructure provision in the largest cities of Uzbekistan (see Figure 3.7 and Box 3.1). While this estimate can be considered conservative, as it assumes a slowdown in urban population growth and does not account for the operating and maintenance costs of infrastructure, it suggests that half the cost of additional infrastructure can be saved by moving to a higher-density growth model. 22  Calculations based on current agricultural yields in Uzbekistan: https://knoema.com/atlas/Uzbekistan/topics/Agriculture/Crops-Production- Yield/Wheat-yield . THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT? 51 Figure 3.8 Costs of urban infrastructure provision under different models of accommodating population growth 3.500 Total infrastructure investment needed 3.000 2.500 (million $ US) 2.000 1.500 1.000 500 0 Scenario 1: low density expansion Scenario 2: medium density expansion Scenario 3: densification SOURCE: Original elaboration using GHSL data and additional assumptions following methodology presented in appendix H. The goal, however, should not be to limit people’s options to high-rise housing, but rather to gradually increase the average density by exploring options for efficient land use and mixed-use development. First, it is important to admit that building high-rise apartment blocks is much more expensive than build- ing individual homes on greenfield sites, which partially counterbalances the savings on infrastructure provision. So, the idea is not to replace all new housing construction with high-rises. It is rather to change the mix of new housing types, introducing more high-rise and medium-high-rise residential buildings. This will, on one hand, reduce the amount of greenfield land required and the cost of infrastructure investments, and, on the other, it will give households more options (in terms of type of housing and location). Chapter 3 has shown that the development outcomes of urbanization in Uzbekistan are distorted by factor market inefficiencies and suboptimal urban growth patterns. First, this chapter showed that limited population mobility likely leads to underutilization of local economic potential and fuels urban sprawl. It then presented evidence that urban sprawl in Uzbekistan produces inefficient urban forms that are limiting urban productivity, and in the future would result in loss of arable land and vast additional costs for infra- structure provision. The next chapter will turn toward uncovering factors that lead to these distorted and inefficient urbanization patterns. 52 THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT? Box 3.1 Box 3.1 Estimated costs of urban expansion For estimation of costs of urban expansion, agglomerations surrounding the 10 largest cities of Uzbekistan were considered: Tashkent, Namangan, Nukus, Bukhara, Samarkand, Karshi, Andijan, Kokand, and Fergana and Margilan (though, together, these last two might be said to make up their own agglomeration). Only costs of water, sewerage, road, and power infrastructure were considered. The average estimated capital costs, per capita, of infrastructure provision were based on assessments con- ducted in the Kyrgyz Republic in 2015 (Amerasekera et al. 2015); however, all costs were divided by the standard correction coefficient of 2.1823 to capture the fact that most cities in the Kyrgyz Republic are located in mountainous areas, while most cities in Uzbekistan are in flatlands. Other than that, due to the similar context of the two countries, the cost estimates were considered applicable to Uzbekistan. Figure B3.1.1 Population growth Figure B3.1.2 Three scenarios of forecasts for the 10 selected accommodating population growth in the agglomerations largest agglomerations of Uzbekistan 16000000 100% 0,2 14000000 0,3 80% 0,2 12000000 60% 0,8 10000000 0,5 40% 8000000 0,6 6000000 20% 0,1 0,2 4000000 0,1 0% 2000000 Scenario 1 Scenario 2 Scenario 3 0 One-story housing Three-story Nine-story 2010 2019 2030 2040 with land plot housing and more SOURCE: Original elaboration based on GHS-POP and SOURCE: Original elaboration. Uzbekistan Statistics Agency data. The following methodology was used to generate population growth forecasts. First of all, the population growth in all cities and urban-type settlements within the boundary of each of the agglomerations between 2015 and 2019 was calculated using the official data of the State Com- mittee on Statistics of Uzbekistan. Then it was assumed that the same annual average popula- tion growth rate would persist in each of the urban areas until 2030, between 2030 and 2040 the growth rate would decrease by 25 percent, and in the following decade it would decrease by 50 percent from the current level. This forecast suggests a population increase of 62 percent across all major cities combined by 2050 (Figure B3.1.1). This can be considered a relatively conservative 23  The correction rate was recommended during consultations with World Bank water engineers working in Uzbekistan. THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT? 53 estimate, because the ongoing liberalization of mobility restrictions can actually lead to an ac- Box 3.1 celeration of population growth in agglomerations of large cities, rather than to a slowdown. Finally, three scenarios of accommodating this population growth were considered (summarized in Figure B3.1.2):   SCENARIO 1 — the status quo scenario, wherein 60 percent of population growth will be accommodated through low-rise expansion (one story with a land plot, at an average density of 25 people per hectare), and the rest would be split between medium- and high-density developments.   SCENARIO 2 — low-density expansion still accommodates a fifth of the population growth, while 50 percent of the population increase is accommodated by medium-density devel- opment (three stories, without land plots, averaging 100 people per hectare), and the rest is housed in high-rise buildings.   SCENARIO 3 — low-density expansion is limited to 10 percent of population growth, 10 percent more is housed in medium-density areas, and 80 percent is absorbed by high- rise, high-density areas (nine stories and higher, at a minimum density of 150 people per hectare). The resulting forecasts suggest that substantial capital investment savings will be achieved across the portfolio of urban agglomerations, with the largest savings in the most populous of them: Tashkent, Fergana/Margilan, and Namangan (Figure B3.1.3). This estimate does not ac- count for operational expenses, which would also be higher in lower-density areas. Figure B3.1.3 Costs of infrastructure provision costs in the largest agglomerations of Uzbekistan across different growth scenarios 1.000.000.000,00 900.000.000,00 800.000.000,00 700.000.000,00 600.000.000,00 500.000.000,00 400.000.000,00 300.000.000,00 200.000.000,00 100.000.000,00 0,00 Namangan Fergana Tashkent Nukus Bukhara Samarkand Karshi Andijan Kokand Margilan city Scenario 1: Scenario 2: Scenario 3: low density expansion medium density expansion densification SOURCE: Original elaboration based on GHS-POP and Uzbekistan Statistics Agency data. 54 THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT? Chapter 4 Key factors that distort urbanization in Uzbekistan THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT? 55 Chapter 4 Key factors that distort urbanization in Uzbekistan The previous chapter presented a detailed view of the process and results of urbanization in Uzbeki- stan. This chapter will focus on the root causes of the weak development outcomes of the urbanization process. First, the chapter looks at critical structural constraints that restrain factor markets, focus- ing specifically on barriers to population mobility. Then, using case studies of three cities the chapter demonstrates that an array of policy conditions inhibits the more efficient utilization of urban land, and enables sprawl as the main model of urban growth. Among these, inefficiencies of land markets and the urban land-use planning system are key. The chapter then considers urban governance. It shows, first, that while recent administrative and budgetary reforms intend to empower subnational governments, for now, municipalities remain heavily dependent on regional and national governments, which have lit- tle chance to predict their revenues and plan medium- and long-term investments. Using the examples of two cities, it shows that with their current capacity municipalities have limited ability to foster local economic development. 4.1 Critical structural constraints to the development potential of urbanization The economic performance of urban areas cannot be separated from the macrostructural conditions of the country at large. Macrostructural conditions shape the development of the national economy in a substantial way and affect every local economy within the country. Currency exchange rate fluctu- ations affect all domestic businesses, particularly importers and exporters, no matter where they are located. Administrative barriers such as difficulties obtaining permits and licenses of various types result in additional costs of doing business, or hinder access to product markets, or the free flow of investment and people, and restrict the ability of all cities to reach their full potential. Despite Uzbekistan making substantial leaps forward in pro-market reforms over the past several years, substantial national-level barriers to private sector driven growth remain. For this reason, it is important to acknowledge the criti- cal importance of continued structural reforms for improving cities’ economic performance. Uzbekistan has been on a path of aggressive market liberalization reforms since 2017, but despite im- pressive achievements, significant challenges persist. The most critical reforms over this time include the unification of exchange rates, liberalization of administrated price controls, and removal of barriers to the flow of goods. These have included the reduction of custom charges, reform and deregulation of the banking sector, a reduction in the tax burden on businesses (particularly the reduction of the pay- roll tax rate, which had kept a lot of employment off the books), and a reduction of the administrative costs of doing business. Despite important successes brought about by these reforms, the transition to a system that enables the efficient operation of markets is incomplete. Key issues that still need to be addressed include overall lack of transparency in the legal sphere and a high degree of state partici- pation in the economy, and persistence of tariff and nontariff barriers to market access and mobility of population and capital (World Bank, forthcoming). While reforms in these areas have been progressing, slower than desirable progress is being made. Critical ongoing reforms include restructuring and privat- izing selected state enterprises (including commercial banks that dominate the financial market), the administrative reform (which is supposed to professionalize the public service and decentralize powers and resources), and the continued easing of the propiska regulations (Yusupov 2020). The success of THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT? 57 these reforms would be critical for the further development of Uzbekistan overall, and the sustainable and productive growth of its cities specifically. The incomplete structural reforms are a key cause of the inefficiency of factor markets, which is imped- ing the process of structural transformation and urbanization. As was discussed in chapters 2 and 3, structural transformation and urbanization are related processes that depend on seamless reallocation of factors of production to different industries and locations. Evidence presented in this report, suggests that these markets do not function efficiently in Uzbekistan. Decision-makers in Tashkent still have sub- stantial say in how land, labor, and capital in the economy are used. Without markets and an enabling environment to let markets guide reallocation of factors of production, resources have remained siloed within sectors and locations over the course of several decades (World Bank, forthcoming), thus holding the structural transformation and related rise of urban economies back in various ways.   Land: Strict controls over land ownership and land-use are still limiting the transfer of almost all nonresidential land rights from less to more efficient users and uses. Most nonresidential land re- mains under public ownership and its transaction and change of use are complicated (nonagricul- tural land reforms are discussed later in this chapter). For example, in agriculture, land allocation to strategic crops (wheat and cotton) remains regulated by the government, thus restricting the ability of farmers to choose the most productive use of land, including legally pursuing its conver- sion to urban uses (World Bank 2019).   Capital: Uzbekistan’s capital markets are at a nascent stage of development. Only 22 percent of firms have access to loans (compared to 38 percent in the Europe and Central Asia region). Mean- while, 85 percent of commercial banks are state owned, meaning their assets are in danger of be- ing allocated following nonmarket principles—for example, to fit the next industrial policy priority, or prioritizing state-owned enterprises (SOEs) over private firms. Furthermore, commercial banks make up 95 percent of total financial sector assets, signaling that capital markets are almost nonexist- ent. Stock and bond markets are underdeveloped, and different forms of equity investment and nontraditional borrowing are lacking. This leaves almost no financing options for the businesses and individuals overlooked by state-owned commercial banks (World Bank, forthcoming).   Labor: As suggested earlier in this report, very low rates of internal migration and limited growth of urban areas in Uzbekistan can at least partially be explained by the constraints of propiska reg- ulations. Evidence from China and its hokou registration system shows that migration restrictions are harmful to urban development. Au and Henderson (2006) provide evidence of how these re- strictions had a negative effect on Chinese GDP in the mid-1990s, by reducing productivity in urban areas due to the lack of sufficient agglomeration effects (Au and Henderson 2006a). Most cities in the country are undersized (Au and Henderson 2006b). Although in Uzbekistan, the most critical constraint to domestic labor mobility—the “propiska” system—is being relaxed, overall cost of liv- ing, and specifically housing affordability will continue to constrain internal migration. Persistence of the state’s role in setting salaries in SOEs (which dominate several major industries) disconnects earnings from the productivity of sectors and areas, and reduces incentives to change jobs and location (Au and Henderson 2006b). The impact of barriers to labor mobility and proposed policy changes to address this are the focus of discussion in the rest of this section. 58 THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT? Figure 4 .1 The effect of propiska restrictions on population growth across the country .04 U Unweighted average population Tashkent city Diff. T-U p=0.4407 All other cities .03 Diff U-TC p=0.0153 T Cities in Tashkent region* Diff T-TC p=0.9411 TC growth rate .02 .01 0 2000 2005 2010 2015 2020 Year SOURCE: State Committee on Statistics of Uzbekistan. NOTE: *Excluding Tashkent city. The population growth dynamics observed in Tashkent in the past two decades can be closely mapped against the changes in the propiska regime, showcasing the system’s continued impact on population distribution. Tashkent city has been among the slowest-growing administrative areas in the entire coun- try in the past two decades despite its high productivity, as propiska was enforced particularly strictly (Muradova and Seitz 2020). Figure 4.1 provides some additional analytical evidence on the effects of propiska. First of all, it is interesting to note how the population growth rate of Tashkent city and other cities of the Tashkent region was considerably lower than in other Uzbek cities at the turn of the century, as restrictions on moving to the capital were tightened after a series of terrorist attacks in 1999 (left side of Figure 4.1) (Hojaqizi 2008). However, as the decade approached its end, the population started grow- ing faster in Tashkent city and the Tashkent region, matching the growth rate of cities in the rest of the country. In 2011, there was a spike in population growth due to mass regularization of unregistered citizens across the country. These regularizations preceded a new set of regulations in 2012 that was meant to, once again, strengthen enforcement of migration restrictions—this time focusing not only on the admin- istrative area of the capital but the entire Tashkent region as well. 24 In the following years, the population growth rate in Tashkent and the cities of the Tashkent region once again fell below that of the rest of the country (middle part of Figure 4.1). However, as some of the restrictions embedded in the propiska system have been lifted since 2016, population growth in Tashkent city has accelerated (right side of Figure 4.1). These dynamics suggest that the propiska system constituted a serious barrier to migration, and small changes in its strictness and enforcement generated large responses in terms of population flows (or at least made legal migration more prominent). Similar conclusions can be drawn from Box 4.1, showing that Uzbekistan has a much lower level of urban primacy than other former Soviet republics, which appears to be associated with the retention of a more restrictive propiska regime than in other countries. The propiska system was a major constraint in the past and still affects citizens’ lives. Despite the fact that in recent years some of the limitations imposed by the propiska system on unregistered dwellers have been lifted, survey data show that many respondents still face significant hurdles when trying to 24  Lipatov 2018; Gazeta.ru: https://www.gazeta.uz/ru/2018/12/06/registration/ THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT? 59 access health, education, and social services (Muradova and Seitz 2020). Migrants unable to obtain the registration certificate to settle in a larger city may nonetheless move but stay unregistered. Since they are not recorded in the official population data, the effective population of urban areas can often be un- derestimated. This phenomenon is particularly common in the country’s largest cities. Muradova and Seitz (2020) estimate that Tashkent city alone accounts for around 15 percent of the country’s illegal dwellers or unregistered residents. The housing market represents an additional barrier to migration. While propiska historically repre- sented a significant barrier to migration (and still does to some extent), another hurdle is the housing market. Low supply of new housing and housing prices that are unaffordable for many potential migrants, particularly in the city of Tashkent, prevent people from relocating. Seitz (2020) reports that in 2018 the housing supply in Uzbekistan stood at less than 1 housing unit per household (0.943), with supply further constrained in the capital, where it was 0.862 unit per household. As a result, housing prices have skyrock- eted in cities (based on estimates, due to the lack of official records) and are largely unaffordable to most of the rural population. Seitz (2020) reports that housing in Uzbekistan is more expensive for the average Uzbek than housing is in New York City for the average American. Supply of housing is quite inelastic as construction is hindered by several (legal and nonlegal) obstacles, some of which will be discussed later in this chapter. As propiska restrictions are starting to be lifted, we can expect a substantial increase in migration and in urban population growth. Allowing rural-urban migration in Uzbekistan would, based on glob- al experience, unleash large aggregate productivity gains (Bryan and Morten 2019; Tombe and Zhu 2019). The progressive loosening of propiska restrictions in recent years represents an unequivocal first step and the available data seem to show its effectiveness (Figure 4.1). But further action is required to address other barriers to migration—specifically insufficient urban infrastructure and service and poor housing affordability (Seitz 2020). As the migration barriers are being lifted, it is critical to recognize their historic impact and what it means for the next phase of urbanization. Over time, migration restrictions have shaped spatial distri- bution of the population and labor force across the territory of Uzbekistan in a manner that is likely quite different from results that a market-driven process would have produced. As barriers are lifted and mar- ket incentives become stronger, the system will start self-correcting, meaning that a lot of people will be migrating across the country, most likely to the largest cities, and above all to Tashkent. This process should be welcome, as it carries the potential to unlock the development opportunities of leveraging agglomeration forces. However, to be able to make the most of this opportunity Uzbek cities need to be ready to accommodate population inflows with infrastructure, housing, services, job opportunities, and appropriate public management capacity. Continued structural reforms, including further easing of propiska, are the foundational requirement for unlocking sustainable and productive urbanization. While this section predominantly focused on internal population migration barriers and their impact, a similar logic applies to land and capital markets. Re- forms required for development and efficient operation of these markets remain essential. Macrostructural reforms would not themselves guarantee that cities become locomotives of economic and social devel- opment, but in their absence, policy interventions directly targeting urban challenges will likely remain insufficient. Without a predictable and stable environment for market forces to guide the allocation of labor, land, and capital, urban development will remain distorted and will fail to produce the development results envisioned. 60 THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT? Box 4.1 What the size of major cities in post-Soviet countries tells us about the impact of the propiska regime Available data do not allow a precise estimation of the detrimental effect of the registration system (propiska) on the size of Tashkent. However, as the system was originally devised and deployed in the Soviet Union (Buckley 1995), many of the former Soviet republics still maintain different versions of the regulation that are enforced with varying levels of severity (Chekirova 2019). As Uzbekistan is one of the countries that has kept implementing the system most dutifully, a simple comparison between former Soviet republics could shed light on its effect on migration. While the differences across countries may be due to a multiplicity of confounding factors, the analysis provides interesting and novel pieces of evidence. First, Uzbekistan is the only country that sees a marked negative trend in terms of urban prima- cy—meaning that Tashkent’s share of population is getting smaller over time, while in other coun- tries the share of population in the largest cities is growing (a far less pronounced negative trend can be seen in Azerbaijan, Figure B4.1.1). This trend predates the dissolution of the Soviet Union. Since 1960, the share of Uzbekistan’s total population living in Tashkent has more than halved— likely the effect of limited migration in combination with higher fertility rates in rural areas. Most of the decline occurred after independence, at the same time when neighbors such as Kazakhstan and the Kyrgyz Republic saw marked increases in primacy. A simple model that predicts urban primacy based on five variables (a country’s gross domes- tic product, total land mass, population, level of urbanization, and share of arable land) shows that Uzbekistan has by far the most undersized primary city and also the one with the largest discrepancy between the actual and predicted size, as the model would expect Tashkent to be roughly double in size (Figure B4.1.2). This result once again suggests that migration barriers and most probably propiska restrictions have had a substantial effect on the spatial distribution of population in Uzbekistan. Figure B4.1.1 Share of people living in the largest city (urban primacy) of former Soviet countries 60 Share of population in largest city Georgia Armenia 50 Azerbaijan Kyrgyz Republic 40 Tajikstan 30 Uzbekistan Turkmenistan 20 Belarus Kazakhstan Russia 10 Ukraine Dissolution of the USSR 0 1960 1965 1970 1975 1980 1985 1990 1995 2000 2005 2010 2015 2020 SOURCE: WDI. THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT? 61 Box 4.1 Figure B4.1.2 Actual and predicted size of the largest city UKR RUS UZB KAZ TKM BLR TJK AZE KGZ GEO ARM 10 20 30 40 50 60 Share of total urban population in largest city Actual Predicted SOURCE: Original elaboration based on WDI. 4.2. Policies and other conditions leading to inefficient urban growth Sprawl is one of the key visible manifestations of the mismanagement of urban growth in Uzbekistan. As shown in chapter 3, Uzbek cities respond to population growth by sprawling rather than densifying, and they do so much more than comparator cities in ECA. In turn, this produces inefficient urban forms, makes cities less dense and compact, more polluting, services and jobs more difficult to access, and productivity-boosting agglomeration effects weaker. But what causes sprawl to be the main pattern of urban growth in Uzbekistan? It is possible that the slow structural transformation of the economy has limited the incentives for densification because lacking services industries are important for increasing the value of city centers and justifying building up. Cultural traditions that lead to greater attachment to single-family homes might also have had an impact. But it is also clear that policy has directly contrib- uted to this pattern of urban growth. This section uses in-depth case studies of several Uzbek cities and their spatial growth to unpack this in more detail. A closer look at selected cities and their spatial evolution shows that urban expansion, at least in part, is a result of regulations and the inefficiencies of government services and systems. Further investiga- tion of the experience of selected cities helps us understand what conditions affect this inefficient pattern of urban growth. Three cities—Namangan (a large city), Karshi (a smaller regional center), and Kasansay (a smaller town)—were selected for an in-depth study of sprawl and its causes. 25 The case studies were based on review of urban planning documents, field visits, and interviews with national and local govern- ment officials and private actors. 25  City selection criteria included: (1) notable size and shape of built-up footprint expansion, which was calculated using the Global Human Settlement (GHS) built-up grid data; (2) cities of different size and different places in the national urban hierarchy; and (3) cities with different patterns of built-up footprint expansion. 62 THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT? To understand the factors affecting the observed growth pattern it is important to conduct analysis on two levels: (1) direct constraints and incentives leading to sprawl, and (2) systematic issues that produce these more specific conditions. On the surface, the conditions that affect this pattern of growth can be broken down into two groups: those that prevent densification and those that encourage sprawl. This section will start by breaking down some of these conditions using the cases of Karshi, Namangan, and Kasanasay to illustrate their effects. The later part of this section will consider two fundamental sys- temic issues that underpin the persistence of urban sprawl in Uzbekistan: underdeveloped land markets and an inadequate system of urban planning and land use regulation. 4.2.1. What conditions are preventing densification? Densification requires a favorable economic and regulatory environment, but in Uzbekistan a number of important conditions work against it. The most fundamental conditions required for building tall26 are financial in nature. The main question is whether capital investment that is needed to build tall is justified by the value the structure will generate in a particular location. But even if such investment is justified economically, it should be enabled by sufficient infrastructure and favorable regulations. The examples of Uzbek cities show that even if building tall might be financially justified in some cases, infrastructure and regulations create barriers. The most direct barriers to the densification of Uzbek cities are inadequate construction regulations. So-called ShNKs (urban construction norms and regulations) and SanPiNs (sanitary regulations and norms) contain outdated norms regulating planning of residential and commercial areas. These documents are relevant for settlements of all types and sizes and contain several provisions limiting densification. For example, ShNK  2.07.01-03, 27 which is one of the key documents regulating construction in urban areas, includes among others, the following norms:   Maximum floor-area-ration (FAR) requirements for housing depending on the type of structures— from 1,200 square meters (m2) per hectare for individual housing (maximum FAR of 0.12), to 6,700 m2 per hectare for 12-level urban blocks (maximum FAR of 0.67).   Requirements to reserve land for playgrounds, open-air workout spaces, and other facilities for public use. The recommended size of such facilities is also regulated based on the number of res- idents.   Minimal building separation distances to ensure exposure of apartments to the sunlight during the daytime, free flow of fresh air, visual privacy, accessibility for emergency vehicles, and so on—min- imum 15 meters for buildings of up to 3 floors, 40–45 meters for 9+ floor buildings.   Increased distances between high-rise buildings in seismic hazard zones—that is, the separation distance between high rises should be at least twice the height of the tallest building. These and other existing regulations and guidelines were based on planning approaches that assume the use of outdated construction and ventilation technologies. In general, none of the modern global ur- ban centers, including those located in seismic hazard zones like Japan, would achieve their current level of density with such construction regulations . Outdated urban planning documents prevent denser developments in Uzbek cities. Before 2016, when the government initiated a massive revision and update of city-level planning documents (“general 26  While building tall(er) is not always directly associated with densification (e.g., tall buildings can be built very sparsely, utilizing a small portion of the land), in most cases building tall(er) is a critical component of densification. 27  Available at: https://mc.uz/gradostroitelnye-normy/ (in Russian and Uzbek). THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT? 63 plans”), many Uzbek cities had not had their general plans updated since the 1980s. 28 Plans of that era directed the actions of the only developer—the state—and policy makers did not see the benefits of densification in urban areas and specifically of urban centers. High-rise housing in the Soviet Union was a cheap and quick way of providing decent accommodation and basic services with access to places of employment (often in manufacturing) for the labor force. In the absence of a land market, locations of new high-rise developments were defined by planners, and this often meant that high-rise residential neighborhoods appeared on the outskirts of cities on greenfield sites, where they were cheaper to build. In a general sense, this can be interpreted as a major example of land misallocation in the absence of price signals and incentives (Bertaud and Ranaud 1995). While some of these plans are still used, even though most are being replaced, their general rationale is still present in new urban planning documents, thus continuing to limit opportunities for densification. The availability and capacity of basic infrastructure, both in larger and medium-sized cities, pose another major barrier to the densification of central areas. Basic infrastructure in many Uzbek cities was constructed in 1960–80 with a view that most multistory buildings would rise to 4–5 floors and some to 9–16 floors in selected locations, while most urban developments would remain low-rise. Since then, population growth and technological development have produced a need for greater density, but infra- structure investment has not been keeping up. Despite regular investments in rehabilitation, infrastructure networks in Uzbek cities today are inadequate and cannot support high-rise development. Furthermore, urban “general plans,” outdated in their content or logic, often suggest further expansion of infrastructure networks, instead of increasing their capacity to enable greater density in urban centers. This is most viv- idly showcased by Karshi, where the infrastructure deficiency is the main constraint on densification, yet the new general plan still falls short of recognizing this challenge and prioritizing investments to address it. Instead, it essentially legitimizes further sprawl and prioritizes further expansion of infrastructure net- works to accommodate it (Box 4.2). In addition to enabling inefficient urban form, this approach obliges the public sector to handle higher costs of infrastructure provision, as upgrading networks is usually sub- stantially cheaper than building new ones. The availability and quality of urban infrastructure reflect multiple systematic challenges related to the widespread practice of infrastructure management and funding. While this topic deserves a sep- arate report, it is worth highlighting that the poor state of infrastructure affecting cities is to a large extent a result of the inefficient national system for infrastructure funding, cost recovery, and manage- ment. Today tariffs for utilities are set by authorities below the cost recovery level. Insufficient incomes of state-owned utility companies are father exacerbated by collection inefficiency. There are no official subsidies to utility companies, so they systematically run at a budget deficit, which results in systematic underspending on infrastructure maintenance and upgrading. Dilapidated underserviced infrastructure brings additional losses to utility companies due to leaks and breakdowns. This produces a vicious cy- cle of “poor maintenance—higher costs—lower spending in maintenance” (Izvorski et al. 2019). This brief illustration suggests that while investment in urban infrastructure is urgently needed, it will not offer a sustainable solution to some of the issues discussed here without a more systematic reform of utility management and cost recovery. 28  Despite the drive to update general plans initiated by the Decree of the Cabinet of Ministers No 3, dated January 5, 2016, and then the Decree of the President of the Republic of Uzbekistan No. УП-2595, dated April 26, 2000, there are still cities with general plans dating back to the 1980s, for example, Mulgan in the Kashkadarya region. 64 THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT? Box 4.2 Karshi—a large city where expansion is driven by poor infrastructure and long-term failures of urban planning Karshi is an important urban center in southern Uzbekistan, and an administrative center of the Kashkadarya region with a population of around 270,000 (State Committee on Statistics of Uz- bekistan 2019). During the Soviet period, Karshi was developed into an important industrial and service center. The city was also acting as a main center of construction and maintenance of vast irrigation infrastructure developed in the Kashkadarya region to increase cotton production. Karshi is a key transport node located at the crossroads of the western, eastern, and southern parts of Uzbekistan. It is also part of the Central Asia Regional Economic Cooperation corridor 6 connecting Europe and the Middle East and South Asia Figure B4.2.1 Map of built-up area expansion in and around Karshi SOURCE: Analysis based on data from GHS-BUILT (2018) European Union, using data from Open Street Maps. Urban planning documents for the city have not been updated for a long time, while weak en- forcement allowed the city to sprawl beyond its administrative boundary. The current general plan of Karshi was adopted in 2013 for the period until 2030. Before that the city had to follow a general plan endorsed in 1986 and last updated in 1989 with an effectiveness period of 1986–2005. Karshi, however, failed to meet the key projected targets—its population increased from 129,000 THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT? 65 Box 4.2 in 1986 to 212,000 in 2005 instead of the 300,000 prescribed by the general plan. But the city’s built-up area expanded beyond expectations—by 2013 the city footprint reached 8,501 hectares, vastly exceeding the 4,340 hectares documented in 1986 and the ceiling of 7,551 hectares allo- cated for development by 2005. Some part of population growth in this territory was accommodated by a number of small urban settlements and rural mahallas around Karshi, particularly in the west and northwest of the city. To justify connecting to existing infrastructure networks of the bigger city (electricity, natural gas, and water, if available), peri-urban settlements became de facto expansions of the city. This fur- ther contributed to a sprawling agglomeration with a highly inefficient urban form. Such a pattern of urban growth and expansion is partially explained by a major infrastructure backlog. The insufficient capacity of infrastructure, which limits high-rise developments, is most acutely felt in the water supply and sanitation sector. As confirmed by local developers and archi- tects, the water supply system in Karshi is partially dilapidated and could not sustain the pressure needed to deliver water to the upper floors of buildings. The sewerage network covers only 46.5 percent of the population, but it has already reached its maximum capacity. Unable to service land for high-rise development, local administration resorts to allowing low-rise expansion to ad- dress the growing demand for housing. Water supply challenges are extremely complex due to the limited availability of water in the area. Water for Karshi residents is delivered from water intakes near Kitab via a 198-kilometer-long pipe- line that is in need of major rehabilitation and upgrading. Households and businesses in Karshi, specifically those occupying buildings taller than two floors, are using special tanks to accumulate water and to use pumps to deliver water when supply is interrupted. Even the flagship project of Karshi municipality—the Karshi City Project—which envisaged the construction of a dozen high-rise housing blocks of 14–16 stories, relies on a similar solution. Ultimately, such infrastructure shortages pose additional costs for construction and maintenance, which often mean that building tall is not justified. The case of Karshi presents a legitimate question: is it even worth creating conditions for the further settlement of an arid region, which will only increase ecosystem pressures? In any case, Karshi offers a general lesson: insufficient infrastructure services in an area where population and housing demand are growing leave only one avenue to accommodate that demand—sprawl. Karshi’s current plan hardly solves the issues that led it to sprawl. A general plan for 2013–30 documented the sprawl that had already happened in the previous years and endorsed further expansion. The total area of Karshi would increase from 8,501 to 17,696 hectares, while the pop- ulation would grow up to 430,000. These targets include merging Karshi with 29 nearby settle- ments with a total population of 109,000. Investment plans provided in the general plan prioritize expansion of networks to cover the underserved low-density suburbs, rather than providing infra- structure that would allow for densification in central areas. Thus, the general plan is held back by the failings of urban planning of the past, as policy makers still need to ensure a basic level of infrastructure and service provision. But more worryingly, it falls short of proposing a vision for a more sustainable pathway of urban growth and development. SOURCE: Based on interviews with local and national authorities, local developers, and architects/urban planners. Data about the master plan were provided by the Ministry of Construction and its local department, unless another source is indicated. 66 THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT? 4.2.2 What conditions promote sprawl? Since the late 1980s, demand for housing for Uzbekistan’s fast-growing population was addressed mostly through the provision of land for individual housing construction—thus producing sprawl. In 1988, the right to grant land plots for housing construction was provided to kolhozes (Soviet collective farms) and local administrations. Land plots were provided for lifelong inheritable possession both in ur- ban settlements and in rural areas. 29 Sizes of land plots varied from 300 m2 to 600 m2 in cities and from 700 m2 to 1,500 m2 in villages. In 1994, this policy was adopted by the Government of Uzbekistan30 and special commissions under local hokimiyats were granted a right to allocate land plots for housing construction, which could often include agricultural land in peri-urban areas. Since the 1990s, individual housing construction on the edge of cities became the only means of ad- dressing growing housing demand—which fueled the sprawl. In the 1990s, the government almost fully retreated from the housing construction sector and many Soviet-period construction giants went bank- rupt, leading to a dearth of new housing. But the population continued growing, and so did housing demand. Additionally, the underdeveloped land market, and a lack of land cadastre, restricted reuse of brownfield sites in cities. Thus, often the only way local governments could satisfy growing housing demand was through allowing building on greenfield sites at the edges of urban areas. While this was not fully legal, it was allowed due to practical absence of state capacity for development control, and general lack of commitment to the implementation of outdated general plans. This also created an ad- ditional income stream for local governments—thus incentivizing them to allow this process. To some extent, local governments’ allocation of land for individual housing construction became a substitute for failed national housing policy. The households that were allocated land for construction (usually through administrative means and informal payments—but not official market transactions) built houses them- selves—often gradually, in stages, over a long period of time, with quality varying based on the house- hold’s financial capacity. Perhaps, at the time, allocating land for housing was a necessary evil; however, it clearly contributed to sprawl and locked cities in inefficient growth pathways. Restricted mortgage credit access and high interest rates are further compounding problems of housing affordability, making semi-formal individual housing development the only solution for many. Market prices for new apartments in cities are in the range of US$250–500 per m2. Mortgage rates vary between 14 percent and 28 percent. Even at 14 percent, the monthly payment for an average apart- ment of 50–70 m2 will be just slightly less than the national average salary (US$250). In smaller towns like Kasansay (Box 4.4), an additional problem even for those who could afford it is that people who work in the informal sector or get remittances from relatives (which is a large portion of the population in such towns) cannot qualify for mortgage loans. Furthermore, as illustrated by the case of Kasansay, banks re- main skeptical of small business employment and prefer those employed in the government or SOEs, yet such jobs are hard to come by in smaller towns. As a result, there is very little effective demand in smaller towns, and thus little incentive for developers to build. According to media reports, the government is making efforts to address this gap by expanding subsidies to households to cover the first mortgage payment. It has dedicated the equivalent of US$24 million in 2020 and US$90 million in 2021 to support- ing 23,460 families. 31 President Mirziyoev also instructed that the central government and local hokimi- yats develop a set of measures to limit the inflation of the costs of housing construction. 32 It remains to 29  Decree of the Council of Ministers of Uzbek SSR, “About measures for further improvement of organization and management of individual housing construction in Uzbek SSR,” No. 263, dated July 28, 1988. 30  Decree of the President of the Republic of Uzbekistan “About improvement of efficiency of land use,” No. УП-1009, dated November 24, 1994; Decree of the Cabinet of Ministers, “About measures for improvement of efficiency of land use”, No. 575, dated November 29, 1994. 31  https://www.gazeta.uz/ru/2021/03/12/ipoteka/. 32  https://www.gazeta.uz/ru/2021/03/03/lodging/. THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT? 67 be seen whether this will have an impact. But so long as housing remains unaffordable, attempting to build a home themselves on the edge of the city will remain the most promising option for many families. In this context a restriction on allocating agricultural land for development, which the government intro- duced in 202133 as a means for stopping sprawl, cuts off the only outlet for satisfying housing demand in many urban areas. Resettlement regulations and practices also promote sprawl, particularly where a large share of mult- inuclear households live in individual housing. According to the current resettlement regulations in Uz- bekistan, 34 a homeowner who is being involuntarily resettled for the public good is free to choose from several means of compensation for housing and land—that is, full financial compensation, equal housing (same living space, number of rooms, etc.) and partial financial compensation, or partial financial com- pensation and a new land plot for the construction of individual housing. And while the land market is underdeveloped, and purchasing land from current owners is complicated, this procedure is followed for most multi-unit residential developments in city centers. These resettlement regulations specifically entitle the current owner to receive an alternative land plot for housing construction alongside another form of compensation in case the expropriated land has been obtained via a land auction (which is the main mechanism for allocating land for private housing construction by hokimyats over the past 25 years). 35 Legislation also states that a so-called minimal social norm should be considered when new housing is provided. This norm particularly benefits multinuclear households since each family in the household can request separate houses and/or land plots for con- structing houses as a replacement for the one housing unit they are giving up. This ultimately means that if the government attempts to expropriate land in order to redevelop a low-rise residential area in a city center, the city would have to allow individual housing construction on the edges to satisfy the rights of people eligible for resettlement. The area dedicated to such resettlement would often exceed the total area of the land that was redeveloped. The example of Namangan (Box 4.3) demonstrates how the combination of an underdeveloped land market and a lack of timely resettlement solutions leaves the city no choice but to sprawl further. While fair resettlement is critical, it is worth considering two directions: (1) how expropriation regulation can be adjusted to limit sprawl, without preventing fair compensation, and (2) how a market-oriented mechanism of voluntary resettlement can be deployed. 36 Moreover, revenues from building and selling new low-den- sity housing are not sufficient for recouping the high cost of resettlement. Hokimyats governing peri-urban areas are in fact incentivized to allow build-up expansion toward the major city. Smaller urban and rural settlements in proximity to bigger cities in recent years tended to grow toward the cities, which was often allowed by potentially illegitimate and unplanned land allo- cation. Land in proximity to existing settlements and along major roads was in high demand for housing construction due to easier access to basic infrastructure (e.g., electricity), and better access to urban jobs and services. Often the peri-urban district hokimyats granted these in-demand lands for private use, despite being in breach of land use regulations, as this generated income and satisfied real eco- nomic demand. The land allocated to private farmers and dehkan (household) farms was often illegit- 33  The restriction was introduced under implementation of the Decree of the President of Republic of Uzbekistan, “On additional measures to improve the system of use and protection of agricultural lands,” adopted on February 24, 2021, https://lex.uz/ru/docs/5306359. 34  As provided by the previous and current editions of the Housing Code (articles 27–31, 42) and Land Code (article 41), and decrees of the president and cabinet of ministers adopted in different years to specify details of the resettlement policy. 35  Temporary provision, approved by the State committee on state property, dated January 26, 1995; CoM Decree No. 146, dated May 25, 2011; CoM Decree No. 63, dated January 28, 2019. 36  While practically all countries have land expropriation by government as a necessary legal instrument, practices are clearly shifting toward using voluntary acquisitions as a preferred approach (e.g., in Sweden and the United States). Besides, land acquisition for private housing or commercial redevelopment using expropriation is often challenge in courts or not allowed, because such redevelopment does not constitute “public needs.” 68 THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT? imately used to build individual housing, but this was later often legitimized by local authorities to pre- vent community discontent and allow households to get official connections to infrastructure (electricity, water, and natural gas, if available). 37 In addition to income from land allocation this gave peri-urban hokimyats a rationale to seek connection to the infrastructure networks of larger cities, making it easier and cheaper to service their population. This process accelerated the merger of smaller settlements into larger agglomerations, as was discussed in chapter 3. Centralized development of urban planning documents for cities contributes to this pattern of growth. As per Uzbek legislation, supervision and updates of general plans for cities should be done by dedi- cated design institutions (Uzshaharsozlikliti and Toshkentboshplanliti) at least every 10 years and be fi- nanced by the local authorities themselves. However, the municipalities never have incentives or budgets to do so, which let the general plans lapse and their implementation go unsupervised. Additionally, urban planning documentation for cities required approval of the Cabinet of Ministers and Oliy Majlis (national parliament). In contrast to these complex centralized processes, procedures to develop and endorse planning documentation for urban-type settlements (UTSs) and rural settlements is comparatively simple and gives local hokimiyats some flexibility to decide on the expansion of settlements (as per provisions of ShNK 1.03.02-04). This extra flexibility has enabled peri-urban sprawl—responding to otherwise unsat- isfied demand for land for housing and economic activities—as described above. City hokimiyats tend to support expansion of UTSs and rural settlements toward their territory, as it becomes a de facto way of accommodating housing demand from the city itself. Furthermore, this urban growth pattern was technically in line with the existing urban planning norms and standards. Specifically, ShNK  2.07.01-03 provides directions to prioritize individual housing devel- opment for small- and medium-sized cities, low-rise housing (2–3 levels) for all cities, and medium-rise housing (4–5 levels) for big cities—thus providing legal justification for this form of urban growth. It also (1) contains recommendations on the maximum use of available greenfield sites within the existing urban area for new housing construction to avoid demolition—thus creating a disincentive for changing the essence of the urban form and incentivizing infill development; and (2) allows for the expansion of the ter- ritory of settlements proportional to projected population growth to ensure sufficient housing and other facilities. 38 Box 4.3 Namangan—a large city stuck in an inefficient growth pattern Namangan is a big and fast-growing city captured in a trap of inefficient spatial develop- ment patterns due to a lack of timely and efficient interventions. Key propositions of previous and current general plans on restructuring the urban core and transport network were never implemented in full due to high costs of resettlement, among other issues. However, dynamic economic development of the city and a subsequent increase of property values, coupled with low density of redevelopment dictated by existing norms, made the involuntary resettlement of people occupying low-rise housing that dominate the city center even more expensive and complicated. Unfortunately, many larger cities in Uzbekistan, given the current trends, could face similar challenges in the future. 37  It is hard to say how common this specific process was, but this is supported by anecdotal evidence collected from different sources during report preparation. 38  ShNK 2.07.01-03; additional guidelines are provided in other ShNKs as well. THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT? 69 Box 4.3 Namangan is the second-largest city in Uzbekistan, with more than 630,000 residents according to official statistics. It is an administrative center of the Namangan region and one of the biggest consumer service centers in the Uzbek part of the Fergana Valley, attracting new residents and many day-time visitors from the neighboring smaller cities and villages. The official population of Namangan increased by 62 percent in 2000–20, which is one of the highest growth rates among Uzbekistan’s big cities. However, the city is struggling to accommodate this growth and manage increasing traffic and congestion and address infrastructure backlog and poor urban services. Figure B4.3.1 Map of built-up area expansion in and around Namangan SOURCE: Based on data from GHS-BUILT (2018) European Union, using data from Open Street Maps. The city was developing without a full set of urban planning documentation until recently. The previous general plan was approved in 1986 for the period until 2005, but it was hardly followed by the local authorities. The current general plan was approved in 2007 despite the fact that it was developed in 1999–2000 for the period until 2025. Another key planning document identifying the administrative boundary of the city was approved only in 2019. Lack of land use planning and enforcement resulted in sprawl in peri-urban areas and on agri- cultural lands. Namangan and neighboring municipalities were allocating land without a clear 70 THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT? Box 4.3 understanding of what is a part of the city and what it is not. Smaller urban settlements and rural mahallas (neighborhood wards) around Namangan expanded toward the city until they became a part of it. Recent supervision of the general plan’s implementation confirmed an increase in Namangan’s total land area of up to 17,884 hectares in 2019, which is almost 94 percent larger than it was in 1999 (9,219 hectares) and 30 percent larger than the old projec- tions for 2035 (13,840 hectares). Recent updates to the general plan propose further expansion. While these land allocations do not directly represent the buildup area of the agglomeration, the decision to keep expanding the territory of the city points at the fact the urban sprawl has exceeded predictions. Transport and basic infrastructure networks of Namangan are lagging behind. The city’s gener- al plans of 1986 and 2000 proposed considerable restructuring of its transport networks. How- ever, due to the high financial and social costs of resettlement, authorities were reluctant to implement those plans. Around 85 percent of housing in Namangan is individual housing with a notable number of multinuclear households. This implies additional costs of resettlement since each family in the household is entitled to receive separate housing or financial compensation and a land plot for individual housing construction. Nowadays the density of the road network in Namangan is only 0.9–1.1 kilometers (km) per 1 square kilometer (km2), 39 which is only half as high as the level recommended by the national construction norms and regulations and far below international comparators (as shown in Figure 3.4, the average in cities of the Europe and Central Asia region is 10–14 km of roads per km2). The old section of Namangan, extending to the south from the city center, is filled with densely packed individual homes. Narrow and curvy inner streets complicate any infrastructure upgrading efforts, while residents suffer from a lack of urban services. Existing demand for urban housing in Namangan remains unaddressed due to lack of supply at affordable prices. The government reduced the allocation of land for individual housing to the minimum, but demand for housing is high. The urban core of Namangan was densified to some extent with infill development in recent years. However, opportunities for infill development are being exhausted. Due to the high cost of resettlement, local developers prefer to implement new projects outside the urban core, in the peri-urban areas. This further promotes sprawl and traffic congestion, while key structural issues in the city remain unresolved. Namangan missed opportunities to restructure its urban core and get prepared for further growth. Any future efforts to optimize urban structure will come with higher costs for Namangan. The gen- eral plans in the past provided a development vision for the city, but the vision was not supported with necessary policy decisions and investments, as demonstrated by the case of the failed road network development. Other big regional centers in Uzbekistan could face similar challenges in the coming years if urban growth continues without targeted efforts to reshape urban cores and get cities prepared for smart densification. SOURCE: Based on interviews with local and national authorities, local developers, and architects/urban planners. Data on the master plan of Namangan was provided by the Ministry of Construction and its local department unless another source is indicated. 39  Data from Namangan Hokimyat (assessment of road density might be conservative compared to Open Street Maps–based assessments that are informed by spatial imagery that may count approach roads and roads that are not recognized officially). THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT? 71 Table B4.3.1 Key characteristics of Namangan, and plans Box 4.3 Actual condition, Projected Actual condition, Indicators Units 1999 for 2025 2020 Population thousands 370.3 630.0 626.1 Total area hectares 9,318.8 13,840.0 17,884.0 Population density p/ha 39.7 45.5 35.0 Housing, total m2 3,154.6 11,341.3 9,705.5 Individual m2 61.9 60.0 85.3 2–5 stories m2 38.0 35.0 14.0 7–9 stories m2 0.1 5.0 0.7 Housing per capita m2 8.5 18.0 15.5 SOURCE: Data provided by the Ministry of Construction. The table shows the situation documented in 1999 as part of the formulation of a new general plan, projections of the general plan for 2000–25, and the situation documented in 2019–20 during recent revisions of the general plan. 4.2.3 Underdeveloped land markets are impeding the efficiency of land use in cities Underdeveloped land markets have direct impact on urban growth and sprawl. While the importance of land markets for compact urban development might not be intuitive, they are the most critical compo- nents. Essentially, without price signals, it is almost impossible for cities to ensure that land is used in the most productive and efficient way. Alain Bertaud40 summarizes how the absence of urban land markets affects urban forms in very practical terms, based on his experience of working in post-Soviet Russia and in China, in his recent seminal book Order Without Design (see Box 4.4) Box 4.4 Role of land markets in urban transformation In a market economy, when the potential rent of a lot is higher than for its current use, the owner of the lot has a strong incentive to sell or redevelop the land for a more profitable use. In this way low-rise buildings are transformed into high-rise buildings, and warehouses into office buildings. The increased price of the land under the new use pays for the demolition and relocation of the obsolete building. The land use transformation process is triggered by land prices. In the command economy,41 there are no price signals, so an obsolete land use is likely to remain in effect for a very long time. Let us take the example of a factory built a long time ago near a city central business district, on a what would now be a very desirable piece of land for a department store or an office building because of its accessibility. The state firm owning the land use rights cannot move its factory to a different part of town that would be more convenient for operating the factory, because the land occupied by the factory has no value. 40  In the past, a lead urban planner at the World Bank. 41  While Uzbekistan today cannot be classified as a command economy, this example still applies amid ongoing weaknesses in the land market. 72 THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT? This has consequences for the structure of cities. The newest buildings are always found in areas Box 4.4 newly developed in the suburbs. In Russia, for instance, factories built in the nineteenth century or in the first half of the twentieth century found themselves in what is now the downtown area. High-rise residential buildings are found in the periphery of cities, while low-rise buildings are found closer to the center. Thus, high population densities are found in the suburbs, where land values will be the lowest if located in a market economy, and low densities are found close to the city center, where land values would be the highest. SOURCE: Quote from Bertaud (2018). The sprawl of Uzbek cities is in large part a reflection of land market inefficiencies, and while recently updated legislation promises improvement, it needs to be fully implemented to have an impact. The urban land market in Uzbekistan has advanced since Soviet times, when all land was under public own- ership and only an informal market existed. As of now, most of the land is still owned by the state, as only nonagricultural land plots occupied by individual houses, or by some of the industrial and commercial real estate properties, are available for privatization. Legal and institutional frameworks for the land sector, land administration, and management systems are yet underdeveloped and therefore several issues stemming from this situation continue to constrain the development of formal urban land markets. So, while the example described in Box 4.4 does not apply in full, there are aspects of it that describe the situation in Uzbekistan. On the most basic level, the small share of privately owned urban land impedes the process of transferring land to its most efficient use. Further, due to inefficiencies and regulatory lim- itations of land administration and land use planning systems (discussed in detail later), and an overall deficient investment climate, some real estate and most land transactions are conducted informally, which means that information about land values is limited and market signals that should drive investors to underutilized land are weak. 42 The ongoing land reform has made substantial advances, yet it still has some important gaps constrain- ing the development of land markets. Land reform has been one of the key priorities of the Government of Uzbekistan in recent years. Land privatization was introduced in legislation in 2019 (Republic of Uzbekistan 2019), and was further reconfirmed and expanded by the new law signed by the President on November 15, 2021 (Republic of Uzbekistan 2021; Box 4.5). The key decision to advance privatization and thus to simplify the transactions of land and real properties in urban areas is a major step toward enabling market forces to aid the more efficient utilization of urban land. The nonagricultural land privatization legislation simplifies the process of obtaining land rights, and re- duces the legal complexity of land tenure. One of the big achievements of this reform is the simplification of the terminology of land tenure typology and the introduction of an exclusive list of only three types of land tenure: ownership, lease, and continuous use. The previous legal restrictions on sale, donation, ex- change, and use of land as collateral have been removed. The law also provides a provision to transform previously privatized nonagricultural land that has been in continuous possession into ownership. This leg- islation sets a critical framework for operation and further development of the urban land market, though critical implementation challenges remain. 42  Based on interviews with government and local officials; no precise data on forms of urban land ownership are available. THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT? 73 The institutional framework of the land sector in Uzbekistan has recently undergone substantial changes and needs further adjustments and capacity building to enable efficient operation of the market. Land administration and management institutions and practices are critical for transparent, effective, and fair fulfilment of functions and processes that enable land market development. Critical changes that were in development since December 2017 were finally introduced on September 7, 2020, with the release of two decisions by the President of the Republic of Uzbekistan: (1) Decree of the Pres- ident abolishing Goskomzemgeodezcadastre (GKZGDK), formerly responsible for majority of the land management functions; and (2) Resolution No. PP-4819 creating a new state institution—the Chamber of State Cadastres (CSC). Additionally, the new Cadastre Agency (CA) was established under the State Tax Committee. As a result, there are now three main public stakeholders of the land administration system in Uzbekistan with distinct and well-defined responsibilities: (1) the Ministry of Agriculture, (2) the CA, and (3) the CSC. The Ministry of Agriculture has been delegated the tasks, functions, and powers of GKZGDK to monitor agricultural land and crops. The new CA is responsible for quality control and supervision of cadastral and geodetic works conducted by the CSC and implementation of the National Spatial Data Infrastructure (NSDI). The 206 district offices of the CA are responsible for local land administra- tion, quality control, and supervision of CSC regional offices’ functions. The CSC manages a real estate cadastre and registration, mapping, and spatial surveys as well as some land use monitoring functions. The CSC maintains and manages four independent registries covering: land, residential buildings, non- residential buildings, and mortgages. Development and modernization of a unified real estate cadastre, a critical tool for land market operation, is under way, but still has a long way to go. A comprehensive inventory of cadastral and registration records has been conducted in the past two years along with their digitization. As of Sep- tember 1, 2021, the records of the CA/CSC indicate that there are around 8 million cadastral objects, 35 percent of which are not yet registered, and around 8 percent of the total objects contain no cadastral information. This means that rights associated with a significant share of land assets cannot be easily transferred yet. Box 4.5 Summary of Law No. 728 on Privatization of Non-Agricultural Land Parcels The Law on Privatization of Non-Agricultural Land Parcels No. 728, signed on November 15, 2021 (which replaced the previous law of August 13, 2019), regulates the privatization process of state nonagricultural land in Uzbekistan, including vacant land, individual housing construction, land in permanent use, or lifetime inheritable possession. Nonstate legal entities and physical persons of Uzbekistan are eligible to participate in the privatization process. The law regulating privatization of state nonagricultural land is a significant step toward devel- opment of the land market. A new edition of the privatization law, signed on November 15, 2021, introduces the fundamental principle of integrity of buildings and the land under them. It makes possible privatization of land for entrepreneurial purposes and allows privatization upon payment of corresponding fees, which are based on market value in case of new plots and are based on multiples of the land tax rate for plots already in possession. The new law provides a possibility to privatize already held land plots situated in the areas with no approved and published spatial planning documentation if the land use category remains unchanged. 74 THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT? Land with mineral deposits, land plots under apartment buildings and adjoining land plots, land Box 4.5 plots within strategic economic zones, and state-owned strategic and protected objects are not subject to privatization. The law prohibits privatization by foreign citizens and companies with foreign stakeholders. The revenues from land privatization go to the national budget of Uzbekistan. The law provides for two forms of privatization:   Buyout of those who have land in permanent use or lifetime inheritable possession, whereby the cost of the land is determined by the Cabinet of Ministers of the Republic of Uzbekistan. The title of ownership is issued by the State Assets Management Agency.   Plot sale through an open electronic auction, whereby vacant land is selected by the local municipality/state assets management or cadastral agency, with the starting price deter- mined by market value, including the cost of existing structures, if available. The regulations defining the application of this law are being elaborated. A pilot privatization project in the Syrdarya region highlighted the shortcomings of the current legislative, institution- al, and operational setup. The main identified shortcomings included: (1) absence of updated general plans for settlements and territories; (2) lack of integration between data in the state urban cadastre, state land cadastre, and the registry; (3) absence of the known market values of land assets; (4) no provision to privatize plots within small industrial zones; and (5) lack of interest among local authorities in putting up free land plots for privatization. Unfortunately, several of these shortcomings (3–5) have not been addressed by the new law signed on November 15, 2021. There seems also to be an overlap in responsibilities for the desig- nation of land plots for privatization between the local authorities, State Assets Management Agency, and CA. 4.2.4 Urban planning system is inadequate for the needs of cities The centrally planned economy approach to urban land use regulation is inadequate for the devel- opment of land market conditions, and unlike with land reforms, there has been little progress in this area so far. The previous section has demonstrated that land markets are critical for the productive and efficient use of urban space. Conversely, a city’s efficiency and productivity are hindered if the urban land use planning system distorts, rather than enables, the functioning of the market. As evidenced by multiple regulatory measures discussed above, the urban planning system in Uzbekistan today mostly resembles that of a planned economy, and the philosophy behind the system is, to a large extent, at the origin of its systematic flaws. In the Soviet period, development norms and guidelines replaced land markets for al- locating land (Bertaud 2018). The purpose of planning was to exercise control through centralized norms, rather than creating the conditions for market forces to lead the way in urban development at the local level, while ensuring that public goods were provided. The prescriptive approach to urban planning still restricts the flexibility required to adapt the use of land to evolving economic activity. In the long run, rigid land use categorization results in obsolete urban space, which has a negative effect on a city’s economy. While these challenges are starting to be recognized, compared to highly advanced land reforms, urban THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT? 75 land use planning reforms are only starting to be discussed and prepared. It is important that the reforms address the underlying issues causing inefficiencies in urban planning, rather than just solving the surface problem or updating the lapsed planning documents. One of the main carryovers from the Soviet planning system is a high level of central control. Uzbekistan inherited the Soviet urban planning system and construction regulations, and its centralized, top-down characteristics persist today. The process of spatial and urban planning is specified by the urban con- struction code and land code, which are laws developed by the Ministry of Construction and approved by the national parliament. According to the design of the highly centralized land use planning approach, the spatial development and the land use planning documents are expected to follow a strict hierarchy (Figure 4.2). But due to the absence of some of the higher-level documents this very centralized system does not work as intended. For example, while the general plan of a specific city may refer to the planning direction at the national and regional levels, some of these guiding documents are either outdated, or were never completed. This becomes a source of inconsistencies in the system, leaving individual cities without clear guidance and direction, and revealing the weaknesses of this highly centralized approach.43 Furthermore, plans for larger cities require the approval of the central government and the parliament, and the faculty to formulate these plans is restricted to certified national design institutes. The formulation process is complex and very resource intensive, which makes it difficult and expensive for municipalities. In this context, local governments are discouraged from developing their planning capacities. Ultimately the centralized system often leads to weak or deficient urban land use planning at the local level. Figure 4.2 The hierarchy of land use planning documents in Uzbekistan General scheme of location of population National • Scheme territorial planning of Uzbekistan level • Sectoral schemes of territorial development (industries infrastructure, etc) • ... Territorial plan of region Region and • Territorial plan of district (-5) district level • Territorial plan of urban agglomeration • ... General plan of citv (settlement) • Administrative border of city (settlement) City level • Detailed plan of selected territories in city (settlement) • Detailed construction plans • ... Only selected state-owned design institutions are eligible to develop urban plannine documents The current general plans are ineffective guides for urban development, and weak implementation and enforcement undermine them even further. The general plans of Uzbek cities today are mostly generic and lack the analytical foundation that would enable them to respond to local context and link spatial 43  Key national-level documents focus on national territorial planning and development, including by sector. Regional-level documents focus on plans at the district and municipal levels. 76 THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT? and economic development. The absence of participatory input in the plan development process means that it is often disconnected from the real needs of the population and the private sector. This means that urban development planning and regulation are to a significant extent disengaged from the local context, thus leaving space for unplanned growth. Moreover, insufficient flexibility of general plans and their long-term nature are in intrinsic conflict with rapidly changing economic needs and make unplanned land development practically unavoidable. Development control regulations, which are also determined at the central level, are unprepared to guide the market. This fundamental disconnect between the system of urban planning and land use planning, on the one hand, and the need for private investment in urban development, on the other hand, poses a question as to what extent the general plans—in their current form—are implementable. Even simple monitoring of the implementation of general plans is hindered by a lack of transparency in land allocation. A large number of illegal decisions at the local level (Kun.uz 2021) indicates the gravity of this issue. The current approach to urban planning is not forward looking. Rather than anticipating growth and pro- posing a vision for spatial development of a city, these plans largely document the reality on the ground. In the period between 2011 and 2014, the guidelines for the formulation of general plans were revised several times,44 and planning documentation of 96 cities, 87 UTSs, and 550 settlement areas was updated be- tween 2014 and 2018. By 2022, the process should be completed with an additional 23 cities and 994 UTSs expected to update their general plans.45 However, the methodology applied in the updates still follows old-fashioned planning standards. The core of the urban construction and sanitary norms (ShNKs and SanPiNs) are still based on approaches from the Soviet period. In particular, these norms—by not permit- ting higher land use densities—do not allow the utilization of a full economic value of urban land for both public and private benefits. Under these regulations, and largely following the Soviet planning philosophy, the plans continue to accommodate and incentivize urban sprawl. Particularly, the examples of Karshi, Namangan, and Kasansay confirm that general plans avoid recommending redevelopment approaches that would allow land uses in urban cores to shift according to market demand. Instead, they prescribe the new development of low-rise housing. Spatial data are insufficient for urban planning to be an evidence-based practice. The scarcity of spatial data at the local level—for example, on the availability of serviced land tracts, vulnerable and environ- mentally sensitive areas, household characteristics, property ownership, development control applica- bility to plots, and transport modelling, among others—impairs the effectiveness of urban development plans. Inaccurate forecasts, as in the case of Kasansay (see Box 4.6), where the previous general plan estimated greater population growth and territorial expansion rates than occurred, hinder decision-mak- ers and planners from having a complete picture of demographic and spatial trends. The poor quality of population data (until the results of the first census in 30 years are available sometime in 2022) can deter the quality of urban planning. Forecasting will be of fundamental importance to anticipate the impact of migration on land, housing, and infrastructure needs as mobility restrictions are relaxed. In a context of urbanization, inaccurate forecasting and piecemeal decision-making on infrastructure might result in a lock-in to inefficient spatial patterns for generations. In addition, hard-to-obtain geospatial information impedes developers from making realistic business plans and constrains private sector investment and development of the economy overall. There is a disconnect between general plans and infrastructure investment. Currently, the objectives set in general plans are decoupled from the programming of infrastructure. There is no explicit formal link between the infrastructure needs projected in general plans and government investment programs 44  Decrees of the Cabinet of Ministers No. PKM-146, dated May 25, 2011; No. PKM-54, dated February 25, 2013; No. PKM-140, dated June 3, 2014. 45  Decree of the President of the Republic of Uzbekistan, No. PP-3502, dated February 2, 2018. THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT? 77 for infrastructure development. This creates an uncertain setting where the proposed infrastructure net- works may be drawn up in planning documents, they remain undeveloped, and therefore maps have limited resemblance to what is really built. Even for updated general plans, there is no guarantee that central or local budgets will be available to meet required infrastructure investments. The cost of infra- structure is factored to some extent in the price of land for lease, but it is unlikely that provisions would be enough to offset the cost of construction and maintenance. Different infrastructure assets are built and managed by different state companies and government entities§, and current coordination levels do not contribute to clarity. 46 The planning and construction-permitting process is inefficient, and building codes need to be up- dated. A property developer is unable to know what the buildable area of the potential investment is, and therefore if a proposed real estate development is viable, until the project obtains an Architectural and Planning Task (APZ in Russian), a planning permit issued by the local department of the Ministry of Construction following a case-by-case assessment. The APZ, which enables the developer to start pre- paring construction documents, provides parameters such as the borders of the plot, plot occupation, building use, and maximum height. This may create diseconomies, since knowing the buildable area beforehand would allow developers to make a more targeted business model for the proposed devel- opment. Besides, the process of obtaining APZs in other former Soviet countries (e.g., Russia, the Kyrgyz Republic) had been associated with high corruption, until it was replaced by a more transparent system of legal zoning. Building codes include nationwide measures for fire safety, construction in seismic re- gions, and other forms of protection from geohazards. A number of these codes are over 20 years old and need to be updated with new technologies and standards (UNECE 2015). Box 4.6 Kasansay’s ongoing expansion into agricultural land The lack of a comprehensive approach to spatial planning and development in Kasansay has left it lacking an urban core. While the urban expansion of Kasansay was moderate compared to the national average since 2000, several smaller urban settlements and rural mahallas (neighborhood wards) on the periphery grew dynamically with low-rise, individual housing. Today, Kasansay is the center of a small agglomeration across administrative boundaries with a total population ap- proaching 100,000, but it looks and feels more like a large village. Kasansay city is the administrative center of Kasansay district in the Namangan region with a pop- ulation of 54,500. Kasansay is physically merged with four urban-type settlements and two rural mahallas with a total population of around 37,000. The population of these settlements grew faster than that of Kasansay and substantially contributed to both population growth and the expansion of the built-up area on this territory. In 2010–19 the population of Kasansay increased by 14.6 per- cent, while the total population of neighboring settlements grew by 22.9 percent, as per data of the State Committee on Statistics of Uzbekistan. 46  Interviews with representatives of select government agencies and municipalities. 78 THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT? Figure B4.6.1 The footprint of low-rise housing expanded consid- Box 4.6 Map of expansion in Kasansay erably, but was not captured by existing planning documentation as it happened outside the admin- istrative boundary of Kasansay. Most of the expan- sion happened on the fringes of surrounding smaller settlements. The previous general plan of Kasansay was developed in 1990 for the period until 2010 and envisaged a gradual expansion of administrative territory of up to 2,432 hectares and an increase in population of up to 70,000. However, in 2019 the city remained well below these targets—its territory was around 1,356 hectares, or only 56 percent of the projections. At the same time, neighboring rural and urban settlements were expanding on the fringes of Kasansay (figure B4.6.1). The recently developed, general plan for 2020–40 proposes the merging of Kasansay with six neighbor- ing settlements, allows further sprawl, but proposes some densification. The current territory of the city will grow by 1,815 hectares. This includes 731 hectares of greenfield on the south and southwest that will be developed as a new industrial zone and mixed-use SOURCE: Analysis based on Global Human area consisting of 1–2 story residential and com- Settlement Layer and Open Street Maps. mercial facilities. Along selected main streets in the eastern and western parts of the existing built-up area, 2–4 story housing is proposed that implies demolitions and resettlement. The proposed plan would reduce the share of individual housing from the current 94 percent to 65 percent if implemented. The recently introduced strict limitations on allocation of land for individual housing construction are stimulating a demand for apartments in the urban core. Most of Kasansay is low rise but filled with relatively densely built individual housing with very limited opportunities for further infill de- velopment. Since resettlement in the center of Kasansay is complicated and too expensive to implement through expropriation, local developers are favoring greenfield development on the north and south edges of the city, which could further contribute to the sprawl. Local population cannot afford apartments in newly built high-rise buildings, partly because peo- ple cannot access mortgage loans, which limits the commercial incentives for high-rise develop- ment. Only a few modern urban housing blocks were constructed in recent years in Kasansay. According to local developers, the cost of new housing is not affordable for most of the local population and is on the rise due to the increasing prices of construction materials. At the same time, developers working in Kasansay are small and cannot enjoy economies of scale and the mitigation of increased construction costs. Few clients can pay cash while mortgage credit is not THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT? 79 available for many local residents due to high rates and banks’ requirements for income verifica- Box 4.6 tions. Around 70 percent of Kasansay residents are self-employed, employed by micro and small firms (involved in local trade and related activities), or participate in labor migration and cannot confirm that their incomes are high enough to cover the monthly mortgage payment. Moreover, local banks are reluctant to provide mortgages to local entrepreneurs and the self-employed, even if their incomes are confirmed by tax authorities. Banks prefer government sector employees (i.e., of hokimiyat, educational, and health care institutions) as clients since their monthly incomes are stable and can be easily confirmed. SOURCE: Based on interviews with local and national authorities, local developers, and architects/urban planners. Data on the master plan of Kasansay were provided by the Ministry of Construction and its local department unless another source is indicated. Table B4.6.1 Key parameters of previous and current general plans of Kasansay Projected Actual condition Projected Expected change Indicators Units for 2010 in 2019 for 2040 in 2019–40 (%) Population thousands 70.0 54.6 117.0 214.3 Total area hectares 2,432.0 1,356.0 3,171.0 233.8 Population density p/h 29.9 40.2 35.9 89.3 Housing, total m2 1 126.2 616.4 2 106.0 341.7 Individual m2 699.4 577.6 1,371.0 237.4 2 levels m2 80.0 26.5 332.7 1,255.4 3 levels m2 31.5 - 155.8 - 4 levels m2 313.1 12.3 246.4 1,998.7 Housing per capita m2 16.0 11.3 18.0 159.3 SOURCE: Data provided by the Ministry of Construction. NOTE: The table shows projections provided by the general plan of 1990, actual conditions documented in 2019 during formulation of the new general plan, and some key target parameters of the new draft general plan for 2020–40. 4.3 Local government shortcomings that impede potential Weakness of local governments in Uzbekistan aggravates challenges of managing urban growth in Uz- bekistan. Cases of selected cites show that municipal governments (city and district hokimyats) have limited capacities and budgets to improve local infrastructure or service provision or implement compre- hensive programs to aid local economic development and to manage growth. Development tools and incentives available to local governments are mostly limited to allocation of land and provision of access to existing infrastructure networks. Their budget does not allow for substantial capital investments, and 80 THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT? the budgetary system limits their ability to plan expenditure for the future. Recently initiated reforms of the local finance system and interbudget relations are gradually improving the situation; however, for now the reforms have not granted local governments any additional ability to predict and plan their revenues and spending obligations in the medium and long run and manage them more independently. Despite ongoing reforms, the governance system in Uzbekistan remains highly centralized with the Pres- idential Administration and Cabinet of Ministers having a strong impact on local development issues. The structure of the governance system in Uzbekistan includes: the national government; a second tier that includes the government of the autonomous Republic of Karakalpakstan and 12 regional hokimiyats and the hokimiyat of the City of Tashkent; a third tier that includes hokimiyats of cities and districts. The president directly appoints the hokims (governors) of the regions, who in turn appoint hokims of districts and cities (Izvorski et al. 2019). Current legislation does not provide clear delineation of powers between regional, city, and district-level governments. The law on local government indicates similar functions and responsibilities of hokims of regions, districts, and cities (Republic of Uzbekistan 1993). Local hokims imple- ment these functions in their respective territories and coordinate with regional authorities to make sure local developments are following the overall regional development strategy and plans. In reality, this may lead to the concentration of decision-making power at the regional level as local governments often lack capacity and budgets to come up with any initiatives of their own. Grassroot elements of the governance system in Uzbekistan is represented by urban and rural mahallas —self-governance institutions in communities of cities, towns, settlements, and villages. The mahalla has a wide range of responsibilities delegated by the government, mostly related to assisting various government agencies in implementing their functions on the ground; however, its core functions are re- lated to social protection and community mobilization. Mahallas are responsible for allocations of social support allowances to residents and conducting community engagement activities, and supporting im- plementation of selected government programs (e.g., promotion of entrepreneurship) and several other neighborhood-level functions. While mahallas are formally independent, they operate under the oversight of local hokimiyats, who are financing their activities. Recently, mahallas have been at the center of the new participatory budgeting experiment, whereby the residents of a mahalla can decide what allocated funding can be spent on.47 The rest of this section will focus on two overarching issues through which local governance can be linked to poor urban management: (1) inadequate local government finance, and (2) limited capacity and self-suf- ficiency of local governments. 4.3.1. Local government finance challenges Since 2018 the government is gradually advancing the public finance reform with an aim to create a more decentralized, efficient, and accountable public finance system—but results are not clear so far. Key ongoing reforms include (1) revision of distribution of local and national taxes and formulation of clear rules for interbudgetary transfers to provide local governments with the ability to plan beyond a one-year horizon; (2) expanding powers to local governments in planning and spending, and generation of addi- tional revenues for local budgets; and (3) promoting wider citizens’ participation in making decisions on budget allocations, both through direct consultations at the grassroots level and via local assemblies. While the fiscal reform is not yet finished, the latest relevant document is the government-approved distri- bution taxes to different budgets for 2021 that was passed as a part of the budget law for 2021 (see table 4.1; Republic of Uzbekistan 2021b). The proposed allocation is a notable improvement relative to the pre- 47  All tax and nontax incomes collected or generated by hokimiyats in excess of approved budget parameters. THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT? 81 vious system, whereby regional and local governments had to negotiate distribution of budget funds and with higher-up authorities on an annual basis. The proposed distribution was valid for 2021 and extended for 2022 with minor revisions (Republic of Uzbekistan 2022). Additionally, the practice of negotiating shared taxes with each municipality every year was continued in 2021 with different rules applied to different na- tional and regional taxes to cover the needs of district/city budgets (see Box 4.7 summarizing experiences from Akhangaran and Piskent). Thus, at the moment, despite some improvements, the local government’s revenues remain at the mercy of higher levels of government, and local hokimyats have no ability to pre- dict and plan their revenues and expenditures. Table 4.1 Proposed distribution of tax and nontax incomes among budgets of different levels of government in 2021 Republic of Karakalpakstan National budget Districts and cities and regions Shared taxes* and nontax incomes Local taxes and nontax incomes    Value added tax   Corporate property tax   Personal property tax   Excise taxes (except for   Corporate land use tax   Personal land use tax gasoline, diesel fuel, and liquefied natural gas)   Corporate income tax**   Personal income tax paid by citizens renting property   Custom duties   Personal income tax***   Personal income tax paid by   Tax for subsoil use (except for   Water use tax paid by individual entrepreneurs (single construction materials) electricity-generating stations tax payment)   Other incomes and nontax   Excise tax for gasoline, diesel   Water use tax (excluding taxes revenues fuel, and liquefied natural gas paid by electricity-generating   Sales tax stations)   Government charges and fines   Tax for subsoil use (construction materials production only)   Fees for the right to retail trade in certain types of goods   Fees for the purchase and temporary import of vehicles   Revenues from sale of property expropriated by state, dividends on shares of local governments in various assets   Revenues from rent and privatization of government assets   Excise tax on mobile phone services   Excise tax on alcoholic beverages SOURCE: Republic of Uzbekistan 2021b. * In Uzbekistan shared taxes revenues are first directed to the budget of the designated level (national, or regional, or subnational), but can then be redistributed (shared) with the lower levels of government (regional, subnational, or local). ** 100% for all regions except for the Tashkent region (45%) and Tashkent city (5%). *** 100% for all regions except for Navoi region (62%), Tashkent region (45%), and Tashkent city (5%). 82 THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT? Despite an increased number of revenues transferred to subnational budgets, there is still a mismatch between the limited resources and large spending mandates of subnational (regions) and local (dis- tricts and cities) governments. For example, since 2018 the following revenue types were fully transferred to the subnational budgets—excise tax for gasoline, diesel fuel tax, government charges, locally paid fines, and personal and corporate taxes (with some exceptions, Table 4.1). Subnational and local govern- ments also received a right to keep all tax revenues generated locally beyond planned targets. However, the central government still holds the major taxes—that is, corporate income tax, personal income tax, value added tax, and subsoil use tax—which together were projected to generate around 77 percent of national budget incomes in 2021. Thus, the mismatch between financial obligations and revenues of the subnational governments remains vast. Such disbalances between revenues and obligations of subna- tional budgets are known to have been challenging for countries that attempted devolution reforms in the past (Ngigi and Busolo 2019). Imbalances between incomes and the spending obligations of subnational governments are ad- dressed through equalizing interbudgetary transfers. The practice of interbudgetary transfers was introduced in 2019. In 2020 such transfers covered around 41.3 percent of all spending of subnational government budgets (Figure 4.3). For 2021 interbudgetary transfers were projected at the level of 26 percent of total spending obligations of subnational governments budgets (Table 4.2). According to the current budget code, transfers can take many forms, which have implications for how local gov- ernments can use them. Forms and volumes of transfers are defined by the Ministry of Finance and regional finance departments on an annual basis depending on forecasted incomes and the spend- ing of subnational governments. The Ministry of Finance allocates transfers from central to regional budgets, while regional authorities further distribute to districts and cities. A considerable share of such transfers is provided to finance the social services, particularly social allowances and salaries of service providers. Overall, the system is very complicated, and leaves very little authority and control over revenues to the local governments. Most capital investments at the local level are still covered by the central budget via various devel- opment programs implemented by the sectoral ministries and dedicated extrabudgetary funds. For example, the amount allocated for development programs, capital investments, and transfers to ex- trabudgetary funds is around UBZ 38.0 trillion (US$3.6 billion, or almost 25 percent of the national budget in 2021; Republic of Uzbekistan 2021b), which is 3.7 times higher than the amount allocated as transfers to local governments. Such investment programs are planned on the basis of requests from the regional and local levels and needs assessments and planning done by the central agencies. However, actual amounts allocated to the local governments may vary greatly from amounts requested, and often cover only a part of investment required, without any guarantee that the rest of the money will be provided in the subsequent years. This means, for example, that only 5 km of the requested 25 km of road are funded, and there are no guarantees that the rest will be provided in the foreseeable future. Dependance of local governments on the national budgets for infrastructure investments is exasperated by the overall weak- ness of the intrastate cost-recovery system (discussed earlier) and general limited practice of attracting private investors to infrastructure projects in Uzbekistan. The legislation regulating public-private part- nerships was first passed in 2019 and substantially amended in 2021 (Dentos 2021). While these initiatives (mostly for regional- or national-level transport [Perzadayeva 2020] and energy [Nhede 2020] infrastruc- ture) are being discussed, the practice of public-private partnerships remains new to the country and its application for urban infrastructure projects remains distant. THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT? 83 Figure 4.3 Revenues and spending of subnational governments in 2020 (UBZ billion) Revenues of sub-national government budgets, Spending of sub-national government budgets, 2020 (UBZ bIn) 2020 (UBZ bIn) 1 809,5 970,9 89,2 432,9 3 714,9 817,7 1 353,9 3 343,2 478,8 4 209,3 560,4 2 386,7 27 703,0 47 190,7 1 974,3 9 145,4 4 789,4 9 368,7 1 266,4 28 182,1 CORPORATE INCOME TAX PROPERTY TAXES SOCIAL EXPENDITURES COSTS OF RUNNING SELF-GOVERNMENT BODIES SALES TAX LAND TAXES ECONOMIC EXPENDITURES CENTRALIZED INVESTMENTS RESERVE FUNDS PERSONAL INCOME TAX TAX FOR SUBSOIL USE OF REP. OF KARAKALPAKSTAN, COSTS OF RUNNING TASHKENT CITY AND REGIONS VAT WATER USE TAX GOVERNMENT BODIES EXCISE TAXES OTHERS OTHERS SERVICING PUBLIC DEBT SOURCE: Ministry of Finance 2020. NOTE: UBZ = Uzbek soms; VAT = value added tax. Table 4.2 Incomes, expenditures, and equalizing interbudgetary transfers planned for 2021 (combining regional and local governments in each region; in UBZ billion) Transfers as share of Regions Incomes Expenditures Transfers expenditures (%) Republic of Karakalpakstan 1,614.2 2,915.6 1,301.4 44.6 Andijan region 2,352.2 3,228.8 876.7 27.2 Bukhara region 1,935.4 2,375.4 440.0 18.5 Jizzakh region 1,212.9 1,702.9 490.0 28.8 Kashkadarya region 2,570.3 4,147.6 1,577.3 38.0 Navoi region 1,535.8 1,535.8 — — Namangan region 1,970.6 3,254.5 1,284.0 39.5 Samarkand region 2,657.0 3,759.7 1,102.7 29.3 Surkhandarya region 1,754.7 3,048.3 1,293.6 42.4 Syrdarya region 745.7 1,279.9 534.3 41.7 Tashkent region 3,009.8 3,009.8 — — Fergana region 2,904.9 3,815.0 910.1 23.9 Khorezm region 1,442.7 2,075.6 632.9 30.5 Tashkent city 4,068.6 4,068.6 — — Total 29,774.7 40,217.6 10,443.0 26.0 SOURCE: Republic of Uzbekistan 2021b 84 THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT? The announced transformations of the local government system and national public finance system are ambitious, but while reforms are under way a lot of damaging uncertainties remain the reality for local and regional authorities. The improvements to the budgetary process are multiple: (1) the number of taxes going directly to regional and local budgets has been increased; (2) subnational governments have been granted greater flexibility in spending budget surpluses (extra revenues)48 to address the needs of the citizens, improve service delivery, and enhance the overall livability in settlements under their authority; and (3) the system of interbudgetary transfers is being piloted. However, major issues persist. Subnational revenues are still insufficient to cover obligations, and transfers and shared taxes are not distributed on the basis of transparent rules or formulas and are left to the discretion of high- er-level authorities. In these conditions local governments have no clear understanding of medium-term revenue flows, and thus their financial planning remains limited to a one-year horizon, and their ability to implement capital investments by themselves to support local economy and well-being remains cur- tailed. Furthermore, there is no clarity on the distribution of spending responsibilities at different levels of the government, including capital investments and maintenance costs of vital infrastructure such as roads, water supply and sanitation, energy supply, and so on. Box 4.7 Local budget revenues in Akhangaran city and Piskent district During the case study exercise, the municipal finance situation of Akhangaran city and Piskent district were reviewed to better understand the capacities of these cities to maintain and devel- op critical local infrastructure and services. Representatives of local authorities confirmed that while more flexibility was provided in spending budget surplus, incomes of local budgets were far from being sufficient and predictable enough to finance capital investments and major infra- structure maintenance costs. Both Akhangaran and Piskent received transfers from the central and regional budgets mostly in the form of shared taxes in 2019-21. Since the system of interbudgetary transfers is still in pilot mode, the shares of national taxes and overall transfers provided are subject to annual revision by the Ministry of Finance and the Finance Department of Tashkent region hokimiyat as well. Most strikingly, the revenue structure of the two cities is very different, reflecting differences in their context and administrative organization (Akhangaran is an urban district, and Piskent is a town in a mostly rural municipality). This results in governments of two towns of similar size, facing rather different budget incentives: Akhangaran’s revenues in the first quarter of 2021 mostly came from corporate land use tax and cor- porate property tax, which were shared with the regional hokimiyat. This makes the city dependent on the region, despite its vibrant economy, as all local revenues (taxes, fines and fees, incomes from local bazars, etc.) make up only around 26 percent of Akhangaran’s budget. Agriculture dominates the economy of Piskent district, and thus a considerable share of local budget incomes—around 77 percent—are collected from water use tax payments (not to be con- fused with water use fees paid to the water utility), which are local. The total share of all local taxes and payments in Piskent’s budget is around 84 percent, while the contribution of shared regional taxes is only 15 percent. Largely, this means that the local budget has little to gain from 48  All tax and nontax incomes collected or generated by the hokimiyat in excess of approved budget parameters. THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT? 85 the growth of nonagricultural sectors, as growing tax incomes will mostly be shared with the Box 4.7 region, and will likely appear small compared to the established stream of water tax payments from agricultural producers. Table B4.7.1 Distribution of budget incomes in Akhangaran city and Piskent district, 2021 Q1 (%) Taxes and obligatory payments Akhangaran Piskent Personal income tax paid by individual entrepreneurs 8.7 0.6 Personal income tax paid by citizens renting property 1.6 - Personal property tax 3.5 1.1 Personal land use tax 3.2 3.0 Water use tax 5.4 77.6 Tax for subsoil use 0.1 1.8 Fines and fees 1.0 - Other incomes and nontax revenues 3.0 0.3 Total - Local (district / city) taxes 26.6 84.4 Corporate land use tax 11.9 5.3 Corporate property tax 41.3 0.9 Excise tax for gasoline, diesel fuel, and liquefied natural gas 4.3 0.7 Excise tax on mobile services 3.6 3.1 Excise tax on alcoholic beverages 8.5 4.7 Sales tax - 0.5 Total - Shared regional taxes 69.5 15.2 Personal income tax 2.7 0.4 Corporate income tax 1,1 0.1 Total - Shared national taxes 3.9 0.5 Total 100.0 100.0 SOURCE: Information provided by the municipalities of Akhangaran city and Piskent district. 4.3.2 Local government capacity constraints While no comprehensive assessments of local government capacities are available, existing evidence suggests that local governments lack the capacity to address key local development challenges. In addition to financial resources and clearly defined obligations and powers, local governance institutions need improved competences and skills to use powers and resources for advancing local development. Case studies of Akhangaran and Piskent show that breaking the path dependencies that often define 86 THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT? the local development trajectory requires a concerted policy effort that local governments do not have the capacity to undertake. Akhangaran, an overperformer, and Piskent, a laggard, were selected because they showed very different economic dynamism even though both are a one-hour drive from Tashkent city (Figure 4.4). The key conclusion of the investigation is that crucial natural and historic char- acteristics of the selected cities are the main factors defining the observed trajectory of economic de- velopment, while government policy has very little impact on the outcomes. In general, in Uzbekistan’s context the success of most small- and medium-size cities is defined by sev- eral exogenous factors, rather than successful policy intervention. Based on the research conducted for this report, key factors that define the success of a local economy include: (1) manufacturing capac- ities and supporting infrastructure inherited from Soviet times; (2) endowments of natural resources that attract extractive industries (even if their role is not as dominant as it might seem (see appendix F)49; and (3) good location near major roads, attracting transit traffic that supports retail trade and consumer ser- vices. While there are success stories of smaller cities managing to leverage other local advantages to promote economic growth, they are mostly exceptions that prove the rule. Figure 4.4 Location and economic performance of Piskent and Akhangaran SOURCE: Original elaboration based on data from VIIRS, Open Street Maps, and Yandex Maps. 49  The benefits of the extractive industry for local development are not discussed here in detail; appendix F provides further insights into their role. THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT? 87 The presence of a Soviet-era industrial facility often gives a city a platform for success. Towns that have large Soviet factories have the advantage of relatively well-developed infrastructure and a stable supply of services needed for industrial enterprises and their labor force. Some of those towns did not manage to repurpose their industrial assets to the market reality (as did towns in Russia and Kazakhstan), but those lucky enough to adjust to a changing environment (like Akhangaran) stand to benefit. The presence of sev- eral major production facilities, particularly a cement plant, and relatively well-developed infrastructure, have allowed Akhangaran to establish diversified production of construction materials and attract other manufacturers (Box 4.8). Location on a major transport corridor can support successful trade and services clusters. The stra- tegic location of Akhangaran along the major transport corridor connecting the Fergana Valley with Tashkent city and the rest of Uzbekistan have fueled the development of retail trade and a hospitality industry serving people in transit. Piskent, at the same time, located around 4 km away from the same road, struggles to capture any transit traffic and develop its trade and service sectors (Box 4.9). Box 4.8 Akhangaran’s need for greater infrastructure provision A good location and the presence of Soviet legacy manufacturing complexes contributed to the recent success of Akhangaran’s local economy, but the local government had relatively little to do with it. The city appeared among the few well-performing medium-sized cities, demonstrat- ing greater economic results compared to other cities in Uzbekistan.50 However, lack of capacity to further expand infrastructure and service provision are stopping the city from building on its success. The hokimiyat of Akhangaran city holds overall responsibility for development of the city, but it has limited capacity to fulfill such basic functions as the provision of serviced land, not to mention more sophisticated forms of business support. The local economy benefits from national government programs, specifically the industrial zone. However, poor implementation of the pro- posed zone project creates another deterrent for development. Akhangaran is a city with a population of 38,400 located around 60 kilometers (km) southwest of Tashkent. Almalyk and Angren, two important industrial centers, are located nearby. A major cement producing factory has been operating in the city since 1961. In the past, the city also had plants producing asbestos slate and asbestos-cement pipes, plastics and linoleum, and plumb- ing products. However, old industrial capacities, except for cement production, did not survive the transition to a market economy. Nowadays, Akhangaran Cement is among the biggest ce- ment producers in the country and a major investor in the city, with a new US$160 million project to upgrade and expand cement production. Its location along the A-373 road allowed Akhangaran to develop retail trade and the hospitality industry servicing the transit traffic. A cluster of restaurants, shops, and bed and breakfasts (B&Bs) is developing along the road connecting Fergana Valley with Tashkent and the rest of the coun- try. It is notable that this sector is dominated by businessmen and workers from the Namangan region and Buka district, located around 35 km from Akhangaran, both well known for their strong entrepreneurial skills. 50  Using productivity measures calculated based on night-time lights VIIRS data. 88 THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT? Box 4.8 Figure B 4.8.1 Share of Akhangaran city in the industrial production of Tashkent region, 2020 (% of total) Others 12,2 Electricity, natural gas, heating - Electric equipment and goods 0,7 Leather and leather products 33,4 Medical products - Light industry 1,5 Chemical and petrochemical 0,1 Construction materials 21,3 Metal works 2,3 Machinery and equipment 0,7 Food products 11,3 Total 5,6 5,0 10,0 15,0 20,0 25,0 30,0 35,0 40,0 SOURCE: Statistics Department of Tashkent region, 2020. Relatively well-developed infrastructure and a skilled workforce inherited from the Soviet period have allowed Akhangaran to become a regional industrial center. A free economic zone (FEZ) was established in 2012 in the northern part of Akhangaran with a total area of 720 hectares and around 90 lots for investment projects. The FEZ in Akhangaran is a part of the FEZ “Angren” and enjoys similar benefits. Akhangaran has among the greatest industry output per capita in the Tashkent region and nationally. Its industrial production in 2020 totaled UBZ 3,629.3 billion (US$440 million), or around 5.6 percent of the Tashkent region’s total, yet it had only 1.4 percent of the total population. The city is a regional leader in the production of leather, construction materials, and food products. Access to land and infrastructure, and proposed tax preferences are the main attractors for businesses in Akhangaran. As confirmed during interviews, a number of companies in Akhanga- ran were relocated from Tashkent city due to the opportunity to get serviced land without los- ing access to major national markets (e.g., Tashkent, Namangan). In addition, the FEZ offered exemptions from land, property, and water excise taxes for a period of 3, 5, 7, or 10 years, de- pending on the amount of investments, as well as exemption from select custom duties and a special foreign trade regime. On the one hand, this confirms the positive impact of the national support schemes for the local economies. On the other hand, it showcases the limited aggre- gate impact of the national place-based support schemes, as people relocated from other cities to enjoy the tax benefits. THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT? 89 Box 4.8 The administration of the FEZ did not fulfill all its promises to investors, and this has limited its contribution to the local economy. Some planned private investments were cancelled and some were delayed or moved to different locations since benefit packages were altered (e.g., foreign trade benefits were removed recently), while infrastructure investments were never completed in full. For example, as per the initial plan, the FEZ should accommodate three leather pro- cessing enterprises and connect them to a centralized sewage system if they share the cost of construction of a dedicated local wastewater treatment facility. Two companies dropped their investment plans at some point. The third company, which had already advanced with construction, had no choice but to start construction of the wastewater treatment facility on its own. However, no connection to the centralized system has been provided yet and the plant is using a special flusher truck to remove treated water. This radically increases the cost of production. Recently, the government decided to relocate five leather processing plants out of Tashkent because of the limited capacities of the sewage system and the pollution they pro- duce. 51 Akhangaran could be an ideal place for the creation of a leather-processing cluster if proper infrastructure is provided in the FEZ. In general, while the city is getting some benefits from the FEZ and is interested in attracting more investments, it has neither the power nor resources to resolve the issues that block investments. The municipality of Akhangaran, even if it wants, has very few avenues to build on existing ad- vantages and further promote local industry and service sectors. Establishing free economic zones (FEZs) and small industrial zones (SIZs) is a flagship local devel- opment policy of the national government, but their implementation has been far from perfect. As confirmed by the examples of Akhangaran and Piskent, access to serviced land paired with some tax incentives is a strong incentive that can attract investors to medium-size cities in a favorable location. However, while industrial zones supported by the national government have had an important impact on the development of both Akhangaran and Piskent, the shortcomings in their design and implementation meant that their effectiveness was not maximized. For instance, both cities struggled with failures to deliver infrastructure that was planned for the zones. As per information from the Ministry of Investment and Foreign Trade, there are 23 FEZs with total investments of US$5.3 billion and 348 SIZs with docu- mented investments of UBZ 16.7 trillion, but approximately half of the FEZs and 60 SIZs have no adequate infrastructure provision. The Ministry of Investment and Foreign Trade announced that UBZ 3.4 trillion will be allocated to infrastructure projects to support FEZs and SIZs in 2021–22. 52 However, in Akhangaran previous delays in resolving infrastructure issues have already resulted in the withdrawal of investors and delayed implementation of projects. Beyond FEZ/SIZs, the local administrations tend to struggle to make the most of economic development opportunities, as showcased by Piskent. The town has an SIZ (which is having a limited impact due to infrastructure insufficiency), and benefits from proximity to Tashkent and Tashkent sea. However, the local administration does not have a strategy or vision for scaling up the manufacturing sector or recreational and related consumer services through leveraging the potential of the reservoir shore. This illustrates the 51  Available at: https://www.gazeta.uz/ru/2021/05/15/leather-production/ (in Russian). 52  Available at: https://mift.uz/ru/news/tadbirkorlar-istaklarining-amalij-ruebi-itisodij-zonalar-infratuzilmasi-andaj-jahshilanadi (in Russian). 90 THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT? common observation that the local governments lack capacity to develop and implement local develop- ment initiatives and remain totally dependent on national business and industry support programs that may not be tailored to local needs (Box 4.9). Box 4.9 Piskent’s struggle to overcome path dependency The Piskent economy’s limited progress shows that breaking path dependency requires compre- hensive and well-coordinated efforts at the national and local levels, as well as sufficient capac- ities at the municipal level. Despite its location near Tashkent city, Piskent city is not capturing the potential benefits of being close to the biggest market in the country. While there is a small industrial zone (SIZ) located in Piskent, the town remains mostly a minor trade and local service hub in an agricultural district, and infrastructure limitations are among its major shortcomings. Figure B4.9.1 Share of Piskent district in the industrial production of Tashkent region, 2020 (% of total) Others Electricity, natural gas, heating Electric equipment and goods Leather and leather products Medical products Light industry Chemical and petrochemical Construction materials Metal works Machinery and equipment Food products Total 0,5 1,0 1,5 2,0 2,5 3,0 3,5 SOURCE: Statistics Department of Tashkent region, 2020. Piskent is a city with a population of 35,000 and is the administrative center of Piskent district. It is located around 60 kilometers (km) to the south from Tashkent, not far from the Tuyabuguz water reservoir also known as the “Tashkent sea.” The economy of Piskent district is mostly agricultural. The share of industry is marginal, while services involve mostly low value-added activities often provided informally. Most of Piskent’s residents are engaged in various activities related to agri- cultural production. The special economic zone established in Piskent as a part of the national government support program has had a limited impact. The SIZ in Piskent was established in 2017; however, it is not operating to its full capacity due to infrastructure issues. Only two enterprises are conducting some activities in the Piskent SIZ, which occupies 4.7 hectares located several kilometers from the THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT? 91 city. While the SIZ has water and electricity connections, it has no proper supply of natural gas Box 4.9 or a sewerage treatment facility. As a result, the wool processing plant operating in Piskent SIZ is outsourcing the wool washing work to another plant in Jizzakh to keep its output up and fulfill export obligations and secure the right for the land and tax preferences. The entrepreneur has no other choice since the capacity of the local sewerage treatment facility is low and the SIZ is not connected to the sewerage system of the town. Trade and services in Piskent are also underdeveloped due to its location away from major trans- port routes and strong competition from neighboring cities. Piskent is not capturing major transit flows since it is located several kilometers away from the main road. It also faces strong compe- tition in retail trade from Buka city, one of the biggest trade centers in Tashkent region, which is located only 15 km away. Piskent could benefit from development of tourism and recreational facilities along the reservoir shore; however, due to the total absence of infrastructure this area is not attractive for private investors. At the same time, the other shore of the Tashkent sea is popular among residents of Tashkent city as a weekend getaway, since it is relatively well connected and provides various opportunities to enjoy water-related recreational activities (mostly provided informally or at a very small scale). Despite the presence of these assets, the municipality of Piskent lacks the vision, capacity, and resources to even attempt development of a recreation cluster, which could be a major contributor to the economy. This chapter has discussed some of the factors that shape urbanization outcomes in Uzbekistan. Na- tional structural constraints, and most visibly barriers to population mobility, keep urban areas from meeting their economic potential. Weak land markets and an inefficient land use planning system en- able urban growth that produces inefficient urban forms and makes cities unlivable and unsustainable. Finally, weak local governments struggle to take the steps needed to maximize local potential. The final chapter will try to propose key policy priorities for ensuring a transition to more sustainable and produc- tive urbanization in Uzbekistan. 92 THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT? Chapter 5 Policy priorities for sustainable and productive urbanization THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT? 93 Chapter 5 Policy priorities for sustainable and productive urbanization To maximize the development benefits of urbanization, Uzbekistan needs to enable market forces and market signals to become the guiding light and the engine of urban development. Key findings of this report point to the fact that the mobility of factors of production in Uzbekistan is restricted and the man- agement of urban growth is inefficient, all of which can be connected to weak institutions and governance systems. This points to a need for reforms that should enable market forces to propel urbanization, while focusing the attention of the public sector on addressing market failures and ensuring sufficient provision of public goods and services. 5.1 National urban and territorial development strategy A national vision and strategy for urbanization and territorial development are required to coordinate policy action at all levels of governments. The declaration of a national target for achieving a higher level of urbanization is, without doubt, an important signal. However, the target of 60 percent urbaniza- tion is not sufficient. To advance toward sustainable and effective urbanization, the national government should provide a much more detailed vision that links urbanization to critical economic development and population well-being. Furthermore, the government should outline the obstacles that need to be overcome to achieve this vision, and propose policies to address them. Finally, the government should build an institutional structure that would help implement proposed policies through coordinated work among government agencies. This section explains why a national urbanization and territorial develop- ment strategy is needed, highlights its critical characteristics, and discusses its place within the overall framework of the national policy effort related to urbanization. The Government of Uzbekistan has embraced the importance of urbanization. In his address to Oliy Majlis on December 28, 2018, President Mizioyev declared the target of achieving 60 percent urbanization by 2030 (President of Uzbekistan 2018). In many ways, this was a critical turning point for the country—it signaled the acknowledgement of the critical importance of urbanization as a driver of development. It underpinned the goals set out in a “Strategy of Action on Five Priority Areas of Development of the Republic of Uzbeki- stan for 2017–2021,” issued in February 2017, which aimed to develop Uzbekistan through the modernization and diversification of the economy’s leading sectors to ensure the optimal and effective use of their eco- nomic potential (Republic of Uzbekistan 2016). The presidential decree elaborated the urbanization vision by giving priority to housing policy, land tenure security, elimination of migration barriers, support to eco- nomic growth of larger cities, reshaping of medium-size cities, and establishing a network of satellite cities. This declaration was followed by important policy steps including the establishment of the Agency of Urbanization under the Ministry of Economy and Industry (later renamed the Ministry of Economic Devel- opment and Poverty Reduction, MEDPR) (Eurasian Research Institute 2019). However, the Government of Uzbekistan’s vision for urbanization is yet to be detailed and grounded in evidence. Despite important declarations and plans to develop and adopt the Urbanization Strategy, little beyond headline ambitions has been clarified. It is not clear why the level of urbanization was set as the main target, and what steps will be taken to achieve it. Certain policy measures that keep coming up THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT? 95 in the debate—like development of satellite cities and turning large rural settlements into cities—appear to be designed to directly contribute to reaching the 60 percent urbanization target but lack clear analytical underpinning and seem to sidestep some of the key issues of urban development and growth in Uzbekistan discussed in this report. Thus, it is recommended that a strategic policy document be developed to guide the coordination of urban development policy across line ministries and different levels of government. Experience of countries from around the world has demonstrated the importance of a strategic vision for territorial development and urbanization. From Canada (Blake 2003) to Indonesia (Ministry of Nation- al Development Planning 2016), from South Africa (Department of Housing of the Republic of South Africa 1997) to Albania (Ministry of Urban Development of Albania 2014), to Uzbekistan’s neighbors like Russia (Government of Russian Federation 2018) and Kazakhstan (Office of the Prime Minster of Kazakhstan 2019), countries around the world have been adopting (in different formats and levels of formality) strategic documents for urban and territorial development. Furthermore, according to the United Nations Human Settlements Programme (UN-HABITAT 2015), 56 percent out of 162 countries considered have explicit na- tional urban policy documents. This signals the wide-reaching recognition of the critical importance of formulating strategic urban and territorial development priorities to coordinate policy development and implementation. This also means that substantial international experience has been accumulated in the field of strategic national urban policy development and implementation. Figure 5.1 National Urban Policy Uzbekistan should borrow from international development process best practice when developing its urbaniza- tion strategy. UN-HABITAT offers the following definition of national urban policy: “A coher- ent set of decisions derived through a delib- erate government-led process of coordinat- Acupuncture Participation ing and rallying various actors for a common projects Feasibility Diagnosis vision and goal that will promote more trans- formative, productive, inclusive and resil- ient urban development for the long term”. Monitoring & Formulation The approach to the National Urban Policy Evaluation prompted by UN-HABITAT is grounded in in- ternational best practice and is 100 percent Implementation relevant to Uzbekistan. It proposes a clear framework for policy development, and gives guidance on the process of strategy prepara- SOURCE: Capacity UN-HABITAT 2015 development tion (UN-HABITAT 2015). Critically, this approach emphasizes the importance of robust diagnostic exercises that should provide a clear understanding of challenges, and justifications for proposed policy priorities (UN-HABITAT 2019). According to the UN-HABITAT approach the diagnostic phase should precede the formulation of policy, thus ensuring that policy ideas are grounded in evidence, rather than evidence being produced to fit the policy ideas (Figure 5.1). Following best practice, the national urbanization strategy should be consistent with the key develop- ment objectives of Uzbekistan and reflect the reality on the ground. A robust analytical underpinning should help the Government of Uzbekistan to design policy interventions that address the main causes of underwhelming urbanization. A more robust analytical approach can help policy ideas widely discussed 96 THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT? today be more impactful in their development outcomes. One example is the idea of reclassifying rural areas as cities. Changing the status of rural settlements that by their size and function have grown to an urban size and density, as has been proposed in Uzbekistan, can be a tool for allocating additional resources to assure adequate infrastructure and services provision, and additional powers to guarantee adequate level of public management (Zokirov 2020). However, if this reclassification is seen as an easy way to achieve the target level of urbanization, it misses the opportunity for achieving important devel- opment results. Identifying satellite cities, 53 on the other hand, can be useful as a tool for managing urban growth in the largest metropolitan areas and as a means for improving governance and connectivity across a sprawling urban agglomeration. However, it might be understood as a release valve for urban growth, which does not apply to Uzbekistan. Given the relatively low level of concentration of population and the unfulfilled economic potential of larger cities (as was demonstrated in chapter 3), cities of Uzbek- istan need to focus on accommodating sustainable growth, rather than redirecting it to satellite towns. Overall, a move from sporadic policy initiatives to a strategic evidence-based process will help ensure that policy proposals target the critical issues, and maximize the development benefits. As Uzbekistan details its vision for urbanization, it should consider relaxing the prioritization of spa- tial equality of economic development. Uzbekistan’s neighbor, Kazakhstan, which shares its Soviet past, has in recent years adopted a territorial development vision that directly accepts the spatial inequal- ity of economic activity and prioritizes the development and growth of major agglomerations, offering them additional measures of national support (Office of the Prime Minster of Kazakhstan 2019). This vision breaks with the Soviet planning tradition of targeting territorial equality and is much more consistent with the national target of building an innovation-driven economy in Uzbekistan, since global experience shows that innovation is concentrated in cities (Paunov et al. 2019). Uzbekistan may consider following this example and deprioritizing the goal of equalizing spatial development declared in Uzbekistan’s current development strategy (Republic of Uzbekistan 2016). In fact, countries across the world have demonstrat- ed that rapid phases of economic growth are associated with territorial divergence (World Bank 2008). In sum, economic activity will continue to be concentrated in the most productive cities, while the role of policy is to ensure that people in areas that are lagging have a decent quality of life and retain access to services, public goods, and livelihood opportunities. The urbanization strategy should be underpinned by an institutional structure that would ensure the implementation of its vision. At the top level, this implies cross-sectoral policy coordination. The strategy should provide guidance and help coordinate the activities of the MEDPR, Ministry of Investment and For- eign Trade, Ministry of Construction, Ministry of Transport, and others. All these line agencies should use the strategy’s guiding principles for prioritizing infrastructure investments and structural reforms. Howev- er, an even more important aspect of the institutional structure is developing the capacity of subnational and local governments, who will be directly responsible for defining local policy goals and implementing programs that contribute to the implementation of the national vision. The next section will specifically focus on the importance of subnational governments and their empowerment for the success of territorial development efforts. The urbanization strategy is critical for coordinating policy action in an inherently multisectoral policy area. To be successful it needs to fulfill several key criteria:   Be rooted in data and analytics and a realistic understanding of urbanization trends and challenges.   Formulate a vision for urbanization and territorial development not just in terms of target numbers (level of urbanization) but in terms of its key qualities, such as availability of services infrastructure and housing, economic development indicators, other criteria related to quality of life, and so on. 53  As mentioned in the Decree of the President of Uzbekistan, “On measures for optimization of urbanization process,” No. УП-5623, dated January 10, 2019. THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT? 97   Identify priority reforms required and propose timelines and targets for their implementation.   Offer an institutional framework for strategy implementation including key responsibilities for each of the line ministries, agencies, and subnational authorities, and suggest a mechanism for coordi- nation between them. 5.2 Empowering subnational governments It will be extremely difficult to implement the urbanization and territorial development strategy without further empowerment of local governments. Chapter 4 offered a preview of the current state of local governance in Uzbekistan. The local authorities have very limited powers, resources, and capacities; struggle to provide services and infrastructure; and play a very passive role in overseeing development initiatives stemming from the national government. However, global experience suggests that successful urban and territorial development strategies rely on subnational and local governments taking a lead- ership role. While this is a subject that deserves more detailed investigation, this section sets out some preliminary thinking about the importance of empowering local governments and offers some policy ideas gleaned from international best practices. One of the main common denominators of international best practice in territorial development is that the success of cities and regions requires empowered local governments that hold sufficient powers and have adequate capacity. Decentralization of powers and public finance may be a global trend in recent decades, as it is seen as a tool to overcome failures of the national government in implementing territorial development policy (Pike et al. 2012). However, decentralization is not guaranteed to be good for urban and local economic development. The fragmentation of local government territories may pro- duce diseconomies of scale in policy making and provision of services (Rodríguez-Pose and Bwire 2004). Global evidence suggests that the main critical criteria for decentralizing authority and public resources is the quality of institutions and the collaboration between jurisdictions. Thus, the best examples of terri- torial development policies often combine decentralization of power with incentives and capacity-build- ing efforts (Toboso and Scorsone 2010). Multiple countries have achieved progress in local economic development through empowering sub- national authorities to develop and implement local development strategies while providing them support and guidance. Examples of successful territorial development efforts from the European Union (EU), Australia, and Canada demonstrate the critical importance of subnational governments’ leader- ship in defining local development priorities. The EU cohesion policy delegates responsibility for local development strategies to regions. Australia and Canada both have regional development agencies, charged with identifying needs and market failures in each region and designing and carrying out pol- icy responses (World Bank 2020). These experiences in summary make a strong case for elevating the capacity of local governments, so that they can contribute substantively to the creation of the national strategic framework for urban and territorial development, and the achievement of the urbanization vision. For regional and local strategies to become a functional part of the territorial development framework, sustained capacity support from the national authorities is required. Providing detailed guidance and hands-on support for subnational strategy development through a dedicated national institution can enhance the quality of subnational strategic planning. During the early and mid-1960s, the Economic Development Administration was established by the US government. Till today, this provides an e-tool 98 THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT? called the Comprehensive Economic Development Strategy (CEDS) to support subnational governments in building strategy-driven plans for regional economic development. This helps regional governments build evidence-based strategic documents, and also promotes the participatory planning processes and engagement of community leaders and the private sector in the process (World Bank 2020). In addition to carrying the strategic vision of urbanization, strong local governments can be a pivotal part of the system for designing and implementing effective and impactful place-based policy. Uz- bekistan is running an extensive program of special economic zones and small industrial zones and has a track record of spatially targeted support for other industries, including tourism and local trade and services along major transportation corridors. A place-based policy can be a useful tool that contributes to the growing of the national economy, rather than leading to a relocation of economic activity in space, only if it targets specific local market failures—factors that prevent a specific location from leveraging the comparative advantage that it holds (Duranton and Venables 2018). Identifying such market failures and addressing them requires detailed understanding of the local context, the opportunities that the given location holds, and the constraints that hold it back. Fulfilling these criteria is only possible with the en- gagement of strong local governments. Only they can understand local context and design a targeted intervention specific to local issues (even if they need technical support to do so). National place-based policies often (even if they achieve some success) fail to maximize impact, because they are rarely ad- justed to specific local needs, which is part of the problem with the industrial zones in Akhangaran and Piskent. Uzbekistan has started moving toward empowerment of local authorities. As discussed in chapter 4 (section 4.3), the government recently started providing financial and managerial tools to local gov- ernments to increase the level of decentralization. This includes the possibility of local administrations retaining a larger share of locally generated revenues. 54 But, while the administrative and budgetary re- forms are ongoing, both limited powers and resources and the low capacity of subnational governments are major restrictions for sustainable and productive urban development. At the moment, subnational governments in Uzbekistan are limited in their ability to have an impact on cities’ development pathways. The following key issues are critical, as showcased by examples covered in chapters 3 and 4 of this report:   Limited authority and capacity. The cases of urban expansion (chapter 3) and local economic development of Uzbek cities (chapter 4) clearly show that local authorities have relatively little say over defining local priorities, and oftentimes lack the tools to address the issues that they are aware of. This is showcased by the inability of municipalities to provide infrastructure that is proposed by planning documents (as in Namangan, Box 4.3) or to propose a strategy to leverage underutilized economic potential (the case of the recreation industry in Piskent — Box 4.9). These examples provide a simple illustration of the fact that local authorities are hamstrung by a system in which they have very limited powers and resources, and struggle to address local development challenges.   Lack of predictable revenue streams. While the recent reform has made it possible for local gov- ernments to retain additional taxes, the impact of this remains limited, because the legislation keeps evolving, and the structure of revenues and the share of taxes that different municipalities can retain can change over time depending on decisions made at the regional or national level. 54  Presidential Decree from June 7, 2017, on the possibilities of local administrations keeping locally generated revenues; and Cabinet of Ministers Decree from June 29, 2017, on the collection of reserves of local taxes. THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT? 99 As a result, municipalities cannot rely on predictable revenue streams and plan their future invest- ment programs based on them. The Ministry of Finance in collaboration with the World Bank has planned a study that should help to establish clear rules and guidelines for tax revenue sharing, which would advance the understanding of how the system of intergovernmental fiscal relation- ships should be changed to support capacity building of local governments and to provide the incentives needed to advance local development in a manner consistent with the national devel- opment strategy.   Confused incentives. In addition to the complex structure of local revenues, local government incentives are distorted due to the centralization of the governance system. For instance, local governments are often required to contribute to target indicators for the region, which may not reflect local needs. One of the examples encountered during data collection for this report are the incentives related to export businesses. As regional hokimyats are held accountable for fulfilling export targets by the national government, they pass on the requirements to local governments to attract and support exporting firms (and only direct exporters are counted), which end up getting preferential treatment when it comes to land and infrastructure provision. Attracting exporters, however, may not be the same as maximizing potential and leveraging the competitive advan- tages of individual cities, which may gain greater productivity in industries that target local mar- kets, or by housing a firm that contributes to the upstream part of a supply chain of an exported final product. Further administrative reforms require a comprehensive review of current local government authority, capacity, financial resources, and incentives. This report presents only sporadic observations about the critical agenda of subnational government empowerment. It is critical to carry out a deeper investigation of local government authority and fiscal and technical capacities and incentives, which should be con- sidered in relation to their development challenges and needs. The World Bank would be ready to sup- port the Government of Uzbekistan with such an analysis to contribute to developing the administrative reform urgently needed to ensure the transition to sustainable urbanization. In the meantime, this section highlights several important policy approaches that the government may want to consider for improving subnational government capacity. The decentralization of powers and public finance should be synchronized and accompanied by a systematic capacity-building effort. The experience of multiple countries shows that decentralization reforms tend to fail if they do not strike the right balance between powers, finance, and capacity. Local governments that receive additional responsibilities without adequate funding tend to fail at service provision, thus putting local development in jeopardy (Ngigi and Busolo 2019). Similarly, local govern- ments that have powers but not the capacity to use them efficiently, will at best misuse them. Thus, it is critical for the administration to strike the right balance between devolving powers and responsi- bilities, delegating resources, and building capacities. One of the policy proposals to be considered is leveraging the National Academy of Public Service to develop a series of technical training that local government representatives would need to attend before they use the additional funding resources or additional powers. Financial incentives and competitions are tools Uzbekistan may consider for encouraging capacity building of subnational governments. Nations all over the world have successfully used policy tools that link the delegation of power and financial resources to various conditions relating to capacity. These can be implemented as asymmetric decentralization (only granting powers to governments that meet certain criteria), or on a level of individual investments in the form of competitions (only well-prepared 100 THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT? local development projects get funded from the central budget). International experiences with these approaches include:   City Deals in the United Kingdom—Manchester and municipalities in its agglomeration were re- warded for being relatively advanced in building institutions for collaborative metropolitan gov- ernance (OECD 2015) (Box 5.1).   A similar approach based on the idea of the asymmetric decentralization of responsibilities and financial resources was used in Colombia. Each territory could apply for responsibility for a certain task, and its application was evaluated based on the level of development in the relevant area, and the characteristics of specific regional institutions that were related to the task considered in the application. This model has proven to be popular—in 2017 the government received 11 applica- tions for tasks, 3 of which were approved (World Bank 2020).   Russia has introduced a practice of holding competitions for the best project proposals in a se- lected category. The prize for the best project is funding for its implementation (Box 5.1). Box 5.1 International experience of competitions (Russia) and asymmetric devolution of powers (United Kingdom) as drivers of local capacity building To make rapid progress in building the capacity of local governments, Uzbekistan might want to consider combining support and incentives. International experience offers various examples of schemes that provide incentives to subnational governments to incentivize capacity building. This Box offers two examples—development competitions in Russia and the asymmetric decen- tralization scheme from the United Kingdom. Starting in 2015–17, the Russian territorial development management has used open competi- tions as a tool to optimize funding allocation and incentivize capacity building.  The Ministry of Construction of the Russian Federation holds a countrywide competition for the best projects for public space improvement in small towns and historical settle- ments. This competition calls for integrated concepts to develop urban spaces. Up to 400 municipalities participate annually and 80 winners are selected from the participants to receive funding to implement these improvement projects.  The Ministry of Economic Development of the Moscow Region holds a growth points com- petition in which all municipalities in the region participate. The purpose of the competi- tion is to prepare comprehensive town development programs; winners receive funding to develop their concepts into full-fledged strategies. This experience points to the following key lessons for adopting this practice in Uzbekistan: (1) competitions allow for identification of local governments with the highest capacity and poten- tial and direct funding to support them; and (2) participation in competitions is in itself valuable education for local authorities. In 2011, the UK government announced City Deals. This was a program that gave the largest cities an opportunity to access additional powers and revenue streams in exchange for devel- THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT? 101 Box 5.1 oping economic development plans and showing the track record of strong governance. Among the criteria for approving city deals, the emphasis was on the record of cross-jurisdictional gov- ernance and engagement of the private sector. The Great Manchester city deal, signed in July 2012, was one of the most ambitious, and set the standard for other urban areas to follow. It included establishment of a revolving infrastructure fund allowing Greater Manchester to “earn back” a portion of the additional tax revenue that resulted from investment in infrastructure. The deal also included apprenticeship, skills development, and housing initiatives. The revolving fund alone allowed for the return and local reinvestment of about £30 million from taxes paid from the national budget to the region (OECD 2015). Empowering local governments is not a technical problem that can be solved with a click of a but- ton; rather, an iterative approach needs to be adopted. Different models of decentralization fit different country contexts. Both South Africa and Indonesia have had success in building the capacity of local governments, each using quite a different approach. Indonesia focused on devolving specific functions related to service delivery and poverty alleviation and making sure that local governments have fiscal and technical capacity to fulfil these tasks (Nasution 2016). South Africa’s reform prioritized the self-sufficiency of local governments, which were encouraged to expand their own revenues and capacity to lead local development. It also focused on building plat- forms for negotiations and consensus building between levels of government and creating a system of transfers that assured equal provision of basic services across the highly unequal and diverse country (Feinstein 2015). The experiences of these two countries over the past 20 years, and broader international best prac- tice suggests that building capable local governments is a long, iterative process. It would be hard to define all the parameters of a system that would work in Uzbekistan at once. Distribution of authority, responsibilities, revenues, and spending powers would have to be gradually adjusted as intermediate results of the reform become evident. It is thus useful to approach local government empowerment us- ing the problem-driven iterative adaptation method for reform implementation developed by Andrews, Pritchett, and Woolcock (2016). This approach suggests that complex policy problems in most cases cannot be solved simply through adopting a solution from elsewhere and instead require the gradual introduction of best practices that may have emerged within the local context, or can be identified elsewhere. And most importantly this approach implies multiple iterations, whereby small changes to the system are tested, then assessed and either adopted more broadly or adjusted in search of better outcomes. The empowerment of subnational governments could be a major cornerstone of Uzbekistan’s national urbanization policy, complementing the broader structural reform agenda. This section has tried to make a case for the critical importance of local governments in the formation of a national policy framework for urbanization, the implementation of a strategic vision of urbanization, and the efficiency of place-based policies. In the next section we turn to a different, essential part of the policy agenda for sustainable ur- banization—critical structural reforms. 102 THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT? 5.3 Critical structural reforms Continued structural reforms are important for enabling and strengthening the role of markets in guiding urbanization and urban growth, and thus for unlocking the full potential of urbanization. When weaknesses in legal and regulatory frameworks constrain factor markets, the presence of the public sector in the economy remains a distortion to private initiative, the legal environment does not inspire confidence in investors, and the provision of critical public services remains insufficient, it is hard to expect cities to become drivers of development. Thus, continuing critical structural reforms focused on liberalizing labor and capital markets, reducing and optimizing the role of SOEs, modernizing and consolidating critical legislation, and improving access to education and health care at the national level are all critical for urbanization, even if they are not covered in detail in this report (World Bank, forthcoming). Without successful advances in these critical reforms, the success of urbanization as a driver of development will be at best limited. While advances across the spectrum of structural reforms would be important for sustainable and effective urbanization, several are of the highest direct relevance, and first among these is address- ing barriers to migration. This report has focused on the negative impact of the propiska regime, and showed its significant impact on the distribution of the urban population and on development outcomes of urbanization. Thus, the ongoing relaxation of the restrictions to mobility and access to services that propiska entails are welcome. As argued in chapter 4, the reform should advance and be accompa- nied by policies addressing other migration barriers and incentives distortions, specifically housing af- fordability, accessibility of urban services, and centralized wage regulations in certain industries. Easing cross-country mobility may mean that more people will migrate to cities, and this would increase pres- sure related to managing urban growth and providing services and infrastructure. But such challenges should not be excuses for depriving people of the fundamental right to move to places where they see better opportunities, and often confining them to poverty. The possibility of increased migration to cities should be seen as an opportunity to advance development and elevate quality of life, and a motivation to advance policy action in areas that are critical to prepare urban areas to accommodate a growing population by providing livable and sustainable urban environments. Better management of cities and urban growth in a manner that enables and guides markets to de- liver the best outcomes for people while ensuring provision of public services, requires building an enabling system including development of urban land markets and urban land use planning system. Empowered local governments, which were discussed in the previous section, are critically important for implementing the national development vision locally, and for addressing the specific needs and challenges of every city. But local governments cannot be effective without critical systems that en- able market forces to drive transformation of urban spaces, and grant the public sector the capacity to efficiently guide the market and ensure accessibility of public goods and services. Chapter 4 already dis- cussed the critical importance of land markets, land management, and urban land-use planning systems for unlocking the full potential of cities. The rest of this section will provide an in-depth discussion of further steps needed to address current challenges and advance structural reforms in these two critical areas. 5.3.1 Developing land markets The ongoing land reform has made substantial advances; however, its full implementation requires sustained effort from the government of Uzbekistan. A lot of work is still required to ensure that a func- tional land market is built on a solid foundation. First, further harmonization and improvements to the legislation are needed. Second, the complicated work of building a land administration and manage- THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT? 103 ment system needs to continue to ensure reliable and efficient operation of the land market. Finally, ad- ditional efforts are needed to enable value-based land taxation and land value capture to become im- portant tools for enhancing local government capacity and addressing local development challenges. Further strengthening the new land legislation The legislative advances related to land privatization and land tenure are major achievements, but further improvements to legislation are needed to ensure a strong legal foundation for the urban land markets. Section 4.2.3 provided a detailed discussion of important changes to land legislation made in Uzbekistan recently (for details see Box 4.5). Enabling the privatization of urban land and simplifying and clarifying land tenure regimes are major advances that lay the foundation for land market development and private investment mobilization. However, some shortcomings still need to be addressed. Some of the required changes include: (1) allowing for privatization of land under multiapartment buildings, (2) creating mechanisms for foreign individuals and/or organizations (or companies with partial foreign ownership) to participate or contribute to the privatization process, and so on. The land privatization experience of Georgia could be a source of useful lessons for Uzbekistan (see Box 5.2). Box 5.2 Privatization of nonagricultural (urban) land in Georgia Private ownership of land and buildings were nonexistent during the Soviet period in Georgia. It emerged only after independence in 1991, when the country started transitioning from a planned to a market economy. Emergence of private ownership of nonagricultural land in Georgia was preceded by the pri- vatization of apartments and buildings. In February 1992 the Cabinet of Ministers enacted de- cree #107, according to which apartments and individual houses could be transferred into private ownership free of charge. The Civil Code adopted in November 1997 considered the buildings’ structure permanently attached to the land as an integral part of the real property; however, land privatization was delayed due to the lack of legislation for it. Enactment of two laws: (1) on dec- laration of nonagricultural land used by individuals and privately owned organizations as private property, and (2) on management and alienation of state-owned nonagricultural land in Novem- ber 1998, ensured establishment of private ownership on nonagricultural land and supported the inception of a formal urban land market in Georgia. The State declared lawfully used land parcels that were attached to individual houses as well as multistorey apartment buildings as private. Lawfully used land was defined as previously allocated by state and registered in Bureau of Technical Inventory – thus, after adoption of the aforementioned legislation, the owners of houses automatically acquired the ownership of land plots. The methodology for allocation of land attached to the buildings was developed for cases where the supporting documents verifying the land area were absent from the cadastre. With regard to privatization of land under industrial buildings lawfully used by individuals and private entities, the land was transferred from state into private ownership following the pay- ment of annual land tax before January 1, 1999, and twice after 1999. These lands were declared as privately owned and could be formally registered upon submission of payment receipts. 104 THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT? Currently the privatization of state-owned urban land in Georgia is administered by the Ministry Box 5.2 of Economy and Sustainable Development and the local self-government bodies through e-auc- tions and direct sale with conditions. The chart demonstrates area of privatized land by the State and Municipalities. Figure B5.2.1 Land privatization in Georgia (ha per year) 45000,00 40000,00 35000,00 30000,00 25000,00 20000,00 15000,00 10000,00 5000,00 0,00 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 State Municipal SOURCE: National Agency of Public Registry of Georgia (2021) Later, in 2007, the Government of Georgia adopted legislation that abolished the previous law and defined the legalization fee as five times the value of annual property tax (set by the local government at the time of application submission). Currently the privatization of state-owned urban land in Georgia is administered by the Ministry of Economy and Sustainable Development and the local self-government bodies through e-auc- tions and direct sale with conditions. Building the land management system One of the key priorities in the initial phase of land market development should be enhancement of the National Land Management System (LMS) to enable interinstitutional collaboration in the implementa- tion of the land reform. Figure 5.2 presents the overall framework for LMS development. This system can be seen as a set of principles governing land rights, land use and development, the policies articulated by government for their implementation, and the procedures and tools, including the information tech- nology platforms employed by the government, to achieve those policies. Specifically, LMS includes land THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT? 105 and real property registration; cadastral survey and mapping; land adjudication; physical planning; and land and real estate valuation and taxation. Traditionally, the main objective of a LMS was to support development and functioning of land markets. Currently, it is evolving to recognize its broader function of serving economic development, environmental management, and social stability. Environmental, social, and public service provision constraints are today seen as a part of the LMS’s land resources allocation and redistribution functions. A comprehensive, unified, and secure LMS is vital to economic advancement in developing countries (Dale and McLaughlin 2000) and an important pillar for land markets and national economies (Dale and Baldwin 1999). Figure 5.2 Global perspective of a modern land management system Sustainable Development Economic, Social & Environmental Efficient Land Market Effective Land Use Management Land Tenure Land Value Land Use Land Development Titles, Mortgages & Easemerts Asessment of land value Policies and Spatial planning Construction planning and Permits Secure legal rights Collection of property tax Control of land use Regulation and implementation Land Institutional Policies Framework Land Information Cadastral and Topographic Data Geospatial Data Infrastructures SOURCE: Enemark, Williamson, and Wallace 2005. The work in rebuilding the institutional framework for land administration and management (discussed in section 4.2) is an important advancement in LMS development, but it is far from done. The remaining challenges include organizing the coordination and collaboration within the new institutional framework, finalizing inventory and registration of all real properties and land within private and public custodianship, and implementing the Integrated Information System for Cadastre and Real Estate Registration (IISRPRC) across all the entities of the Cadastre Agency (CA) (currently 206 offices). The new institutional framework of land administration lays the foundation for development of the IIS- RPRC. In addition to the functions of cadastral registration, the CSC will also carry out research work, maintain a unified address registry, maintain and develop a network of permanent global satellite posi- tioning stations, create and develop a national Geographic Information System (GIS) and National Spatial Data Infrastructure (NSDI), as well as the IISRPRC. The CSC was created using the capacities of the former National Center for State Cadastres, Geodesy, and Cartography and territorial enterprises of land ad- ministration and real estate cadastre and their branches, and thus retains their infrastructure and human 106 THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT? capacity. The ongoing project “Modernization of Real Property and Cadastre Registration,” implemented by the CA and supported by the World Bank, aims to implement the IISRPRC across the country by the end of 2022. More than 7.5 million cadastral documents have been digitized as a part of the project and represent around 85-90 percent of all real properties across the country. The remaining plots would have to be surveyed and registered to provide Uzbekistan with a comprehensive cadastre. Recent government announcements laid out the technical feasibility foundations for a modern digital LMS built according to good international standards and practices and require immediate implemen- tation to provide support to the ongoing land reform. 55 With the purpose of further development of digital technologies in all branches of the economy and social sphere, the Decree of the President No. UP-6079 was passed to “approve programs for the digital transformation of regions and industries in 2020-22.” The resolution of the Cabinet of Ministers No. 732 further develops this agenda and specifically introduces the use of a GIS for the town-planning cadastre and its integration with the national GIS and the electronic government system . Two decrees of the Cabinet of Ministers aimed at introducing geo- information technologies were issued and assigned a significant role in GIS development to the CA.56 According to these documents the CA now holds the responsibility to: (1) by November 1, 2020, launch an online geoportal for a trial; (2) by December 1, 2020, ensure the entry of all agricultural maps into the on- line geoportal and its integration with the national GIS; and (3) by January 1, 2021, complete the launch of the online geoportal. Furthermore, Resolution No. 110 of the Cabinet of Ministers, adopted on February 26, 2021, created a legal base for standardization in the field of spatial data in accordance with international standards and technical regulations like INSPIRE, the Land Administration Domain Model, Open Geo- graphic Consortium, and so on. While all these measures represent important advances, it is important that their implementation continues to be supervised closely, and that necessary adjustments are made to ensure achievement of desired results. Further progress in the area of spatial technology will aid not only the land market but urban devel- opment more broadly. The main challenges with respect to data and information would be to organize an interoperable and continuously updated spatial data infrastructure, accessible to public and private sector stakeholders, and provide aerial photography, cadastral, valuation, town planning, and other data enabling delivery of public services and providing accurate and comprehensive information for decision-making. This will allow private actors to evaluate risks better and will unlock investments in urban areas thus stimulating the growth of land markets. Additionally, as discussed in the next section, these developments would help address important gaps affecting the land use planning system today. Laying the foundation for value-based taxation and other land value capture tools The next frontier of land administration is implementation of market-based valuation of land and tran- sition to value-based land and property taxation. While the first goal of improving the legislation and the land administration systems is to allow for easy land market transactions needed for optimizing urban land use, further down the line it can enable a powerful tool of urban governance and revenue generation —unified (combining land and buildings) value-based property taxation. On the one hand, such tax will al- low local administration to be beneficiaries of economic development, as it will capture a part of growing 55  Resolution of the President of the Republic of Uzbekistan, “On measures to digitalize the activities of the judiciary,” No. PP-4818, dated September 3, 2020, provides: (1) implementation of the mechanism for direct entry by the court (judge) into the automated information system “Notary” of information on the imposition of a ban on the alienation of real estate, and (2) providing courts with the opportunity to receive online information about real estate objects, their owners, and ownership history from the State Register of Rights to Real Estate Objects. 56  Resolution of the Cabinet of Ministers of the Republic of Uzbekistan, “On measures for the implementation of the geographic information-analytical system of the city of Tashkent in the framework of the ‘Digital Tashkent’ program,” No. 443, dated July 16, 2020; and Resolution of the Cabinet of Ministers of the Republic of Uzbekistan, “On measures for the further development of the electronic government system, as well as on the introduction of the order of electronic reporting of state bodies and organizations before population on their activities,” No. 444, dated July 16, 2020. THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT? 107 property values that it will bring. This will also utilize property value increases enabled by public invest- ments in infrastructure and services. On the other hand, value-based property taxation creates a cost to suboptimal utilization of land in prime locations, and thus incentivizes change of use toward a higher-value one. Other tools may include developer charges (impact fees), intensification of uses of government-use land, and a multitude of other instruments (Kaw and Kaganova, forthcoming). However, while these goals should be recognized, for the time being they should be secondary to the development of land legislation and a land administration institutional framework. This is because they are critical for establishing a grow- ing and active formal land and real estate market, without which value-based taxation and more advance land value capture tools will not become available. Advances toward introducing a value-based land taxation system are being made already. To move toward market-based valuation and taxation, Uzbekistan adopted legislation and started piloting mar- ket-based valuation methodologies and tools and elaborating corresponding national standards. The Decree No. UP-6121 of the President of the Republic of Uzbekistan dated December 3, 2020, “On fur- ther improvement of resource taxes and property tax” introduces a mechanism for transitioning to a value-based property tax: (1) in the first phase (2021–23) for housing real assets (apartments, individual dwellings, country houses), as well as land plots occupied by these objects; and (2) in the second phase (2022–24) for individual nonresidential real property objects, as well as land plots occupied by these objects. It should also be noted that several pieces of legislation issued by the local authorities in the field of regulation of land relations were also adopted during the same period. 57 It is expected that the roadmap for further improvement of taxation of real estate assets, introducing one real property tax instead of property tax and land tax will be finalized in the near future. The CA developed during 2021 a draft methodology for determining the cadastral value of immovable properties based on their mar- ket value, improving the information system to create a complete and reliable database of real estate assets to each owner (user), and conducting a high-quality appraisal of real estate objects based on their market value. Institutional reforms aimed at expanding and strengthening the role of the private sector in land valuation are also under way. By the Resolution of December 23, 2020, No. 805, the Cab- inet of Ministers approved new administrative regulations on the provision of government services for the conducting of qualification exams for appraisers and realtors and the issuance of a qualification certificates. The Modernization of Real Property Registration and Cadastre Project implemented with support of the World Bank is advancing the review and piloting of draft methodology in selected areas around Tashkent in 2022. The current dynamism of the property market in Uzbekistan indicates that, hopefully, once mecha- nisms such as market-based valuation and taxation for land and real property are successfully op- erationalized, they could rapidly start contributing to optimize urban land use. Currently, the annual average volume of real estate transactions is around 630,000 based on the data received from the CA between 2018 and 2021. Such a volume of transactions depicts an already quite dynamic real estate market, probably skewed toward transfers of apartments in multiunit residential buildings, rather than land plots. However, the volume of transactions suggests that as legislative and institutional issues are addressed, the active market will quickly facilitate all of the abovementioned benefits of optimization of urban land use to the extent permitted by land use regulations (which, in turn, indicates how critical 57  Decision of the khokim (mayor) of the city of Tashkent, “On the approval of the minimum price for the right to permanent use of a land plot in the city of Tashkent,” No. 112, dated February 5, 2021; Decision of the khokim of the city of Tashkent, “On the minimum rental rates for the use of real estate in the city of Tashkent,” No. 146, dated February 11, 2021; Decision of the Takhiatash District Kengash (Council) of People’s Deputies, “On the establishment of tax rates for non-agricultural land in cities and settlements,” No. 23/72, February 26, 2021; and Decision of the khokim of the city of Tashkent, “On approval of the minimum price of inheritable life possession of land for individual housing construction in the city of Tashkent,” No. 467, dated April 21, 2021; and others. 108 THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT? the reform of land use planning is, a topic that will be tackled in the next section). An ongoing pilot col- laboration project with Rossreestr (the State Agency in charge of land cadastre, land valuation, and registration systems in the Russian Federation) should help the CA to switch to market-based valuation and taxation systems in the foreseeable future. Key reform priorities to advance land market development for the benefit of sustainable urbanization. Based on the discussion presented above, the following key priorities for continuing land reform in Uz- bekistan can be identified:   In terms of remaining major legislative and institutional gaps the following issues need to be priorities in the next phase of land reform:   Comprehensively revise the unified land code to incorporate all the recent changes per- taining to the land sector.   Prepare and operationalize regulations for the privatization of nonagricultural land.   Prepare and operationalize law and regulations on land market-based valuation for nona- gricultural land and corresponding national standards and requirements for practitioners.   It is critically important that the process of development of the land cadaster is complete. The priority in this area should be to operationalize a transparent and publicly accessible IISPRC and finalize registration of all land parcels/properties in the country, including state and municipal properties to ensure security of tenure and support further development of land and real property markets. An estimated 1.8 million parcels are not yet registered.   Spatial data infrastructure development needs to be advanced and access to spatial data should be eased. This may be achieved through implementation of the National Spatial Data In- frastructure Strategy and concept. This would be critical for supporting efficient territorial planning and development, and for improving interagency data-sharing, consultation, and coordination. The World Bank will continue collaboration with the Government of Uzbekistan to advance these re- forms. To ensure alignment of the objectives of the reform between the relevant stakeholders and to synchronize sequencing of the activities to be implemented, the MEDPR is working on a comprehensive roadmap for agricultural and nonagricultural land reform for Uzbekistan. Additionally, an interministerial working group for development of the roadmap led by the MEDPR and involving relevant public stake- holders such as ministries, institutions, representatives of the city governments, and agencies is being set up (a draft outline of the proposed road map is included in appendix K). 5.3.2 Reforming urban land use planning The urban land use planning 58 system in Uzbekistan needs to be reshaped. Urban planning in Uzbekistan today can be characterized as an unfit and underutilized asset, characterized by several data, methodol- ogy, and process efficiency constraints. As explained in section 4.2, inertias from the centralized economy period limit the forward-looking capacity of the urban planning process, which does not take into account market dynamics and is disconnected from infrastructure investment programming. These characteristics hamper the optimum utilization of urban space to meet the demands of urbanization and call for a holistic reform of the urban planning system. A key enabler of urbanization value for Uzbekistan would be a fit-for- 58  The terms urban land use planning and urban planning are used interchangeably, and may be considered synonyms. THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT? 109 purpose urban land use planning system that has a clear strategic intent and qualitative and quantitative objectives, taps market potential, and is well coordinated with land administration. Fundamentally the urban planning system needs to transition to an approach suitable to market con- ditions. A shift that overcomes inertias in the current planning practice is required to unlock the potential of urbanization. This shift should start with the understanding that urban land use planning is not just an exercise of producing new general plans. The modernization needs to be systemic and address both the approach to urban planning and the process whereby its instruments are produced. Modernization should also be supported by an update in institutional capacity and skills. The rest of this section breaks the discussion of the needed system reforms into two parts: (1) revisiting foundational principles of urban land use planning, and (2) streamlining the process of urban land use planning. Revisiting foundational principles of urban land use planning Urban planning documents should evolve to be seen as “market guides,” rather than “rule books.” The prevalent urban planning approach in the Soviet period, which to a large extent shapes the develop- ment of Uzbekistan’s cities today, was not conceived with the market in mind. Under the planned econ- omy, the state was the only actor, and urban planning was a government action instrument epitomized by the “general plan,” traditionally more of a rule book or even a government action plan than a vision for development of urban land outlining key parameters of utilization of space. In the market economy, the role of urban planning is to guide the activity of multiple actors and to provide organized flexibility to update land and property uses according to shifting economic and social demands and conditions. Certainty and transparency are essential for urban land development to serve the purpose of increas- ing prosperity and livability of the city. Beyond the general plan, urban planning must be conceived as a system with several parts that work together to generate certainty and flexibility, respond to actual needs of people and businesses, and mobilize investment in support of urbanization objectives. Urban land use planning as a system must provide consistent solutions from national-scale to lo- cal-level land use planning. Such a system would comprise, upstream, a national urban strategy and territorial development documents which define a system of settlements and a high-level preferred spatial structure. Strategic and binding urban planning documents would further articulate the vision of spatial development for a specific settlement and provide guidance and parameters for future de- velopment regarding infrastructure and land use. Downstream, urban planning documents would be anchored by zoning regulations, also referred to as development control regulations, which are applied through construction permitting. Relevance to local context and consistent implementation of plans are critical, to generate certainty, and these elements require a supporting legal framework that clarifies the scope and function of each planning instrument, which eventually adds value individually and col- lectively. A coherent urban planning system does not exist in isolation, but coordinates with economic development policy so as to be mutually supportive and complementary. Box 5.3 illustrates an example from the Republic of Korea, where the national territorial plan evolves in alignment with the economic development vision. The system transposes national objectives to urban planning instruments through a chain of instruments with an increasing level of detail. This efficiency and functionality offer valuable lessons. 110 THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT? Box 5.3 Korea’s spatial development process: Providing a vision while addressing evolving qualities of a place The planning system in the Republic of Korea has a clear hierarchical structure. The Comprehensive National Territorial Plan (CNTP), revised once every five years, outlines the vision for the development of the whole country, and sets out objectives with relevant development strategies as well as spa- tial strategies to achieve the overall vision. The subsequent plans at provincial and city levels are determined and based on the vision and strategies set out in the CNTP. Spatial planning, a key component in the CNTP, regional development plans, and comprehen- sive plans at the city level, seeks to balance regional development, enhance the competitive- ness of regions, and pursue the environmentally friendly management of land. Various regional development plans are carried out to attain such balance. The purposes of spatial planning are multifaceted with everchanging needs and demands while sustaining a high quality of living. A key task is to identify the critical relations among many spatial components that are likely to shape the future economic, social, political, and environmental qualities of a region, with a per- suasive and mobilizing capability to facilitate multilevel participation among all the key actors including the private sector. These plans evolve with changing demands and priorities based on opportunities that defined achievement of the vision declared by the CNTP. For example, as illustrated in Figure B5.3.2, when industrialization was the CNTP’s goal in the 1960s, spatial planning focused on providing housing within cities and critical transportation, which allowed a flow of people and goods as a means to attract economic growth. By contrast, in more recent years there has been a shift in focus to high-tech industries and local economic self-reliance, promoted through regionally specialized development. For example new towns are be planned on the periphery of metropolises to address housing shortages. Figure B5.3.1 The structure of the urban planning system Comprehensive National Framework Action National Territory Territorial Plans Regional plans Sector plans Comprehensive • Metropolitan city development plans • Metropolitan city development plans province plans • Capital area development plans • Capital area development plans • Special region development plans • Special region development plans • Other regional development plans • Other regional development plans required by law required by law Comprehensive city-level plans (Urban plans) Comprehensive National Metropolitan city plans Territorial Plan Basic urban plans Urban management plans (Applicable nationwide) Plans by administrative unit Urban areas Managed areas Farms and forests Nature preserves THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT? 111 Figure B5.3.2 The historic evolution of the urban planning system in Korea Box 5.3 Lanch Acceleration Housing Expansion of Balanced national of industrialization of industrialization supply housing supply development today Economic Electronic & High-tech development Light industry Heavy industry High-tech & Financing industry Automoble industry industry Housing Small scale 2nd New Town supply Provision of housing within cities 1st New Town land development Sejong & Inovation city Transportation Expressway and industrial railroad construction Primary network expansion & subway construction Eco-friendly transport Source: OECD 2017. Urban planning needs to be connected to other policy tools, metropolitan in scope, and adjusted over time. To guide urban development, local governments need to think of urban planning not as a single instrument but as part of a policy toolbox together with the programming of infrastructure investments and property taxation incentives and disincentives. Local governments also need to have opportunities and incentives to adjust the territorial scope of urban planning from administrative municipal bounda- ries to functional urban areas, thus addressing peri-urban development from an integrated perspec- tive and generating synergies between municipalities. As external forces evolve, cities grow and core challenges change, urban planning needs to fine-tune its strategic direction over time, as for example, in the case of the Seoul Master Plan, which developed from a functionally segregated to a polycentric structure (Box 5.4). Box 5.4 Seoul’s urban master plan: Adjusting a spatial strategy to reach urbanization goals The 2030 Seoul Urban Master Plan presents a long-term direction for which a polycentric spatial structure is the fundamental approach. The polycentric structure is explicitly intended to control sprawl in order to increase productivity, reduce mobility demands, cut emissions, and improve access to green spaces and affordable housing. The approach guides statutory plans that de- 112 THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT? Box 5.4 termine specifically how land is used. The polycentric structure represents a shift from func- tionally segregated zoning, which separates uses such as residential, industrial, and commer- cial, and proposes a spatial pattern that mixes compatible land uses based on transit-oriented nodes. Previous approaches to contain sprawl, such as green belts, had generated negative externalities beyond the belt areas, such as higher land and housing costs, longer commutes, and increased public infrastructure spending. SOURCE: Seoul Solution, https://www.seoulsolution.kr/. Engaged and capable local governments are an essential part of the updated approach to urban land use planning. In the market economy, urban plans will not be implemented by the government alone but to a large extent by the private sector. The role of local and national administrations is central in realiz- ing the objectives of urbanization in Uzbekistan, but it is especially important that local governments are engaged in it. Urban planning departments at the local level need to be equipped with a set of skills to understand the granularity of location value, know how to work with market demand, establish synergies between sectors, and act as coordination and convergence agents between public and private actors. They must provide solutions to evolving challenges with agility while enabling municipalities to implement the national vision. Urban regulations should rationally support development rather than constrain it. Zoning regulations, also referred to as development control regulations, determine—in a legally binding and transparent way— two sets of critical information: (1) which types of land uses are permitted in particular areas (“zones”) and (2) which building parameters developers and landowners must follow on their land sites. The segrega- tion of uses (for example, “residential,” “commercial,” or “industrial”), characteristic of the 20th century, can be associated with a more dispersed spatial pattern of development. Today, it is widely embraced that regulations with a clearly defined mix of uses, create synergies between compatible activities (that is, excluding those which are noxious or hazardous) and can be associated to compact development which has positive effects on productivity and livability. Building parameters in zoning regulations apply to land parcels and buildings on them, and generally include maximum buildable areas, plot occupation, heights, setbacks, and other parameters. The floor area ratio (FAR) determines the buildable area for a plot of land considering factors such as site accessibility and infrastructure availability and has a strong impact on urban density. In market economies, FAR and location are key viability factors for private sector real estate investment in residential, commercial, and industrial real estate assets. One way to interpret FAR is to see it is as a parameter that captures the conversion from land to capital in providing floor space (Bertaud 2018). FAR values are established at the local level (i.e., zone, block, or parcel rather than generic as is the case for maximum FAR values that exist in Uzbekistan, as discussed in section 4.2). The public availability of binding information on zoning regulations is essential. It provides explicit intelligence for developers to prepare business cases and for municipalities to guide urbanization toward compact urban patterns. Compared with the current system, updated, binding and transparent zoning regulations will limit the risk of piecemeal and inconsistent urban development and reduce the cases of subjective interpretation that may arise from unclear regulations. This would contribute to generating certainty and level the playing field, which is key for a commercially viable and competitive real estate sector, while safeguarding public interest. THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT? 113 Streamlining the process of urban planning Alongside revisiting the foundational principles that shape the urban land use planning in Uzbeki- stan, the practice of developing and implementing plans and regulations needs to be streamlined. This means understanding the core functions and priorities of the land use planning system, making sure that priority issues are resolved first and the legacy of the most critical mistakes of the past are addressed, and ensuring that the system and the people operating in it have capacity to fulfill these critical func- tions, and the bureaucracy surrounding it does not deter it. In other words, while revisiting and moderniz- ing foundational principles of urban planning, authors of this reform need to keep a focus on the practi- calities of the operation of the system as well. Delineation of public and private land is a critical task for enabling market-driven and sustainable urban development. Identifying all publicly owned land (including vacant land) and registering it as gov- ernment-owned land is a critical first step of prudent land management that safeguards public wealth concentrated in this land. The second step is to decide which of this land can be subject to privatization and which should be retained for current and future public needs. Such delineation must take a long-term view and consider public needs for at least the next 15 years. In particular, sufficient corridors should be reserved for future roads. Delineation of public land into land that can be privatized and land that needs to be retained public should be a joint effort by land managers and urban planners; local government entities must play a prominent role in this process. Results of the delineation must be publicly available. Privatization of vacant land also should be planned strategically instead of quick dispositions and should be performed through competitive transparent procedures. Requiring and enforcing competitive, trans- parent privatization of land rights by local government entities will curb corruption currently associated with land allocation (Kun.uz 2021). However, this would not solve the problem that what developers or landowners want to build based on market demand often differs from what general plans and Architec- tural and Planning Task (APZs in Russian) allow them to build. For addressing this critical issue, the system of land use planning needs modernization through introduction of legal zoning and modernization of other city-level urban plans as discussed in previous sections. For areas where delineation of private and public land has not been done preemptively, land read- justment approaches can be used to improve service delivery. For unplanned urban development in contexts of rapid urbanization (as in the case of Namangan, discussed in chapter 4), land readjustment has shown potential as a process to improve areas that have grown without sufficient infrastructure, as in the case of the Gangnam district of Seoul (Box 5.5). Land readjustment consists of a negotiation between governments or other project initiators and landowners in areas with poor access to basic infrastruc- ture who contribute a part of their plot, forming a redevelopment area. The project initiator replots land boundaries and lays out infrastructure networks such as roads and water supply and introduces public facilities and open spaces in exchange for a portion of the land. Each landowner receives a serviced plot that has higher value and can remain in the same location with improved living conditions. Governments can upgrade entire areas without engaging in disruptive, costly, and time-consuming expropriation pro- cesses. 114 THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT? Box 5.5 Land readjustment in Japan allows cities to expand in a planned manner and to retrofit areas developed with insufficient infrastructure In Japan, land readjustment (LR) is an important method to develop and improve public facilities, enhance land usability and, in general, realize target spatial development patterns in growing and large cities. This includes residential area development in areas of peri-urban expansion, but also urban densification in core areas, as well as integrated urban development linked to public transport facilities, and comprises assets such as serviced residential and commercial land, roads and public transport corridors, and public spaces. To date, about one-third of the country’s total urban area has been created and redeveloped through LR. In the LR process, the land rights of residents in a project site are converted into rights to a part of a new construction and a share of the joint ownership of the plot in which it sits. Land is replot- ted to reshape land plots and create public spaces and reserve lands to recover development cost. Reliability, essential for the operability of LR, is afforded by a law specific to LR, related laws such as the Urban Planning Law and the Urban Redevelopment Law, and other regulations and guidelines. Each replotted land parcel is legally connected with the original land right by the des- ignated replotting plan. A transparent land exchange and valuation method allows maintaining fairness to land rights owners, which is central to the success of LR schemes. Projects can be initiated by public entities (central and local government and other state or pub- licly owned corporations), a landowner or landowners’ cooperatives, and corporations. Govern- ments provide subsidies for the construction of roads in the project site and provide technical and financial support in project preparation and implementation. LR implementers can receive incentives such as tax exemption for the real estate registration tax for replotted land and for the sale of reserve lands, and LR cooperatives and government corporations can receive reductions in corporate tax and income tax. SOURCE: World Bank 2017. Strengthening market understanding and technical skills at the local level is a critical part of stream- lining the system. Putting the updated approach and processes to work requires a solid understanding of land economics, market dynamics, and an ability to anticipate their impact on the demand of land and floor area. Strengthening the skills of local officials so that plans and zoning regulations do not be- come a constraint for these dynamics should be a foremost point in the process update. Planners working in municipalities would be required to combine these skills with a command of what the overall objectives of urban development are, and that every asset that is built in a city must support them. Required tech- nical and management aptitudes include the capacity to prepare normative urban plans and zoning regulations and enforce them through the permitting process and compliance check inspections. Clear guidelines of what can be built, and where, would enable local officials to overcome the inertia of case- by-case approvals, which are otherwise likely to lead to a piecemeal pattern defined by a myriad of spatially fragmented and formally incoherent property development projects. Construction permitting should be streamlined, but accelerating the permitting process does not mean relaxing requirements. Overall, the permitting process needs to be streamlined to limit the time THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT? 115 and cost to complete all formalities in the planning permit, and the quality control and safety mecha- nisms in the construction permitting system. Introduction of legal zoning/development controls, moving away from the practice of issuing bespoke APZs for every land plot, and updating of the building codes would all contribute to simplifying the permitting process. Streamlining the processing time to obtain construction permits would be beneficial for a dynamic real estate and construction sector. However, it is important to point out that cutting down permitting process time does not mean lowering require- ments that determine what can be built and how it should be built, as this may contribute to haphaz- ard and unsafe construction, as in recent cases of collapsing constructions. This process is above all a question of clarity and transparency: clarity in how the application will be evaluated and transparency around the availability of information so that developers know what can be built beforehand. These principles should be reflected in the language of requisites, the format of applications, and in the exter- nal and internal communication of urban planning and development control regulations so that they are known by interested parties and the public. Another critical component of implementation of plans is the system of enforcement of development regimes. Issuing a construction permit in accordance with zoning rules does not guarantee that devel- opment follows land use planning regulations. Further impartial compliance checks during and after con- struction are required. Such work is not possible without professionals trained and certified to carry out compliance checks. In rapid urbanization contexts, poor enforcement may render urban planning efforts futile as has occurred in cases in Vietnam (see Box 5.6). Box 5.6 A general plan is not enough: Negative consequences of lack of enforcement in Vietnam Although a city may be in possession of an updated general plan, this may not be enough to seize urbanization opportunities. A disconnection with market dynamics, the overprescription of gen- eralities and a lack of specific regulations, and a weak construction permit system and compli- ance inspection process may result in the emergence of inefficient spatial urban growth patterns. In Vietnam, a significant number of buildings have been built both in central urban areas and on the periphery of cities, including Hanoi and Ho Chi Minh City, without proper scrutiny. Construction on an ad hoc basis in the inner part of cities, where land supply is scarce, has produced dense neighborhoods with narrow streets, without drainage or sewerage systems, and with scarce pub- lic space. However, most of the development after 2000 occurred on the urban fringes, where land is cheaper and higher returns could be obtained by developers. The construction of new urban areas has contributed to the overconsumption of agricultural and environmentally sensitive land. Lack of scrutiny of proposed developments resulted in poorly designed developments being built with low-quality construction and insufficient services. SOURCE: World Bank 2018. Key reform priorities to leverage urban planning as an urbanization enabler Once land is urbanized, it is not possible to reverse the process in the short or mid term, and it can be a very costly affair to rectify it. An outdated urban planning approach and the obliviousness of general planning to market forces may place urbanizing countries (including Uzbekistan) on a trajectory that locks them into inefficient patterns for decades. For urbanization to reach its socioeconomic potential in Uzbek- 116 THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT? istan, a series of structural reforms in the realm of how urban land is planned and used, the deployment of infrastructure networks, and coordinated taxation policies will require sustained effort and actions. Specifically, in the area of spatial and urban land use planning, the following are key steps to follow:   Understand market dynamics across the urban development process. Unlocking the potential of the market will not only come from releasing land for development to the private sector. For all parties to work toward attaining overall urbanization objectives requires planners to understand the trends that affect demand, anticipate how these trends will manifest spatially, and develop municipal skills to engage and negotiate with the private sector. Completing a transition to ad- equate planning for market conditions is not possible without a change in the mindset of stake- holders across the whole system and will require taking a number of steps, starting with overall methodological guidance and corresponding training for local officials.   Focus on the systemic modernization of the urban planning system, not just updating general plans. The urban planning system must work like clockwork to avoid inefficient spatial patterns that hamper productivity and quality of life. In addition to enabling the attainment of the an- nounced quantitative urbanization goal, the urban planning system has a central contribution to make to the qualitative aspects of urbanization. To deliver on such potential, an updated method- ological guidance for the development of planning documents must provide coherence across all elements of the urban planning system. This includes, upstream, an evidence-based assessment of constraints of the current system and an up-to-date national hierarchy including an urban pol- icy and its associated spatial framework; midstream, strategic and normative urban plans; and downstream, zoning or development control regulations and permitting processes.   Decentralize the urban planning system and build local capacity. A key aspect of fully attain- ing urbanization benefits is that urban plans and regulations respond as effectively as possible to local challenges. Progressively transitioning to increased discretion at the local government level would initially require greater coordination with the subnational branches of the Ministry of Construction in parallel with building hands-on urban planning and management skills in munic- ipalities. Capacity development should equip technical staff with practical knowledge on how to achieve policy objectives through normative urban plans and development regulations. In addi- tion to the technical ability to draft these instruments, it is critical for local governments to develop a body of human resources to run the permitting and compliance check processes.   Generate certainty through coordinated infrastructure investments and urban planning. The urban planning system sketches, at various scales, a preferred spatial structure for a city, but it is infrastructure networks such as transport, water supply, and energy that physically materialize the spatial structure for decades. Implementing infrastructure investment as planned assures that public commitments made in binding planning instruments are predictable. Such consisten- cy becomes a real guide for home buyers, developers, and investors, and thus will have a strong impact on the shape of cities. Urban planning, including development control regulations, infra- structure programming, and property taxation policy need to work closely together for urbaniza- tion to reach its full potential. Coordinately, these levers can adjust decisions on accessibility to land and services to meet changing demand, to avoid mismatches between asset location and population distribution, and, overall, to guide compact, density-efficient development to where it makes most sense.   Make development control explicit and its application reliable. Zoning regulations and permit- ting processes are more effective in guiding development toward qualitative and quantitative THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT? 117 urbanization objectives when they are as explicit and location specific as possible. This allows urban planning instruments to tap on the contribution of public and private sectors toward shared objectives not by constraining development but by setting a clear direction and unambiguous rules for the land and property development required to accommodate the increase in urban population and economic activity. Zoning regulations and permitting should be modernized si- multaneously and in an integrated manner as it is essential that they are coordinated and un- equivocally support upstream and midstream urban planning instruments and infrastructure in- vestment decisions. 118 THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT? Chapter 6 Conclusion THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT? 119 Chapter 6 Conclusion It is critical that the Government of Uzbekistan undertake a systematic policy effort to enable mar- ket-driven urbanization and urban growth, so that the country does not miss the once-in-history de- velopment opportunity offered by urbanization. This report offers a high-level overview of the urban- ization process in Uzbekistan and provides substantial evidence that Uzbekistan’s current pathway to urbanization is cause for concern. Cities do not act as the engines of economy and of structural trans- formation and fail to provide a high quality of life. These worrying results are a product of two issues: (1) distorted factor markets that do not let talent and capital be applied in the manner and locations that are most productive; and (2) poor management of urban areas and specifically their growth, resulting in the emergence of sprawling and fragmented agglomerations with flailing infrastructure and missing ser- vices. Both issues are rooted in a lack of key conditions that enable and encourage market forces to drive urbanization, and policies and institutions that can provide public goods and address market failures to ensure access to opportunities and improved well-being for all. Among factor market disruptions, migration barriers have the most direct effect on urbanization. From earlier research we know that internal migration rates in Uzbekistan are very low. This report shows that this has resulted in potentially inefficient distribution of population across the country and its cities. Tash- kent is the clear productivity outlier among Uzbek cities, which should make it a major magnet for people. Yet it accounts for a smaller share of population than its peer cities in comparable countries, which sug- gests a missed opportunity to provide more people with a better quality of life and bring more workers to more productive employment. Analysis shows that in Uzbekistan cities close to large population centers (or markets) do not grow faster than others, a relationship that holds across the world. This suggests that for some reason these places fail to attract people. Barriers to migration, most specifically the propiska household registration system, has a lot to do with this outcome. The relaxation of the propiska regime, which is underway, together with work to address other mobility barriers (e.g., housing affordability) is a step toward unlocking the country’s urban potential. But it will not be enough. Migration might unlock great opportunities, but to leverage these, cities will need to be productive and livable. Thus, while mi- gration barriers are being lifted, along with other factor market distortions (e.g., financial market develop- ment), it is critical to focus on ensuring that cities are ready to provide the conditions needed for growth and improved well-being. The failures of urban management are best captured by the unproductive sprawl of urban agglomera- tions, which if sustained will limit the productivity and sustainability of Uzbek cities. There are different ways cities can grow. Global experience suggests that cities that are sustainable and productive den- sify and grow tall over time. These cities are also cheaper to service and provide with infrastructure. This is not what is happening in Uzbekistan. Results of analysis presented in this report suggest that Uzbek cites grow through sprawling. The agglomerations of Uzbek cities today have shapes, land use patterns, and road network structures that are less efficient than comparators, which most likely contributes to the limited growth of productivity and well-being. The predicted future expansion shows that if nothing changes, a lot of arable land will be lost to urban expansion, and the cost of providing residents of new urban neighborhoods with infrastructure and services will be much higher. This form of urban growth is to a large extent a result of policy failures. When infrastructure networks are underdeveloped and dilapidat- THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT? 121 ed, urban land use planning documents are outdated and not implemented, land administration system is not fully functional. As a result, for example, high-rise housing in central areas of cities is extremely difficult. But as housing demand keeps rising with population growth, cities tend to sprawl by allotting land for the building of individual private homes. In Uzbekistan, these conditions are further exacerbated by weaknesses of local governments that do not have powers, resources, or capacity to tackle key local issues, and only play a passive role in implementing locally targeted national government development programs. This report provides an overview of the urbanization process, but further detailed analysis will be needed to inform specific policy action. At this stage, the following key reform priorities have been identified:   Continue the advancement of critical structural reforms, with a specific focus on land market development and rebuilding the urban land use planning system. Structural reforms are need- ed to provide foundational conditions for market-driven urbanization, such as the reform of the household registration propiska system, housing reform, and improved provision of urban infra- structure and services. But as cities face the need to accommodate growing rural-urban migra- tion, urban land use planning reforms will arguably be of utmost importance.   The formation of land markets will help attract private investment to cities, optimize land use, and create new opportunities for local governments. While legislative breakthroughs on land privatization have spearheaded land reform, its full implementation is critical. A robust land management system needs to be built, with a focus on continued legislation updates, building the institutional framework and the robust information system. This is no easy task—but once it is achieved, it will on the one hand aid the land market in directing the capital investments to the most valuable land and help optimize urban land use; and on the other hand, it will provide the public sector with all the critical information and new tools (like value-based property taxation) to manage urban growth better.   The transition to an urban land use planning system suited to the market is also critical to ensure the transition to a more sustainable and productive form of urbanization. The urban land use planning system of Uzbekistan is derived from a tradition of centralized planning in the Soviet Union—it fails to meet the needs of the current moment. The planning documents that Uzbek cities use are inadequate, their implementation and enforcement is weak, and the rules that guide them are outdated. This creates chaotic, inefficient urban growth, and makes it difficult to create urban spaces that are attractive and productive. The reform should start the reframing of the core ideology of the system and changing urban plans from “rule books” into “market guides.” But it should also address the practice of centralized urban planning from the national capital, to ensure that adequate technical capacity is available at a local level; to link plans to capital investments managed by local authorities; and to ensure that development control regulations are enforced transparently.   Make a comprehensive policy effort to empower subnational governments and put them in the leading position with regard to local development. No territorial development effort can be fully implemented from a central location. Global experience shows that subnational governments are better positioned to define local challenges and development priorities and implement the policy programs and investments. However, this is only true when they have the powers, the finances, and the capacities that enable them to have an impact. Uzbekistan remains a highly centralized system. Local governments’ powers and resources are limited, planning documents for cities are 122 THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT? developed in Tashkent, and industrial zones are managed and funded from the capital. Local governments cannot predict their future revenues and plan around them. They get confusing in- centives and limited technical assistance or training. Empowering subnational governments will start with successful implementation of ongoing administrative reforms but will need to go beyond them. The policy effort would need to address the challenges of limited technical capacity of local governments, and the confused incentives that they are facing. Building strong local governments is not a problem that can be solved overnight. Most probably an iterative adoptive approach would be required, and innovative tools and incentives should be considered to advance it. But most importantly it is important to recognize, that without focusing on subnational governments’ capacity, the national effort to support development of cities will likely fail.   Develop and adopt an urbanization and territorial development strategy. It is incredibly impor- tant that at the national level, urbanization is recognized as an important agenda item. However, the target of reaching the official 60 percent urbanization level is not enough. To define priorities for the multisectoral policy agenda related to sustainable urbanization, a more detailed national strategy for urbanization is needed. Like multiple countries from around the world, including Cen- tral Asian neighbors, Uzbekistan should develop a strategy rooted in a deep analytical under- standing of current and future urbanization trends, offering the vision not only of the quantity but also the quality of urbanization and proposing how it can be achieved. The Government of Uzbekistan has a lot of work to do, but the benefits of sustainable and productive urbanization are worth it. There is still time to act—still time to create conditions under which people concentrate in places that provide more opportunity, and thus contribute to greater economic growth, still time to incentivize cities to grow denser rather than vaster, still time to make sure that subnational governments are becoming local development champions. But time is short, and decisive action needs to start now. It will not be easy and will require further detailed technical analysis of specific issues re- lating to housing and land markets, local government finance and capacity building, and governance of and management of agglomerations, among others. But it will pay off when cities emerge as the true champions that pave Uzbekistan’s way to a sustainable and prosperous future. THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT? 123 References THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT? 125 References Addison, D. M., & Stewart, B. (2015) Nighttime lights revisited: the use of nighttime lights data as a proxy for economic variables. World Bank Policy Research Working Paper, (7496). Adler, M. W., Pasidis, I., Levkovich, O., Lembcke, A. C., & Ahrend, R. (2020) Roads, market access and regional economic development. Amerasekera, A et al (2015) Building Urban Resilience to Disasters in the Kyrgyz Republic: Urban Planning, Development and Expansion in the Kyrgyz Republic, ICF International. Andrews, M., Pritchett, L., Woolcock, M. (2016) Doing Iterative and Adaptive Work – Working Papers, Center for International Development at Harvard University Au, C. C., & Henderson, J. V. (2006a) How migration restrictions limit agglomeration and productivity in China. Journal of development economics, 80(2), 350-388. Au, C. C., & Henderson, J. V. (2006b) Are Chinese cities too small?. The Review of Economic Studies, 73(3), 549-576. Bell, Martin, and Elin Charles-Edwards (2013) Cross-national comparisons of internal migration: an update of global patterns and trends. Bertaud, A. (2018) Order Without Design. Cambridge: MIT Press. Bertaud, A., Ranaud, B (1995) Cities Without Land Markets: Location and Land Use in the Socialist City, World Bank. Bertaud, A., & Richardson, H. W. (2004) Transit and density: Atlanta, the United States and western Europe. Urban Sprawl in Western Europe and the United States. London: Ashgate, 293-310.. Birch, Eugenie L. (2009) The Urban and Regional Planning Reader. Routledge. Blake, R. (2003) Regional and Rural Development Strategies in Canada: The Search For Solutions.: Royal Commission on Renewing and Strengthening Our Place in Canada Bleakley, H., & Lin, J. (2012) Portage and path dependence.  The quarterly journal of economics,  127(2), 587-644. Hein, C., & Schubert, D. (2021) Resilience and Path Dependence: A Comparative Study of the Port Cities of London, Hamburg, and Philadelphia. Journal of Urban history, 47(2), 389-419. Bleakley, H., & Lin, J. (2015) History and the Sizes of Cities. American Economic Review, 105(5), 558-63. Bruegman, R., (2005) Sprawl: A Compact History. University of Chicago press, London. Bryan, G., & Morten, M. (2019) The aggregate productivity effects of internal migration: Evidence from Indonesia. Journal of Political Economy, 127(5), 2229-2268. Tombe, T., & Zhu, X. (2019) Trade, migration, and productivity: A quantitative analysis of china. American Economic Review, 109(5), 1843-72. THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT? 127 Bryan, G., Glaeser, E., & Tsivanidis, N. (2020) Cities in the Developing World. Annual Review of Economics, 12, 273-297. Buckley, C. (1995) The Myth of Managed Migration: Migration Control and Market in the Soviet Period. Slavic Review, 54(4), 896-916. Chekirova, A. (2019) Undocumented Citizens: The Legacies Of The Propiska System In A Post-Soviet City (Doctoral dissertation, University of Illinois at Chicago). Ciccone A., and R. E. Hall (1996) Productivity and the Density of Economic Activity. American Economic Review 86 (1): 54–70. Combes, P. P., & Gobillon, L. (2015) The empirics of agglomeration economies. In Handbook of regional and urban economics (Vol. 5, pp. 247-348). Elsevier. Dale, Peter and Richard Baldwin (1999) Lessons Learnt from the Emerging Land Markets in Central and Eastern Europe. Quo Vadis - International Conference Dale, P., & McLaughlin, J. (2000) Formalizing Property Rights. In Land Administration.: Oxford University Press. Retrieved 25 Oct. 2021 de Roo, G., and D. Miller (2000) Compact Cities and Sustainable Urban Development: A Critical Assessment of Politics and Plans from an International Perspective. Aldershot, U.K.: Ashgate. Dentos (2021) Uzbekistan enhances public-private partnership legislation, https://www.dentons.com/en/insights/articles/2021/february/5/uzbekistan-enhances-public-private- partnership-legislation Department of Housing of Republic of South Africa (1997) Urban Development Framework. Dijkstra, L.,; E., Hamilton; S., Lall; S., Wahba (2020a) How do we define cities, towns, and rural areas? – World Vank Blogs: Sustainable Cities. https://blogs.worldbank.org/sustainablecities/how-do-we-define-cities-towns-and-rural-areas Dijkstra, L., Florczyk, A. J., Freire, S., Kemper, T., Melchiorri, M., Pesaresi, M., & Schiavina, M. (2020b) Applying the degree of urbanisation to the globe: A new harmonised definition reveals a different picture of global urbanisation. Journal of Urban Economics, 103312. Duranton, Gilles and Diego Puga (2004) Microfoundations of urban agglomeration economies, in J Vernon Henderson and Jacques François Thisse (editors), Handbook of Regional and Urban Economics: Volume 4, Cities and Geography, Elsevier North Holland, Amsterdam, pages 2063−2117; Duranton G., Venables A. (2018) Place-Based Policy for Development, World Bank Group. Ellis, Peter; Roberts, Mark (2016) Leveraging Urbanization in South Asia : Managing Spatial Transformation for Prosperity and Livability. Washington, DC: World Bank. Elvidge, C. D., Baugh, K., Zhizhin, M., Hsu, F. C., & Ghosh, T. (2017) VIIRS night-time lights. International Journal of Remote Sensing,  38(21), 5860-5879. See Elvidge et al. (2017) for more information on data processing. 128 THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT? Enemark, S., Williamson, I., Wallace J. (2005) Building modern land administration systems in developed economies, Journal of Spatial Science, 50:2, 51-68, DOI: 10.1080/14498596.2005.9635049 Eurasian Research Institute (2019) Weekly e-bulletin #218: NEW URBANIZATION POLICY OF UZBEKISTAN. European Commission (2019) Global Human Settlement Layer: Urban Centre Database 2015. https://ghsl.jrc.ec.europa.eu/ghs_stat_ucdb2015mt_r2019a.php Fallah, B., M. Partridge, and M. Olfert (2011) Urban Sprawl and Productivity: Evidence from US Metropolitan Areas.” Papers in Regional Science 90 (3): 451–73. Fay, M., Opal, C. (2000) Urbanization without Growth: A not-so-uncommon Phenomenon. Feinstein, A. (2015) Decentralization: The South Africa Experience – Global Partners, Governance. Ferré, Céline, Francisco HG Ferreira, and Peter Lanjouw (2012) Is There a Metropolitan Bias? The relationship between poverty and city size in a selection of developing countries. The World Bank Economic Review 26.3: 351-382 Ferreyra, M. and Roberts M. (ed), (2017) Raising the Bar: for Productive Cities in Latin America and the Caribbean, World Bank. Florczyk, Aneta; Corbane, Christina; Schiavina, Marcello; Pesaresi, Martino; Maffenini, Luca; Melchiorri, Michele; Politis, Panagiotis; Sabo, Filip; Freire, Sergio; Ehrlich, Daniele; Kemper, Thomas; Tommasi, Pierpaolo; Airaghi, Donato; Zanchetta, Luigi (2019a) GHS Urban Centre Database 2015, multitemporal and multidimensional attributes, R2019A. European Commission, Joint Research Centre (JRC) [Dataset] PID:  https://data.jrc.ec.europa.eu/dataset/53473144-b88c-44bc-b4a3-4583ed1f547e Giacomin, David J., and David M. Levinson (2015) Road Network Circuity in Metropolitan Areas. Environment and Planning B: Planning and Design 42 (6): 1040–53. Glaeser, E. L. (1998) Are Cities Dying? Journal of Economic Perspectives 12 (2): 139–60. Glaeser, E. L. (2008) Cities, agglomeration, and spatial equilibrium. Oxford University Press. Glaeser, E. L. (2011) Cities, productivity, and quality of life. Science, 333(6042), 592-594. Glaeser, E. L., & Gottlieb, J. D. (2006) Urban resurgence and the consumer city. Urban studies, 43(8), 1275-1299. Glaeser, E. L., Cutler, D. (2021) Survival of the City: Living and Thriving in the Age of Isolation. Glaeser, E. L., Kolko, J., & Saiz, A. (2001) Consumer city. Journal of economic geography, 1(1), 27-50. Government of Russian Federation (2018) Strategy of Spatial Development of Russian Federation until 2025 [in Russia]. Graham, D. J., & Gibbons, S. (2019) Quantifying Wider Economic Impacts of agglomeration for transport appraisal: Existing evidence and future directions. Economics of Transportation, 19, 100121. Gyourko, J., Mayer, C., & Sinai, T. (2013) Superstar cities. American Economic Journal: Economic Policy, 5(4), 167-99. THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT? 129 Hale, T., Petherick, A., Phillips, T., & Webster, S. (2021a) Variation in government responses to COVID-19. Blavatnik school of government working paper, 32, 2021-06. Hale, T., Angrist, N., Goldszmidt, R., Kira, B., Petherick, A., Phillips, T., ... & Tatlow, H. (2021b) A global panel database of pandemic policies (Oxford COVID-19 Government Response Tracker). Nature Human Behaviour, 5(4), 529-538. Handbury, J., & Weinstein, D. E. (2015) Goods prices and availability in cities.  The Review of Economic Studies, 82(1), 258-296. Hansen, V. (2012) The Silk Road. Oxford University Press. Harari, Mariaflavia (2020) Cities in Bad Shape: Urban Geometry in India. American Economic Review, 110 (8): 2377-2421. Henderson, J.V. (2010) “Cities and development”, Journal of Regional Science Vol 50, No 1, pages 515–540. Henderson, J. V., & Wang, H. G. (2007 Urbanization and city growth: The role of institutions.  Regional Science and Urban Economics, 37(3), 283-313. Henderson, V., Storeygard, A., & Weil, D. N. (2011) A bright idea for measuring economic growth. American Economic Review, 101(3), 194-99. Henderson, J. V., Storeygard, A., & Weil, D. N. (2012) Measuring economic growth from outer space. American economic review, 102(2), 994-1028. Henderson, J. V., Squires, T., Storeygard, A., & Weil, D. (2018) The global distribution of economic activity: nature, history, and the role of trade. The Quarterly Journal of Economics, 133(1), 357-406. Hojaqizi, G. (2008) Citizenship and Ethnicity: Old Propiska and New Citizenship in Post-Soviet Uzbekistan. Inner Asia, 10(2), 305-322. Huber, S., & Rust, C. (2016) Calculate travel time and distance with OpenStreetMap data using the Open Source Routing Machine (OSRM). The Stata Journal, 16(2), 416-423. Izvorski, Ivailo V.; Trushin, Eskender; Appiah-Koranteng, Alex; Varoudakis, Aristomene; Islam, Roumeen; Safarov, Maksudjon; Lord, David; Motohashi, Mitsunori; Ihsan, Ahya; Zorya, Sergiy; Hawkesworth, Ian; Philipsen, Ferry; James, Sebastian; Anadolu, Elvira; Teixeira, Janssen; Karakulah, Kenan (2019) Uzbekistan Public Expenditure Review. World Bank, Washington, DC. © World Bank. https://openknowledge.worldbank.org/handle/10986/33371 License: CC BY 3.0 IGO. Kaw, J.K., Kaganova, O. (2022, forthcoming) Urban Planning and Land Management in the MENA region: Policy and Research. – In Kaw and Lall (Editors). Vibrant Cities: How Can Cities Support Jobs and Growth – World Bank. Krugman, P. (1991a) Increasing returns and economic geography. Journal of political economy, 99(3), 483-499. Krugman, P. (1991b) History and industry location: the case of the manufacturing belt.  The American Economic Review, 81(2), 80-83. Kummu, M., Taka, M., & Guillaume, J. H. (2018) Gridded global datasets for gross domestic product and Human Development Index over 1990–2015. Scientific data, 5(1), 1-15. 130 THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT? Kun.uz, April 24 (2021) Khokimiyats illegally allocated 106,000 hectares of land in 2020, Prosecutor General. https://kun.uz/en/news/2021/04/24/khokimiyats-illegally-allocated-106-thousand-hectares-of-land-in- 2020-prosecutor-generals-office Lall, S. V., Lebrand, M., & Soppelsa, M. E. (2021) The Evolution of City Form: Evidence from Satellite Data. Policy Research Working Paper No. 9618. World Bank, Washington, DC. Leyk, S., Gaughan, A. E., Adamo, S. B., Sherbinin, A. D., Balk, D., Freire, S., ... & Pesaresi, M. (2019) The spatial allocation of population: a review of large-scale gridded population data products and their fitness for use. Earth System Science Data, 11(3), 1385-1409. Lipatov, Sh. (2018) Republic Tashkent, how propiska system diminishes the rights of citizen. [In Russian]; Gazeta.ru https://www.gazeta.uz/ru/2018/12/06/registration/ Litman, T. (2015) Analysis of Public Policies That Unintentionally Encourage and Subsidize Urban Sprawl. Victoria Transport Policy Institute. Supporting paper commissioned by LSE Cities at the London School of Economics and Political Science, on behalf of the Global Commission on the Economy and Climate (www. newclimateeconomy.net) for the New Climate Economy Cities Program. Marconcini, M., Gorelick, N., Metz-Marconcini, A., and Esch, T. (2018) Mapping the Global Settlement Growth from 1985 to 2015 - the World Settlement Footprint Evolution Dataset, vol. 2018. McKinsey & Co (2013) How to Make a City Great. Michaels, Guy, Ferdinand Rauch, Stephen J. Redding (May 2012) Urbanization and Structural Transformation, The Quarterly Journal of Economics, Volume 127, Issue 2, Pages 535–586. Ministry of Finance (2019) Budget for the citizens: Implementation of the budget for 2020. Ministry of National Development Planning of Indonesia (2016) National Urban Development Policy: toward Sustainable Urban development. https://www.unescap.org/sites/default/files/Arifin%20Rudyanto%20-%20Quito.pdf. Ministry of Urban Development of Albania, National Territorial Planning Agency of Albania (2014) National General Territorial Development Plan 2015-2030. Muradova, S., & Seitz, W. (2020) Compulsory Residence Registration in Uzbekistan. Implementation, Recommendations, and International Comparisons. Nasution, A. (2016) Government Decentralization Program in Indonesia – ADBI Working Paper Series, Asian Development Bank Institute. Ngigi. S, Busolo, D.N. (2019) Devolution in Kenya: The Good, The Bad and the Ugly. - Public Policy and Administration Research Nhede, N. (2020) Uzbekistan shares details of new public-private partnership energy project, - Smart Energy International https://www.smart-energy.com/industry-sectors/finance-investment/uzbekistan-shares-details-of- new-public-private-partnership-energy-project/ THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT? 131 OECD (2015) Local Economic Leadership. OECD (2017) Land Use Planning Systems in OECD. Office of Prime Minster of Republic of Kazakhstan (2019) Forecast Scheme of territorial development of Kazakhstan until 2030. [in Russian]. Olsson, O., & Hibbs Jr, D. A. (2005) Biogeography and long-run economic development.  European Economic Review, 49(4), 909-938. Oxford Economics (2021) Global Cities Database. Paunov, C., et al. (2019) On the concentration of innovation in top cities in the digital age, OECD Science, Technology and Industry Policy Papers, No. 85, OECD Publishing, Paris, https://doi.org/10.1787/f184732a-en. Perzadayeva, S. (2020) Public-Private Partnership In Uzbekistan: Development Of Road Projects, - Conventus Law https://www.conventuslaw.com/report/kazakhstan-public-private-partnership-in/ Pike, A., Rodríguez-Pose, A., Tomaney, J., Torrisi, G. and Tselios, V., (2012) In search of the ‘economic dividend’ of devolution: spatial disparities, spatial economic policy, and decentralisation in the UK. Environment and Planning C: Government and Policy, 30(1), pp.10-28.0 President of Uzbekistan (2018) Address by the President of the Republic of Uzbekistan Shavkat Mirziyoyev to Oliy Majlis. Quintero, L. E., & Restrepo, P. (2019) City Decline in an Urbanizing World-Preliminary. Republic of Uzbekistan (1993) Law on Local Public Administration, adopted on September 2, 1993, https://www.lex.uz/acts/112168. Republic of Uzbekistan (2016) Strategy of action on five priority areas of development of the Republic of Uzbekistan for 2017-2021 Republic of Uzbekistan (2019) The Law on Privatization of Non-agricultural Land https://lex.uz/ru/docs/4471686 [in Russian] Republic of Uzbekistan (2021a) Law 728 on privatization of non-agricultural land, signed by the President on November 15, 2021 Republic of Uzbekistan (2021b) Law on National Budget of the Republic of Uzbekistan for 2021 https://lex.uz/ru/docs/5186047. Republic of Uzbekistan (2022) Law on National Budget of the Republic of Uzbekistan for 2022 https://www.lex.uz/uz/docs/5801129. Roback, J. (1982) Wages, rents, and the quality of life. Journal of political Economy, 90(6), 1257-1278. Rodríguez-Pose, A., & Bwire, A. (2004) The economic (in) efficiency of devolution. Environment and Planning A, 36(11), 1907-1928. 132 THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT? Sala-i-Martin, X. (2006) The world distribution of income: falling poverty and… convergence, period. The Quarterly Journal of Economics, 121(2), 351-397. Satterthwaite, David (2007) The transition to a predominantly urban world and its underpinnings, Human Settlements Working Paper Series, Urban Change 4, IIED, London, pages 28–31. Schiavina, Marcello; Freire, Sergio; MacManus, Kytt (2019) GHS population grid multitemporal (1975, 1990, 2000, 2015) R2019A. European Commission, Joint Research Centre (JRC) DOI:  10.2905/42E8BE89-54FF- 464E-BE7B-BF9E64DA5218 PID:  http://data.europa.eu/89h/0c6b9751-a71f-4062-830b-43c9f432370f Seitz, W. (2019a) International migration and household well-being: Evidence from Uzbekistan. The World Bank. Seitz, W. (2020). Free Movement and Affordable Housing: Public Preferences for Reform in Uzbekistan. The World Bank. Seitz, W. (2019b) Where They Live: District-Level Measures of Poverty , Average Consumption and the Middle Class in Central Asia. – Working Paper Series, World Bank Group. Seitz, W. (2020) Free Movement and Affordable Housing: Public Preferences for Reform in Uzbekistan. The World Bank. Shorrocks, A. F. (2013) Decomposition procedures for distributional analysis: a unified framework based on the Shapley value. Journal of Economic Inequality, 11(1), 99. Squires, G. D. (2002) Sprawl: Causes and Consequences and Policy Responses. Washington, DC: The Urban Institute. State Committee on Statistics of Uzbekistan (2019) Storper, Michael (2010) Agglomeration, trade and spatial development: bringing dynamics back in, Journal of Regional Science Vol 50, No 1, pages 313−342. Tewari, M., S. Alder, and M. Roberts (2016) Patterns of India’s Urban and Spatial Development in the Post- Reform Period: An Empirical Analysis. Working paper. Toboso, F., & Scorsone, E. (2010) How much power to tax do regional governments enjoy in Spain since the 1996 and 2001 Reforms?. Regional & Federal Studies, 20(2), 157-174. Tripathy, Pratyush, and Amit Kumar (2019) Monitoring and Modelling Spatio-Temporal Urban Growth of Delhi Using Cellular Automata and Geoinformatics. Cities, vol. 90, July 2019, pp. 52–63. ScienceDirect, doi:10.1016/j.cities.2019.01.021. Turok I, McGranahan G. (2013) Urbanization and economic growth: the arguments and evidence for Africa and Asia. Environment and Urbanization. 2013;25(2):465-482. UN (2018a) Revision of World Urbanization Prospects. UN (2018b) World Urbanization Prospects 2018; UNECE (2015) Country Profiles on Housing and Land Management: Uzbekistan. THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT? 133 UN-HABITAT (2015) NATIONAL URBAN POLICY: A Guiding Framework, -United Nations Human Settlements Programme UN-Habitat (2019) How to Formulate a National Urban Policy: A practical Guide. Vishwanath, T., Lall, S. V., Dowall, D., Lozano-Gracia, N., Sharma, S., & Wang, H. G. (2013) Urbanization beyond municipal boundaries: nurturing metropolitan economies and connecting peri-urban areas in India (No. 75734). White, R., and G. Engelen (2000) High-Resolution Integrated Modelling of the Spatial Dynamics of Urban and Regional Systems. World Bank (2008) World Development Report 2009: Reshaping Economic Geography. World Bank (2015) Competitive Cities for Jobs and Growth. What? Who? And How. World Bank (2017) Case Study. Land Readjustment in Japan. https://collaboration.worldbank.org/content/usergenerated/asi/cloud/attachments/sites/collaboration- for-development/en/groups/tod-cop/documents/jcr:content/content/primary/blog/case_study_land_ re-4U95/Land_Readjustment_Japan.pdf World Bank (2018) Vietnam’s Urbanization at a Crossroads; Urban Gateway. Vietnam cities fail to follow urban development plans. http://www.urbangateway.org/content/news/vietnam-cities-fail-follow-urban-development-plans. World Bank (2019) Farm Restructuring in Uzbekistan: How does it go and what is next? https://documents1.worldbank.org/curated/en/686761549308557243/pdf/134322-WP-P162303-PUBLIC- Report-Farm-Restructuring-in-Uzbekistan-eng.pd World Bank (2020) Territorial Development in Argentina: using differentiated policies to reduce disparities and spur economic growth. World Bank (2021) World Development Indicators. World Bank (2022) World Development Indicators. World Bank (forthcoming) Towards a Prosperous and Inclusive Future: the Second Systematic Country Diagnostic for Uzbekistan. World Health Organization (2019) Paraguay and Uzbekistan certified as malaria-free Weekly Epidemiological Record, 94(10), 117-120. Yusupov Y. (2020) Economic reforms in Uzbekistan: Achievements, problems, perspectives. The Foreign Policy Center. Zokirov, S. (2020) Large Villages Are An Important Source Of Urban Formation. - European Journal of Molecular & Clinical Medicine. Volume 07, Issue 03,2020. 134 THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT? Appendices THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT? 135 Appendix A Urban sprawl in Uzbekistan, additional analysis Uzbek cities are sprawling more than their comparators and densifying less. Figure A.1 plots the growth rate of the urban agglomerations from 2000 to 2015 (from GHS-POP data) 59 of cities with populations above 50,000 in Europe and Central Asia (ECA) (excluding high-income ones) against their growth rate of population over the same period of time. The graph shows that the fastest-growing agglomerations in Uzbekistan (blue points further to the right) are among those with the larger area sprawl amo ng comparators (higher on the graph). Indeed, when zeroing in on the evolution of density of urban areas in Central Asian (and densification is the only alternative to expansion when accommodating urban growth), Uzbek cities appear to be densifying less than their regional comparators (Figure A.2 shows that light blue points signifying Uzbek cities are mostly located lower than dark blue points representing cities of comparator countries). 60 This pattern is particularly pertinent to the largest cities in Uzbekistan. Tashkent, for example, saw a mere 2.2 percent increase in density in the 15-year period, vis-à-vis a 10.9 percent increase in population. At the same time, cities like Ashgabat, Almaty, and Dushanbe have wit- nessed increases in density ranging from 26.3 percent to 36.3 percent. Figure A.1 Population growth and urban sprawl (built-up area expansion) in cities of Uzbekistan and the ECA region (non-high income) Growth rate of urban extent 2000-2015 Uzbekistan Central Asia ECA (World Bank) Declinig cities Growing cities Growth rate of population 2000-2015 SOURCE: Original elaboration based on GHS-POP 2000 and 2015 data using the EC definition of urban clusters. 59  Florczyk et al. 2019; there are different alternatives. 60  However, the difference between the average density growth in cities in Central Asia (13.7 percent) and Uzbekistan (10.7 percent) is not statistically significant. THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT? 137 Figure A.2 Change of density in cities of Uzbekistan and Central Asia Growth rate of population density 2000-2015 Nur-Sultan Log population in 2015 Uzbekistan Central Asia SOURCE: Original elaboration based on GHS-POP 2000 and 2015 data using the EC definition of urban clusters. As discussed in chapter 2 urban sprawl results in cities merging into large agglomerations and urban regions. Additional evidence of Uzbek cities’ tendencies to expand rather than densify is provided by the reality of cities merging into large agglomerations which can be seen on built-up area maps. Figure A.3 depicts how it looks on the map using the example of the Karshi agglomeration. And the process of urban areas merging was more prevalent in Uzbekistan than in other countries in recent years. Figure A.4 plots the number of agglomerations in the country in 2000 and then in 2015. Uzbekistan is a clear outlier (note that as the axes are in logarithms, distances on the right of the graph represent larger quantities) as 134 urban areas that were identified in 2000 merged to form only 62 in 2015. This translates into an average of 2.16 urban agglomerations in 2000 for each of the 2015 agglomerations (the largest in the region), com- pared to an ECA median of 1.25. This can be interpreted as the number of self-standing urban agglomera- tions in Uzbekistan halving over the 15-year period, due to urban sprawl that forms larger agglomerations. Other Central Asian countries have similar trends, as Turkmenistan and Tajikistan exhibit the second- and third-largest averages in ECA, 1.86 and 1.59, respectively. 138 THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT? Figure A.3 Map of extensive urban area forming Karshi SOURCE: Original elaboration based on WorldPop and OSM. Figure A.4 Merging urban agglomerations: Number of urban clusters in 2000 and 2015 6 Uzbekistan Central Asia 5 2010 urban clusters (log) Other ECA 4 3 2 1 45 degree line 0 0 1 2 3 4 5 6 2015 urban clusters (log) SOURCE: Original elaboration based on GHS-POP 2000 and 2015. NOTE: ECA = Europe and Central Asia. THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT? 139 As cities in Uzbekistan grow, they tend to densify less than in other countries. Table A.1 provides the re- sults of regressions looking at the effect of population increases on density using a panel of cities, similar to Lall, Lebrand, and Soppelsa (2021). The coefficients imply that density of cities changes less as popula- tion increases in Uzbek cities compared to other cities in ECA. In Uzbekistan, doubling a city size yields an increase in density that is 9 percentage points lower compared to the mean response in ECA (column 1). When measuring the effect within cities (column 2), Uzbek cities still display a lower elasticity than the ECA mean, but the difference reduces to 1.45 percentage points.61 These estimates suggest that in Uzbekistan cities tend to accommodate population growth more through Table A.1 Population elasticity of density (1) (2) Between estimator Within estimator Log population 0.340*** 0.522*** (0.016) (0.023) Uzbekistan # Log population -0.0900** -0.0145 (0.039) (0.065) Fixed effects Country and year City and year R^2 0.579 Observations 1,834 1,834 NOTE: The model in column (2)—Within estimator—includes a dummy for each city in the sample which captures the effect of city-level time invariant characteristics that could affect population density. In the first column, the relationship between population and density is mostly explained using the variation between cities (how cities differ from each other), while in the second using the variation within cities (how cities differ from themselves at a different point in time). 61  However, the coefficient is not statistically different from zero at the 10 percent significance level. 140 THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT? Appendix B Using night-time lights to measure city productivity Canonical measures of urban productivity rely on either wages or total-factor productivity (Combes and Gobillon 2015), which require either detailed firm-level data or administrative data covering the working population. The collection of such data involves established practices in the hands of agencies with ad- vanced capabilities and is not attainable for many developing countries across the world. In this situation, night-time light emissions can come to the rescue. Following a seminal paper by Henderson, Storeygard, and Weil (2011), night-time light emissions are widely used in economics to proxy economic variables in the absence of data from national accounts, as they are strongly correlated with economic activity (Hender- son, Storeygard, and Weil 2012). Figure B.1 shows the positive relationship between night-time lights per capita and gross domestic product (GDP) per capita in Global Human Settlement (GHS) urban centers.62 Throughout this work, the night-lights data used is the time series of the annual global Visible Infrared Im- aging Radiometer Suite captured radiance measured in nano watts per square centimeter per steradian (nW/cm2/s) produced by the Earth Observation Group from monthly cloud-free average radiance grids spanning 2012 to 2020.63 Productivity is measured as the residual of regressing the natural logarithm of the sum of night-lights radiance in the city extent on the natural logarithm of the population. In other words, it represents the supplemental radiance (economic activity) that cannot be explained by the sheer number of people living in the area. When computing, the elasticity of density productivity is instead directly measured through night-time lights emissions per capita that despite being a looser measure offers the advantage of a clearer interpretation of the estimated elasticities. Moreover, as shown in Figure B.2, the two variables are strongly correlated. Figure B.1 Night-time lights per capita and GDP per capita in cities Uzbekistan Log night-time lights per capita ECA (World Bank) Rest of the world Log city GDP per capita 2015 SOURCE: GHS-UCDB 2015. 62  The GHS urban centers’ database includes the average night-time radiance from VIIRS (2015). See also Florczyk et al. (2019). 63  See Elvidge et al. (2017). In this work, the time series is restricted to 2013–19 as the 2012 data are incomplete and there are no available population data for 2020. The data used are the yearly median per cell and based on the updated methodology (Version 2) by the EOG, which employs outlier removal to discard biomass burning pixels and isolate the background. See Elvidge et al. (2017) for more information on data processing. THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT? 141 Figure B.2 Comparing different night-time lights-based measures of productivity Uzbekistan cities (2013-2019) Log night-time lights per capita Residuals of regressing log NTL on log population SOURCE: EOG VIIRS night-time lights 2013–19, Yandex Maps boundaries, State Committee on Statistics of Uzbekistan. NOTE: NTL = night-time lights. 142 THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT? Appendix C Analysis of productivity elasticity of Uzbek cities Larger cities are expected to achieve higher productivity through sharing, matching, and learning, but this relationship does not hold for cities in Uzbekistan. Using administrative population data and relative boundaries, 64 it is possible to estimate elasticity of productivity of approximately -0.67. 65 Fur- ther validation is achieved by using the lagged value of log population density as an instrument that this produces and estimated elasticity of urban productivity to density for Uzbekistan of -0.17. Adding dummies for each city interacted with the explanatory variable (log population density) allows estimat- ing city-level elasticities. Figure C.1 plots cities’ density elasticities of productivity against the natural logarithm of population in 2018. It is evident how many cities in the country display negative elasticities, meaning that they fail to capitalize the benefits from densification. Naturally, these estimates need to be interpreted with caution as night-time lights per capita represent an imperfect proxy for productivity and would need to be validated once granular wage or industry data are available. Nonetheless, they present a much needed and otherwise lacking assessment of the competitiveness of cities in Uzbek- istan. Figure C.2 shows the absence of a clear relationship between the average population growth rate in the period of interest (2013–19) and the estimated city-level elasticity, suggesting that cities that would benefit more from densification are not necessarily growing relatively faster than the others. Figure C.1 City-level agglomeration elasticities in Uzbekistan Uzbekistan cities Agglomeration elasticity Uzbekistan agglomeration elasticity (-0.6652) Log population in 2018 SOURCE: EOG VIIRS night-time lights 2013–19, Yandex Maps boundaries, State Committee on Statistics of Uzbekistan. 64  The administrative boundaries are downloaded from Yandex Maps (https://yandex.com/maps). For a discussion of the different definitions of cities, see Box 2.1 in chapter 2. 65  The estimate is obtained using year and city fixed effects. THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT? 143 Figure C.2 Population growth and agglomeration elasticity at the city level Average population growth (2013-2019) Uzbekistan cities Median population growth (2013-2019) Agglomeration elasticity in 2013 SOURCE: EOG VIIRS night-time lights 2013–19, Yandex Maps boundaries, State Committee on Statistics of Uzbekistan. 144 THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT? Appendix D Effect of adopting a different definition of cities and urban areas on perceptions of Uzbekistan’s urban system As growing cities in Uzbekistan expand rather than densify, they merge into large agglomerations with low population density. These agglomerations in most cases create spaces that are functionally interre- lated—they have one labor market (people commute across the areas for work), one housing market, and one market for public and private services. If this definition of urban areas based on urban built-up foot- prints is adopted, it changes the understanding of geography of individual cities and of the understanding of the structure of the system of cities. For instance, one may think that Fergana valley has dozens of small and large urban settlements, but a closer look reveals that it essentially is covered by an extensive urban fabric that crosses administrative boundaries of cities, and makes up just a handful of separate functional urban agglomerations (Figures D.1 and D.2). Figure D.1 Fergana Valley’s administrative Figure D.2 Fergana Valley’s urban boundaries of cities and built-up areas agglomerations SOURCE: Original calculations using GHSL, WorldPop, and administrative boundaries from yandex.ru/maps. NOTE: GHSL = Global Human Settlement Layer. This form of urban expansion happens across the country and changes the way we think of the geog- raphy of cities in Uzbekistan. Figure D.3 demonstrates the alternative map of urban areas in Uzbekistan which is not defined by 20–30 large cities, but by over a dozen sprawling agglomerations, which can be grouped into seven large city-regions (Figure D.4). Furthermore, when a different definition of cities is ap- plied, it changes the understanding of the structure of the urban system. Relative to the system defined by administrative boundaries, the system defined on the basis of built-up area looks substantially more concentrated (Figure D.5). It also highlights that the country now has multiple secondary cities that, while big themselves, lag far behind the leaders, and thus fail to become gravitational centers that attract population. THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT? 145 Figure D.3 Uzbekistan’s functional urban areas SOURCE: Original calculations using WorldPop database data. Figure D.4 Uzbekistan’s urban regions SOURCE: Original calculations using the WorldPop database. 146 THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT? Figure D.5 Zipf’s curve representing Uzbekistan’s urban system 7,0 6,5 6,0 5,5 5,0 4,5 4,0 3,5 0 0,5 1,0 1,5 2,0 2,5 Log Ranking of City Uzbekistan Uzbekistan Agglomerations (using GHS POP) SOURCE: Original calculations based on data from the Statistical Agency of Uzbekistan and GHS-POP. NOTE: GHSL = Global Human Settlement Layer. THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT? 147 Appendix E Tashkent relative to its comparators Tashkent is poorer than the closest Figure E.1 Key indicators of Tashkent’s growth relative to its cities in ECA and the world. This ap- closest comparators, 1990–2015 pendix (and Figure E.1) presents ad- ditional evidence on Tashkent, and benchmarks it to the most similar cities Tashkent 2.7M 247 km2 2573 $ 6097 per km2 $ $ in ECA and the world selected using 1990 2015 algorithms and exploiting the richness of the GHS Urban Centre Database 3.2M 284 km 2 5254$ 6265 per km2 (Florczyk et al. 2019). The selected (+16.9%) (+15.1%) (+104.1%) (+2.8%) comparators for Tashkent are Xi’an, ECA comparators Jinan, and Algiers in the unrestricted sample and Kyiv, Izmir, and Baku in the Kyiv UKR Izmir TUR Baku AZE ECA one. Analysis shows that Tashkent 2.4M 236 km2 2.5M 144 km2 1.3M 181 km2 is the city with the lowest GDP per capita among comparators. Howev- 2.7M 287 km2 2.8M 171 km2 2M 200 km2 (+14.8%) (+21.4%) (+10.9%) (+18.8%) (+48.6%) (+10.7%) er, Tashkent has been growing signifi- $ 8943 $ 5299 per km2 10946 $ 10454 per km2 6105 $ 4888 per km2 $$ $ $$ cantly faster as it started from a much lower income level in 1990. Nonethe- less, the two closest comparators in 5285 per km2 7915 per km2 5908 per km2 the world, the Chinese cities of Xi’an 11207 $ (-0.2%) (-24.3%) 10064 $ (+64.8%) (+11.1%) (+25.3%) 17564 $ and Jinan, started from even lower (+60.5%) levels of income and have outpaced Tashkent in GDP per capita growth by International comparators a factor of 7. Xian CHN Jinan CHN Algiers DZA Tashkent accommodated population 3.5M 263 km2 2.4M 234 km2 2.5M 243 km2 growth through increasing built-up area. Population density in Tashkent 335 km2 308 km2 290 km2 has not changed in the 25-year pe- 5.1M (+27.7% 3 5M (+43.8%) (+31.6%) 3.9M (+52.7%) (+19.6%) (+46.6%) riod of interest (1990–2015) as popu- 939$ 7314 per km2 1407$ 6758 per km2 7663 $ 7154 per km2 $ $ $$ lation growth has been matched by $ $ an equivalent growth in built-up area. This is in line with most comparators, 8127 per km2 (+11.1%) 6542 per km2 (-3.2%) 5306 per km2 (-25.8%) 9111$ but Baku and Xi’an which saw larger 7566 $ (+705.5%) 11609 $ (+18.9%) (+725%) increases in density. Izmir and Algiers, instead, witnessed significant de- Population (200,000) Built-up area (20 km2) $ GDP per capita PPP 2011 USD Population density creases in population density which Source: GHS Urban Centre Database 2015 are expected for higher-income cit- ies as residents demand more space. SOURCE: GHS-UCDB 2015. In Algiers, in particular, while overall NOTE: ECA = Europe and Central Asia; GDP = gross domestic product; population density has fallen, density km2 = square kilometer; M = million. THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT? 149 over built-up area has increased since population growth (52.7 percent) has outpaced built-up area growth (19.6 percent) suggesting that unlike in Tashkent the increase in population has been met by increasing density in existing built-up area (for example, through high-rise buildings). The same can be said about Baku, Xi’an, and Jinan, despite the latter also witnessing an overall decrease in density. 150 THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT? Appendix F Analysis of the relationship of population growth to access to markets in the cities of Uzbekistan Access to markets usually predicts population growth in cities. The Urban Economics literature that fo- cuses on supply-side agglomeration suggests that cities are consumption centers, so that larger markets offering more consumption options should be relatively more attractive (Glaeser, Kolko, and Saiz 2001). These residents migrate to these cities, resulting in them growing faster than those that are more remote. There are different ways market access can be defined and measured. Market access can be distin- guished as external (proximity to large markets) and internal (size of the urban place’s own market). The analysis utilizes a commonly used measure of external market access based on a distance-weighted population density measure, 66 as in Henderson and Wang (2017), 67 which is also commonly known as a gravity equation (see Box F.1). The internal market is instead proxied by the logarithm of the urban place’s population. Table F.1 Market access and population growth Dependent variable: Population growth (1) (2) (3) (4) (5) (6) -4.39e-08* -0.000000179 External market access (0.000) (0.000) External -3.78e-08* -0.000000242 market access (regional) (0.000) (0.000) External -3.78e-08* -0.000000203 market access (50 km) (0.000) (0.000) -0.0413*** -1.072*** -0.0413*** -1.072*** -0.0413*** -1.072*** Internal market (0.007) (0.076) (0.007) (0.076) (0.007) (0.076) Region and Urban place Region and Urban place Region and Urban place Fixed effects year and year year and year year and year R^2 0.122 0.821 0.122 0.821 0.122 0.821 Observations 2,078 2,078 2,078 2,078 2,078 2,078 NOTE: * p < 0.10, ** p < 0.05, *** p < 0.01. Robust standard errors clustered at urban place level in parentheses. km = kilometer; OLS = ordinary least squares. 66  Population density–based measures are commonly used in the Urban Economics and Economic Geography literatures to proxy market size as they correlate with urban amenities. See Glaeser and Gottlieb (2006). 67  See Combes and Gobillon (2015) for an extensive review of empirical studies using measures of market access. THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT? 151 In Uzbekistan, such a relationship does not exist. Indeed, Table F.1 shows that the effect of market access on population growth is approximately 0, while the size of the internal market (city size) is a neg- ative predictor of population growth. Using different versions of the external market access measure, the results consistently show the absence of any effect of external market access on the population growth of Uzbek urban places, and that the results are mostly driven by the size of the internal market. The absence of a positive effect of the external market access on population growth is in line with what was observed in other ECA countries by Quintero and Restrepo (2019). However, their model shows that it is a feature of (mostly Eastern European) declining cities in countries with a shrinking population, while these trends are not present in Uzbekistan, where both total and urban population have been constantly growing. Instead, better performing Northern, Western, and Central European cities display a positive relationship between market access on one side and growth in GDP, population, and employment on the other (Adler et al. 2020). Smaller urban places grow faster than larger ones. The second and possibly most surprising result of the analysis is that there is a negative and statistically significant effect of the size of the internal market on population growth—meaning that smaller urban areas tend to grow faster on average. The result holds true for both urban-type settlements (UTSs) and cities (Table F.2). The results are also robust to excluding outliers from the sample. There are two potential interpretations of this puzzle. On the one hand, it is pos- sible that larger urban places are in fact relatively less attractive than smaller ones and a larger internal market in Uzbekistan is not a good proxy for higher service accessibility and more urban amenities. Ac- cordingly, it is also possible that while larger urban places, and cities above all, are indeed more attrac- tive, property prices do not reflect the relative value of cities across the country as in a Rosen-Roback framework (Roback 1982), but are even higher in larger cities so that despite offering higher amenities they are relatively less attractive due to the trade-off with housing costs. On the other hand, the results may just be pointing out the absence of migration (disused in section 2.2 of this report) and capture the differ- ent fertility rates between peri-urban areas and larger urban places. Table F.2 Market access and population growth decomposed by type of urban places Full sample Excluding outliers UTSs Cities UTSs Cities -0.000000168 0.00000173 -0.000000211 0.000000737 External market access (0.000) (0.000) (0.000) (0.000) -1.069*** -1.118*** -0.973*** -1.067*** Internal market (0.079) (0.061) (0.017) (0.045) Urban place and Urban place and Urban place and Urban place and Fixed effects year year year year R^2 0.849 0.439 0.857 0.914 Observations 1,840 238 1,706 230 NOTE: * p < 0.10, ** p < 0.05, *** p < 0.01. Robust standard errors clustered at urban place level in parentheses. UTS = urban-type settlement. 152 THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT? Box F.1 Measuring market access using different measures The external market access for urban place u is measured as , the sum of the ratios between the population of each urban place i and the geodetic distance between u and i. The basic formulation of the equation as used in Tables F.1, F.2, and F.3 assumes an exponent γ = 1 at the denominator. In Table F.1, two variations of the market access variable are used. The first one (columns 3 and 4) restricts places i to be in the same region as u, while the second one (columns 5 and 6) restricts the distance between u and i to be no larger than 50 kilometers (km), thus focusing only on market access in a 50 km radius around the urban place. Moreover γ can be increased or reduced to modify the relative weight assigned to places further away. Also, driving distance can be used instead of geodetic distance. Table F.3 employs dif- ferent strategies to measure external market access, namely: increasing or reducing the weight assigned to places further away (columns 1–3); computing market access to cities only (column 4); using driving distance in minutes from the Open-Source Routing Machine using OpenStreetMap data (column 5) (Huber and Rust 2016); and restricting to places in a one-hour driving distance radius around the urban place (column 6). The results provide further robustness to the estimates from Table F.1 results, showing that most of the variation in population growth is explained by the size of the internal market, while the effect of the size of the external market is negligible. Table F.3 Alternative measures of market access (1) (2) (3) (4) (5) (6) γ = 0.3 γ = 1.5 γ=2 City Driv Dist 1 hour -2.24e-08 -3.36e-10 4.82e-11 -0.00000118 -0.0000470 -0.0000501 External market access (0.000) (0.000) (0.000) (0.000) (0.000) (0.000) Internal market -1.072*** -1.072*** -1.072*** -1.072*** -1.071*** -1.072*** (0.076) (0.076) (0.076) (0.077) (0.077) (0.076) Urban place Urban place Urban place Urban place Urban place Urban place Fixed effects and year and year and year and year and year and year R^2 0.821 0.821 0.821 0.821 0.821 0.821 Observations 2,078 2,078 2,078 2,078 2,078 2,078 NOTE: * p < 0.10, ** p < 0.05, *** p < 0.01. Robust standard errors clustered at urban place level in parentheses. Internal and external market size do not matter even when accounting for administrative boundaries’ changes. One of the limitations of the data used for the analysis is that they fail to capture the effect of redistricting and changing administrative boundaries. For example, as part of a UTS is phagocytized by an expanding larger city or urban place we will see the population of the former drastically decline, while THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT? 153 the second will receive an immediate population boost. While this phenomenon is negligible in most countries, it is particularly relevant in Uzbekistan where the government has been very active in adjust- ing boundaries of urban settlements in recent years. Accordingly, the result of a negative not statistically significant effect of the external market access could be driven by the fact that UTSs that are closer to larger cities (and hence larger external markets) are more prone to have areas (and population) carved out to be added to the expanding city. A first attempt to control for this is to exclude all outliers from the sample as shown in Table F.4. 68 A more sophisticated approach (Table F.3, where columns 1–3 focus on UTSs only and columns 4–6 on the whole sample) consists in controlling for the growth rate of nearby urban places as abnormal growth could signal the existence of a government intervention, as well as looking at the proximity to cities with a focus on those that experienced abnormal growth. The regres- sion coefficients are very stable across models and suggest that administrative boundaries’ changes are not driving the market access results, but only the size of the internal market. Finally, a diametrically different strategy consists in eschewing administrative data and focusing on a different proxy for market size as discussed in Box F.2. Table F.4 Nearby population changes and population growth in Uzbekistan’s UTSs UTSs only Cities and UTSs (1) (2) (3) (4) (5) (6) 0.000501 -0.0167 -0.000117 -0.0337 Nearest city outlier=1 (0.010) (0.054) (0.009) (0.053) Nearest city out- 0.00585 0.0105 lier=1 # Distance nearest city (0.022) (0.021) -2.857 -3.619 -7.749** -9.404** Growth rate nearby (2.536) (2.765) (3.362) (3.806) -0.000000168 -0.000000147 -0.000000145 -0.000000179 -0.000000127 -0.000000124 External market access (0.000) (0.000) (0.000) (0.000) (0.000) (0.000) -1.069*** -1.068*** -1.067*** -1.072*** -1.068*** -1.066*** Internal market (0.079) (0.079) (0.079) (0.077) (0.076) (0.077) Urban place Urban place Urban place Urban place Urban place Urban place Fixed effects and year and year and year and year and year and year R^2 0.849 0.849 0.849 0.821 0.822 0.822 Observations 1,840 1,840 1,840 2,078 2,078 2,078 NOTE: * p < 0.10, ** p < 0.05, *** p < 0.01. Robust standard errors clustered at UTS level in parentheses. UTS = urban-type settlement. 68  The urban places defined as outliers are those that in a given time period have a population growth rate that is larger (smaller) than 1.5 interquartile ranges than the third (first) quartile of the population growth rates distribution in the same time period. 154 THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT? Box F.2 Overcoming data limitations: Market access using night-time lights It is also possible that, regardless of changing administrative boundaries, administrative popu- lation data do not provide correct estimates due to the phenomenon of dwellers that are una- ble to obtain the propiska but nonetheless choose to live in cities bearing the costs and risks of being unregistered, as documented by Seitz (2020). Night-time lights provide a solution to this potential issue that has a double advantage. On the one hand, night-time lights capture the real extent of economic activity regardless of what is actually recorded; on the other, they represent a better measure of market access as they perform well as a proxy for economic activity (Hen- derson, Storeygard, and Weil 2012).69 Moreover, these data are used in combination with urban agglomerations with populations larger than 5,000 computed from GHS-POP, using the degree of urbanization definition developed by the European Commission that allows to capture urban agglomerations that are not officially recorded as urban areas. The results provide a similar pic- ture and are robust to using different modelling techniques: exploiting the yearly variation pro- vided by the night-time light series 2013–19 (columns 1–2 in Table F.5) or focusing on the change from the starting year to the final one (column 3).70 These results point to the absence of the effect of market access size to be the correct interpretation rather than administrative population data failing to capture the effective size of local markets. Table F.5 Market access and population growth using night-time lights Panel data Overall change (1) (2) (3) -0.163*** 0.329 -0.355 External market access (0.047) (0.242) (0.256) -0.0488*** -0.758*** -0.178*** Internal market (0.006) (0.063) (0.030) Fixed effects Region and year Urban place and year Region and year R^2 0.218 0.420 0.314 Observations 2,088 2,094 348 NOTE: * p < 0.10, ** p < 0.05, *** p < 0.01. Robust standard errors clustered at urban agglomeration level in parentheses. 69  Therefore, night-time lights since the seminal paper by Henderson, Storeygard, and Weil (2011) have been widely used in Economics to measure economic growth for countries with poor national income accounts and at the subnational level (including city level) where a breakdown of these statistics does not exist. 70  Addison and Stewart (2015) point out that while the most commonly used night-time lights data, the Defense Meteorological Satellite Program (DMSP), perform well in capturing economic activity in a static framework, they are not as reliable when looking at changes over time. One of the reasons for that is that DMSP sensors become saturated and data in the brightest cities are top-coded (meaning that there would be no difference between the brightest areas in New York City, London, and Tashkent). For this reason, this work utilizes the newer Visible Infrared Imaging Radiometer Suite, which covers a shorter time span (2013–19) but offers higher resolution and accuracy. THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT? 155 Appendix G Analysis of historic locations of human settlements in Uzbekistan The geography of human settlements in Uzbekistan is a reason to believe in the potential of cities as development drivers. Unlike in other countries in ECA, the location of Uzbek cities does not seem to have been excessively distorted by Soviet planning as it mainly depends on natural fundamentals. As a result, cities are in the correct locations and the inefficiencies in the urban system can be mainly ascribed to mi- gration restrictions and housing market conditions which can be corrected through policy. This observa- tion paired with the realization that reforms in these areas are forthcoming brings to a positive conclusion that steering the course of urbanization in Uzbekistan toward productivity, livability, and sustainability is well within the government’s possibilities. The historical determinants of the location of Uzbek cities shed a light on the factors that affected the geographical patterns of human settlements in the country. While many factors affect the decisions of humans to move across space, a common trend across the world and epochs has been for many to crowd in small extents of space that are commonly known as cities (Glaeser 2008). This process dramatically accelerated with the Industrial Revolution as cities became centers of industrial specialization (Krugman 1991a). Nowadays, urbanization is a force to be reckoned with, above all, in the developing world where it happens largely uncontrolled (Bryan, Glaeser, and Tsivanidis 2020). Nonetheless, much of this urbaniza- tion happens in sites that have seen constant human presence long before any industrialization. This is particularly relevant in a country like Uzbekistan, host of one of the oldest civilizations in the world, whose cities are among the first to be historically recorded. Samarkand, for example, is thought to date to the eighth/seventh century BC,71 and other cities such as Bukhara, Khiva, and Shahrisabz share an equally continued history and are all recognized as World Heritage Sites by the United Nations Educational, Sci- entific and Cultural Organization.72 Therefore, any discussion on the urban system of Uzbekistan cannot ignore the historical roots of cities in the country and leads with an investigation on the original determi- nants of city location. The location of cities is path dependent. A large literature in Urban Economics has identified the key role of natural determinants in the presence of cities and economic activity. While the mechanisms vary case by case, cities tend to persist in their original sites well after the original location advantage has disappeared (Bleakley and Lin 2015). Cities may also rise and persist in locations without strong funda- mentals, as in the case of the manufacturing belt in the United States (Krugman 1991b), or be the result of government planning, as the monoindustrial towns in the Soviet Union, or planned capital cities such as Canberra, Ottawa, Washington, DC, and, more recently, Brasilia and Nursultan. Nonetheless, most cities will be found in the presence of strong location fundamentals (favorable conditions for human settle- ment), that can take the form of agricultural suitability of the surrounding areas (Olsson and Hibbs 2005), or accessibility to trade routes (Bleakley and Lin 2012; Hein and Schubert 2021). 71  https://whc.unesco.org/en/list/603/ 72  https://en.unesco.org/countries/uzbekistan. THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT? 157 Agricultural fundamentals are among the main determinants of cities’ location in Uzbekistan setting it apart from the rest of Central Asia. To investigate the role of natural fundamentals in affecting the Uzbek system of cities by determining cities’ location, this work follows the approach of other flagship World Bank reports (Ferreyra and Roberts 2018) to replicate the seminal paper by Henderson et al. (2018) who exploit night­light emissions to trace human presence consistently across the whole globe, overcoming the limitations given by the lack of reliable administrative data for many countries in the world. Table G.1 repli- cates the model in table II of Henderson et al. (2018) using country fixed effects looking at how much of the variation in human activity—proxied by night-time lights radiance—is explained by different sets of natural fundamentals.73 While column (a) reports the worldwide estimates seen in Henderson et al. (2018), columns (b) and (c) benchmark Central Asia74 and Uzbekistan against the rest of the world. It is evident that most of the variation in human activity in Uzbekistan is explained by agricultural variables, with a lower influence of base and trade covariates than in the rest of the world. Moreover, the high adjusted R-squared (0.613) indicates that natural fundamentals have a strong predictive power in Uzbekistan, suggesting that the location of human settlements is largely path dependent. The same pattern is not verified in the rest of Central Asia, where the model has poorer predictive pow- er. It is possible that in these countries (Kazakhstan in particular could be driving the results as it covers a significantly larger land area), the role of other factors, such as Soviet planning and state-led urban development more recently (as in the case of Nursultan—the capital of Kazakhstan that was only a small town in Soviet times) was stronger in shaping the urban system. This is an important finding highlighting that Uzbek cities tend to be in locations determined by natural factors similar to most countries globally, and that the influence of the planned economy on human settlements was not as strong as in neighbor- ing former Soviet republics. This is a positive result that offers hope that the existing urban system could provide a good foundation for productive and sustainable urbanization in the future. Table G.1 Adjusted R-squared results for relationship between log night-time lights and base, agriculture, and trade fundamentals (a) (b) (c) All countries Central Asia Uzbekistan (1) Base + FE 0.355 0.060 0.061 (2) Base + Agriculture + FE 0.566 0.221 0.595 (3) Base + Trade + FE 0.369 0.079 0.209 (4) All variables 0.576 0.233 0.613 NOTE: FE = fixed effects. 73  The explanatory variables are divided into three groups: base (ruggedness and malaria incidence), agricultural (biomes, average temperature, average precipitation, length of growing day, land suitability, latitude, and elevation), and trade (being on the coast, distance from the coast, being in the proximity of a harbor, river, and lake). For the replication exercise, the original data from Henderson et al. (2018) are used. The unit of observation is cells of a ¼-degree, covering approximately 770 square kilometers (km2) each for a sample size of 242,184 land cells of which 729 are in Uzbekistan. See Henderson et al. (2018) for a detailed explanation of the variables used and methodology. 74  Central Asia excludes Uzbekistan, so it is comprised of Kazakhstan, Kyrgyzstan, Tajikistan, and Turkmenistan. 158 THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT? Agricultural suitability variables have the most impact in defining the location of human settlements in Uzbekistan. Decomposing the individual variables’ contributions to the overall R-squared as in Hen- derson et al. (2018),75 the length of the growing period, and the agricultural suitability of the land are the variables that provide the most substantial contributions to the location of cities, followed by latitude (see table G.2). Table G.3 shows instead the coefficients measuring the effect of the covariates in the same model, which also gives an idea of how the agricultural variables affect the location of human ac- tivity and, hence, cities. Land suitability, latitude, and elevation all have a statistically significant effect at the 1 percent significance level and the sign is what is expected: land suitability positively predicts economic activity, while higher and northern (higher latitude) areas have lower population. The same is true for the length of the growing period which also provides the highest contribution to the model, although the effect is not statistically significant as the associated standard error is high, possibly due to the effect of other confounding variables.76 Interestingly, malaria transmission seems to also be a factor in the spatial distribution of human activity in Uzbekistan.77 As expected, trade variables add little explanatory power to the model.78 However, the results also show a significant positive effect of the distance to the coast. This may seem odd as proximity to the coast represents a favorable position for trade.79 However, as Uzbekistan is landlocked it is likely to be capturing the effect of an omitted trade variable, as discussed in Table G.2. Table G.2 Variables’ contribution (Shapley values) (a) (b) Central Asia Uzbekistan Rugged 0.002 0.014 Malaria 0.012 0.050 Temperature 0.007 0.044 Precipitation 0.031 0.045 Growing period 0.074 0.123 Land suitability 0.010 0.120 Latitude 0.007 0.106 Elevation 0.005 0.025 Distance to coast 0.003 0.016 Within 25 km from lake 0.005 0.006 Biomes 0.058 0.069 NOTE: km = kilometer. 75  This is done computing Shapley values following the approach developed by Shorrocks (2013). Shapley values represent the average marginal contributions of individual regressors to the overall R-squared. 76  When introducing the distance to the Silk Road (see Box G.1) in the model (columns 3 and 4), the positive coefficient for land suitability also becomes not statistically significant, suggesting that the effects of these three variables are correlated across the country and within regions. 77  This is not fully surprising as despite malaria being first eradicated in 1961, it made its way back from Afghanistan and Tajikistan, thus suggesting that historically it could easily be a factor affecting the location where people settled in the region. It was eradicated again in 2010, and in 2018 the World Health Organization officially certified Uzbekistan as malaria free (World Health Organization 2019). 78  Not all trade covariates are included because of collinearity when restricting to cells in Central Asia and Uzbekistan. 79  The distance to the coast variable has indeed had a negative and significant effect at the global level—see table I in Henderson et al. (2018). THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT? 159 Natural determinants also explain the location of economic activity within regions, not only across the country. The location of cities and economic activity in Uzbekistan is undoubtedly related to the natural advantages in terms of agriculture offered by their sites. However, it is still unclear whether the model presented is only capturing the differences between regions in the country. Indeed, as mentioned above, the north is mostly deserted while the Ferghana valley in the southeast is one of the most fertile areas in the whole region. Column (4) of Table G.3 investigates this by adding regional fixed effects to the regression in column (3). Most coefficients are consistent across models, suggesting that natural fundamentals affected the historical distribution of population within regions as well. Table G.3 Natural determinants of economic activity (1) (2) (3) (4) -60.19 *** -61.67 *** -54.55 *** -43.85*** Malaria transmission (14.278) (14.634) (12.608) (11.259) 0.0161 0.00190 0.0182 0.0222 Length of growing period, days (0.018) (0.017) (0.015) (0.016) 2.913 *** 2.338 ** 1.494 1.133 Land suitability (1.092) (1.126) (1.062) (0.788) -0.909 *** -0.759 *** -0.497 *** -0.664*** Latitude (0.198) (0.210) (0.181) (0.159) -3.147 *** -3.380*** -2.374*** -1.983*** Elevation, km (0.743) (0.721) (0.663) (0.720) 3.036*** -0.485 1.114 -4.127 Distance to nearest coast, 000s km (0.821) (2.157) (2.071) (3.769) -0.262 ** -0.215* -0.363 Longitude (0.122) (0.116) (0.264) -0.0107 *** -0.00903*** Distance to Silk Road, km (0.002) (0.002) Regional FE ✓ R2 0.619 0.624 0.659 0.708 Observations 729 729 729 729 NOTE: FE = fixed effects; km = kilometer. Box G.1 The impact of the Silk Road on the location of cities in Uzbekistan While one of the reasons for the concentration of the population in the Tashkent and Fergana re- gions in the southeast is certainly the more favorable conditions in terms of agricultural suitability compared to the largely desertic northwest, columns (1) and (2) of Table G.3 show that variables 160 THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT? capturing the regional position in the country such as distance to coast and longitude have nonetheless a significant effect. This points to another location fundamental that is correlated to these variables and whose effect is not fully captured by the agricultural covariates. Column (3) introduces the distance to the ancient track of the Silk Road, a collection of arteries, some of which date back to the second century BC, connecting South and Central Asia to the Middle East and Europe, which passed through Uzbekistan and had one of its most prominent hubs in Samar- kand (Hansen 2012). The regression coefficient shows that areas of the country further away from the Silk Road have lower economic activity, implying that cities are concentrated on the ancient commercial track. This can be also seen in Figure G.1. Column (4) reveals that the effect exists not only across the country but also within regions. While it is likely that cities closer to the Silk Road have benefitted from it, the results cannot be interpreted as the causal effect of proximity to the early international trade route in the absence of a natural experiment as the track of the ancient road arteries may be endogenous and depend on the site of the most prominent cities at the time, whose location could have been influenced by other natural fundamentals such as agricultural characteristics. Even if distinguishing between the two mechanisms is not possible, the evidence points out to the role of the Silk Road in generating path dependence in Uzbekistan and suggests that participation in international trade shaped the urban system of the country to some extent, despite it being land locked and not being able to engage in maritime trade. Figure G.1 Map of the historic tract of the Silk Road, and current night-time lights emissions in Uzbekistan SOURCE: Original elaboration using data from the Harvard Centre for Geographic analysis and the Earth Observation Group (EOG). THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT? 161 Appendix H Analysis of the link between urban shape and productivity Improving urban form presents an opportunity for reaching higher productivity for Uzbek cities. Fig- ure H.1 reports the coefficient of regressing productivity80 on four urban shape metrics (circuity, fullness, street density, and sprawl) for cities with a population of over 50,000 in Uzbekistan and ECA, as well as for only the largest cities in ECA (over 500,000). All coefficients show the expected sign when significant except sprawl, which is associated with higher productivity in the first two regressions. However, con- sistent with Harari (2020), larger and often more productive cities tend to be more sprawling so that the positive relationship between sprawl and productivity may be capturing differences in city sizes. Indeed, when restricting the sample to the 85 cities in the region with populations of over 500,000, the coeffi- cient for sprawl is negative and statistically significant. For all urban shapes metrics, the relationship with productivity is stronger in Uzbekistan than in the rest of ECA. Despite this, these findings cannot be interpreted as causal evidence as city shapes may be correlated with other factors affecting productivity, since cities in Uzbekistan lag behind comparators in many of the introduced urban form metrics, this points out to a large untapped potential in terms of productivity from improving spatial structure of cities. Figure H.1 Results of regression analysis of effects of various urban shape and structure indicators on productivity or urban areas in Uzbekistan and the ECA region p = 0.142 p = 0.975 Uzbekistan >50k Circuity ECA >50k p = 0.000 ECA >500k (N=85) p = 0.000 Fullness p = 0.768 p = 0.036 Log p = 0.139 street dens p = 0.092 p = 0.000 p = 0.000 Sprawl p = 0.015 -5 0 5 10 SOURCE: Authors’ elaboration using GHS-POP 2015 data based on EC definition of urban clusters. NOTE: ECA = Europe and Central Asia. 80  Once again measured as the residual from regressing the natural logarithm of the sum of night-time lights radiance in the city area over the logarithm of city population. THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT? 163 Appendix I Urban expansion forecast methodology Urban expansion forecasting was done using a cellular automata (CA) model. CA models are rule based, highly adaptable, and have been widely used for urban growth modelling (Tripathy and Kumar 2019). CA models consist of a grid space, and a set of states which characterize the grid cells. There is a definition of a neighborhood of cells and a set of transition rules to determine the state of each cell depending on the state of its neighbors. Finally, there is a sequence of time steps where each cell gets updated (White and Engelen 2000). Urban expansion forecasts were applied independently on several cities in Uzbekistan using the same model. For each city, a grid covering the built-up area as of year 2015 with a 1.5 km buffer was used. Each cell was 30 m resolution. The model consisted of a sequence of discrete time steps, where each time step represented a year, with all cells updated simultaneously. Each cell has a state of being either built or unbuilt. Using a logistic regression technique, a transition probability matrix was produced. Each cell was populated with its built-up state using World Settlement Footprint (WSF) Evolution data (Marconcini et al. 2018). WSF Evolution maps settlement growth from 1985 to 2015. A 5 X 5 neighborhood was used and for each cell the number of neighboring cells that were already built were counted. Also, the number of neighboring cells that were built up within the previous five years were counted. Using these variables, the logistic regression model was fitted, and the predicted probabilities were calculated to determine how likely a cell was to transition to being built. Using these probabilities, a nonuniform random choice was applied to determine the state of each cell. The transition probability matrix and all cells would get updated for each time step. In addition to using predicted probabilities to determine transition states, two restriction layers were used. The European Space Agency’s Climate Change Initiative (ESA CCI) landcover dataset81 was used to restrict growth in areas that contained water and OpenStreetMap data was used to restrict growth where various types of infrastructure existed including highways, rivers, canals railways, parks, ceme- teries, and airports. As cells in the CA model transitioned from unbuilt to built, the ESA CCI landcover dataset was used to calculate the amount of arable land projected to be lost. The model was run from 2015 to 2050. 81  https://www.esa-landcover-cci.org/ THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT? 165 Appendix J Analysis of specialization and productivity of cities in Uzbekistan A brief analysis of specialization of urban economies reveals that contrary to anecdotal evidence, ex- tractive areas are not among the most productive and do not generate more output per capita. Uz- bekistan’s industrial production is reliant on the extractive industry, which accounts for over 10 percent of industrial output in the country. This leads to a hypothesis that (like in other post-Soviet countries including Russia and Kazakhstan) locations specializing in extractive industries should be the leaders of productivity. However, upon scrutinizing industrial data from Uzbek districts (tumani in Uzbek)82 it is evident how districts that are specialized in extractive industries83 are not more productive than the others (Figure J.1) and are not associated with higher output per capita (Figure J.2). While output per capita does increase with the local share of extractive industries for high levels of specialization, the districts with the highest per capita production in the country do not specialize in extractives. Figure J.1 Extractive specialization and output per capita Log industrial production per capita (UZS 000's) Local share of extractive industry SOURCE: Authors’ elaboration based on GHS-POP 2015, Yandex Maps, and data from the Ministry of Economic Development and Poverty Reduction, EOG VIIRS night-time lights 2013–19. UZS = Uzbek soms. 82  The choice of districts as the unit of analysis when looking at the regional industrial composition is constrained by the fact that production data is not available at a more granular level. 83  Specialization is defined as either local specialization (the highest contribution in percentage to the overall districts’ output in UBZ billions) or national specialization, using a location quotient: the highest ratio between the local specialization share and the contribution share of the specific industry at national level. THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT? 167 Figure J.2 Extractive specialization and productivity of local economies Specialized extractive (local) Specialized extractive (national) Other specialization Productivity (residuals) Log Population SOURCE: Authors’ elaboration based on GHS-POP 2015, Yandex Maps, and data from the Ministry of Economic Development and Poverty Reduction, EOG VIIRS night-time lights 2013–19. The relationship between specialization and productivity in districts of Uzbekistan is weak, but dis- tinction between cities and the rest indicates the possible emergence of agglomeration economies. Figure J.3 plots the local specialization of Uzbek districts against productivity (measured as the residual of regressing the logarithm of the sum of night-time lights in the district on logarithm of its population). There is no apparent relationship between increased specialization and productivity, with areas heavily specialized in oil processing, coal, metallurgy, and utilities among the least productive. The results are driven by a negative relationship between specialization and productivity in city-districts (called shahari in Uzbek) which is less surprising as production tends to be more diversified in cities, and productivity benefits tend to be extracted from interindustry spillovers (Figure J.4). For other districts (that include ru- ral areas, smaller cities, and properties of the agglomerations) there is a mild positive relationship. While these initial results should not be overinterpreted, they suggest that larger cities tend to benefit from having more industrial diversity (consistent with Jacobian interindustry agglomeration economies), and other areas tend to benefit from specialization (consistent with Marshallian intraindustry agglomeration economies). This is consistent with urban economic theory. While these effects are weak, they confirm that despite all of the disruptions to urban economies described in this report, agglomeration econo- mies exist and have an effect in Uzbekistan, and thus further liberalization of factor markets and removal of urban development bottlenecks will unlock further productivity growth in cities. 168 THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT? Figure J.3 Districts’ specialization and productivity Chemical Food, drink, tobacco Oil processing, coal, Productivity (residuals) metalurgy Trade Extractive Machines, equipment, transport Pharma Transport Finance Final products from metal, wood, rubber Textile Utilities Local share of highest specialization industry Figure J.4 Specialization and productivity by district type District City Chemical Food, drink, tobacco Productivity (residuals) Oil processing, coal, metalurgy Trade Extractive Machines, equipment, transport Pharma Transport Finance Final products from metal, wood, rubber Textile Utilities Local share of highest specialization industry SOURCE: Authors’ elaboration based on GHS-POP 2015, Yandex Maps, and data from the Ministry of Economic Development and Poverty Reduction. THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT? 169 Appendix K Draft outline of a comprehensive land reform roadmap for Uzbekistan (to be discussed and revised together with the government Working Group, initial draft as of November 30, 2021) INSTITUTION # PILLAR ISSUES TO BE COVERED IN CHARGE   Mapping and functional analysis of land administration institutions; harmonization of functions to ensure effective and efficient institutional structure   Update and input accurate cadastral and legal rights data to the Integrated Information System for Real Property Registration and Cadastre Modernization of real   Make sure all land and properties are registered on property administration one platform (urban, agriculture, forest, nature reserve, system and institution Cadastre private, state land, leases, etc.) 1 and completion of Agency registration of national   Improve land services (registration of transactions cadastre including transfer, mortgages, subdivision, etc.) as part of the smart public services and e-government initiatives   Effectively implement open coordinate system for use for all mapping functions, including land administration   Developing and implementing the mechanism of establishing and registration of servitudes and encumbrances of rights to the land plots   Introduction of private ownership and tradable lease rights MoEDPR   Development and approval of policies to support 2 Agriculture land reform and Ministry agriculture land markets of Agriculture   Ensure access to credit for small farmers   Advocacy and awareness of the reform policies   Development of legal and regulatory framework for spatial planning, as well as its execution and monitoring MoEDPR, Nonagriculture land   Update/develop master/land use/zoning plans for key 3 Ministry of reform areas of the country Construction   Allow privatization of land under buildings and improve regulations to streamline processes THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT? 171 INSTITUTION # PILLAR ISSUES TO BE COVERED IN CHARGE   Establishment of legal and regulatory framework for land and property valuation in line with international valuation standards based on market values   Capacity development of land and property appraisal institutions; provide training for potential property valuers/ appraisers   Undertake mass valuation for key areas Property valuation and Ministry 4   Development of fair and equitable land and property taxation of Finance taxation with a goal to correct market distortions and discourage holding of unused land and properties   Development of an effective administration and enforcement mechanisms for land and property taxation   Interoperability of tax and land administration systems as well as other systems.   Streamlining tax appeal system   Complete full inventory of all public land and properties (all land and properties that are not private)   Demarcation and protection of forest, water, and environmental and culturally sensitive areas   Development of transparent and effective policies on public land alienation mechanisms and on valuation and pricing of public property, i.e., for investment, public- State Property private partnership projects, distribution for poor farmers, 5 Public land management to support social housing, etc. Agency   Division of state property into republican and municipal   Demarcation and protection of infrastructure objects (pipelines, roads, railways, communications, etc.) and registration of protection zones as encumbrances of rights to the land plots   Development of policies on intensification/effective use of land and property utilized by government entities   Analysis and development of laws, regulations, spatial standards, and institutional coordination to support the National Spatial Data Infrastructure Land information system,   Development of policies that regulate access to land access to information and MoEDPR, and property data, including cadastral and ownership 6 property data, openness Cadastre data, sharing of land and property information among of the data government agencies, private sector, and citizens with consideration of privacy and liability issues   Interoperability of land administration data with different public registries, civil registry, animal registry, etc. NOTE: MoEDPR = Ministry of Economic Development and Poverty Reduction. 172 THE TIME IS NOW. HOW CAN UZBEKISTAN LEVERAGE URBANIZATION AS A DRIVER OF SUSTAINABLE DEVELOPMENT?