Policy Research Working Paper 11247 When Does a Village Become a Town? Revisiting Pakistan’s Urbanization Using Satellite Data Oscar Barriga Cabanillas Marziya Farooq Moritz Meyer Christina Wieser Poverty and Equity Global Department Office of the Chief Economist October 2025 Policy Research Working Paper 11247 Abstract This study revisits Pakistan’s level of urbanization using sat- by the official classifications. The discrepancy between func- ellite imagery and the Degree of Urbanization methodology. tional and administrative classifications of urban areas has While official statistics report that 39 percent of the popula- important fiscal and planning implications. Misclassified tion resides in urban areas, this analysis reveals that the true areas reduce property tax revenues and undermine the figure is closer to 88 percent. The substantial discrepancy planning and provision of critical public services. Moreover, arises from Pakistan’s reliance on administrative boundaries misclassification distorts spatial socioeconomic indicators, that do not reflect actual population density or settlement masking the true extent of urban-rural disparities and patterns. The findings indicate that secondary cities and complicating the design of effective, evidence-based public peri-urban areas—not megacities—are the primary drivers policy. of recent urban expansion and are systematically overlooked This paper is a product of the Office of the Poverty and Equity Global Department. It is part of a larger effort by the World Bank to provide open access to its research and make a contribution to development policy discussions around the world. Policy Research Working Papers are also posted on the Web at http://www.worldbank.org/prwp. The authors may be contacted at obarriga@worldbank.org. The Policy Research Working Paper Series disseminates the findings of work in progress to encourage the exchange of ideas about development issues. An objective of the series is to get the findings out quickly, even if the presentations are less than fully polished. The papers carry the names of the authors and should be cited accordingly. The findings, interpretations, and conclusions expressed in this paper are entirely those of the authors. They do not necessarily represent the views of the International Bank for Reconstruction and Development/World Bank and its affiliated organizations, or those of the Executive Directors of the World Bank or the governments they represent. Produced by the Research Support Team When Does a Village Become a Town? Revisiting Pakistan’s Urbanization Using Satellite Data1 Oscar Barriga Cabanillas Marziya Farooq Moritz Meyer Christina Wieser Authorized for distribution by Salman Zaidi, Practice Manager, Poverty and Equity Global Department, World Bank Group Keywords: Urbanization, Pakistan, land use. JEL codes: P25, R14. 1 This paper is a product of the Poverty and Equity Global Practice. It is part of a larger effort by the World Bank to provide open access to its research and make a contribution to development policy discussions around the world. Policy Research Working Papers are also posted on the Web at http://www.worldbank.org/prwp. Oscar Barriga-Cabanillas (corresponding author) may be contacted at obarriga@worldbank.org. The authors want to thank Zishan Karim, Shahnaz Arshad, Gailius J. Draugelis, and Vasudha Thawakar for their insightful comments, and to Lydia Teinfalt for her research assistance on an initial draft and data compilation. We declare that we have no relevant or material financial interests that relate to the research described in this paper. The findings, interpretations, and conclusions expressed in this work do not necessarily reflect the views of The World Bank Group or any affiliated organizations, 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. Introduction Pakistan’s official classification of what areas are urban is determined by municipal committees. These have the authority to designate areas as urban; however, this decision is not based on objective measures like population density, service access levels, or economic function. The last use of objective metrics to define urban centers was in 1972. Since then, the rapid population growth and spatial concentration have significantly transformed the country's urban landscape compared to 50 years ago. Urban classifications shape fiscal policy, planning, and statistical reporting, yet Pakistan’s definitions have remained unchanged since 1972. We leverage satellite imagery to revisit the distribution of the country’s population between urban and rural areas. For this, we use the Degree of Urbanization (DoU) methodology, which uses remotely sensed data on population count and density to provide a high- resolution view of current urbanization levels. A key advantage of our approach is that it is collected and calculated independently of pre-defined administrative boundaries, providing a more nuanced view of the urban landscape. 2 Our results show that, while official statistics indicate that 39 percent of Pakistan’s population is urban, when considering population density and concentration measures, this figure rises to 88 percent. This occurs as official classification captures the largest urban centers but fails to recognize the urban character of growing peri-urban areas that have emerged from rural areas and now host 42 percent of the population. Additionally, we discuss how mirroring the urban expansion of Pakistan with the development of “megacities” overlooks the significant urbanization occurring in growing peri-urban areas. While the expansion of cities like Lahore and Karachi is a notable indicator of the country’s increasing urbanization, a large part of the urban population growth occurs in many intermediate and small urban centers. Classifying urban areas as rural has practical policy implications. In particular, it decreases the capacity to deliver services, collect taxes, and have statistical visibility of urban populations. Since property taxes are only levied for urban buildings, this misclassification reduces the financial resources available to meet the needs of an expanding urban population, thereby limiting the potential of urban centers to act as engines of growth. This implies that the classification of an area as urban is not solely a technical decision, but it is also shaped by political economy considerations. Additionally, as urban populations generally exhibit better socioeconomic indicators than rural ones, misclassifying a significant portion of urban households as rural artificially inflates rural living standards statistics and complicates the understanding of price differentials between rural and urban areas. The paper is structured as follows. Section 1 presents the different approaches available to classify an area as urban and introduces the DoU methodology. Section 2 provides the results of Pakistan’s urbanization rate, and its evolution. Section 3 concludes with a discussion of the policy implications of the misclassification of urban areas. 2 This paper provides foundational data to understand urbanization levels in Pakistan and explores the benefits of redefining the urban-rural classification. A public GitHub repository (search for “worldbank/DoU_Urban” on github.com) complements this note and provides access to the results, along with a fully reproducible package that can be applied in other countries. 2 Identifying urban areas in Pakistan While the classification of urban areas varies across the world, few countries rely solely on administrative definitions. There is large heterogeneity in how countries classify an area as urban. However, three in four countries use a combination of population size, economic factors, and urban characteristics to identify urban areas. What is considered urban varies considerably across countries; however, there are four primary criteria to classify an area as urban, which include administrative boundaries, population size or density, economic characteristics, and urban characteristics. Table 1 provides a summary of how the countries use different criteria across the world. 3 Table 1: Different methods used to define “Urban” in different countries Source: United Nations Department of Economic and Social Affairs (2018). In this paper, we use the DoU methodology to calculate the share of urban and rural population in Pakistan. The DoU, unlike traditional administrative definitions, provides a standardized and globally comparable measure of urbanization. It uses two key metrics of population count and population density. The Global Human Settlement Layer (GHSL) provides the base layers for applying the DoU methodology to classify urban areas at a resolution of 1 square kilometer grid cell. Moreover, due to the flexibility and 3 In more detail: (i) Administrative Criteria: 121 of 233 countries use an administrative definition. Within this group, 59 countries rely solely on administrative designations, with Pakistan falling into this category. (ii) Population Size or Density: In 108 countries, population size or density determines whether an area is urban. For 37 countries, these demographic characteristics are the only criteria. However, thresholds on population span from as little as 200 to up to 50,000 inhabitants. (iii) Economic Characteristics: 38 countries include economic features in their criteria for defining urban areas. However, no country solely uses this to define urban areas. (iv) Urban Characteristics: 69 countries consider functional characteristics, such as paved roads, water supply, sewerage systems, and electric lighting. In eight cases, these are the sole criteria for classification. 3 global applicability of the DoU framework, the United Nations Statistical Commission has recommended it for cross-national urban comparisons (Dijkstra et al., 2021). 4 The main advantage of the DoU is that it relies on remote sensing data, making it independent of national census cycles and administrative boundaries, allowing for international comparisons. Additionally, it can be disaggregated at any level, independently of political and administrative boundaries, and provides a more nuanced view of rural and urban areas instead of a binary category. Moving away from binary classification or urban and rural recognizes that a significant share of households live in areas that lie somewhere between purely rural and high-density cities. Table 2 displays the names and technical terms for each DoU category. These categories rank, in ascending order of population density, from Mostly Uninhabited and Dispersed Rural Areas (codes 11 and 12) to Dense Towns and Cities (codes 23 and 30). The seven categories in the DoU can also be aggregated to three distinct levels of urbanization; Rural, Urban Cluster, or Urban Center (Code 3). Table 2: Degree of Urbanization categories of population centers Population (Lower Bound) Area type Category Topological constraints Density total Concentration Urban Center High-density continuous urban area, City 1,500 50,000 (High Density) cell has at least 50,000 inhabitants. Dense Town 1,500 5,000 Moderate-density areas Moderately dense, more than 3 km Urban Cluster Semi Dense 300 5,000 from a dense urban cluster or urban (Moderate Town center. Density) Suburban/Peri- Urban cluster within 2 km of a dense 300 5,000 Urban Areas urban cluster or urban center. Villages 300 500 Clusters of lower-density cells Rural Grid Dispersed Cell (Low 50 None None Rural Areas Density) Mostly None None Population > 0 and land >= 50% Uninhabited Source: European Commission Joint Research Centre (EU-JRC) and The Center for International Earth Science Information Network (CIESIN). 4 Remote sensing still relies on census data to calibrate population density. In Pakistan, it uses 1998 census data and 2010 population estimates. 4 Results Pakistan: An urban country The last time Pakistan’s official classification of urban areas reflected on-the-ground realities was in 1972. Since then, the country’s urban landscape has significantly changed. Before 1972, the classification of urban areas was based on on-the-ground realities such as population count and objective urban characteristics such as roads and public service provision. However, after 1972, the responsibility shifted to provincial and municipal committees, which have no formal obligation to consider population concentration or other objective metrics when delimiting urban areas, nor are they required to periodically revise urban boundaries. 5 The evolution of the classification of urban areas has responded to local political economy considerations where the interplay of on-the-ground dynamics, taxation pressures, and administrative boundaries have played a role. 6 Building on the preceding discussion of the limitations of the current urban-rural classification, we apply the DoU to quantify the extent of urbanization in Pakistan. Our analysis finds that Pakistan’s urban population is nearly twice the size reported in official statistics, with the administrative definition only accounting for the largest cities. Based on the DoU classification, 88 percent of the population lives in an area with urban characteristics, with 46 percent living in areas considered high-density cities and about 42 percent concentrating in urban centers of moderate density. These numbers contrast with the 39 percent of the urban population reported in official statistics, revealing a significant gap between classifications based on political and technical considerations (Figure 1). It is not possible to conduct a one-to-one comparison between the DoU categories and official areas classified as urban since highly spatially disaggregated census data is not available. However, robustness tests at the Tehsil level, the most disaggregated data available from official sources, show that the DoU overlaps with primary and secondary urban centers. Mirroring this result, official data overestimates the rural population by a factor of five. Statistics from the 2023 Housing and Population Census indicate that about 61 percent of the population is officially classified as rural. Yet, according to the DoU rural categories, dispersed areas and villages account for 8 and 4 percent of the population, respectively. The discrepancy with official rural figures arises because population centers categorized as dense towns (11 percent), semi-dense towns (2 percent), and suburban/peri-urban areas (29 percent) are considered rural in official statistics. In reality, these areas house a significant number of households, have high population density, and are integrated into economic urban networks. 5 Prior to 1972, areas were classified as urban if they were inhabited by a minimum of 5,000 people or if census commissioners determined that the area had “urban characteristics” such as roads, sanitation, schools, etc. The responsibility to classify an area as urban is shared between the provincial and municipal governments, with the former having the additional capacity to impose differential tax rates within urban areas. 6 In the 1981 census, urban areas were defined as metropolitan corporations, municipal corporations, municipal committees, or cantonments. This new definition reclassified 72 urban places from the 1972 census as rural, affecting 5.7 percent of the urban population. Additionally, 1,462 places with a population over 5,000 were also classified as rural. These changes reflect the political economy dynamics behind the classification of urban and rural areas. 5 Figure 1: Comparing urban population shares between the DoU and official estimates 100% City 80% 39 Dense Town Share of Population living in each DoU 46 Urban (PBS) Semi Dense 60% Town Rural (PBS) Suburban/Peri- 11 Urban 40% 2 Village 61 29 Dispersed Rural 20% Area 4 Mostly Uninhabited Area 8 0% 1 PBS (2023) GHSL (2020) Source: Authors’ calculations using GHSL data created through a partnership between the European Commission Joint Research Centre (EU-JRC) and The Center for International Earth Science Information Network (CIESIN) and PBS Census 2023. The DoU also facilitates cross-national comparisons, as it provides a consistent criteria. Application of the DoU reveals that, despite variation across urban typologies, the proportion of the population residing in urban areas exceeds 70 percent in all examined countries. Notably, Bangladesh and the Arab Republic of Egypt exhibit urban shares as high as 98 percent, while Mexico records a lower bound of 77 percent. These findings underscore the extent to which traditional administrative definitions may systematically underestimate urbanization levels relative to functionally derived metrics (Figure 2). 6 Figure 2 DoU across the World in 2020 (share of the population) 100% Share of Population living in each DoU 80% 39 48 55 54 67 64 60% 14 13 5 40% 13 6 16 2 11 2 25 8 7 19 1 20% 6 31 6 10 1 8 6 8 10 10 10 7 8 0% 2 2 1 3 2 Pakistan Bangladesh Brazil Egypt India Mexico Mostly Uninhabited Area Rural Dispersed Area Village Suburban/Peri Urban Area Semi Dense Town DenseTowns City Source: Authors’ calculations using GHSL data created through a partnership between the European Commission Joint Research Centre (EU-JRC) and The Center for International Earth Science Information Network (CIESIN) and PBS Census 2023. Across all provinces, the divergence between the official urban classification and the DoU persists, with official statistics consistently underrepresenting the extent of urbanization. In Balochistan, Punjab, and Sindh, the share of the urban population in official figures aligns with the DoU (categories Urban PBS and city in Figure 3). Interestingly, despite its clear urban character, the underrepresentation of urban areas is also evident in Islamabad, where 47 percent of the population is officially considered urban, a share that is almost half of the 90 percent estimated by the DoU. A similar situation occurs in KP, where the DoU classification for the population share living in a city is almost three times the 15 percent reported in official data. Additionally, the DoU reveals a varied urban landscape within provinces. While in Islamabad most of the urban areas are classified as a dense city, in other provinces, urban areas present a more mixed combination of suburban and peri-urban towns. 7 Figure 3: Comparing provincial urbanization rates between the DoU and official estimates 100% 15 31 32 80% 41 41 43 47 Share of Population living in each DoU 54 54 11 60% 5 89 17 12 10 85 2 7 40% 3 13 2 69 59 53 28 33 22 46 20% 24 4 2 3 3 7 6 7 9 0% 5 0 2 1 1 1 0 PBS DoU PBS DoU PBS DoU PBS DoU PBS DoU Balochistan Islamabad Khyber Punjab Sindh Pakhtunkhwa Mostly Uninhabited Area Dispersed Rural Area Village Suburban/Peri-Urban Semi Dense Town Dense Town Source: Authors’ calculations using GHSL data created through a partnership between the European Commission Joint Research Centre (EU-JRC) and The Center for International Earth Science Information Network (CIESIN) and PBS Census 2023. Pakistan’s population has more than doubled over the past forty years, accompanied by a steady migration out of rural areas—dynamics largely missed by official statistics. Since 1980 the population in the country has more than doubled, going from 78 million to 222 million people in 2020. This growing population has gravitated toward small and large urban centers due to economic dynamics where urban areas offer better labor opportunities and, in many cases, better access to services. Relying on official statistics, the share of people living in urban areas barely moved from 28 to 39 percent during the same period. Once more, the DoU estimates indicate not only a more pronounced increase in urban dwellers (from 61 to 88 percent) but also provide a more nuanced view. Most of this growth occurred in urban and peri-urban areas, which saw their share of the population increase from 9 to 29 percent (Figure 4). An additional consequence of Pakistan’s large population growth is that, despite the rural population share falling by two-thirds since 1980 accordingly to the DoU, rural areas still concentrate a similar number of people. While about 39 percent of the population lived in rural areas in 1980, this share fell to 12 percent by 2020. Yet, due to high birth rates, the number of people living in rural areas has only slightly decreased over the past forty years, from about 30 million to close to 27 million people. This steady number of people living in rural areas reflects a unique demographic phenomenon where rapid urbanization has not resulted in a complete exodus to urban centers, with rural areas still requiring significant public resource investments. 8 Figure 4: Evolution of population shares in each Degree of Urbanization 100% Share of Population living in each DoU 80% 35 37 43 46 46 46 60% 13 13 5 13 13 13 11 6 8 4 2 40% 3 3 15 17 18 21 24 29 12 20% 8 18 6 5 15 4 11 10 9 8 0% 4 2 2 1 1 1 1980 1990 2000 2010 2015 2020 Mostly Uninhabited Area Dispersed Rural Area Village Suburban/Peri-Urban Semi Dense Town Dense Town City Source: Authors’ calculations using GHSL data created through a partnership between the (EU-JRC) and The Center for International Earth Science Information Network (CIESIN). The urbanization process has transformed Pakistan’s landscape, with new urban areas emerging and well- established urban centers expanding into surrounding towns. Over the past 20 years, the landscape in Pakistan has significantly changed. Since the early 2000s, a rising share of the rural population has found work outside agriculture, leading to the transformation of areas previously considered rural into new urban centers. Meanwhile, well-established urban centers expanded and created a ring of suburban areas in their periphery. Table 3 presents the spatial evolution of Pakistan’s landscape, with the left column showing the DoU classification for each 1 square kilometer grid cell of land in the country in 2000, and the rows indicating its status in 2020. During this period, about 62 percent of the area in the rural dispersed areas category remained unchanged. However, a significant percentage transformed into villages (4.2 percent) and suburban and peri-urban districts (27 percent). The shift has been more dramatic for areas classified as villages in 2000, where half transformed into suburban and peri-urban areas, and for dense towns, where about 17 percent turned into new “secondary cities” (World Bank, 2024). 9 Table 3: Urbanizing space: Percentage of grids in each DoU category in 2000 and 2020 Status in the year 2020 Suburban, Mostly Rural Semi Dense Village Peri- City Total Uninhabited Dispersed Dense Town Urban Mostly 91.9 6.3 0.2 1.4 0.1 0.0 0.0 100 Uninhabited Status in the year 2000 Rural Dispersed 4.5 62.1 4.2 26.8 2.0 0.1 0.3 100 Village 0.0 5.8 32.0 49.6 11.1 1.5 0.0 100 Suburban and 0.0 0.9 0.1 87.1 0.3 5.9 5.7 100 Peri-Urban Semi Dense 0.0 3.5 1.3 69.9 20.6 4.6 0.0 100 Town Dense Town 0.0 0.0 0.0 3.9 0.4 78.4 17.2 100 City 0.0 0.0 0.0 0.4 0.0 0.3 99.3 100 Source: Authors’ calculations using GHSL data created through a partnership between the European Commission Joint Research Centre (EU-JRC) and The Center for International Earth Science Information Network (CIESIN) Note: Initial and final percent of DoU category for each of the 1 km2 grid cells in Pakistan between 2000 and 2020. An in-depth analysis of two urban areas shows the spatial expansion of urban centers in Pakistan is consistent with growing urban cores, which absorb nearby rural or semi-urban areas. This showcases how the urban expansion dynamics surpass administrative boundaries. The DoU approach provides a highly granular view of settlement evolution. Using Sialkot as an example, the DoU showcases that around 97 percent of the district’s population lives in areas with urban characteristics, and around 56 percent resides in areas classified as city. Over the past twenty years, the urban core has rapidly expanded outward into previously peri-urban and rural areas. Meanwhile, previously peripheral rural areas, even if not part of the city, have lost their rural character as their population density increased significantly. A similar situation occurred in Lahore, where around 87 percent of the district’s population lives in areas with urban characteristics and around 44 percent of the population resides in areas classified as city. In both cases this spatial expansion has occurred beyond the boundaries of administrative limits. For instance, Lahore and Sialkot urban areas include two and four Tehsils, respectively. 10 Figure 5: Spatial evolution of urban centers, the case of Sialkot and Lahore 2000 2020 Sialkot Lahore Source: Authors’ calculations using GHSL data created through a partnership between the (EU-JRC) and The Center for International Earth Science Information Network (CIESIN). The lack precise geocoded information in official household surveys makes it impossible to reconstruct welfare and socioeconomic characteristics at the different DoU levels. However, when using satellite- based measure of wealth, we find that urban areas show profound wealth disparities. To assess wealth, we use the high resolution of the RWI, 7 which serves as a proxy for living standards and illustrates that urban areas have higher living standards than rural ones (Figure 6), with cities in particular reporting higher wealth levels than any other urban clusters. However, for each DoU, there is substantial variation in living standards. For instance, even though cities provide better living standards than dense or semi- dense towns, many locations within cities report dire living conditions. Among urban areas, proximity to main urban centers seems to explain living standard differences. Semi-dense towns—often developing as endogenous towns and situated at least 3 kilometers from the urban centers—struggle to provide better living conditions for residents compared with semi-urban and peri-urban areas adjacent to cities. In the case of rural areas, their wealth levels vary very little even as they grow from very low-density areas into villages. 7It was developed by META Data for Good initiative, and as of 2024, the RWI was available at a 2.4 kilometer-squared resolution for over 135 low- and middle-income countries. 11 Figure 6: Relative Wealth Index by DoU – Rural (left), Urban Cluster, and Cities (right) Source: Authors’ calculations using RWI and DoU. Note: The histograms display individual pixel values of RWI weighted by population. Lower values refer to lower wealth levels. The dark line shows the national distribution. Implications of an outdated classification The empirical findings underscore a growing mismatch between Pakistan’s de facto urban development and its de jure administrative classifications. This divergence has substantial consequences for the design and targeting of spatial policy. We describe three dimensions where the discrepancy between an area's administrative classification as urban and its actual reality constrains the implementation of public policies. First, there are tax and fiscal implications as areas classified as rural do not pay property taxes. In Pakistan, property taxes are only levied on urban buildings; thus, considering urban areas as rural has a significant implication on the fiscal resources available at the local level. The International Growth Center estimates that the entirety of Punjab—Pakistan’s most populous province, home to more than 125 million people— collected only 0.01 percent of its GDP in property taxes (FY19), about seven times lower than the average in lower-middle-income countries (LMICs). Importantly, this pattern is mirrored across the rest of the country and contributes to significant gaps in the provision of services and infrastructure (Abbas et al., 2023). 8 Only a minority of countries use purely administrative definitions to identify urban areas. Changing how Pakistan determines urban areas to include population density, service access, and other urban characteristics will allow it, as the DoU shows, to account for a varied urban landscape. Recognizing the existence of areas between dense cities and rural villages can help to establish a staggered expansion of the areas subject to property taxes. Updating the urban classification could increase property taxes sevenfold, and new technologies can help modernize cadaster systems. Besides supporting the 8Moreover, the low tax rate on urban property can help explain real estate being one of the main investment tools in Pakistan, which distorts investment incentives. 12 reclassification of what areas are urban, satellite data offers additional possibilities to identify properties and update the cadasters. 9 Second, an incorrect classification of urban areas has negative implications for the capacity to provide services. While being exempt from property taxes acts as an incentive for local authorities to maintain areas with an unmistakable urban character as rural, it diminishes their capacity to mobilize resources for public investments in infrastructure. The lack of investment limits the capacity to provide high-quality public services and constrains cities' capacity to act as engines of growth. The mismatch between on-the- ground realities and the official classification of urban areas is more pronounced outside the traditional large urban centers, including mid and secondary cities. These have experienced significant population growth, which calls for a growing infrastructure that is being hampered, in part, by the lack of local fiscal resources. Density brings new urban planning challenges. Urban areas have growing infrastructure needs that require large investments and planning, often across administrative boundaries. As urban centers expand, the delivery of public services also requires growing capabilities for planning and administering larger and more complex settlements. Water provides a telling example of how the misclassification of a rural area as urban has significant implications for the capacity to provide services. For instance, rural settlements must plan for the provision of only 40 percent of the available drinking water per person of urban ones, which implies that, even in the presence of appropriate planning, households living in a urban areas that are considered rural will be underserved. 10 The challenges of planning appropriate levels of service delivery are further compounded by the fact that urban growth is not limited by administrative boundaries. Proper planning requires recognizing that urban growth expands beyond administrative boundaries, with the efficient planning of distribution networks for services such as water and transportation requiring accounting for population density and cross- administrative boundaries. Analyzing metropolitan agglomerations independently of administrative boundaries could improve planning and make resource allocation more efficient (Dijkstra et al., 2020). Moreover, when not properly managed, density can lead to “sterile agglomeration”, a situation where people are concentrated in certain areas without gains in living standards and productivity. While density can improve service delivery efficiency, it imposes new demands on urban governance and infrastructure planning. “Sterile agglomeration” (Grover et al., 2022) occurs when the population is concentrated without significant productivity gains. Productive and inclusive urbanization and spatial development require continuous investments in human and physical capital, good local governance of the local administration and adequate local financing. Urban growth puts pressure on public service delivery, requiring the expansion of infrastructure and new administrative capacity. The per capita cost of providing public services is lower in locations with higher population density, particularly for those who rely on centralized infrastructure, such as networks, to provide sanitation and drinking water. However, 9 Some countries have already used satellite-based technology, with positive results. For example, the Arusha City Council in Tanzania leveraged satellite data to identify taxpayers’ properties, leading to a tripling of the number of eligible taxpayers (Fjeldstad et al., 2017). 10 According to National Drinking Water Quality Standards, access means that at least 45 and 120 liters per capita per day of drinking water are available for rural and urban areas, respectively. 13 agglomeration does not improve living standards or productivity where urban planning and investment are lacking. Classic examples include the lack of transportation networks that create negative congestion loops and the layout of public service networks, which becomes more costly and inefficient if implemented after urban expansion occurs. In the case of Pakistan, urban slums—or “Katchi Abaadis"— represent examples of failures of urban planning that lead to high population density but low living standards. Not investing in rapidly urbanizing areas early will lead to “sterile agglomeration”, areas that are congested but not more productive. Finally, an artificial, outdated, and binary urban/rural divide masks the true extent of the gaps in living conditions across space, making it difficult to design relevant public policy and creating statistical invisibility. Misclassification introduces bias into spatial socioeconomic indicators. Given that urban areas typically exhibit better welfare outcomes, the inclusion of peri-urban settlements under rural categories may inflate rural averages and obscure true disparities. This complicates the measurement of spatial inequality, leading to underestimating the actual size of the rural and urban development gap. Importantly, not correctly measuring the extent of these gaps limits the information on the effective interventions and resources needed to close them. Examples of social indicators that are probably affected include gaps in educational attainment, child stunting, and price differentials between officially classified urban and rural areas. 14 References Abbas, A., Ahsan, N., Amin, M. A., Best, M., Callen, M., Cheema, A., Farooqui, A., Khan, A., Khan Mohmand, S., & Masud, O. (2023). Property tax utilization and equity in Punjab: Policy challenges. International Growth Centre (IGC). Chen, R., Yan, H., Liu, F., Du, W., & Yang, Y. (2020). Multiple global population datasets: differences and spatial distribution characteristics. ISPRS International Journal of Geo-Information, 9(11), 637. Dijkstra, L., Florczyk, A. J., Freire, S., Kemper, T., Melchiorri, M., Pesaresi, M., & Schiavina, M. (2020). Applying the Degree of Urbanisation to the globe: A new harmonised definition reveals a different picture of global urbanisation. Journal of Urban Economics, 125, 103312. European Commission. (2023). GHSL Data Package 2023 (JRC133256). Publications Office of the European Union. Fjeldstad, O.-H., Ali, M., & Goodfellow, T. (2017). Taxing the urban boom: Property taxation in Africa (CMI Insight 1). Chr. Michelsen Institute. Grover, A., Lall, S.V., and Maloney, W.F. (2022). Place, Productivity, and Prosperity – Revisiting Spatially Targeted Policies for Regional Development. Washington, DC: The World Bank. Javed, N., & Khan, M. (2018). Redefining Urban: A case for multi-criteria definition of urban areas in Pakistan. The Urban Gazette. Melchiorri, M., Florczyk, A., Freire, S., Schiavina, M., Pesaresi, M., & Kemper, T. (2018). Unveiling 25 Years of Planetary Urbanization with Remote Sensing: Perspectives from the Global Human Settlement Layer. Remote Sensing, 10(5), 768. Pakistan Bureau of Statistics. (2020). Pakistan social & living standards measurement survey (PSLM) 2018- 19 national/provincial (social report). Government of Pakistan. Pakistan Bureau of Statistics. (2023). Key findings report: Population and housing census 2023. Ministry of Planning, Development and Special Initiatives, Government of Pakistan. Rabiee, M., & Dahmm, H. (2020). Leaving no one off the map: A guide for gridded population data for sustainable development. Sustainable Development Solutions Network. United Nations, Department of Economic and Social Affairs, Population Division. (2018). World urbanization prospects: The 2018 revision (ST/ESA/SER.A/420). United Nations. World Bank. (2024). Realizing the potential of Pakistan’s secondary cities (English). World Bank Group. 15