Measuring the Local Economic Impacts of Nature-Based Tourism in Uganda Supported by © 2025 International Bank for Reconstruction and Development / The World Bank ​ 1818 H Street NW​ Washington DC 20433​ Telephone: 202-473-1000​ Internet: www.worldbank.org 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, completeness, or currency of the data included in this work and does not assume responsibility for any errors, omissions, or discrepancies in the information, or liability with respect to the use of or failure to use the information, methods, processes, or conclusions set forth. 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 construed or 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 The material in this work is subject to copyright. Because The World Bank encourages dissemination of its knowledge, this work may be reproduced, in whole or in part, for noncommercial purposes as long as full attribution to this work is given. Any queries on rights and licenses, including subsidiary rights, should be addressed to World Bank Publications, The World Bank Group, 1818 H Street NW, Washington, DC 20433, USA; fax: 202-522-2625; e-mail: pubrights@worldbank.org. Cover image: © Gudkovandrey / Adobe Stock Design: Sergio Moreno Tellez (WBG) Editor: Mark Mattson Acknowledgements This report was written by Justin Kagin (Founder, Kagin’s Consulting and Consultant, World Bank Group [WBG]), Vanessa Satur (Consultant, WBG), and Wendy Li (Environment and Tourism Specialist, WBG). The LEWIE-LITE technical work and analysis was carried out by Ed Taylor (Distinguished Professor Emeritus, University of California, Davis and Consultant, WBG), Justin Kagin, and Mateusz Filipski (Associate Professor, University of Georgia and Consultant, WBG). Urvashi Narain (Program Leader, WBG) and Lisa Farroway (Senior Environmental Specialist, WBG) provided technical oversight and guidance. Sara Enders Estupinan (Consultant, WBG) helped coordinate the production of this report. Data collection and field work was conducted by Laterite. The authors are thankful for the support of colleagues from the Government of Uganda for this work: Samuel John Mwandha (Executive Director, Uganda Wildlife Authority), Tom Okello (Executive Director, National Forest Authority), Edgar Buhanga (Project Officer, Uganda Wildlife Authority), Leone Candia (Principal Wildlife Officer, Ministry of Tourism, Wildlife and Antiquities), Richard Kapere (Manager, Conservation Planning, Uganda Wildlife Authority), Vanice Mirembe (Manager, Community Awareness and Conservation, Uganda Wildlife Authority), and Dorcus Twesigomwe (Business Development Manager, Uganda Wildlife Authority). The team also thanks the chief wardens and staff of Queen Elizabeth National Park, Bwindi Impenetrable National Park, and local government officials. Technical inputs, review, and overall collaboration and facilitation of field work were provided by Stephen Ling (Lead Environmental Specialist, WBG), Nicholas Zmijewski (Senior Environmental Engineer, WBG), and Lesya Verheijen (Senior Operations Officer, WBG) from the Uganda Investing in Forests and Protected Areas for Climate-Smart Development Project. The team also thanks Barbara Amaso (Statistician, Uganda Wildlife Authority), Ojok Denis Rodney (Senior Statistician, Ministry of Tourism, Wildlife and Antiquities), and their entire team for their technical contributions. Peer review was provided by Jessie McComb (Senior Private Sector Specialist, Tourism, WBG) and Mokshana Wijeyeratne (Senior Environmental Specialist, WBG). The team thanks Keith Hansen (Country Director, Uganda) and Rosemary Mukami Kariuki (Country Manager, Uganda, WBG) for their support. Many other World Bank colleagues participated in LEWIE-LITE workshops and offered insights to improve the methodology and analysis. This report was funded by PROBLUE, PROGREEN, the Swedish International Development Agency via the Uganda Multi-Donor Trust Fund, and the Global Environment Facility- financed Global Wildlife Program. Table of Contents Abbreviations and Acronyms / ix Glossary of Terms / ix Executive Summary / 1 How Was the Study Done? / 2 What Did the Study Find? / 4 What Recommendations and Lessons Can Be Drawn from the Study? / 7 SECTION 1. Introduction / 8 Direct and Indirect Impacts of Tourism / 9 Nature-based Tourism in Uganda and the LEWIE-LITE Pilot / 11 Structure of this Report / 12 SECTION 2. Methodology / 13 Sampling Design / 16 SECTION 3. and Analysis / 19 Data Collection ​ Data Analysis and the LEWIE-LITE Dashboard / 22 SECTION 4. Descriptive Statistics of Tourist Numbers and Spending at Queen Elizabeth and Bwindi Impenetrable National Parks / 25 SECTION 5. Using the Model to Simulate Impacts of Tourism / 35 Local-Economy Impacts of Tourist Spending ($) / 37 Local-Economy Impacts of Tourist Spending ($) Net of Park Entry Fees / 41 Impacts of Future Growth in Tourism / 44 SECTION 6. Testing the Robustness of Results and Analyzing the Differences in Multipliers Between the Two Protected Areas / 47 SECTION 7. Using the Model to Simulate Changes in Park and Community Revenue Sharing Spending / 51 Impacts of Changes in Park and Community Revenue Sharing / 52 SECTION 8. Using the Model to Simulate Complementary Interventions / 55 Impacts of Increased Demand for Agricultural and Nonagricultural Goods / 56 Impacts of Increased Wage Earnings for Local Workers / 59 Conclusions and Recommendations / 64 Main Results and Findings / 65 Study Limitations / 67 Recommendations / 68 Strengthening the local economic benefits of protected area tourism / 68 Recognizing the economic value of protected areas / 69 Potential further applications / 69 References / 70 Appendixes Appendix A.Questionnaires / 71 Park Entry / 71 Appendix B. Questions and Answers about the LEWIE-LITE Model and Analysis / 82 Appendix C. Calculation of Multipliers / 85 Table of Figures FIGURE ES.1 Direct and Indirect Impacts of Tourist Spending in Protected Areas / 3 FIGURE ES.2 Effects of a $100 Increase in Tourist Spending on the Local Economy Around QENP / 4 FIGURE ES.3 Effects of a $100 Increase in Tourist Spending on the Local Economy Around BINP / 5 FIGURE 1.1 Direct and Indirect Impacts of Tourist Spending in Protected Areas / 10 FIGURE 3.1 Social Accounting Matrix for QENP as shown on the LEWIE-LITE Dashboard  / 23 FIGURE 3.2 LEWIE-LITE Dashboard for QENP Multipliers for $100 of Tourist Spending under the Simulations Tab / 24 FIGURE 5.1 LEWIE-LITE Dashboard of Tourist Spending Multipliers for QENP / 38 FIGURE 5.2 Effects of a $100 Increase in Tourist Spending on the Local Economy Around QENP / 38 FIGURE 5.3 LEWIE-LITE Dashboard of Tourist Spending Multipliers for BINP / 39 FIGURE 5.4 Effects of a $100 Increase in Tourist Spending on the Local Economy Around BINP / 40 FIGURE 5.5 LEWIE-LITE Dashboard of Tourist Spending Multipliers Net of Park Fees for BINP / 42 FIGURE 5.6 Effects of a $100 Increase in Tourist Spending Net of Park Fees on the Local Economy Around BINP / 42 FIGURE 5.7 Effects of a $2.5 Million Increase in Tourist Spending on the Local Economy Around QENP / 45 FIGURE 5.8A Effects of a $5.03 Million Increase in Tourist Spending on the Local Economy Around BINP / 45 FIGURE 5.8B Effects of a $2.09 Million Increase (Net of Park Entry Fees) in Tourist Spending on the Local Economy Around BINP / 46 FIGURE 7.1 Effects of a $100 Increase in Park Spending on the Local Economy Around QENP / 53 FIGURE 7.2 Effects of a $100 Increase in Community Revenue Sharing Spending on the Local Economy Around QENP / 53 FIGURE 7.3 Effects of a $100 Increase in Park Spending on the Local Economy Around BINP / 54 FIGURE 7.4 Effects of a $100 Increase in Community Revenue Sharing Spending on the Local Economy Around BINP / 54 FIGURE 8.1 Effects of a $100 Increase in Local Agricultural Production on the Local Economy Around QENP / 57 FIGURE 8.2 Effects of a $100 Increase in Local Nonagricultural Production on the Local Economy Around QENP / 57 FIGURE 8.3 Effects of a $100 Increase in Local Agricultural Production on the Local Economy Around BINP / 58 FIGURE 8.4 Effects of a $100 Increase in Local Nonagricultural Production on the Local Economy Around BINP / 58 FIGURE 8.5 Effects of a $100 Increase in Earnings for Unskilled Female Workers on the Local Economy Around QENP / 60 FIGURE 8.6 Effects of a $100 Increase in Earnings for Skilled Female Workers on the Local Economy Around QENP / 60 FIGURE 8.7 Effects of a $100 Increase in Earnings for Unskilled Male Workers on the Local Economy Around QENP / 61 FIGURE 8.8 Effects of a $100 Increase in Earnings for Skilled Male Workers on the Local Economy Around QENP / 61 FIGURE 8.9 Effects of a $100 Increase in Earnings for Unskilled Female Workers on the Local Economy Around BINP / 62 FIGURE 8.10 Effects of a $100 Increase in Earnings for Skilled Female Workers on the Local Economy Around BINP / 62 FIGURE 8.11 Effects of a $100 Increase in Earnings for Unskilled Male Workers on the Local Economy Around BINP / 63 FIGURE 8.12 Effects of a $100 Increase in Earnings for Skilled Male Workers on the Local Economy Around BINP / 63 MAP 2.1 Map of Uganda Highlighting Queen Elizabeth and Bwindi Impenetrable National Parks / 15 TABLE 2.1 Summary of Sample Size by Types of Actor / 16 TABLE 2.2 Definitions of Skilled and Unskilled Workers / 18 TABLE 3.1 Summary of Information Collected from Visitors, Local Businesses, (from Data Collection Protected Area Authorities, and Households ​ Instruments) / 21 TABLE 4.1 Number of Visitors and Their Expenditures at QENP and BINP / 27 TABLE 4.2 Expenditure Shares in Tourism Activities, Restaurants, and Hotels or Lodges Surrounding QENP and BINP / 28 TABLE 4.3 Expenditure Shares for Agriculture, Fishing, Retail Services, and Production Businesses Surrounding QENP and BINP / 29 TABLE 4.4 Population, Income, and Expenditures of Poor and Nonpoor Households Surrounding QENP and BINP / 31 TABLE 4.5 Park Budgets and Community Revenue Sharing and Spending in 2022 / 33 TABLE 5.1 Summary of the Impact of a $100 Increase in Tourist Spending at QENP and BINP / 43 TABLE 6.1 Incomes in the Local Economies Around Each Park / 48 TABLE 6.2 Wage and Profit Flows to Households / 49 TABLE 6.3 Leakage Share from Production Sectors and Households / 50 TABLE C.1 Social Accounting Matrix of the Economy in and Surrounding QENP / 86 TABLE C.2 Social Accounting Matrix Multiplier Model of the Economy in and Surrounding QENP / 87 Abbreviations and Acronyms BINP Bwindi Impenetrable National Park COMREVSH Community revenue sharing DCI Data collection instrument G National government GDP Gross domestic product K Capital LEWIE Local Economy-Wide Impact Evaluation LFSK Labor female skilled workers LFUSK Labor female unskilled workers LMSK Labor male skilled workers LMUSK Labor male unskilled workers LocalG Local government LSMS Living standards measurement survey QENP Queen Elizabeth National Park ROW Rest of the world (outside of the local economy) UWA Uganda Wildlife Authority SAM Social Accounting Matrix WBG World Bank Group *All dollar amounts are US dollars unless otherwise indicated. Glossary of Terms Glossary Description Data collection instrument (DCI) A survey used in data collection Income multiplier The income multiplier quantifies the total income that is generated in the local economy for every dollar spent by a tourist, including all indirect or spillover effects Living standards measurement Comprehensive World Bank household survey survey (LSMS) Production multiplier The production multiplier quantifies the total value of goods and services generated in the local economy for every dollar spent by a tourist, including all indirect impacts or spillover effects Protected Area Tourism Local A tool to estimate the direct and indirect economic impacts of tourism in Economy-Wide Impact Evaluation and around protected areas “Lite” (LEWIE-LITE) Social accounting matrix (SAM) Captures flows of all economic transactions that take place within an economy; usually, a matrix representation of national accounts but can also apply to regions or areas Spillover effects Indirect effects that occur after direct impacts ix Executive Summary This study addresses the critical connection between Uganda’s protected areas and tourism and estimates the economic impact of tourism on these sites and their surrounding communities. The primary audience of this report is decision-makers such as the ministry of tourism, protected area management authorities, local authorities, and task teams supporting nature-based tourism. In Uganda, where tens of thousands of tourists visit protected areas annually, there is little information on the economic implications of nature-based tourism. This hinders the ability of tourism authorities, protected area managers, and the government to optimize the economic value of protected areas and their associated benefits. Studies on the economic impact of tourists on protected areas are scarce, and few consider the broader impacts on local economies. Most studies have focused on direct effects, such as those on tourism-related businesses (for example, tour operators, restaurants, and lodges), and overlooked the indirect impacts on other businesses, commercial farmers, and households near protected areas. These indirect or spillover effects determine how tourism influences local economies, especially local production, and helps households that are not directly involved in tourism. Therefore, it is necessary to include them in development plans, policies, and cost-benefit analyses when considering tourism development. To address this knowledge gap and facilitate data-driven decision-making, this study introduces the Protected Area Tourism Local Economy-Wide Impact Evaluation (LEWIE) “Lite” tool—hereinafter referred to as LEWIE-LITE. LEWIE-LITE uses data from economic actors near protected areas to quantify direct and indirect impacts of tourist spending on local economies. The tool supports policies on tourism impacts and informs on park spending, community revenue sharing, and complementary policies for protected areas. 1 How Was the Study Done? LEWIE-LITE was piloted in two of Uganda’s protected areas: Queen Elizabeth National Park (QENP) and Bwindi Impenetrable National Park (BINP). The LEWIE-LITE methodology entails collecting data from actors in the local economy in or around the protected area using data collection instruments (DCIs). The objective is to sample representative numbers and types of visitors, households, and businesses at each protected area. The LEWIE-LITE approach aims to minimize time and resources spent on data collection, which is important for scaling the model. LEWIE-LITE models capture market linkages and direct and indirect impacts of tourist spending around protected areas. Figure ES.1 illustrates these linkages and the general theory behind the models used in this study. The black arrows show direct impacts and the yellow dotted arrows show indirect impacts. The direct impacts begin with tourists spending money on food, lodging, and activities when they visit a protected area. Tourists also pay taxes and fees, including park entry fees to the government. Indirect effects include the flow of wages and profits from tourism businesses into households, which, in turn, spend this income and spread impacts to other businesses and farms. This creates additional rounds of sales, income gains to businesses, flows of profits and wages into local households, and household spending, which increase the local gross domestic product (GDP). Park authorities hire guides and wardens, invest in park improvements and, in some cases, share some park entry fees with local communities. Spending by parks and communities adds to the local economic impacts of nature- based tourism. The sum of all direct and indirect impacts is likely to exceed the amount of money tourists spend. Queen Elizabeth National Park. Photo credit: melissamn / Shutterstock 2 Measuring the Local Economic Impacts of Nature-Based Tourism in Uganda FIGURE ES.1 Direct and Indirect Impacts of Tourist Spending in Protected Areas Park rules limit some human activities; Direct Impact human-wildlife conflicts Indirect Impact TOURISM Tourists spend money Negative on food, lodging, and impact tourist activities Tourists pay taxes and fees Local GDP increases as households Businesses and Tourists come spend, businesses GOVERNMENT households pay taxes to park grow, and income increases Households get wages and profits Government hires guides and wardens, invests in park improvements, and shares Households get revenues with communities wages, shared revenue, etc. Source: World Bank. To build the model, field data were entered into Microsoft Excel. An algorithm was applied to these data to create a local social accounting matrix (SAM) and a SAM multiplier model, which is used by LEWIE-LITE to analyze tourism’s impact on the local economy surrounding protected areas. Finally, simulations were carried out on the local economic impacts of existing tourism, the effects of increases in tourism (for example, from a new investment in the protected area), changes in spending by the park and community revenue sharing projects, and an array of complementary interventions designed to enhance the benefits of tourism for the local economy. A dashboard for each protected area was developed for government stakeholders, providing a user-friendly interface for government to explore local economy impacts, including the simulations carried out for this study. 3 What Did the Study Find? The study provides interesting observations on the local economies surrounding the two protected areas. Tourism impacts both local incomes and local production (business revenues). Protected areas support the local economy through their spending on local labor and local goods and services. For every tourist dollar, the SAM multipliers indicate that local incomes increase by $2.03 in the local economy surrounding QENP and by $0.37 in the local economy of BINP. Local production increases by $5.67 in QENP and $1.20 in BINP. The figures for BINP are lower because most park entry fees and gorilla permits fees are remitted to Uganda Wildlife Authority (UWA) and thus considered a leakage from the local economy. Net of park fees and gorilla permits, the values for BINP would increase to $1.33 (local incomes) and $4.29 (local production). The dashboard can be used to detail the impacts of any amount of tourist spending on different production sectors or activities, household groups, wages by worker group, and community and park revenue. To illustrate this, the impacts of a $100 increase in tourist spending were simulated. The dashboard displays the impact of this increase in tourist spending on production, on incomes, and on labor income, as shown for QENP in figure ES.2 and BINP in figure ES.3. FIGURE ES.2 Effects of a $100 Increase in Tourist Spending on the Local Economy Around QENP E ects of this tourist spending... A. ...ON PRODUCTION B. ...ON INCOMES C. ...ON LABOR INCOME 37.64 37.67 270.02 196.8 Additional Production Value ($) Additional Labor Income ($) 188.8 Additional Income ($) 62.01 8.06 23.8 2.26 14.94 7.32 5.74 Tourism Restaurants Lodges Agricultural Non- Fish Nonpoor Poor Female Male Female Male agricultural unskilled unskilled skilled skilled Tourism-related activities Nontourism-related activities Households Labor categories Source: World Bank data. 4 Measuring the Local Economic Impacts of Nature-Based Tourism in Uganda FIGURE ES.3 Effects of a $100 Increase in Tourist Spending on the Local Economy Around BINP Effects of this tourist spending... A. ...ON PRODUCTION B. ...ON INCOMES C. ...ON LABOR INCOME 54.55 32.41 5.23 4.91 Additional Production Value ($) Additional Labor Income ($) Additional Income ($) 30.68 2.78 2.47 20.15 10.7 4.63 2.19 1.79 Tourism Restaurants Lodges Agricultural Non- Fish Nonpoor Poor Female Male Female Male agricultural unskilled unskilled skilled skilled Tourism-related activities Nontourism-related activities Households Labor categories Source: World Bank data. Poor households are less employed in tourism and nontourism sectors of the local economies surrounding both parks. They receive only 3 percent (QENP) and 13.5 percent (BINP) of total direct and indirect tourism benefits. The small share of income gains to poor households reflects their lack of access to local formal jobs and capital. Investing in park management generates economic benefits for local communities. In 2022, tourism to QENP generated $69.6 million in benefits against a park budget of $3 million, while tourism to BINP generated benefits of $14.4 million against a budget of $2.3 million. Additional tourism growth, particularly of international tourists, can generate even higher local economic benefits. If the number of visitors to the parks increases at prepandemic annual growth rates, then QENP and BINP will generate an additional income to households of $5.1 million and Queen Elizabeth National Park. Photo credit: Pecold / Adobe Stock $2 million, respectively. 5 Measuring the Local Economic Impacts of Nature-Based Tourism in Uganda Increases in community revenue sharing spending can create additional substantial local economic benefits. Uganda has a formal revenue sharing program in which 20 percent of park entry fees and $10 from each gorilla permit are used to fund projects that reduce human-wildlife conflict, build local infrastructure, or support other socio-economic programs. Simulations on the dashboards revealed that a $100 increase in community revenue sharing would lead to a local GDP gain of $245 (QENP) and $120 (BINP). The data also provide valuable insights that may be missed in other tourism sector research. For example, the study found a higher percentage of women workers in tourism-related jobs than in nontourism-related jobs, corroborating the view that tourism worldwide is a valuable job entry point for women. Additionally, model simulations reveal that the impacts of tourist spending are considerably larger on nontourism activities than on tourism activities and highlight the importance of looking beyond tourism activities when evaluating the impacts of tourism in local economies. Bwindi Impenetrable National Park. Photo credit: Gunter Nuyts / Shutterstock 6 Measuring the Local Economic Impacts of Nature-Based Tourism in Uganda What Recommendations and Lessons Can Be Drawn from the Study? Indirect impacts or spillover effects of tourism are an important part of how tourism can impact local economies. LEWIE-LITE simulations indicate that tourism generates higher multipliers in nontourism activities than on tourism activities; these impacts should therefore be considered in country economic development plans, policy design, or cost- benefit studies before designing and implementing new tourism projects. The analysis of the tourism sectors around BINP and QENP suggests several ways for policy makers to maximize the impacts from the sector. For example, local linkages between tourism and other sectors, such as agriculture, fishing, and manufacturing could be strengthened to increase income and production multipliers. Using Bwindi as an example, if more inputs to non-agricultural businesses could be procured locally, the production and income multipliers from tourism would increase. Also, promoting local ownership of businesses and employment of local workers may increase economic benefits in communities surrounding the parks. These kinds of interventions could also address how to increase the employment of women in nontourism activities which is currently at 4 percent (as against a 12 percent of employment of women in tourism activities). Considering the lower proportion of poor households that benefit from tourism compared to nonpoor households, skills and entrepreneurship programs for poor households in tourism and nontourism activities could increase the amount of tourism benefits that they capture. Reviewing Uganda’s national revenue sharing program to direct more benefits to poor households may also make these benefits more equitable. This pilot shows how the LEWIE-LITE model can help address data gaps on the direct and indirect impacts to local economies in and surrounding protected areas. The tool can support government to estimate the costs and benefits of investing in tourism at protected area sites. The model can also simulate local economy impacts of fewer tourists but higher tourist spending and vice versa—or increasing both the number of tourists and how much they spend. One could also use the tool to monitor developments in the tourist sector such as increased demand for ecotourism, or negative impacts like a reduction in local agricultural revenue due to human-wildlife conflicts. The model can further be used to simulate the impacts of expanding Uganda’s community revenue sharing program. While recognizing the limitations of the LEWIE-LITE model’s simplified nature, this research prompts further studies to delve deeper into the results. Technically, the model and dashboard could be adapted to study specific subsectors of the local economy or tourism markets and products. A comparison could be undertaken of the results from LEWIE-LITE against a more comprehensive LEWIE model for protected area tourism (World Bank, 2021) to enhance understanding of direct and indirect effects of protected area tourism on local economies. 7 SECTION 1 Introduction Bwindi Impenetrable National Park. 8 Photo credit: Christie / Adobe Stock Measuring the Local Economic Impacts of Nature-Based Tourism in Uganda A country’s protected areas and tourism are closely connected, and yet, the economic impacts of tourism on the businesses and households in and around protected areas are largely unknown. Protected areas in Uganda are visited by hundreds of thousands of tourists every year, but there are few data on the economic impact of these tourists. Without this information, tourism ministries, park services, communities, and the central and local government are unaware of the economic value of protected areas and the costs and benefits of investing in protected areas and tourism. While there have been some economic and statistical analyses of tourism in Uganda1, studies on the impact of tourist dollars on specific Ugandan protected areas are rare, and only a few studies have estimated the impacts of tourism on local economies surrounding protected areas, such as the World Bank’s Banking on Protected Areas report (2021). Direct and Indirect Impacts of Tourism Studies seldom go beyond direct impacts on tourism businesses—tour operators, restaurants, lodges, souvenir shops, and so on—to include indirect impacts on, for example, commercial farmers whose crops are sold to restaurants, poor and nonpoor households that get income from tourism activities, or expenditures by households and businesses that create local income, production, and employment multipliers. Indirect impacts or spillover effects of tourism are an important part of how tourism affects local economies. These impacts should be considered in country economic development plans, sector development plans, policies, or cost-benefit studies before new tourism projects are undertaken. This study demonstrates a tool that can be widely employed to satisfy these needs. The tool is the Protected Area Tourism Local Economy-Wide Impact Evaluation Lite (LEWIE- LITE) model, which gathers information from economic actors around a protected area and uses it to calculate direct and indirect impacts of tourist dollars on the local economy. LEWIE-LITE models capture market linkages and direct and indirect impacts of tourist spending around protected areas. Figure 1.1 illustrates these linkages and the general theory behind the models. The black arrows show direct impacts and the yellow dotted arrows, indirect impacts. The direct impacts begin with tourists spending money on food, accommodation, shopping and tourist activities. They also pay taxes and fees, important among which are park entry fees that accrue to the government. Not all impacts are necessarily positive. For example, park rules can limit some human activities which can impact negatively on sources of income, while human-wildlife conflict around protected areas can result in losses for communities. These direct effects of tourist spending appear in the top part of the figure. Indirect effects include the flow of wages and profits from tourism businesses into households, which, in turn, spend this income and spread impacts to new businesses 1 See https://www.wavespartnership.org/en/knowledge-center/economic-and-statistical-analysis-tourism-uganda and, https://utb.go.ug/wp-content/uploads/2024/04/UG-Tourism-Stat-and-Econ-Analysis-2020-brief_compressed.pdf 9 Measuring the Local Economic Impacts of Nature-Based Tourism in Uganda and farms. This creates additional rounds of sales, income for businesses, flows of profits and wages into local households, and household spending, which increase local GDPs. As business and household incomes grow, so do tax revenues to governments. Park authorities hire guides and wardens, invest in park improvements and, in some cases, share park entry fees with local communities. Spending by parks and communities adds to the local economic impacts of nature-based tourism. According to LEWIE-LITE, the sum of all direct and indirect impacts is likely to exceed the amount of money tourists spend. The sum of impacts divided by tourist spending gives the multiplier effect of tourist spending on local economies. The model calculates multiplier effects on local production (sales), household income, and employment per tourist and per dollar of tourist spending. LEWIE-LITE can also guide policies to strengthen linkages among local actors, and simulate interventions to strengthen tourism impacts in communities around protected areas. It is easy to use, clear, and generates tables and visuals. LEWIE-LITE builds on the LEWIE models for protected area tourism created for the World Bank’s Banking on Protected Areas report of 2021, but with a simpler, more scalable approach. FIGURE 1.1 Direct and Indirect Impacts of Tourist Spending in Protected Areas Park rules limit some human activities; Direct Impact human-wildlife conflicts Indirect Impact TOURISM Tourists spend money Negative on food, lodging, and impact tourist activities Tourists pay taxes and fees Local GDP increases as households Businesses and Tourists come spend, businesses GOVERNMENT households pay taxes to park grow, and income increases Households get wages and profits Government hires guides and wardens, invests in park improvements, and shares Households get revenues with communities wages, shared revenue, etc. 10 Measuring the Local Economic Impacts of Nature-Based Tourism in Uganda Nature-based Tourism in Uganda and the LEWIE-LITE Pilot Uganda ranks among the top 10 most bio-diverse countries globally, hosting approximately 54 percent of the world’s mountain gorillas, 11 percent of the world’s bird species, 7.8 percent of global mammal diversity, 19 percent of Africa’s amphibian species, and 14 percent of Africa’s reptiles.2 To conserve this rich natural heritage, the government has designated a number of protected areas, which include national parks, wildlife reserves and wildlife sanctuaries, community wildlife areas, central forest reserves, and local forest reserves. These resources and attractions, complemented by an attractive year-round climate, offer a conducive environment for nature-based tourism and is the foundation of Uganda’s tourism sector. In 2019, Uganda received 1.5 million international arrivals. Tourism has been identified as a priority sector within the Third National Development Plan (2020/2021–2024/2025)3 which seeks to increase tourism arrivals and revenues as well as employment in the tourism sector. The World Bank is supporting nature-based tourism in Uganda via investments under the Investing in Forests and Protected Areas for Climate-Smart Development Project. One of the main objectives of the project is to increase benefits to communities from the sustainable management of forests and protected areas. For these reasons, Uganda was selected as a pilot for the LEWIE-LITE methodology. Two Ugandan parks, Queen Elizabeth National Park (QENP) and Bwindi Impenetrable National Park (BINP), were selected with the help of the Uganda Wildlife Authority (UWA) for piloting. Queen Elizabeth National Park was selected because it is one of the most visited parks in the country and also offered an opportunity to assess impacts of resource extraction such as fishing. Bwindi Impenetrable National Park was selected because it offers one of Uganda’s main nature- based tourism attractions, Bwindi Impenetrable National Park. Photo credit: Jane Rix / Shutterstock gorilla trekking. 2 Uganda Tourism Development Master Plan 2014-2024. 3 See https://www.health.go.ug/wp-content/uploads/2020/08/NDP-3-Report.pdf 11 Measuring the Local Economic Impacts of Nature-Based Tourism in Uganda The tool offers an easy-to-use online dashboard for each site. On this dashboard, users such as government (ministries of tourism or environment, protected area managers, etc.) or World Bank teams supporting tourism projects can explore different assumptions about current or anticipated levels of protected area tourism and model the impacts of complementary interventions on local economies surrounding each park. Structure of this Report This report explains the LEWIE-LITE methodology (sections 2 and 3) and provides descriptive statistics on the economic impacts of tourism in each protected area collated through the field surveys (section 4). It also presents and compares results of simulations using the LEWIE-LITE interactive dashboards, and discusses simulated impacts of changes in tourism spending surrounding QENP and BINP (sections 5 and 6), local economy impacts of park and community revenue sharing spending (section 7), and impacts of complementary interventions such as increasing demand for locally produced goods and services (section 8). The final section offers conclusions and recommendations for further work (section 9). Queen Elizabeth National Park. Photo credit: melissamn / Shutterstock 12 SECTION 2 Methodology 13 Bwindi Impenetable National Park. Photo credit: Travel Stock / Shutterstock Measuring the Local Economic Impacts of Nature-Based Tourism in Uganda The two protected areas selected for this project were surveyed separately but with similar DCIs and methods. They are: Queen Elizabeth National Park, renowned for its rich biodiversity, is home to elephants, leopards, buffaloes, hippos, and crocodiles, and famous for its tree-climbing lions. It is situated in the southwest of Uganda (map 2.1) and spans 1,978 square kilometers of savanna, forests, wetlands, and crater lakes. The park received 95,340 visitors in 2022. It is bordered by many communities in the districts of Kasese, Rubirizi, Kamwenge, and Rukungiri and even has a few fishing villages within its boundaries. These communities participate in various economic activities. Some are directly related to tourism, such as hotels and lodges, restaurants, souvenir shops, and tour operators, while others indirectly benefit from these activities by working in or owning retail shops, local services, and small manufacturing enterprises such as furniture making. Those inside the park may not farm or keep livestock but still participate in fishing and local businesses, while those surrounding the park engage in agriculture and other businesses. The population of this local economy is approximately 600,300 people living in 129,589 households.1 Bwindi Impenetrable National Park is also endowed with rich biodiversity, including endangered mountain gorillas and over 350 species of birds. It is a landlocked park located southwest of QENP along the border with the Democratic Republic of Congo. It covers approximately 331 square kilometers of dense rainforest, hills, and steep valleys. The park welcomed 32,628 visitors in 2022. It is bordered by communities in Kanungu, Kisoro, and Kabale districts, which engage in a range of local economic activities including tourism and nontourism-related work. They also farm and keep livestock. The population of this local economy consists of approximately 76,900 people living in 15,276 households.2 1 Estimates from World Bank 2019 population data for subcounties surrounding the parks. 2 Estimates from World Bank 2019 population data for subcounties surrounding the parks. 14 Measuring the Local Economic Impacts of Nature-Based Tourism in Uganda MAP 2.1 Map of Uganda Highlighting Queen Elizabeth and Bwindi Impenetrable National Parks 0 50 PROTECTED AREAS Kilometers DISTRICT BOUNDARIES INTERNATIONAL BOUNDARIES SOUTH S UDAN NATIONAL CAPITAL DEM. REP. UGANDA O F CO NGO t er lb Lake A Kwania ke La Lake Kyoga Queen Elizabeth KENYA National Park KAMPALA Lake George Lake Edward L a k e Vi c t o r i a Bwindi Impenetrable National Park IBRD 47998 | APRIL 2024 TANZ ANIA This map was produced by the Cartography Unit of the World Bank Group. The boundaries, colors, denominations and any other information shown on this map do not imply, on the part of RWANDA the World Bank Group, any judgment on the legal status of any territory, or any endorsement or acceptance of such boundaries. 15 Measuring the Local Economic Impacts of Nature-Based Tourism in Uganda Sampling Design The LEWIE-LITE methodology entails collecting a small set of data from key local economy actors, using DCIs designed for this purpose. Definitions of the local economy vary, reflecting the structure of economies and markets as well as the regional interest of studies. For this study, a 10-kilometer distance from each park’s boundary (and, in the case of QENP, in the park itself) was used to define the sampling area for the two local economies. Similar criteria are used in other local-economy models; for example, the Economic Impact of Giving Land to Refugees (Zhu et al. 2023) and the Banking on Protected Areas studies. The fieldwork objective was to gather data from the parks, local government, visitors, and businesses (by type) at each site, using DCIs programmed into tablets. Table 2.1 summarizes the samples by types of actor. TABLE 2.1 Summary of Sample Size by Types of Actor ACTOR OR ENTITY SAMPLE Park manager Budget and entrance fees to QENP and BINP Revenue sharing projects for each of the parks Tourists 200 tourists targeted and randomly selected from various park gates and key tourist locations (QENP) and gorilla trekking debriefing and graduation points (BINP) 100 questionnaires per site Hotels, lodges, and resorts 5 hotels or lodges and 5 all-inclusive resorts selected randomly from a list of accommodation compiled locally through consultations 10 questionnaires per site Restaurants 20 restaurants selected randomly 10 restaurants per site Other tourism-related 30 units selected randomly businesses 15 units per site, including 5 units per business type (tour operators or guides, souvenir shops, and equipment rental) Other nonagricultural 30 units selected randomly based on a list of enterprises businesses 15 units by site, 5 from each of 3 categories: retail businesses (small and large stores), other services (repairs, hairdressers, beauty salons, and so on), and nonservice businesses (carpentry shops, food processors, and so on) Commercial farmers 20 commercial farmers selected randomly 10 farmers per site Commercial fishers 10 fishers interviewed (only for QENP) Tour operators in the capital 10 tour operators surveyed in the capital, Kampala Source: World Bank. 16 Measuring the Local Economic Impacts of Nature-Based Tourism in Uganda The information was collected as follows: 1. Protected areas QENP and BINP are managed by UWA. Data were collected through face-to-face interviews at each site in which park staff were asked about park budgets and revenue sharing revenues. 2. Revenue sharing program Uganda, through UWA, has an arrangement through which 20 percent of all park entry fees and $10 from each gorilla trek permit are reinvested into community programs to improve local livelihoods or address human-wildlife conflict. The beneficiaries are communities bordering the national parks. Data were obtained by interviewing authorities for each park. 3. Tourists Tourists were interviewed at the two national parks. The study considered three types of tourists: nationals, foreigners with residency in Uganda, and foreigners visiting Uganda for tourism. In total, 200 tourists (100 at each site) were randomly surveyed. In the case of a group3, only one member was interviewed. 4. Hotels, lodges, and resorts Information was gathered from 10 hotels/ lodges/resorts from each site. Through local consultations with park and government officials, lists of hotels, lodges, and resorts were compiled and a random selection were chosen for interviews, comprising five small and five medium-to-large establishments. 5. Restaurants Interviews were conducted with 20 restaurants, 10 at each site, randomly selected. 6. Other tourism-related businesses Besides lodges and restaurants, 30 tourism-related businesses, 15 from each site, were randomly selected. Data were collected from three types of tourism businesses: tour operators or guides (excursions and so on), souvenir shops, and equipment rental stores. Five interviews were conducted for each type of business per park. 7. Other nonagricultural businesses Thirty nonagricultural businesses, not related to tourism, 15 from each site, were randomly selected. Data were collected through face-to-face interviews with three types of nontourism businesses: retail trade (small and large grocery stores), other services (auto repairs, transport, hairdressers, beauty salons, and so on), and nonservice establishments (for example, carpentry shops and food processing companies). Five interviews were carried out for each type of business per park. 8. Commercial farmers Twenty farmers, 10 from each site, were randomly selected and interviewed. 3 Either a family or tour group. 17 Measuring the Local Economic Impacts of Nature-Based Tourism in Uganda 9. Commercial fishers As natural resource users, 10 fishers were randomly selected and interviewed at QENP (Bwindi has no fishing). 10. Households The LEWIE-LITE model also simulates the effects of tourism and other benefits on household incomes. Households surrounding a park may supply labor, produce crops and livestock, run businesses, and in some cases, receive profits from tourist firms and operations. Household data were extracted from the most recent World Bank Living Standards Measurement Survey (LSMS) for Uganda, conducted in 2019–20. Location data were used to identify households from the districts around each park. The districts identified in the LSMS (which were the same or comparable to the districts surrounding the parks where surveys were conducted) and used for BINP were Rubanda, Kisoro, and Kanungu. For QENP, Mitooma, Ibanda, Rukungiri, Kamwenge, Kasese, and Bushenyi were used. Using World Bank 2018–19 absolute poverty lines, these households were separated into poor and nonpoor categories. 11. Tour operators Many protected area visitors purchase travel packages from tour operators outside the local economy, typically in Kampala, the capital. Ten tour operators based in Kampala were surveyed. These data are important because tourists who buy package deals generally do not know how costs are allocated to local spending by sector (lodges, tour operators, meals, etc.). To get this information, tour operators in Kampala were asked what share of the package price went to businesses around the two parks. Additionally, respondents were asked for the percentages spent on skilled and unskilled male and female workers. The definitions were agreed on with local experts (table 2.2). TABLE 2.2 Definitions of Skilled and Unskilled Workers LEVEL PERSONNEL SKILLED OR UNSKILLED Level 1 Executive managers Skilled For example, institutional heads, administrative and Regardless of degree or years of financial directors, and human resource directors experience Level 2 Managers or employees who have authority over Skilled others Regardless of degree or years of For example, accountants and managers experience Level 3 Cleaning staff Skilled Servers If (and only if) have diploma or Janitors qualification Bartenders Gardeners Unskilled Kitchen staff If they have experience (regardless of Guides the number of years) but do not have a diploma or qualification Source: World Bank. 18 SECTION 3 Data Collection ​ and Analysis 19 Queen Elizabeth National Park. Photo credit: HartSmith / Adobe Stock Measuring the Local Economic Impacts of Nature-Based Tourism in Uganda Data were collected for the LEWIE-LITE survey from February 6 to March 17, 2023, including travel. Five enumerators were mobilized through a local data collection firm, Laterite, to carry out the field survey. The survey was timed to capture the largest number of tourists as well as the dynamics of the local economy during the tourism high season. Ideally, this exercise would be repeated multiple times a year to capture the economic impacts across high and low tourist seasons. A goal of the LEWIE-LITE approach is to minimize time and resources spent on data collection, which is essential for scalability of the model. This means surveying an adequate sample of tourists to obtain reasonable estimates of their spending patterns, combined with interviews of local tourism and nontourism businesses. Keeping the information gathered to a minimum meant that interviews could be carried out quickly. Table 3.1 summarizes the information gathered from visitors and businesses (see appendix A for the short questionnaires used for each type of actor or entity), and that were collected from national park authorities and households (from the LSMS data for the region surrounding each park). This is the minimum data set needed for LEWIE-LITE to model local economic activity directly or indirectly connected to protected area tourism. The visitor survey (table 3.1a) asked respondents about how much money they spent on each category of goods and services listed on the visitor DCI: lodging, restaurant meals, souvenirs, and so on. The goal of the business survey (b) is to capture broadly, yet comprehensively, all activities in the local economy that may benefit directly or indirectly from tourism. Most of the questions concern percentages of total revenue that businesses spent on intermediate inputs and labor, purchased (or hired) locally or outside the local economy (“local imports,” for purposes of the model). Businesses were also asked to “ballpark” a typical profit share for enterprises in their activity and to estimate what share of these businesses are locally owned. Data from park budgets (c) were used to calculate protected area spending on local and outside goods and services, including wages for different worker groups. They also provided the information needed to calculate the amount of park entry fees shared with local communities, which comprises the community revenue sharing budget. Interviews with community revenue sharing staff provided information on how much of this revenue was spent on local and outside goods and services, including wages to different worker groups.1 The spending categories for community revenue sharing and park spending were the same, as given in 3.1c. The LSMS data were used to calculate average per capita income, remittances, government transfer incomes, and other variables for each household group (d), as well as household budget shares and goods and services purchased locally and outside the local economy. 1 Since there were several community projects, the study took the largest one from each park to estimate the shares of expenditures and wages to labor, and applied those shares to the total amount. 20 TABLE 3.1 Summary of Information Collected from Visitors, Local Businesses, Protected Area Authorities, and Households ​ (from Data Collection Instruments) A. Visitors (from Visitor Survey) A. VISITORS i. B. BUSINESSES Number of multi-day tourists (adults and children >5) (in the whole zone of impact) C. PROTECTED AREA D. HOUSEHOLDS (FROM ​ Average ii. VISITOR stay (days) SURVEY) (FROM BUSINESS SURVEY) (FROM NATIONAL PARK ACCOUNTS) (FROM HOUSEHOLD SURVEY) iii. Average stay (nights) Numberiv. ofAverage nightly multi-day price per tourists room (total, (adults About and including taxes, double what occupancy, including percentage of monthly and other fees) resortspending goes to Total expected annual park budget 2022 What is the population of the communities constituting this local v. Expected number of single-day tourists (adults and children >5, no lodging) children >5) (in the whole zone of impact) each of the following: vi. Expected spending per person per day while visiting this protected area, on: economy (number of people) 1. Park entry Average stay (days) 2. Salaries and wages for male unskilled workers Local restaurants (food and drink) Total expected park entry fees 2022 How many households are in this local economy (number of 3. Guides and tours households) 4. Souvenirs/handicrafts Average stay (nights) 5. Retail shops, local markets Salaries and wages for male skilled workers (machine What share of park entry fees are being assigned About what is the average annual per-capita income of households 6. Other operators, supervisors, receptionists, accountants, etc.) to community revenue sharing in this region ($) B. Businesses (from Business Survey) Average About what i. nightly percentage price per room of monthly goes to each (total, spendingSalaries andof the following: wages for female unskilled workers Difference (net transfer from government to parks) About what is the average annual government transfers (e.g., social 1. Salaries and wages for male unskilled workers including taxes,2. double occupancy, Salaries and wages for male skilled workers (machine operators, supervisors, receptionists, accountants, etc.) cash transfers) to households in this region including resort3. and other fees) Salaries and wages for female unskilled workers 6. Locally produced agricultural products (fruits, vegetables, meats) 4. Salaries and wages for female skilled workers (machine operators, supervisors, receptionists, accountants, etc.) Expected number tourists Salaries and wages female skilled workers (machine 7. TotalLocally produced fish community or other revenue natural resources sharing budget About what percentage of household income comes from: Crop purchases 5. of single-day from local farmers or animal from productsfor local ranchers 8. Services (laundry, maintenance, construction, repairs) from local providers 6. (adults and children Purchases >5, from tourism no lodging) operators, supervisors, receptionists, accountants, etc.) 9. Purchases from local stores and other businesses 7. Local fish 10. Purchases made outside the local economy 8. Expected spending perServices person (machine per day maintenance,Cropconstruction, purchases repairs) local fromfrom local providers farmers products or animalvii. Wages earned by male unskilled workers in the household Other variables 9. Purchases while visiting this protected from area, local stores and on: other from businesses local ranchers 1. Percentage of salaries and wages paid to local workers 10. Purchases outside the local economy, like merchandise (for stores) or supplies D. Households (From household Survey) Park entry 11. rate (%) Farm tax/fishing business tax Purchases from tourism much Wages earned by male skilled workers in the household i. population What is the How of of thethis communities goes to: this local economy (number budgetconstituting of people) ii. Other variables ii. How many households are in this local economy (number of households) 1. Share Local restaurants (food and drink)of businesses locally owned Local fish wages for male unskilled workers Wages earned by female unskilled workers in the household is the average iii. About whatSalaries andannual per-capita income of households in this region ($) 2. Share of wages paid to local workers iv. About what is the average annual government transfers (e.g., social cash transfers) to households in this region Guides and tours3. Average profit margin Services (machine maintenance, construction, v. repairs) Salaries and wages for male About what percentage of household income comes from: skilled workers Wages eamed by female skilled workers in the household C. Protected Area (from National Park Accounts) from local providers 1. Wages earned by male unskilled workers in the household i. Total expected annual park budget 2022 2. Wages earned by male skilled workers in the household ii. Total expected park entry fees 2022 Souvenirs/handicrafts Purchases from local stores and other businesses 3. Salaries and wages unskilled workers Profits from household-owned farms or businesses or renting property Wages earned forunskilled by female female workers in the household iii. What share of park entry fees are being assigned to community revenue sharing 4. Wages eamed by female skilled workers in the household the household owns iv. Difference (net transfer from government to parks) 5. Profits from household-owned farms or businesses or renting property the household owns v. Total community Retail shops, local markets revenue sharing budget Purchases outside the local economy, like merchandise Salaries and for female skilled workers Migrant remittances (domestic and foreign) 6. Migrant wages (domestic remittances and foreign) vi. How much of this budget goes to: (for stores) or supplies 7. About what percentage of household spending each month is on: 1. Salaries and wages for male unskilled workers 8. Food bought from local grocery stores 2. Salaries and wages for male skilled workers Other tax/fishing 9. Buying food Payment direct of rents from on local land, farmers or your buildings, etc. own farm About what percentage of household spending each month is on: 3. unskilled Salaries and wages for female Farm workers business tax rate (%) 10. Buying local fish 4. Salaries and wages for female skilled workers Other variables 11. Buying Locally food and drink produced at local restaurants agricultural products (fruits, Food bought from local grocery stores 5. Payment of rents on land, buildings, etc. 12. Things besides vegetables, food that are sold by people or businesses in your community, including services meats) 13. Things you buy from businesses, etc., in places outside your community Share of businesses locally owned 14. Rental Locally income produced fish or other natural resources Buying food direct from local farmers or your own farm 15. Income tax payments Share of wages paid to local workers Services (laundry, maintenance, construction, Buying local fish repairs) from local providers Average profit margin Purchases from local stores and other businesses Buying food and drink at local restaurants Purchases made outside the local economy Things besides food that are sold by people or businesses in your community, including services Other variables Things you buy from businesses, etc., in places outside your community Percentage of salaries and wages paid to local Rental income workers Income tax payments Source: World Bank data. Measuring the Local Economic Impacts of Nature-Based Tourism in Uganda Data Analysis and the LEWIE- LITE Dashboard After the data are collected, they are entered into structured spreadsheets using Microsoft Excel. An algorithm then uses the data to construct a social accounting matrix (SAM), and from it, a SAM multiplier model, upon which the LEWIE-LITE online dashboard is built. The first SAM was built in 1962 as a matrix representation of national accounts (Stone and Brown 1962). Since then, country-level SAMs have been used widely by the World Bank, the Organisation for Economic Co-operation and Development, and other international organizations, and form the basis for many countries’ computer general equilibrium models. The DCIs, SAM, and SAM multiplier matrices for each protected area are accessed by the “Data” link on the dashboard’s main page menu bar. Figure 3.1 shows a picture of the dashboard with data and a SAM multiplier matrix for QENP. The dashboard can run simulations on the impacts of tourism and other policies on the local economy. When users run a simulation, for example, to estimate the impact of an additional $100 in visitor spending in the protected area on local production, employment, and incomes (the simulation discussed in section 5), the number (in this case, 100) is entered onto the dashboard under: “How much tourist spending ($) do you want to simulate?” The dashboard passes this number to the model algorithm and reports the results in easy-to-visualize figures. Figure 3.2 shows the dashboard display of the multiplier results of tourist spending from the data and a $100 increase simulation, using the “Simulations” tab, for QENP. Bwindi Impenetrable National Park. Photo credit: typepng / Adobe Stock 22 FIGURE 3.1 Social Accounting Matrix for QENP as shown on the LEWIE-LITE Dashboard AGRICU TOURISM NONAGR. FISH LMUSK LMSK LFUSK LFSK K POOR NONPOOR RESTAURANTS LODGES TOURISTS PROT.​ COMREVSH AREA Agricultural 3.11 1.61 1.97 1.59 2.33 2.33 2.34 2.34 2.33 2.61 2.32 2.30 1.95 1.89 1.95 2.16 Tourism 0.00 1.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.03 0.24 0.00 0.00 Nonagricultural 2.90 2.29 3.68 2.31 3.21 3.21 3.21 3.21 3.21 3.19 3.21 2.83 2.68 2.70 3.20 3.45 Fish 0.15 0.12 0.16 1.12 0.16 0.16 0.16 0.16 0.16 0.15 0.16 0.20 0.16 0.15 0.15 0.16 LMUSK 0.52 0.33 0.40 0.42 1.43 0.43 0.43 0.43 0.43 0.47 0.43 0.43 0.38 0.38 0.44 0.70 LMSK 0.52 0.31 0.37 0.29 0.41 1.41 0.42 0.42 0.41 0.45 0.41 0.43 0.40 0.38 0.37 0.43 LFUSK 0.05 0.08 0.07 0.05 0.06 0.06 1.06 0.06 0.06 0.06 0.06 0.06 0.06 0.08 0.27 0.21 LFSK 0.00 0.01 0.00 0.00 0.00 0.00 0.00 1.00 0.00 0.00 0.00 0.01 0.03 0.02 0.04 0.00 K 1.33 1.06 1.07 1.02 1.15 1.15 1.15 1.15 2.15 1.22 1.14 1.34 1.23 1.17 1.01 1.11 Poor 0.07 0.05 0.05 0.05 0.09 0.08 0.13 0.12 0.08 1.06 0.06 0.06 0.06 0.06 0.07 0.08 Nonpoor 2.37 1.74 1.85 1.73 2.96 2.98 2.93 2.94 2.97 2.14 2.99 2.21 2.04 1.97 2.05 2.37 Restaurants 0.05 0.04 0.04 0.04 0.07 0.07 0.06 0.06 0.07 0.05 0.07 1.05 0.05 0.07 0.05 0.05 Lodges 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 1.00 0.62 0.00 0.00 Tourists 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 1.00 0.00 0.00 Protected area 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.06 1.00 0.00 ComRevSh 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.01 0.22 1.00 Source: World Bank data. Note: LMUSK = Labor male unskilled workers; LMSK = Labor male skilled workers; LFUSK = Labor female unskilled workers; LFSK = Labor female skilled workers; K = Capital; ComRevSh = Community revenue sharing; Measuring the Local Economic Impacts of Nature-Based Tourism in Uganda FIGURE 3.2 LEWIE-LITE Dashboard for QENP Multipliers for $100 of Tourist Spending under the Simulations Tab LOCAL ECONOMY WIDE IMPACTS OF TOURIST SPENDING $ You may wish to evaluate different values of tourist spending: total tourist spending attributable to the protected area, change in tourist spending you expect from this project, etc. How much tourist spending ($) do you want to simulate? 100 Effects of this tourism spending on... D. ...ON A. ...ON PRODUCTION B. ...ON INCOMES C. ...ON LABOR INCOME COMMUNITY AND PARK EARNINGS 37.64 37.67 270.02 6.48 196.8 Additional Production Value ($) Additional Labor Income ($) 188.81 Additional Income ($) Additional Income ($) 62.01 8.06 1.4 23.8 2.26 14.94 7.32 5.74 Tourism Restaurants Lodges Agricultural Non- Fish Nonpoor Poor Female Male Female Male Nonpoor Poor agricultural unskilled unskilled skilled skilled Tourism-related activities Nontourism-related activities Households Labor categories Source: World Bank data. 24 SECTION 4 Descriptive Statistics of Tourist Numbers and Spending at Queen Elizabeth and Bwindi Impenetrable National Parks 25 Queen Elizabeth National Park. Photo credit: Jane Rix / Shutterstock Measuring the Local Economic Impacts of Nature-Based Tourism in Uganda Tables 4.1 to 4.5 present the calculations for the LEWIE-LITE model based on the data gathered in the field, and provide a “snapshot” of tourist visits and spending across the local economies surrounding Bwindi Impenetrable National Park and Queen Elizabeth National Park. Table 4.1 shows the number visitors and their spending on goods and services during their visit to each site. The total number of visitors to QENP in 2022 was 95,340. However, of these, 67,736 were multiday tourists (32,106 foreign nonresidents, 3,673 foreign residents, and 31,957 East African residents) while the remaining comprised mainly students. While students could also potentially be multiday tourists, they have a much lower expenditure than tourists whose main purpose of travel is to engage in the specific tourism activities offered by the park. Therefore, for the purposes of this study, it was decided to remove students from the total number of visitors to QENP. BINP had 32,628 visitors, all of whom were multiday tourists (30,440 foreign nonresidents, 341 foreign residents, and 1,818 East African residents). The average stay was 1.9–2.2 nights at QENP and 2.3 days at BINP. Rooms were, on average, more expensive at BINP: $182 per room, compared with $164 at QENP. Entry fees were considerably higher at Bwindi where visitors purchase their entry fee plus a gorilla trekking permit, which includes a park guide. This gives an average park fee of $698 per person per day, versus $32.80 at QENP.1 Bwindi visitors spent more in local restaurants and on guides, tours and souvenirs; while in and around QENP, visitors spent more at retail businesses and local markets and other categories of expenditure. In summary, the average visitor to QENP spent $268 per day, of which 61 percent went to accommodation, 12 percent to park fees, 2.5 percent to local restaurants, 14 percent to guides and tours, 4.6 percent to souvenirs and handicrafts, 2.5 percent to retail shops and local markets, and 3 percent to other goods and services. The relatively low percentage spent in local restaurants (as also in the case of BINP as indicated later in this paragraph) is because most accommodation offer meals in their packages on either a half or full board basis. On the other hand, the average visitor to BINP spent $998 per day, of which 19 percent went to accommodation, 67 percent to park fees, 2 percent to local restaurants, 5 percent to guides and tours, 3 percent to souvenirs and handicrafts, 1 percent to retail shops and local markets, and 3 percent to other goods and services. The disparity in spending between the two parks is mainly due to the costs of the gorilla trekking permit at BINP, particularly for foreign (nonresident and resident) visitors. 1 These are average entry prices for randomly selected tourists who are primarily nonresident foreigners (especially at Bwindi Impenetrable) but include some domestic and foreign resident visitors as well (mainly at Queen Elizabeth National Park). 26 Measuring the Local Economic Impacts of Nature-Based Tourism in Uganda TABLE 4.1 Number of Visitors and Their Expenditures at QENP and BINP VISITOR INFORMATION GATHERED QUEEN ELIZABETH BWINDI IMPENETRABLE NATIONAL PARK NATIONAL PARK Park Number of multi-day tourists (adults and 67,736 32,628 records children > 5) Visitor Average stay (days) 2.2 2.3 surveys Average stay (nights) 1.9 2.3 Average nightly price per room (total, including taxes, double occupancy, including $164 $182 resort and other fees, $) Expected spending in $ per person per day on: Expenditure category Park entry $32.80 $698.00 Local restaurants (food and drink) $6.70 $17.80 Guides and tours $37.00 $55.40 Souvenirs and handicrafts $12.30 $32.20 Retail shops and local markets $6.70 $5.30 Other (including hotel shops) $8.40 $6.90 Total $267.90 $997.60 Source: World Bank data. Attribution can be challenging at some protected area sites. Can one attribute tourist expenditure in the local economy to the existence of the park itself? In the case of QENP and BINP, attribution is straightforward, as the parks are the main tourism attractions in each area so one can assume that all, or almost all, tourist spending in the local economies can be attributed to the parks. Visitor spending is the direct or first-round impact of protected area tourism on the local economy, as illustrated in figure 1.1. The LEWIE-LITE algorithm calculates visitor spending for each expenditure category. It channels park entry fees to the park sector and visitor goods and services to mainly the local tourism business sectors.2 Visitor demands for local goods and services direct more money to the corresponding production activities. Tourism businesses spend this money purchasing intermediate and factor inputs, including hired labor. This transmits impacts to nontourism businesses, which supply other inputs, as well as to households, which receive wage and profit incomes from tourism activities. Wages and profits stimulate household spending, which adds to the local demand for goods and services from nontourism activities and creates 2 This may not be the case for all tourist spending. For example, in many high-end accommodation facilities, wine is perhaps not acquired through a local business but imported and delivered directly to the establishment. 27 Measuring the Local Economic Impacts of Nature-Based Tourism in Uganda new rounds of impacts on nontourism businesses and households. The model adds up these multiple rounds of impacts which converge to local multiplier effects of protected area tourism. Table 4.2 Expenditure Shares in Tourism Activities, Restaurants, and Hotels or Lodges Surrounding QENP and BINP4.2 reports the gross income or sales, percentages of gross income spent on intermediate inputs and wages, and profit margins of tourism-related activities (tourism businesses, restaurants, and hotels and lodges) in BINP and QENP. These were calculated from the interviews with souvenir stores, tour operators and tour equipment rental shops, restaurants, and hotels and lodges surrounding the parks.3 The numbers reveal how these businesses channel income to male and female unskilled and skilled workers; local purchases from commercial farmers, herders, and fishers; nearby retail and service businesses; profits; and nonlocal purchases. TABLE 4.2 Expenditure Shares in Tourism Activities, Restaurants, and Hotels or Lodges Surrounding QENP and BINP QUEEN ELIZABETH NATIONAL PARK BWINDI IMPENETRABLE NATIONAL (AVG. %) PARK (AVG. %) About what percentage of monthly spending goes to TOURIST RESTAURANTS HOTELS/ TOURIST RESTAURANTS HOTELS/ each of the following: BUSINESSES LODGES BUSINESSES LODGES Salaries and wages for male unskilled workers 6% 4% 4% 0% 3% 5% Salaries and wages for male skilled workers 6% 4% 12% 1% 8% 9% (machine operators, supervisors, receptionists) Salaries and wages for female unskilled workers 8% 7% 3% 2% 3% 6% Salaries and wages for female skilled workers 2% 7% 7% 4% 2% 12% (machine operators, supervisors, receptionists) Crop purchases from local farmers or animal 13% 28% 20% 31% 44% 12% products from local ranchers Purchases from tourism activities 0% 0% 4% 0% 0% 4% Local fish 1% 4% 3% 2% 7% 2% Services (machine maintenance, construction, 13% 16% 15% 24% 8% 10% repairs, etc.) from local providers Purchases from local stores and other businesses 14% 13% 13% 19% 13% 12% Purchases outside the local economy, like 35% 15% 14% 16% 9% 26% merchandise (for stores) or supplies Nonfarm tax/meal tax/lodge tax rate (%) 3% 2% 6% 3% 2% 3% Other variables Share of businesses locally owned 71% 100% 36% 100% 90% 17% Share of wages paid to local workers 53% 80% 51% 100% 43% 80% Average profit margin 32% 25% 46% 32% 25% 46% Number of observations 15 10 11 15 10 10 Source: World Bank data. 3 Profit margins in BINP were not calculated but presumed to be similar to businesses in QENP. 28 Measuring the Local Economic Impacts of Nature-Based Tourism in Uganda Table 4.3 reports the same expenditure shares for agricultural and nontourism related businesses. These businesses benefit mainly indirectly from tourists. For example, tourists buy directly from restaurants which purchase food from local farmers. TABLE 4.3 Expenditure Shares for Agriculture, Fishing, Retail Services, and Production Businesses Surrounding QENP and BINP QUEEN ELIZABETH NATIONAL PARK BWINDI IMPENETRABLE NATIONAL (AVG. %) PARK (AVG. %) About what percentage of monthly spending AGRICULTURE FISHING RETAIL/ AGRICULTURE FISHING (NOT RETAIL/ goes to each of the following: SERVICES / AVAILABLE SERVICES / PRODUCTION LOCALLY) PRODUCTION Salaries and wages for male unskilled workers 19% 16% 7% 18% - 1% Salaries and wages for male skilled workers 21% 1% 5% 12% - 5% Salaries and wages for female unskilled 0% 0% 4% 2% - 0% workers Salaries and wages for female skilled workers 0% 0% 0% 1% - 1% Crop purchases from local farmers or animal 19% 10% 27% 14% - 15% products from local ranchers Purchases from tourism activities 0% 0% 0% 0% - 0% Local fish 2% 2% 3% 1% - 2% Services (machine maintenance, construction, 15% 22% 14% 14% - 11% repairs) from local providers Purchases from local stores and other 11% 8% 20% 18% - 21% businesses Purchases outside the local economy, like 14% 36% 16% 16% - 42% merchandise (for stores) or supplies Nonfarm tax/meal tax/lodge tax rate (%) 1% 3% 2% 5% - 2% Other variables Share of businesses locally owned 100% 100% 90% 100% 93% Share of wages paid to local workers 100% 43% 43% 100% - 47% Average profit margin 30% 23% 11% 30% - 11% Number of observations 10 10 15 15 - 15 Source: World Bank data. 29 Measuring the Local Economic Impacts of Nature-Based Tourism in Uganda Tables 4.2 and 4.3 are snapshots of the data input for the model, gathered from the local interviews. They reveal interesting aspects of the local economies surrounding the protected areas. For example, table 4.3 shows limited participation of women as paid employees in agriculture, fishing, and other nontourism activities. The percentages of income (gross sales) that these activities spend on female-worker wages range from 0 percent to 4 percent. However, the percentages going to male-worker wages in these activities are usually much higher—12 percent to 21 percent in the case of agriculture. Female-worker wages are higher in tourism-related activities (table 4.2). It appears that tourism is an important entry point for female workers in the local economy. This raises the question of why women are not being employed much in nontourism sectors. These tables reveal both direct local impacts of business spending as well as leakages out of the local economy, as businesses purchase intermediate inputs in outside markets and send wages and profits to households outside local economies. Most nontourism businesses are locally owned and hire mostly local labor. Nonlocal ownership and hiring are more common among tourism-related businesses. Promoting local ownership and employment of local workers may increase economic benefits in communities surrounding the parks. The model does not explain why local tourism business ownership and local employment are not higher (or how to make them higher), but it does show that local business ownership and employment can inform interventions to increase local economic benefits from protected area tourism. Retail stores have large leakages in these local economies (like most), because a large portion of their merchandise comes from outside markets. The same is true for production activities. For example, carpentry shops purchase many of their inputs from outside markets. At BINP, which is more isolated than QENP, purchases from outside markets make up 42 percent of total retail spending compared with 16 percent at QENP. Households in QENP also spend a large share of their income in local stores (see table 4.3). Because stores at QENP source more of what they sell locally, more money is circulated in the local economy thereby creating larger multipliers. Household spending is an important link in the chain of income and expenditures that can create local income multipliers. Table 4.4 shows the population, income, and expenditures of poor and nonpoor households in the two local economies, which were calculated from LSMS survey data. Generally, households surrounding QENP tend to be better off than those surrounding BINP. This is due to the fact that QENP is located in a busy commercial part of the country, along the main north-south trunk road while BINP is located in mountainous terrain far away from main cities. The average annual per capita income at QENP is $207 for poor households and $682 for nonpoor households, $26 and $174 higher, respectively, than at BINP. The share of poor households is also much higher in villages surrounding BINP: 44 percent, compared with only 11 percent at QENP. In both parks, poor households receive more income from unskilled and skilled female workers than nonpoor households—20 percent and 15 percent compared to 8 percent and 7 percent. 30 Measuring the Local Economic Impacts of Nature-Based Tourism in Uganda TABLE 4.4 Population, Income, and Expenditures of Poor and Nonpoor Households Surrounding QENP and BINP HOUSEHOLD POPULATION, INCOME AND EXPENDITURES QUEEN ELIZABETH BWINDI IMPENETRABLE NATIONAL PARK NATIONAL PARK POOR NONPOOR POOR NONPOOR What is the population of the communities constituting this 70,775 529,525 29,453 47,447 local economy (number of people) How many households are in this local economy 12,522 117,067 4,692 10,584 (number of households) About what is the average annual per-capita income of $207 $682 $181 $508 households in this region About what is the average annual government transfers (e.g., $- $- $- $9 social cash transfers) to households in this region About what percentage of household income comes from... Wages earned by male unskilled workers in the household 10% 8% 10% 8% Wages earned by male skilled workers in the household 5% 8% 5% 8% Wages earned by female unskilled workers in the household 20% 8% 20% 8% Wages earned by female skilled workers in the household 15% 7% 15% 7% Profits from household-owned farms or businesses or renting 21% 25% 6% 24% property the household owns Migrant remittances (domestic and foreign) 7% 3% 1% 4% About what percentage of household spending each month is on... Food bought from local grocery stores 2% 7% 9% 12% Buying food direct from local farmers or your own farm 48% 35% 41% 35% Buying local fish 0% 1% 0% 0% Buying food and drink at local restaurants 0% 2% 0% 1% Things besides food that are sold by people or businesses in 31% 45% 32% 40% your community, including services Things you buy from businesses, etc., in places outside your 1% 3% 4% 2% community Rental income 1% 1% 0% 0% Income tax payments 0% 1% 0% 0% Source: World Bank data. 31 Measuring the Local Economic Impacts of Nature-Based Tourism in Uganda Nonpoor households receive more income from renting property or having their own businesses than poor households. This is especially true in Bwindi where nonpoor households earn 24 percent of their income from farms, businesses, or renting property, while poor households earn only 6 percent. Poor households surrounding the two parks spend from 40–50 percent of their income on purchases from local farms, with the rest of expenditures mainly on other goods and services from local businesses, with little purchased from outside of the local economy. Table 4.5 shows the park budgets and park revenue from entry fees for QENP and BINP. Parks can have various sources of revenue (entry fees, concession fees, research fees, etc.) but, for the purposes of this LEWIE-LITE exercise, park revenue is considered as the total income from park entry fees (and at BINP, gorilla permits). The park budget is determined by the government. At both parks, park revenue exceeds the park budget— substantially, in the case of BINP. Park revenue is transferred to UWA which then shares 20 percent of park entry fees (plus $10 of each gorilla trek permit fee at BINP) with local communities under the revenue sharing program, thereby creating incentives for conservation via this arrangement.4 Communities spend this money on various projects, generally using local labor and materials. 4 The calculation of payments to local community revenue sharing is more complicated at BINP than QENP due to the gorilla fee. At QENP, foreign visitors pay park entry fees of $40 and domestic visitors pay $5.20, of which 20 percent, or $8 and $1.04, respectively, go to local community revenue sharing. At BINP, nearly all visitors are foreigners who pay a $40 entry fee plus a gorilla permit of around $660. There, 20 percent of the park entry fee and $10 per gorilla permit go to community revenue sharing, for an average of $18 per foreign visitor. 32 Measuring the Local Economic Impacts of Nature-Based Tourism in Uganda TABLE 4.5 Park Budgets and Community Revenue Sharing and Spending in 2022 QUEEN ELIZABETH NATIONAL PARK BWINDI IMPENETRABLE NATIONAL PARK PARK BUDGET COMMUNITY PARK BUDGET COMMUNITY REVENUE SHARING REVENUE SHARING Total annual park budget 2022 $3,008,398 $2,276,284 Total park entry fees 2022 $3,127,775 $45,265,300 What share of park entry fees are being assigned to 20% 20% of park community revenue sharing entry fees and $10 of every gorilla permit Amount sent to the national government $119,377 $42,989,016 Park budget as a percent of park entry fees 96% 5% Total Community Revenue Sharing Budget $625,555 $842,262 How much of park and community revenue sharing budgets go to… Salaries and wages for male unskilled workers 0.3% 28.5% 1.7% 7.4% Salaries and wages for male skilled workers 0.0% 3.3% 0.2% 3.5% Salaries and wages for female unskilled workers 37.0% 14.9% 36.8% 32.6% Salaries and wages for female skilled workers 7.5% 0.0% 7.7% 0.0% Payment of rents on land, buildings, etc. 0.1% 0.0% 0.5% 0.0% Locally produced agricultural products (fruits, 2.9% 1.7% 4.6% 10.9% vegetables, meats) Locally produced fish or other natural resources 0.0% 0.0% 0.0% 0.0% Services (laundry, maintenance, construction, repairs) 30.9% 0.2% 14.9% 4.3% from local providers Purchases from local stores and other businesses 13.2% 51.4% 28.0% 41.3% Purchases made outside the local economy 8% 0% 5.8% 0.0% Percentage of salaries and wages paid to local 45% 100% 46% 100% workers Source: World Bank data. 33 Measuring the Local Economic Impacts of Nature-Based Tourism in Uganda Park revenue is higher at BINP than QENP due to the gorilla trekking activity. In 2022, a visitor paid an average of $700 per person for a gorilla trek (a $40 park entry fee plus $660 gorilla permit) at BINP but only $40 to enter QENP. The visitor sample from BINP contained no Ugandan nationals and only two out of 89 visitors surveyed were foreign residents of Uganda. The rest were nonresident foreigners; thus, the average fee for BINP is close to what nonresident foreigners pay. This contrasts with QENP, where 20 out of 98 surveyed were domestic visitors (including two foreign residents). Domestic visitors pay only $5.20 as entry fee at QENP, bringing the average park entry fee among surveyed visitors down to $32.80. Both parks are fully funded by their fees and, in fact, generate a net income to the national government, which can use this money for other purposes, including to help fund parks that receive fewer visitors. BINP generates especially large net revenues through gorilla permit fees. However, local multiplier effects of tourist spending are lower at BINP because most of the gorilla permit fees are paid to UWA and are not spent locally. Community revenue sharing programs create local social assets and generate economic benefits by hiring local workers and buying materials from local businesses. However, local purchases by both parks impact the local economy in ways beyond just the community revenue sharing program: local purchases make up 92 percent of the budget for QENP and 94 percent for BINP.5 5 The percentage of the budget spent on local purchases is calculated as 100 percent, minus the percentage of “purchases made outside the local economy.” 34 SECTION 5 Using the Model to Simulate Impacts of Tourism 35 Queen Elizabeth National Park. Photo credit: Nadine Wagner / Shutterstock Measuring the Local Economic Impacts of Nature-Based Tourism in Uganda To illustrate uses of the model, the two LEWIE-LITE dashboards were used to simulate impacts of changes in tourist spending surrounding Queen Elizabeth National Park and Bwindi Impenetrable National Park (further simulations are shown in sections 7 and 8). The online dashboards generate graphs showing impacts on local production activities (both tourism-related and nontourism-related), incomes of poor and nonpoor households, and wages of male and female skilled and unskilled workers. For this report, the authors compared impacts for an additional $100 in spending across the different sectors (income, production, jobs, wages, skilled and unskilled, and poor and nonpoor). It was determined that this was a reasonable simulation to run because the additional $100 could come from more tourists or the same number of tourists spending more. Other simulations can be undertaken (for example, increasing the number of tourists or increasing the spending of tourists per day; see appendix B for more information on how LEWIE-LITE can be used). Queen Elizabeth National Park. Photo credit: Kylie Nicholson / Adobe Stock 36 Measuring the Local Economic Impacts of Nature-Based Tourism in Uganda Local-Economy Impacts of Tourist Spending ($) The dashboard gives a snapshot of tourist spending at the parks as well as the total multiplier effects of tourist expenditures. These are the changes in gross sales or the total value of local production of goods and services, household incomes, wages, and profits per dollar of tourist spending. They include both direct impacts on tourist industries such as lodges, restaurants, guides, and souvenir shops, and indirect impacts on the local economy. Queen Elizabeth National Park The tourist spending multipliers for QENP are shown in figure 5.1. An additional dollar of tourist spending increases total local production by $5.67 and local income or GDP by $2.03. Most of this income gain, $1.97, accrues to nonpoor households. Poor households’ income rises by $0.06 per dollar spent by park visitors. Of the $2.03 increase in income, $0.86 is worker wages and $1.17 is profits or payments to capital. A more detailed explanation of how these multipliers are calculated is in appendix C. The dashboard can be used to detail the impacts of any amount of tourist spending on different production sectors or activities, household groups, wages by worker group, and community and park revenue. To illustrate this, the impacts of a $100 increase in tourist spending were simulated. The dashboard displays the impact of this increase in tourist spending on production, on incomes, and on labor income, as shown for QENP in figure 5.2. This was done by multiplying the previously mentioned multipliers and others produced by the model by $100 in extra tourist dollars. For example, the increase in income in figure 5.2b is the $100 increase in tourist spending times the multiplier on income, 2.03, giving the total $203 of which approximately $197 would accrue to nonpoor households and $6 to poor households. Although the direct impacts of tourist expenditures are strong, there are indirect impacts that, taken together, exceed the direct impacts of tourist spending. Figure 5.2a shows that the largest impacts are on nonagricultural activities such as retail shops, local services, and other production activities. The large impacts on nonagricultural activities and local agriculture reflect their importance in the local economy. Workers in the tourist industry use a large share of their income to purchase local goods and services. Local business owners and workers in nontourist sectors also now have extra income to spend on these same goods and services, creating additional rounds of indirect impacts. Summing up the six production bars in figure 5.2a, the total impact on all production sectors of $567 exceeds the simulated $100 increase in tourist spending. 37 Measuring the Local Economic Impacts of Nature-Based Tourism in Uganda FIGURE 5.1 LEWIE-LITE Dashboard of Tourist Spending Multipliers for QENP Local-economy impacts of tourist spending ($): For every dollar of tourist spending, the total production multiplier is: WHICH CAN BE SPLIT INTO: For every dollar of tourist spending, the total income multiplier is: WHICH CAN BE SPLIT INTO: OR. ALTERNATIVELY, CAN BE SPLIT INTO: Source: World Bank data. FIGURE 5.2 Effects of a $100 Increase in Tourist Spending on the Local Economy Around QENP Effects of this tourist spending... D. ...ON A. ...ON PRODUCTION B. ...ON INCOMES C. ...ON LABOR INCOME COMMUNITY AND PARK EARNINGS 37.64 37.67 270.02 6.48 196.8 Additional Production Value ($) Additional Labor Income ($) 188.81 Additional Income ($) Additional Income ($) 62.01 8.06 1.4 23.8 2.26 14.94 7.32 5.74 Tourism Restaurants Lodges Agricultural Non- Fish Nonpoor Poor Female Male Female Male Nonpoor Poor agricultural unskilled unskilled skilled skilled Tourism-related activities Nontourism-related activities Households Labor categories Source: World Bank data. 38 Measuring the Local Economic Impacts of Nature-Based Tourism in Uganda Bwindi Impenetrable National Park Production, income, and wage and profit multipliers are smaller at BINP, as shown in figure 5.3. An additional dollar of tourist spending at BINP increases total local production by $1.20 and local income or GDP by $0.37. These impacts are much smaller than at QENP due to the large share of tourist spending that goes to the gorilla permit, most of which leaves the local economy when remitted to central government. This is considered a leakage from the local economy. As at QENP, most of the income gain of $0.32 accrues to nonpoor households while poor households’ income rises by $0.05 per dollar spent by park visitors. Of the $0.37 increase in income, $0.15 goes to workers as wages and $0.22 goes to owners of capital (including local businesses) as profits. The simulated production impacts of an additional $100 of tourist spending in the local economy around BINP are shown in figure 5.4, disaggregated by sector, income impacts by household group, and wage income impacts by worker group. A $100 increase in tourist spending generates $120 in all production sectors where, similar to QENP, the largest production impacts are in nontourism businesses (figure 5.4a). FIGURE 5.3 LEWIE-LITE Dashboard of Tourist Spending Multipliers for BINP Local-economy impacts of tourist spending ($): For every dollar of tourist spending, the total production multiplier is: WHICH CAN BE SPLIT INTO: For every dollar of tourist spending, the total income multiplier is: WHICH CAN BE SPLIT INTO: OR. ALTERNATIVELY, CAN BE SPLIT INTO: 39 Measuring the Local Economic Impacts of Nature-Based Tourism in Uganda FIGURE 5.4 Effects of a $100 Increase in Tourist Spending on the Local Economy Around BINP Effects of this tourist spending... D. ...ON A. ...ON PRODUCTION B. ...ON INCOMES C. ...ON LABOR INCOME COMMUNITY AND PARK EARNINGS 941.7 942.6 162 6755.8 4923.9 Additional Production Value ($) Additional Labor Income ($) Additional Income ($) Additional Income ($) 4723.9 1551.4 201.6 35.1 595.5 373.9 56.6 183.1 143.5 Tourism Restaurants Lodges Agricultural Non- Fish Nonpoor Poor Female Male Female Male Nonpoor Poor agricultural unskilled unskilled skilled skilled Tourism-related activities Nontourism-related activities Households Labor categories Source: World Bank data. Bwindi Impenetrable National Park. Photo credit: Adrian Solumsmo / Adobe Stock 40 Measuring the Local Economic Impacts of Nature-Based Tourism in Uganda Local-Economy Impacts of Tourist Spending ($) Net of Park Entry Fees A large portion of tourist spending at BINP is made up of the park entry fee (including gorilla trek permit fee). Given the large number of visitors that visit the park annually, the total park entry revenue (that is, income from park entry fees and gorilla permits) far exceeds the park’s operational budget. As a result, the surplus is remitted to UWA which then uses this money to support other protected areas in the country. Since this money leaves the local economy, it is considered a leakage and it reduces the production and income multipliers of additional tourist dollars for the local economy. The dashboard provides an adjustment factor that can be used to calculate tourist impact multipliers net of park fees. This answers a different question to the simulation on tourist spending above. Instead of asking how much an extra tourist or extra tourist spending may contribute to the park, it asks how encouraging tourists to spend more money per day, but without any change in park entry fees, might affect the local economy. The adjustment factor is 1/(1-pfs), where pfs is the share of park fees in average tourist spending. Figure 5.5 presents the multipliers for BINP, net of park entry fees. These are larger than those in figure 5.3. It shows that an additional dollar of tourist spending outside of the park entry fees at BINP increases total local production by $4.29 instead of $1.20; and local income or GDP by $1.33 rather than $0.37, more than three times as much. These impacts are still smaller than at QENP but now much higher than when park entry fees were included in the calculation (see figure 5.3). Poor households receive $0.17 per dollar of tourist income and $1.16 goes to nonpoor households. Of the $1.33 increase in income, $0.55 goes to workers as wages and $0.78 goes to owners of capital. The simulated impacts of $100 of tourist spending in the local economy at BINP net of park entry fees are shown in figure 5.6, disaggregated by production impacts, income impacts, wage income impacts, and community and park earnings. As expected, impacts on the different sectors, household groups, and labor categories net of park entry fees are substantially higher given the larger multipliers. An additional $100 of tourist spending increases tour and souvenir sales by around $38 (as opposed to only $10.70 without netting out park fees). Sales of hotel or lodge accommodation increase by $72 instead of only $20. Again, the largest production impacts are on nonagricultural activities, which increase by $195 instead of $55, followed by agricultural production ($110 as opposed to only $31). The largest wage impacts are for male unskilled ($17.57 instead of $4.91) and skilled workers ($18.69 instead of $5.23); wages for unskilled and skilled female workers rise by $8.84 and $9.93 instead of only $2.47 and $2.78, respectively. For both males and females, wage impacts are slightly larger for skilled workers. 41 Measuring the Local Economic Impacts of Nature-Based Tourism in Uganda FIGURE 5.5 LEWIE-LITE Dashboard of Tourist Spending Multipliers Net of Park Fees for BINP Local-economy impacts of tourist spending ($): For every dollar of tourist spending, the total production multiplier is: WHICH CAN BE SPLIT INTO: For every dollar of tourist spending, the total income multiplier is: WHICH CAN BE SPLIT INTO: OR. ALTERNATIVELY, CAN BE SPLIT INTO: FIGURE 5.6 Effects of a $100 Increase in Tourist Spending Net of Park Fees on the Local Economy Around BINP Effects of this tourist spending... A. ...ON PRODUCTION B. ...ON INCOMES C. ...ON LABOR INCOME 18.69 17.57 115.9 195.1 Additional Production Value ($) Additional Labor Income ($) Additional Income ($) 9.93 109.7 8.84 72.08 38.25 16.7 7.85 6.42 Tourism Restaurants Lodges Agricultural Non- Fish Nonpoor Poor Female Male Female Male agricultural unskilled unskilled skilled skilled Tourism-related activities Nontourism-related activities Households Labor categories 42 Measuring the Local Economic Impacts of Nature-Based Tourism in Uganda Table 5.1 summarizes the multiplier impacts of the extra $100 in tourist spending on the economies surrounding QENP and BINP. For BINP, it includes impacts of a $100 increase in tourist spending with and without park entry fees. Again, it is notable that large percentages of gross revenue benefits from tourists go to sectors other than tourism, specifically retail, services, production and, in QENP, agriculture. TABLE 5.1 Summary of the Impact of a $100 Increase in Tourist Spending at QENP and BINP OUTCOME IMPACTS OF TOURIST SPENDING ($) QUEEN ELIZABETH BWINDI IMPENETRABLE NATIONAL PARK BWINDI IMPENETRABLE NATIONAL PARK NATIONAL PARK (WITH PARK ENTRY FEE) (WITHOUT PARK ENTRY FEE) Impacts per $100 of tourist spending Gross revenue from local production       Agriculture $189 $31 $110 Fishing $16 $2 $6 Tourism businesses $24 $11 $38 Retail, services, and production $270 $55 $195 Restaurants $7 $2 $8 Lodges $62 $20 $72 Total production multiplier $567 $120 $429 Payments to:       Labor (wages) $86 $15 $55 Male unskilled labor (wages) $38 $5 $18 Female unskilled labor (wages) $8 $2 $9 Male skilled labor (wages) $38 $5 $19 Female skilled labor (wages) $2 $3 $10 Capital (profits) $117 $22 $78 Income to       Poor households $6 $5 $17 Nonpoor households $197 $32 $116 Total income (GDP) multiplier $203 $37 $133 Park revenue $6 $67   Community revenue sharing $1 $1   Impacts of an additional tourist Average spending per tourist $506 $1,194 $496 Total local GDP impact per tourist $1,028 $442 $660 To poor households $30 $60 $84 To nonpoor households $998 $382 $575 Source: World Bank data. 43 Measuring the Local Economic Impacts of Nature-Based Tourism in Uganda The table also shows total impacts per visitor. A tourist spends $506 on average in and around QENP and $1,194 around BINP ($496 net of park entry fees). Using the income (local GDP) multipliers, this translates to an additional $1,028 ($506 x 2.03) GDP gain per visitor for QENP and $442 for BINP ($1,194 x 0.37 with park entry fees) and $660 ($496 x 1.33 net of park entry fees). When applying the number of tourists that visited each protected area in 2022 (67,7361 in QENP and 32,628 for BINP as per table 4.1), the total amount generated by these tourists was $69.6 million for QENP and $14.4 million for BINP ($21.5 million net of park entry fees). Both amounts exceed the park budgets needed to operate these sites ($3 million and $2.3 million, respectively). Impacts of Future Growth in Tourism The LEWIE-LITE tool was used to estimate the impact of projected growth in protected area tourism for QENP and BINP. COVID-19 was a major shock to Ugandan tourism, causing respective declines of 62 percent and 78 percent from fiscal year 2019/20 to fiscal year 2020/21.2 However, according to UWA data collected for this study, tourism had recovered to prepandemic levels by 2022 (although this includes a growing share of domestic visitors which tend to spend less than foreign visitors). The model used the prepandemic (2012/19) average annual growth rate in visits to QENP and BINP (7.3 percent and 12.9 percent, respectively) to forecast the increase in tourists for 2023/24 and the likely impacts on the local economy. This implied an increase of 4,945 tourists to QENP and 4,209 tourists to BINP. At the average spending per tourist estimated from the survey data, these increases would add $2.5 million and $5.03 million ($2.09 million net of entry fees) in tourist spending at QENP and BINP, respectively. Entering this average additional tourist spending onto the dashboard for tourism impacts, the model gives the local economy impacts shown in figures 5.7 and 5.8.3 For QENP (figure 5.7), the overall impact predicted for the local economy was a $14.2 million increase in local production, a $5.1 million increase in income to households, and a $2.1 million increase in labor income. For BINP (figure 5.8), the impact on the local economy would be a $6.4 million increase in local production, a $2 million increase in income to households, and a $0.8 million increase in labor income. Keeping park entry fees fixed at BINP, a rise in local tourist spending would create an additional $9 million increase in local production, a $2.8 million increase in income to households, and a $1.2 million increase in labor income.4 1 Although QENP received 95,340 visitors in 2022, for purposes of this study, the authors only counted the volume of foreign nonresident, foreign resident, and East African Community resident visitors which totaled 67,736. This is because the remaining number of visitors to the park are students whose expenditure tends to be very low in comparison with tourists 2 Find Ministry of Tourism, Wildlife and Antiquities 2021 statistics at https://docs.google.com/spreadsheets/d/e/2PACX-1vS PEGocHS8SzrstAoSFXREW5Aq5IFwbgANFilmPKOZWNr9SVcyIOE1WoXCstGo4cTSMfDeUSHhw4nR6/pubhtml. 3 Since the model is linear, the overall multipliers are the same as the estimations of the impact of an additional $100 in tourist spending, but the magnitudes are different. 4 Policy makers could also estimate negative impacts or tourism losses by entering a negative value into the tourist spending window. Graphs would show negative magnitudes and impacts for actors in the local economy. 44 Measuring the Local Economic Impacts of Nature-Based Tourism in Uganda Queen Elizabeth National Park FIGURE 5.7 Effects of a $2.5 Million Increase in Tourist Spending on the Local Economy Around QENP How much tourist spending ($) do you want to simulate? 2500 Effects of this tourism spending... D. ...ON A. ...ON PRODUCTION B. ...ON INCOMES C. ...ON LABOR INCOME COMMUNITY AND PARK EARNINGS 941.7 942.6 162.04 6,755.8 4,923.9 Additional Production Value ($) Additional Labor Income ($) Additional Income ($) Additional Income ($) 4,723.9 1,551.4 201.6 35.1 595.5 56.6 373.9 183.1 143.5 Tourism Restaurants Lodges Agricultural Non- Fish Nonpoor Poor Female Male Female Male Nonpoor Poor agricultural unskilled unskilled skilled skilled Tourism-related activities Nontourism-related activities Households Labor categories Source: World Bank data. Bwindi Impenetrable National Park FIGURE 5.8A Effects of a $5.03 Million Increase in Tourist Spending on the Local Economy Around BINP How much tourist spending ($) do you want to simulate? 5030 Effects of this tourism spending... A. ...ON PRODUCTION B. ...ON INCOMES C. ...ON LABOR INCOME 276.99 1,717.87 260.32 2,890.98 Additional Production Value ($) Additional Labor Income ($) Additional Income ($) 147.19 1,625.96 131.06 1,068.14 566.86 248.54 115.33 95.11 Tourism Restaurants Lodges Agricultural Non- Fish Nonpoor Poor Female Male Female Male agricultural unskilled unskilled skilled skilled Tourism-related activities Nontourism-related activities Households Labor categories Source: World Bank data. 45 Measuring the Local Economic Impacts of Nature-Based Tourism in Uganda FIGURE 5.8B Effects of a $2.09 Million Increase (Net of Park Entry Fees) in Tourist Spending on the Local Economy Around BINP How much tourist spending ($) do you want to simulate? 2090 Effects of this tourism spending... A. ...ON PRODUCTION B. ...ON INCOMES C. ...ON LABOR INCOME 4,077.54 390.68 2,422.95 367.17 Additional Production Value ($) Additional Labor Income ($) Additional Income ($) 2,293.31 207.61 184.85 1,506.54 799.52 350.56 164.08 134.14 Tourism Restaurants Lodges Agricultural Non- Fish Nonpoor Poor Female Male Female Male agricultural unskilled unskilled skilled skilled Tourism-related activities Nontourism-related activities Households Labor categories Source: World Bank data. 46 SECTION 6 Testing the Robustness of Results and Analyzing the Differences in Multipliers Between the Two Protected Areas 47 Queen Elizabeth National Park. Photo credit: Kylie Nicholson / Shutterstock Measuring the Local Economic Impacts of Nature-Based Tourism in Uganda The methodology should reasonably represent the local economies, including the actors within them. One way to check this is to ask whether modelled per capita incomes correspond with what they would be expected to be. The SAM and DCIs were used to predict per capita incomes of poor and nonpoor households at the two sites. They are shown in table 6.1. At both sites, poor households have per capita incomes substantially less than $1 per day, whereas nonpoor households have cash incomes more than $1 per day. The per capita incomes of both household groups are slightly lower, on average, near BINP. At both parks, the impacts of tourist spending are considerably larger on nontourism activities, particularly nonagricultural businesses and agricultural production, than on tourism activities. These findings highlight the importance of looking beyond tourism to evaluate the impacts of tourism on local economies. Impacts are shaped and magnified by local market linkages. When visitors spend money on tourism activities, at lodges, or in restaurants, this stimulates tourism businesses’ demand for locally produced goods and services, grows wages, and profits. These wages and profits flow into households, which spend this income in local businesses and create new rounds of impacts. As the cash created directly or indirectly by tourism ripples through local economies, it creates production, income, wage, and profit multipliers. Studies of tourism that ignore nontourism activities and households miss many, if not most, of these impacts. For example, tourists rarely purchase food directly from farmers, yet at both Ugandan parks, an additional dollar of tourist spending has a large positive effect on local farm sales. Also, tourists are not spending much on local retail, services, or production at either national park, but these businesses have the largest benefits from indirect effects of tourist spending (table 5.1). TABLE 6.1 Incomes in the Local Economies Around Each Park QUEEN ELIZABETH NATIONAL PARK HOUSEHOLD GROUP POPULATION TOTAL INCOME PER CAPITA INCOME $ PER DAY Poor 70,775 $9,565,000 $135 $0.37 Nonpoor 529,525 $272,421,000 $514 $1.41 BWINDI IMPENETRABLE NATIONAL PARK HOUSEHOLD GROUP POPULATION TOTAL INCOME PER CAPITA INCOME $ PER DAY Poor 29,453 $3,311,000 $112 $0.31 Nonpoor 47,447 $24,103,228 $508 $1.39 Source: World Bank data. 48 Measuring the Local Economic Impacts of Nature-Based Tourism in Uganda In the tourist spending simulations presented earlier (and those in section 7), the largest production impacts are on nonagricultural sectors. This reflects the importance of nonagricultural production activities in these local economies. Most intermediate inputs purchased from local businesses are nonagricultural goods and services, and households—particularly nonpoor ones—spend more of their income on nonfood than food items. Most wage and profit incomes flow to nonpoor households. This is an expected result because households are classified by income level, and nearly all incomes come from wages or profits. The small share of income gains to poor households reflects their lack of access to local formal jobs and capital. A summary of how wages and profits are channeled into the two household groups at both parks follows in table 6.2. It shows that, at QENP, only 1.9–7.2 percent of incomes flow to poor households. Poor households get larger shares of factor incomes at BINP; however, they are still low compared with the shares going to nonpoor households, especially in the case of profits. The total income and production multipliers are larger at QENP than BINP. It is notable that an additional dollar of tourist spending increases local household income by more than one dollar at QENP. Several factors shape the size of these multipliers. Foremost among these, in the case of BINP, is the high price visitors pay for gorilla permits, and the leakage of this money from the local economy as it is remitted to Government. When the multipliers net of park entry fees are estimated, they increase substantially, but are still less than at QENP. This is because the local economy surrounding BINP is less developed, likely due to its remote setting and rugged terrain, which means more goods and services come from outside markets. Local purchases contribute to multipliers by circulating more cash within the local economy. Purchases from outside markets (which can be thought of as “imports” into the local economy) shift the multiplier effect from local to outside economies. TABLE 6.2 Wage and Profit Flows to Households PARK AND FACTOR PAYMENTS HOUSEHOLD GROUP WAGES BY WORKER GROUP PROFITS MALE UNSKILLED MALE SKILLED FEMALE UNSKILLED FEMALE SKILLED Queen Elizabeth National Park Poor $1,745,000 $874,000 $649,000 $110,000 $3,391,000 Nonpoor $45,099,000 $45,155,000 $8,391,000 $1,663,000 $133,299,000 Percentage to Poor 3.7% 1.9% 7.2% 6.2% 2.5% Bwindi Impenetrable National Park Poor $774,000 $459,000 $603,000 $611,000 $768,000 Nonpoor $2,680,000 $3,178,000 $1,043,000 $1,235,000 $14,327,000 Percentage to Poor 22.4% 12.6% 36.6% 33.1% 5.1% Source: World Bank data. 49 Connections with outside markets through trade can offer many advantages for local economies. For example, it can provide local producers with access to a larger market for the goods they produce, which can be profitable if they are able to compete. It can also give consumers access to lower prices for goods produced more cheaply in other places. Nevertheless, at locations where external markets satisfy a large part of local demand, the local multiplier effects of tourism generally tend to be smaller, because purchasing goods from other places causes income to “leak out” of the local economy. Leakages are larger at BINP, as shown in table 6.3, and this explains the smaller multipliers there. TABLE 6.3 Leakage Share from Production Sectors and Households PARK AND PRODUCTION SECTOR/ACTIVITY HOUSEHOLD GROUP PROTECTED COMMUNITY EXTERNAL ACCOUNT AREA REVENUE SHARING AGRICULTURE TOURISM NONAGRICULTURE FISH RESTAURANTS LODGES POOR NONPOOR Queen Elizabeth National Park G $1,676,000 $401,000 $6,498,000 $458,000 $150,000 $2,158,000 $1,987 $2,336,000 - ROW $24,025,000 $5,050,000 $54,900,000 $5,040,000 $614,000 $5,483,000 $102,000 $8,436,000 $575,000 - TotalExp $241,683,000 $17,693,000 $340,558,000 $18,031,000 $8,229,000 $46,095,000 $9,565,000 $272,421,000 $6,778,000 $1,468,000 Leakage share 10.6% 30.8% 18.0% 30.5% 9.3% 16.6% 1.1% 4.0% 8.5% 0.0% Bwindi Impenetrable National Park G $763,000 $130,000 $883,000 - $20,000 $392,000 - $74,759 $41,131,000 - ROW $2,395,000 $768,000 $15,149,000 $1,244,000 $167,000 $3,140,000 $146,000 $646,763 $129,000 - TotalExp $21,900,000 $7,007,000 $38,724,000 $1,244,000 $1,454,000 $13,203,000 $3,311,000 $24,103,228 $43,611,000 $822,000 Leakage share 14.4% 12.8% 41.4% 100.0% 12.9% 26.8% 4.4% 3.0% 94.6% 0.0% Source: World Bank data. SECTION 7 Using the Model to Simulate Changes in Park and Community Revenue Sharing Spending 51 Queen Elizabeth National Park. Photo credit: Jane Rix / Shutterstock Measuring the Local Economic Impacts of Nature-Based Tourism in Uganda Impacts of Changes in Park and Community Revenue Sharing Governments generally decide on protected area budgets and the share (if any) of park revenue going to community revenue sharing. In some cases, the government subsidizes protected areas by giving them budgets that exceed the entrance fees paid by park visitors. In others, entrance fees exceed the park budget, and some of the park revenue is used to subsidize other parks or nonpark spending. UWA, like many protected area authorities globally, consolidates all revenue received from all protected areas to finance operations for the entire park system. Revenue from park entrance fees and gorilla permits exceeded park budgets at both Queen Elizabeth National Park and Bwindi Impenetrable National Park, and 20 percent of park entrance fees and $10 from every gorilla permit spent on community revenue sharing. While section 5 reported on simulations of increases in tourist spending, the model can also be used to simulate impacts of other types of spending, such as park spending and community revenue sharing spending. To illustrate the use of this feature, the dashboard was used to simulate the impacts of $100 of park and community revenue sharing spending on the local economy around each park, and summarize the impacts on production, household incomes, and wages. At QENP, a $100 increase in park spending (or park budget) leads to a $212 increase in local GDP, or an income multiplier of 2.12 for the local economy (with impacts disaggregated in figure 7.1). In comparison, a $100 increase in community revenue sharing leads to a local GDP multiplier of 2.45 (figure 7.2). At BINP, a $100 increase in park spending leads to a $5.83 increase in local GDP, or a local income multiplier of 0.06 (figure 7.3). A $100 increase in community revenue sharing leads to a local GDP multiplier of 1.2, with impacts disaggregated in figure 7.4. Community revenue sharing income multipliers greater than one signify that economic gains exceed the amount transferred to communities around the parks. In contrast, park revenue spending has a small impact on local incomes at BINP: the local GDP multiplier is only 0.06. This is because most of the revenue is from the gorilla permit fee, most of which is transferred out of the local economy to the UWA consolidated fund. In QENP, on the other hand, park revenue spending has a local GDP multiplier of 2.12, which means that park spending favors local labor and local goods and services, making the multiplier greater than one. This is independent of any additional tourist demand that might result from increased park spending. Thus, park spending to improve the facilities and wages of local park officials generates positive impacts on the local economy, and if it attracts new tourists, there may be additional impacts due to additional demand. The results presented in this report are for single simulations. However, the model can also be used to simulate impacts of changes in more than one variable—for example, an increase in park budget that stimulates new tourism to a protected area. It can also be used to perform a social cost-benefit analysis—that is, one that includes spillover effects in local economies. This would require making some assumptions, ideally backed up by data or experience, about changes in tourist revenue that might result from a larger park budget (for example, one that enables the park to accommodate more visitors). 52 Measuring the Local Economic Impacts of Nature-Based Tourism in Uganda FIGURE 7.1 Effects of a $100 Increase in Park Spending on the Local Economy Around QENP Effects of this park spending on... A. ...ON PRODUCTION B. ...ON INCOMES C. ...ON LABOR INCOME 319.74 204.95 44.17 36.62 Additional Production Value ($) Additional Labor Income ($) Additional Income ($) 195.47 26.52 14.69 3.52 6.97 0 4.53 0 Tourism Restaurants Lodges Agricultural Non- Fish Nonpoor Poor Female Male Female Male agricultural unskilled unskilled skilled skilled Tourism-related activities Nontourism-related activities Households Labor categories Source: World Bank data. FIGURE 7.2 Effects of a $100 Increase in Community Revenue Sharing Spending on the Local Economy Around QENP Effects of this community revenue sharing spending... A. ...ON PRODUCTION B. ...ON INCOMES C. ...ON LABOR INCOME 344.93 237.48 70.08 Additional Production Value ($) Additional Labor Income ($) Additional Income ($) 215.9 42.79 21.37 16.05 7.71 0 5.25 0 0.04 Tourism Restaurants Lodges Agricultural Non- Fish Nonpoor Poor Female Male Female Male agricultural unskilled unskilled skilled skilled Tourism-related activities Nontourism-related activities Households Labor categories Source: World Bank data. 53 Measuring the Local Economic Impacts of Nature-Based Tourism in Uganda FIGURE 7.3 Effects of a $100 Increase in Park Spending on the Local Economy Around BINP Effects of this park spending... A. ...ON PRODUCTION B. ...ON INCOMES C. ...ON LABOR INCOME 10.38 4.78 1.55 Additional Production Value ($) Additional Labor Income ($) Additional Income ($) 0.8 4.73 0.68 1.05 0.26 0 0.03 0 0.22 Tourism Restaurants Lodges Agricultural Non- Fish Nonpoor Poor Female Male Female Male agricultural unskilled unskilled skilled skilled Tourism-related activities Nontourism-related activities Households Labor categories Source: World Bank data. FIGURE 7.4 Effects of a $100 Increase in Community Revenue Sharing Spending on the Local Economy Around BINP Effects of this community revenue sharing spending... A. ...ON PRODUCTION B. ...ON INCOMES C. ...ON LABOR INCOME 188.73 98.03 33.96 Additional Production Value ($) Additional Labor Income ($) Additional Income ($) 19.95 95.47 15.35 21.84 1.51 0 0.56 0 4.14 Tourism Restaurants Lodges Agricultural Non- Fish Nonpoor Poor Female Male Female Male agricultural unskilled unskilled skilled skilled Tourism-related activities Nontourism-related activities Households Labor categories Source: World Bank data. 54 SECTION 8 Using the Model to Simulate Complementary Interventions 55 Bwindi Impenetrable National Park. Photo credit: Travel Stock/Shutterstock Measuring the Local Economic Impacts of Nature-Based Tourism in Uganda Impacts of Increased Demand for Agricultural and Nonagricultural Goods The LEWIE-LITE model can be used to simulate changes in the demand for locally produced agricultural and nonagricultural goods. This is of interest if governments or development agencies wish to design complementary interventions to diversify and grow tourism impacts by increasing local sourcing of goods and services. To illustrate the use of this feature, the dashboard was used to simulate the impacts of a $100 increase in local demand for agricultural and nonagricultural goods on the local economy around each park, summarizing impacts on production, household incomes, and wages.1 Logically, the largest impacts are on the production sector targeted by the intervention (for example, on agricultural production, for programs that seek to increase the local demand for farm products). However, local linkages transmit impacts to other sectors, stimulating their sales as well. Higher agricultural production increases farms’ demand for intermediate inputs, labor, and capital. This creates new wage earnings and profits for households, especially nonpoor households. As households spend their new income, this stimulates local crop and noncrop production activities, which, in turn, creates additional rounds of production, income, and employment gains. Gross sales from nonagricultural activities rise nearly as much as agricultural sales when the demand for agricultural production increases. At Queen Elizabeth National Park, a $100 increase in demand for local agricultural production leads to a local GDP multiplier of $2.43 (with impacts disaggregated as shown in figure 8.1). In comparison, a $100 increase in demand for local nonagricultural production leads to a local GDP multiplier of $1.91 (figure 8.2). For the economy surrounding BINP, the $100 increase in demand for local agricultural production leads to a local GDP multiplier of $1.22 (figure 8.3) and a $100 increase in demand for local nonagricultural production leads to a local GDP multiplier of $0.59 (figure 8.4). In three out of the four cases, increasing local agricultural and nonagricultural production has a multiplier of greater that $1 on local income. The only case where it is less than $1 is an increase in nonagricultural production at BINP. The larger leakages in the local economy surrounding BINP, specifically in the nonagricultural sector, lead to this smaller multiplier effect. Even in this case, local GDP rises by almost 60 cents for every $1 increase in nonagricultural production. 1 There are no impacts on the park and community revenue sharing projects because there are no feedback effects from production activities to their budgets. 56 Measuring the Local Economic Impacts of Nature-Based Tourism in Uganda Queen Elizabeth National Park FIGURE 8.1 Effects of a $100 Increase in Local Agricultural Production on the Local Economy Around QENP Effects of this increase in local agricultural production on... A. ...ON PRODUCTION B. ...ON INCOMES C. ...ON LABOR INCOME 311.12 236.52 52.42 51.92 290.47 Additional Production Value ($) Additional Labor Income ($) Additional Income ($) 5.5 15.21 6.64 0 5.23 0 0.04 Tourism Restaurants Lodges Agricultural Non- Fish Nonpoor Poor Female Male Female Male agricultural unskilled unskilled skilled skilled Tourism-related activities Nontourism-related activities Households Labor categories Source: World Bank data. FIGURE 8.2 Effects of a $100 Increase in Local Nonagricultural Production on the Local Economy Around QENP Effects of increase in local nonagricultural production on... A. ...ON PRODUCTION B. ...ON INCOMES C. ...ON LABOR INCOME 368.29 185.25 39.71 37.18 Additional Production Value ($) Additional Labor Income ($) Additional Income ($) 196.65 6.94 16.16 5.33 0 4.09 0 0.03 Tourism Restaurants Lodges Agricultural Non- Fish Nonpoor Poor Female Male Female Male agricultural unskilled unskilled skilled skilled Tourism-related activities Nontourism-related activities Households Labor categories Source: World Bank data. 57 Measuring the Local Economic Impacts of Nature-Based Tourism in Uganda Bwindi Impenetrable National Park FIGURE 8.3 Effects of a $100 Increase in Local Agricultural Production on the Local Economy Around BINP Effects of this increase in local agricultural production on... A. ...ON PRODUCTION B. ...ON INCOMES C. ...ON LABOR INCOME 189.36 108.64 24.38 155.74 19.04 Additional Production Value ($) Additional Labor Income ($) Additional Income ($) 13.27 2.67 2.02 0 0.62 0 4.31 Tourism Restaurants Lodges Agricultural Non- Fish Nonpoor Poor Female Male Female Male agricultural unskilled unskilled skilled skilled Tourism-related activities Nontourism-related activities Households Labor categories Source: World Bank data. FIGURE 8.4 Effects of a $100 Increase in Local Nonagricultural Production on the Local Economy Around BINP Effects of this increase in local nonagricultural production on... A. ...ON PRODUCTION B. ...ON INCOMES C. ...ON LABOR INCOME 206.74 53.13 9.12 7.86 Additional Production Value ($) Additional Labor Income ($) Additional Income ($) 58.17 1.32 5.67 0.82 0 0.3 0 4.15 Tourism Restaurants Lodges Agricultural Non- Fish Nonpoor Poor Female Male Female Male agricultural unskilled unskilled skilled skilled Tourism-related activities Nontourism-related activities Households Labor categories Source: World Bank data. 58 Measuring the Local Economic Impacts of Nature-Based Tourism in Uganda Impacts of Increased Wage Earnings for Local Workers Another use of the tool could be to simulate impacts of changes in wage earnings for local workers. This is of interest if governments or development agencies wish to intervene to increase local employment, for example, by training and linking workers to tourism and/or nontourism activities. The impacts of a $100 increase in wage earnings for local female and male workers, unskilled and skilled, on the local economy around each park was simulated, summarizing impacts on production, household incomes, and wages.2 Logically, the largest impacts are on the worker group targeted by the intervention (for example, on unskilled female wage earnings, for programs that seek to increase employment opportunities for unskilled female workers). However, local linkages transmit impacts to the other three worker groups, raising their earnings as well. Higher wage earnings increase incomes, mostly in nonpoor households. As these households spend this income, this stimulates local production activities, which, in turn, creates additional rounds of production, income, and employment gains. At QENP, a $100 increase in wage earnings for either unskilled female workers or skilled female workers leads to a GDP multiplier of $3.06 for the local economy. Disaggregated impacts for unskilled female workers are shown in figure 8.5, and in figure 8.6 for skilled female workers. Similarly, a $100 increase in wage earnings for either unskilled male workers or skilled male workers leads to a GDP multiplier of $3.05 for the local economy, reporting practically the same result as for female workers. Disaggregated impacts for unskilled male workers are shown in figure 8.7, and in figure 8.8 for skilled male workers. Results from BINP show lower multipliers. Impacts of a $100 increase in wage earnings for unskilled female workers and skilled female workers are reflected in a local GDP multiplier of $1.84. Disaggregated impacts for unskilled female workers are shown in figure 8.9, and in figure 8.10 for skilled female workers. For male workers, a $100 increase in wage earnings for unskilled male workers and male skilled worker leads to a similar local GDP multiplier of $1.83. Disaggregated impacts for unskilled male workers are shown in figure 8.11, and in figure 8.12 for skilled male workers. 2 Impacts on the park and community revenue sharing are nil because there are no feedback effects from local wage earnings to their budgets. 59 Measuring the Local Economic Impacts of Nature-Based Tourism in Uganda Queen Elizabeth National Park FIGURE 8.5 Effects of a $100 Increase in Earnings for Unskilled Female Workers on the Local Economy Around QENP Effects of increase in earnings for unskilled female workers on... A. ...ON PRODUCTION B. ...ON INCOMES C. ...ON LABOR INCOME 321.02 293.09 106.08 234.47 Additional Production Value ($) Additional Labor Income ($) Additional Income ($) 43.27 41.61 15.94 12.87 6.48 0 0 0.05 Tourism Restaurants Lodges Agricultural Non- Fish Nonpoor Poor Female Male Female Male agricultural unskilled unskilled skilled skilled Tourism-related activities Nontourism-related activities Households Labor categories Source: World Bank data. FIGURE 8.6 Effects of a $100 Increase in Earnings for Skilled Female Workers on the Local Economy Around QENP Effects of this increase in earnings for skilled female workers on... A. ...ON PRODUCTION B. ...ON INCOMES C. ...ON LABOR INCOME 321.04 293.91 100.05 Additional Production Value ($) 234.19 Additional Labor Income ($) Additional Income ($) 43.24 41.57 15.95 6.08 11.91 6.5 0 0 Tourism Restaurants Lodges Agricultural Non- Fish Nonpoor Poor Female Male Female Male agricultural unskilled unskilled skilled skilled Tourism-related activities Nontourism-related activities Households Labor categories Source: World Bank data. 60 Measuring the Local Economic Impacts of Nature-Based Tourism in Uganda FIGURE 8.7 Effects of a $100 Increase in Earnings for Unskilled Male Workers on the Local Economy Around QENP Effects of this increase in earnings for unskilled male workers ... A. ...ON PRODUCTION B. ...ON INCOMES C. ...ON LABOR INCOME 143.15 321.09 296.04 233.48 Additional Production Value ($) Additional Labor Income ($) Additional Income ($) 41.47 9.4 6.08 6.54 15.96 0 0 0.05 Tourism Restaurants Lodges Agricultural Non- Fish Nonpoor Poor Female Male Female Male agricultural unskilled unskilled skilled skilled Tourism-related activities Nontourism-related activities Households Labor categories Source: World Bank data. FIGURE 8.8 Effects of a $100 Increase in Earnings for Skilled Male Workers on the Local Economy Around QENP Effects of this increase in earnings for skilled male workers ... A. ...ON PRODUCTION B. ...ON INCOMES C. ...ON LABOR INCOME 321.13 297.6 141.4 232.96 Additional Production Value ($) Additional Labor Income ($) Additional Income ($) 43.08 6.58 15.97 7.57 6.08 0 0 0.05 Tourism Restaurants Lodges Agricultural Non- Fish Nonpoor Poor Female Male Female Male agricultural unskilled unskilled skilled skilled Tourism-related activities Nontourism-related activities Households Labor categories Source: World Bank data. 61 Measuring the Local Economic Impacts of Nature-Based Tourism in Uganda Bwindi Impenetrable National Park FIGURE 8.9 Effects of a $100 Increase in Earnings for Unskilled Female Workers on the Local Economy Around BINP Effects of this increase in earnings for unskilled female workers on ... A. ...ON PRODUCTION B. ...ON INCOMES C. ...ON LABOR INCOME 177.76 138.26 101.59 Additional Production Value ($) Additional Labor Income ($) Additional Income ($) 111.6 45.32 14.58 13 0.79 4.08 1.58 0 0 Tourism Restaurants Lodges Agricultural Non- Fish Nonpoor Poor Female Male Female Male agricultural unskilled unskilled skilled skilled Tourism-related activities Nontourism-related activities Households Labor categories Source: World Bank data. FIGURE 8.10 Effects of a $100 Increase in Earnings for Skilled Female Workers on the Local Economy Around BINP Effects of this increase in earnings for skilled female workers on ... A. ...ON PRODUCTION B. ...ON INCOMES C. ...ON LABOR INCOME 178.01 141.73 101.59 101.58 Additional Production Value ($) Additional Labor Income ($) Additional Income ($) 111.2 41.8 14.54 12.98 0.81 4.08 0 0 Tourism Restaurants Lodges Agricultural Non- Fish Nonpoor Poor Female Male Female Male agricultural unskilled unskilled skilled skilled Tourism-related activities Nontourism-related activities Households Labor categories Source: World Bank data. 62 Measuring the Local Economic Impacts of Nature-Based Tourism in Uganda FIGURE 8.11 Effects of a $100 Increase in Earnings for Unskilled Male Workers on the Local Economy Around BINP Effects of this increase in earnings to unskilled male workers on ... A. ...ON PRODUCTION B. ...ON INCOMES C. ...ON LABOR INCOME 178.75 152.05 114.4 Additional Production Value ($) Additional Labor Income ($) Additional Income ($) 110.15 31.04 12.9 0.87 4.09 1.57 1.58 0 0 Tourism Restaurants Lodges Agricultural Non- Fish Nonpoor Poor Female Male Female Male agricultural unskilled unskilled skilled skilled Tourism-related activities Nontourism-related activities Households Labor categories Source: World Bank data. FIGURE 8.12 Effects of a $100 Increase in Earnings for Skilled Male Workers on the Local Economy Around BINP Effects of this increase in earnings to skilled male workers on ... A. ...ON PRODUCTION B. ...ON INCOMES C. ...ON LABOR INCOME 179.43 161.49 112.83 Additional Production Value ($) Additional Labor Income ($) Additional Income ($) 109.15 21.2 14.28 0.92 4.09 1.56 1.58 0 0 Tourism Restaurants Lodges Agricultural Non- Fish Nonpoor Poor Female Male Female Male agricultural unskilled unskilled skilled skilled Tourism-related activities Nontourism-related activities Households Labor categories Source: World Bank data. 63 Conclusions and Recommendations 64 Bwindi Impenetrable National Park. Photo credit: Emily Marie Wilson / Shutterstock Measuring the Local Economic Impacts of Nature-Based Tourism in Uganda The findings from these simulations underline the importance of looking beyond the activities and people most directly affected by protected area tourism to also consider indirect impacts on the local economies. For example, when a visitor spends money in a tourism business, lodge, or restaurant, markets transmit the impacts of this spending to other businesses, workers, and owners of capital. Household incomes from wages and profits increase, and most households spend most of their income locally. This creates additional rounds of impacts in local economies. The LEWIE-LITE model captures the total estimated impacts—both direct and indirect—of tourism, park and community revenue sharing spending, and complementary development interventions in local economies. Main Results and Findings Studying and simulating the local economic benefits of tourism offers valuable insights into the sector’s economic contributions to communities, including women, living around protected areas Even before using the model, the survey data provides interesting results that may be missed in other studies of the tourism sector. For example, up to 12 percent of employees across all tourism-related businesses (hotels, restaurants, and other tourism businesses) around Queen Elizabeth National Park and Bwindi Impenetrable National Park are women; however, in nontourism-related businesses, this figure is approximately 4 percent. This reinforces the global finding that tourism is a valuable job entry point for women as compared to other economic sectors. Simulations using the model reveal tourism benefits that go beyond the direct impacts considered by most tourism impact studies. Production multipliers for QENP and BINP are greater than 1 (5.67 and 1.2, respectively). BINP’s production multiplier rises to 4.29 if the model only includes local tourist spending and nets out the high entry fees which go to the national treasury. This means that local production expands by more than one dollar per dollar of increase in tourist spending. This impact is marked in the nonagricultural sector, which includes local grocery stores, salons, taxis, and small-scale production like carpentry shops. Local agriculture in both parks is also stimulated beyond the value of the initial tourist dollar as farmers supply food to restaurants and hotels as well as to households whose income increases thanks to protected area tourist spending. Local income or GDP in the economies surrounding protected areas rises as a result of tourism Local income rises by more than one dollar per dollar spent by tourists at QENP and by 37 cents at BINP. The income multiplier at QENP is 2.03, of which 1.97 goes to nonpoor households and 0.06 goes to poor households. The smaller income multiplier at BINP is mainly because most income from park entry fees and gorilla permits remits to UWA, rather than staying in the local economy. Net of entry fees, the income multiplier to the local economy surrounding BINP rises as high as 1.33. The 37 cent multiplier is not insignificant as it opens the possibility of creating additional benefits through complementary interventions that increase local sourcing of goods and services and employment of local workers, similarly to businesses at QENP. 65 Measuring the Local Economic Impacts of Nature-Based Tourism in Uganda Poor households are less employed in both tourism and nontourism sectors of the local economies surrounding both parks. They receive only 3 percent (QENP) and 13.52 percent (BINP) of total tourism benefits, and this small share reflects their lack of access to local formal jobs and capital. Tourism growth can lead to economic benefits that ripple across local economies and generate high returns on investments for protected area management budgets Using the income multipliers, it is possible to forecast the economic impacts of more tourists. One additional protected area tourist generates an average of $1,028 during their stay at QENP and $442 at BINP (or $660 in local GDP for BINP net of park entry fees). This includes the direct impact of their spending and indirect impacts in the local economies surrounding the protected areas. Multiplying these per-tourist impacts by the number of tourists which visited these protected areas in 2022 (67,736 in QENP and 32,628 in BINP), the total income generated by tourists was approximately $69.6 million at QENP and $14.4 million for BINP—or $21.5 million for BINP outside of park entry fees. These amounts far exceed the current park budgets at these sites, which are $3 million (QENP) and $2.3 million (BINP), and provide revenue for other parks and conservation areas, leading to multiplier impacts in those local economies. If the number of tourists to the parks increases at prepandemic annual growth rates, the protected areas will generate an additional income to households of $5.1 million and $2 million (or $2.8 million net of entry fees), respectively, for the local economies of QENP and BINP. Parks support local economies through their spending on local labor and local goods and services. Every $100 increase in park spending adds $212 to the GDP of communities around QENP and $5.83 to BINP. Park revenue spending has a small impact on local GDP at BINP because most of the park revenue (almost 96 percent) is transferred out of the local economy to UWA. Net of park fees, a $100 increase in park spending would result in $108 increase in GDP in BINP. Parks also support local economies through community revenue sharing. A $100 increase in community revenue sharing spending leads to a local GDP gain of $245 at QENP and $120 at BINP. Furthermore, income multipliers from community revenue sharing spending are also greater than one: every additional dollar spent by the parks increases the benefit to the local economy by more than a dollar. This is in addition to the economic benefits created by tourist spending. These simulations demonstrate the high economic return on government investments in protected areas. Complementary interventions around protected areas can magnify these impacts LEWIE-LITE can simulate impacts of local sourcing of agricultural, fish, and nonagricultural goods and services, and employment, for example through training and linking workers to tourism and/or nontourism activities. Simulations of the impact of $100 of additional wages for female skilled labor generated an additional $306 in local income to households in QENP and $184 in BINP. If a program was to generate additional female employment in the local economy, the multiplier impact on local incomes would be large, in addition to the impact of tourism. The model can be used to calculate multipliers generated by these complementary policies as well as the distribution of impacts across businesses, worker groups, and households. 66 Measuring the Local Economic Impacts of Nature-Based Tourism in Uganda Study Limitations LEWIE-LITE is a useful tool; however, there are some limitations to the model. Further guidance on how LEWIE-LITE works and how it can be used is included in appendix B. The model can explore potential scenarios of future visitation or tourism spend to inform policy discussions, but it cannot design tourism policies or programs It is best to view LEWIE-LITE as a tool to explore the local economic benefits of protected area tourism that are explored in this report, for example, the impacts of increases in tourism and tourist spending, both overall and for different sectors, workers, and household groups. For example, in its current form, the model can show what impacts to expect if the number of tourists or tourist spending increased. It can also be used to simulate local economy impacts of having fewer tourists but higher tourist spending and vice versa—or of increasing both the number of tourists and how much they spend. However, it does not explain how to make more tourists come or spend more, or how to ensure the protected area is able to sustainably accommodate increased tourism and tourist spending. That is a policy design choice, for which it would be necessary to also model the likely effects of different tourism policies on the number of tourists and amount of tourist spending. This goes beyond what LEWIE-LITE is set up to do. The simplified nature of the model supports scalability but brings technical tradeoffs The LEWIE-LITE model cannot explore price changes on local-economy outcomes. It is a fixed-price multiplier model, based on a SAM created for a local economy. Examining price impacts directly would require a more comprehensive LEWIE approach (World Bank 2021). This is a tradeoff that must be made to create a relatively simple and scalable LEWIE-LITE platform. A basic assumption of LEWIE-LITE is that if the local demand for goods or services (including factors like labor) increases, the supply will increase to meet this demand. This assumption is more defensible for an economy with high unemployment and few constraints on local production. High unemployment is common in poor countries, so the availability of workers is not likely to be a constraint. However, worker skills may be. Other production constraints depend on a variety of factors, including the availability of land to grow crops, technological limitations, access to inputs and capital, and market transaction costs. If these constraints are present, it is important to address them as part of tourism development projects via interventions like job training. The model also does not capture nonlinearities in production activities, household spending, and so on. SAM multiplier models are linear by nature. This is the second major tradeoff of using LEWIE-LITE versus more comprehensive LEWIE modeling. Nonlinearities occur if there are diminishing marginal returns to inputs in local production due to technological constraints, or if household spending on goods and services changes as household incomes change. If nonlinearities are important, the LEWIE-LITE model is likely to be better at assessing impacts of relatively small changes (for example, in tourist numbers and spending) than larger ones. 67 Measuring the Local Economic Impacts of Nature-Based Tourism in Uganda Recommendations Based on the above analysis, this report offers several recommendations for policy makers to promote Uganda’s nature-based tourism sector and maximize its contributions to development, poverty reduction, and biodiversity conservation. Strengthening the local economic benefits of protected area tourism 1. Strengthen benefit sharing to reach poorer households Uganda is one of the few countries with a national revenue sharing program from protected areas, in which 20 percent of park entry fees and $10 from each gorilla permit are used to fund projects to address human-wildlife conflict, build local infrastructure, or support other socioeconomic programs. Benefit sharing is a mechanism through which local development can be financed directly from revenue generated by tourism. Sharing revenues from park entry fees with local communities can result in a greater than one- for-one dollar increase in local production, wages, and GDP. With the LEWIE-LITE analysis showing that poor households around QENP and BINP benefit less from tourism than nonpoor households, designing strategic interventions in the revenue sharing program that specifically target the poorest people can help make tourism benefits more equitable. 2. Promote local sourcing and local hiring The economic benefits of tourism are reduced in rural communities with less developed local economies in which goods and services are purchased in urban centers farther away. This means less money is retained in the local economy and smaller economic impacts are generated. Local linkages to sectors such as agriculture, fishing, and manufacturing could be strengthened so that multipliers increase. Using BINP as an example, if local businesses sourced more inputs locally, the production and income multipliers from tourism could be boosted. Because BINP is more isolated than QENP, it would be necessary to address common challenges that affect local sourcing for tourism, such as the quality and quantity of local goods, prices, infrastructure and transport, and communications with producers and vendors. The pilot’s findings prompt further studies to explore how to strengthen local market linkages in BINP and to grow the local economy to reduce leakages and maximize the impacts of tourism. 3. Strengthen employment, training, and entrepreneurship programs Tourism is a significant source of employment, particularly in rural areas, and LEWIE- LITE has shown it as a principal entry point into the job market around QENP and BINP, particularly for women. The analysis also found that nonlocal ownership and hiring are more common in tourism-related businesses than nontourism-related businesses. Tourism requires skills and training, which can increase the types of jobs accessible to workers and promote entrepreneurship. Skills training and entrepreneurship programs for workers from poor households could increase the tourism benefits that poor households capture. The private sector, such as tour operators and hotels, can play a role by providing and expanding employment programs, targeting women and workers from local, poorer communities or households, and offering attractive jobs and fair remuneration. 68 Measuring the Local Economic Impacts of Nature-Based Tourism in Uganda Recognizing the economic value of protected areas 4. Assess and convey economic benefits to policy developers and decision-makers The LEWIE-LITE model, with its simplified data collection and online dashboard, can help to close data gaps on the direct and indirect impacts of tourism on local economies in and around protected areas. It provides clear outputs to convey the economic value of these parks to the government, local communities, donors, and other stakeholders. It can do so in a way that is cost-effective, scalable, and understandable by policy makers and technical advisors from national parks and ministries of tourism, environment, and finance. It is also useful to private sector entities pursuing sustainable tourism and corporate social responsibility targets. The LEWIE-LITE model has been developed for tourism in data-poor contexts such as Uganda and delivers a product similar to the more complex LEWIE but more cheaply. Potential further applications This study identified several areas for potential investigation or further development of LEWIE-LITE. Technically, modifications to LEWIE-LITE and the dashboards, and a comparison of LEWIE-LITE and LEWIE could be undertaken to better understand the direct and indirect effects of tourism on local economies. This includes the use of the tool to estimate the costs and benefits of new investments in tourism at protected area sites. A short DCI could be filled out to detail new investments and an additional activity could be created for protected area tourism development projects in the LEWIE-LITE SAM. The tool could also be used to monitor developments in the tourism sector such as increased demand for ecotourism, or negative impacts like lost agricultural revenue from human-wildlife conflicts. Decreases in tourist spending could just as easily be simulated to quantify losses to local economies if there are negative shocks to tourism. By using the tool to simulate impacts of negative shocks, governments can be more prepared for them and actively try to avoid them (for example, through tourism crisis communications). This report provides snapshots that were modeled near the end of the high season for tourism at two protected areas in Uganda. Ideally, this exercise would be repeated multiple times in a year to capture economic impacts in high and low tourist seasons. Further, the exercise could be carried out with domestic tourists only, and followed by those from selected international markets to capture their different expenditure profiles and impacts these would have on local economies. In this way, the model could help government to develop, market and promote nature-based tourism. Finally, while the pilot showed that economic benefits from tourism flow to communities in two of the most popular national parks in Uganda, applying the tool to other protected areas in the country would allow comparison across sites and help obtain a national average of local production, employment, and income impacts. 69 References IFPRI (International Food Policy Research Institute). 2009. Social Accounting Matrices and Multiplier Analysis: An Introduction with Exercises. Food Security in Practice Technical Guide Series. Washington, DC: IFPRI. Laterite. 2023. PAT-SAM LEWIE Multiplier Data Collection in Uganda. Field report prepared for the World Bank, Kampala. Stone, Richard, and Alan Brown. 1962. “A Computable Model for Economic Growth.” In A Programme for Growth, Volume 1, edited by Richard Stone. Cambridge, England: Cambridge Growth Project. Taylor, Edward J., Mateusz J. Filipski, Mohamad Alloush, Anubhab Gupta, Ruben Irvin Rojas Valdes, and Ernesto Gonzalez-Estrada. 2016. “Economic Impact of Refugees.” Proceedings of the National Academy of the Sciences 113 (27): 7449– 7453. https://www.pnas.org/doi/abs/10.1073/pnas.1604566113. UBOS (Uganda Bureau of Statistics) and MoTWA (Ministry of Tourism, Wildlife and Antiquities). 2023. Uganda Tourism Satellite Account: Measuring the Contribution of Tourism to the Economy of Uganda. UBOS and MoTWA, Kampala. World Bank. 2019. Statistical and Economic Analysis of Uganda’s Tourism Expenditure and Motivation Survey 2019. Analytical report, Washington, DC. World Bank. 2021. Banking on Protected Areas: Promoting Sustainable Protected Area Tourism to Benefit Local Economies. Report, Washington, DC.  Zhu, Heng, Anubhab Gupta, Mateusz Filipski, Jaakko Valli, Ernesto Gonzalez- Estrada, and Edward J. Taylor. 2023. “Economic Impact of Giving Land to Refugees.” American Journal of Agricultural Economics 106 (1): 226–251. https:// onlinelibrary.wiley.com/doi/full/10.1111/ajae.12371#pane-pcw-references. 70 Measuring the Local Economic Impacts of Nature-Based Tourism in Uganda APPENDIX A. Questionnaires Park Entry PARK RECORDS 2022 2023 UNIT Tourists number       What was/is the number of multi-day tourists (adults and children > 5) in the     number park for all of 2022 (2023) What was/is the number of single-day visitors (adults and children > 5, no     number lodging) in the park for all of 2022 (2023) Park budget for 2022 (2023)       What was the total budget for 2022 (2023)     amount What was the total amount received for park entry fees for 2022 (2023)     amount Share of budget       How much of the 2022 (2023) budget went to:       Salaries and wages for male unskilled workers     amount (for example, maintenance workers) Salaries and wages for female unskilled workers     amount (for example, maintenance workers) Salaries and wages for male skilled workers     amount (for example, some guides, wardens, ticket sales, admin) Salaries and wages for female skilled workers     amount (for example, some guides, wardens, ticket sales, admin) Payment of rents on land, buildings, and so on     amount Locally produced agricultural products     amount (for example, fruits, vegetables, meats) Services (for example, laundry, maintenance, construction, repairs) from     amount local providers Purchases from local stores and other businesses     amount Purchases made outside the local economy     amount 71 Measuring the Local Economic Impacts of Nature-Based Tourism in Uganda Community Revenue Sharing COMMUNITY REVENUE SHARING 2022 2023 UNIT What was/is the total annual community revenue sharing budget for 2022     amount (2023) Community revenue sharing budget for 2022 (2023)       What share of the 2022 (2023) community revenue went to:       Salaries and wages for male unskilled workers     % (for example, maintenance workers) Salaries and wages for female unskilled workers     % (for example, maintenance workers) Salaries and wages for male skilled workers     % (for example, guides, wardens, ticket sales, admin) Salaries and wages for female skilled workers     % (for example, guides, wardens, ticket sales, admin) Payment of rents on land, buildings, and so on     % Locally produced agricultural products     % (for example, fruits, vegetables, meats) Services (for example, laundry, maintenance, construction, repairs) from     % local providers Purchases from local stores and other businesses     % Purchases made outside the local economy     % Other     % Please specify the other expense     text Total community revenue sharing budget for 2022     100% 72 Measuring the Local Economic Impacts of Nature-Based Tourism in Uganda Tourists TOURISTS SURVEY   UNIT Tourist identification     What is your gender?   text What is your age range?   text What is your nationality?   text If more than one, please specify the other country   text What is your country of residence?   text If more than one, please specify the other country   text Are you a foreigner, foreign resident of Uganda?   text Expected spending     Are you paying for this trip to {park_name} as part of a package tour (hotel/tour/food and so on)?   text Package fee     Can you estimate your package cost in:     What is the total cost of this package?   amount In your estimate, what is the value of the package that pertains to just {park_name} (lodging/food/   amount park fees/other local activities)? Number of days/ nights     What is the number of days that you’re expecting to stay in {park_name}?   number of days What is the number of nights that you’re expecting to stay in {park_name}?   number of nights Expected spending     In what currency can you estimate your local tourism expenses?   text What is the price per room (including taxes, resort and other fees)?   amount In what category of hotel did you stay?   number Please specify the other category   text Park entry in QENP   amount Park entry in BINP   amount Hotel shops/other hotel amenities   amount Local restaurants (food and drink)   amount Guides and tours   amount Tourist equipment rental and purchases   amount Souvenir/handicrafts   amount Retail shops, local markets, and so on   amount Local transportation (taxi, boda boda, and so on)   amount Other   amount 73 Measuring the Local Economic Impacts of Nature-Based Tourism in Uganda TOURISTS SURVEY   UNIT Please specify the other spending   text Tour Operators (based in Kampala) TOUR OPERATORS SURVEY (PACKAGE TOUR)   UNIT In what currency can you estimate your package tour prices?   text In your estimate, what is the value of the package that pertains to just this park (lodging/food/park   amount fees/other local activities)? Tour operators shares     About what percentage of your expenses for the package tour for ${park_name} goes to:     Hotel room and meals   % Park entry   % Local restaurants (food and drink)   % Guides and tours   % Tourist equipment rental   % Souvenir/handicrafts   % Retail shops, local markets   % Hotel shops/other hotel amenities   % Local transportation   % Other   % Please specify the other category   text After all your expenses, about what percentage of your revenue for the package tour for {park_   % name} goes to savings/profits Tour operator total percentage   100% 74 Measuring the Local Economic Impacts of Nature-Based Tourism in Uganda Lodges HOTELS AND LODGES   UNIT About what percentage of your monthly spending for the hotel or lodge goes to:     Salaries and wages for male unskilled workers   % Salaries and wages for male skilled workers   % Salaries and wages for female unskilled workers   % Salaries and wages for female skilled workers   % Crop purchases from local farmers or animals or animal products from local ranchers   % Value of fish and fish parts purchased from local fishermen   % (only QENP) Local services (for example, laundry, maintenance, construction, and repairs from local providers)   % Tourism products and third-party tour operators   % Purchases from local stores and other businesses   % Purchases you make outside the local economy   % Local rent   % Local tax (plus concession fee if there is one)   % Taxes other   % Other lodge spending   % Please specify the other lodge spending   text Lodge total percentage   100% Other information     What percent of the lodge/hotel is locally owned?   % Share of wages paid to local workers   % After all your monthly expenses, about what percentage of your revenue {park_name} goes to   % savings/profits 75 Measuring the Local Economic Impacts of Nature-Based Tourism in Uganda Restaurants RESTAURANTS   UNIT About what percentage of your monthly spending for the restaurant goes to:     Salaries and wages for male unskilled workers (for example, bussers, dishwashers, others)   % Salaries and wages for female unskilled workers (for example, bussers, dishwashers, others)   % Salaries and wages for male skilled workers (for example, cooks, servers, admin)   % Salaries and wages for female skilled workers (for example, cooks, servers, admin)   % Crop purchases from local farmers or animals or animal products from local ranchers   % Value of fish and fish parts purchased from local fishermen   %​ (only QENP) Services (for example, laundry, maintenance, construction, and repairs) from local providers   % Purchases from local stores and other businesses   % About what percentage of your monthly costs are purchases you make outside the local economy?   % Local rent   % Local taxes   % Other taxes   % Other restaurant spending (specify)   % Please specify the other restaurant spending   text Restaurant total percentage   100% Other information     Share of restaurants locally owned   % Share of wages paid to local workers   % After all your monthly expenses, about what percentage of your revenue {park_name} goes to   % savings/profits 76 Measuring the Local Economic Impacts of Nature-Based Tourism in Uganda Nonagricultural Producers (Tourism-Related Businesses) NONAGRICULTURAL PRODUCERS (TOURISM-RELATED BUSINESSES)   UNIT About what percentage of monthly spending by local nonfarm tourist businesses goes to:     Salaries and wages for male unskilled workers   % Salaries and wages for female unskilled workers   % Salaries and wages for male skilled workers (for example, machine operators, clerks, supervisors)   % Salaries and wages for female skilled workers (for example, machine operators, clerks,   % supervisors) Value of crop purchases from local farmers or animal products from local ranchers   % Value of fish and fish parts purchased from local fishermen   % (only QENP) Services (for example, machine maintenance, construction, and repairs) from local providers   % Purchases from local stores and other businesses   % About what percentage of your monthly costs are purchases you make outside the local economy,   % like merchandise (for stores) or supplies? Local rent   % Nonfarm tax local   % Nonfarm tax other   % Other nonagricultural tourism-related producers spending (specify)   % Please specify the other non-agricultural tourism-related producers spending   text Nonagricultural producers (tourism-related) total percentage   100% Other information     Share of businesses locally owned   % Share of wages paid to local workers   % After all your monthly expense,s about what percentage of your revenue {park_name} goes to   % savings/profits 77 Measuring the Local Economic Impacts of Nature-Based Tourism in Uganda Nonagricultural Producers (Retailers, Other Services, and Other Producers) NONAGRICULTURAL PRODUCERS (RETAILERS, SERVICES, AND PRODUCTION)   UNIT About what percentage of monthly spending by local nonfarm nontourist businesses goes to:     Salaries and wages for male unskilled workers   % Salaries and wages for female unskilled workers   % Salaries and wages for male skilled workers (for example, machine operators, clerks, supervisors)   % Salaries and wages for female skilled workers (for example, machine operators, clerks,   % supervisors) Value of crop purchases from local farmers or animal products from local ranchers   % Value of fish and fish parts purchased from local fishermen   % (only QENP) Services (for example, machine maintenance, construction, repairs) from local providers   % Purchases from local stores and other businesses   % About what percentage of your monthly costs are purchases you make outside the local economy,   % like merchandise (for stores) or supplies? Local rent   % Nonfarm local tax   % Nonfarm tax other   % Other nonagricultural producers’ nontourism-related spending (specify)   % Please specify the other nonagricultural nontourism-related producers spending   text Nonagricultural producers (nontourism) total percentage   100% Other information     Share of businesses locally owned   % Share of wages paid to local workers   % After all your monthly expenses, about what percentage of your revenue {park_name} goes to   % savings/profits 78 Measuring the Local Economic Impacts of Nature-Based Tourism in Uganda Agricultural Producers (Commercial Crop and Livestock Producers) COMMERCIAL FARMERS (CROP PRODUCERS AND LIVESTOCK)   UNIT About what percentage of monthly spending by farms goes to:     Salaries and wages for male unskilled workers   % Salaries and wages for female unskilled workers   % Salaries and wages for male skilled workers (for example, machine operators, supervisors)   % Salaries and wages for female skilled workers (for example, machine operators, supervisors)   % Land rate   % Value from own harvest, livestock or neighbors (for example, seedlings, manure excluding fish or   % fish parts) Value from own harvest, livestock or neighbors (only fish or fish parts)   % Value of animal feed from your own farm or neighbors (excluding fish or fish parts)   % Value of feed from fish or fish parts from your own farm or neighbors   % Services (for example, machine maintenance, construction, repairs, veterinarian) from local   % providers Purchases from local stores and other businesses for your farm   % About what percentage of your monthly costs are purchases of inputs outside the local economy,   % like fertilizer, commercial seed, feed, and chemicals? About what percentage of your monthly costs are purchases of inputs outside the local economy,   % like medicines or vet services from outside, materials? Farm local tax   % Farm tax other   % Other commercial farmers (crop producers) spending (specify)   % Please specify the other commercial farmers (crop producers) spending   text Commercial farmers (crop producers) total percentage   100% Other information     After all your monthly expenses, about what percentage of your revenue {park_name} goes to   % savings/profits Has your crop harvest been negatively impacted by the wildlife from the park?   text What is your estimate of the share of your monthly revenue that has been lost?   % Has your livestock stock been negatively impacted by the wildlife from the park?   text What is your estimate of the share of your monthly revenue that has been lost?   % 79 Measuring the Local Economic Impacts of Nature-Based Tourism in Uganda Resource Extraction Producers (Fishers) COMMERCIAL FISHERS   UNIT About what percentage of monthly spending by fishing operations goes to:     Salaries and wages for male unskilled workers   % Salaries and wages for female unskilled workers   % Salaries and wages for male skilled workers (for example, machine operators, supervisors)   % Salaries and wages for female skilled workers (for example, machine operators, supervisors)   % Crop purchases from local farmers or animals or animal products from local ranchers   % Local fish (for example, fishing bait purchased from other fishers, or if you provide your own bait,   % what share of your monthly spending would you have to use to buy supplies from others?) Services (for example, boat maintenance, construction, repairs) from local providers   % Purchases from local stores and other businesses (for example, nets, lines, hooks, bait)   % About what percentage of your monthly costs are purchases of inputs outside the local economy,   % like nets, lines, hooks, bait, or other supplies (whether imported or bought elsewhere? Fishing local tax   % Fishing tax other   % Other fishers spending (specify)   % Please specify the other fishers spending   text Fish total percentage   100% Other information     Share of fishing operations locally owned   % Share of wages paid to local workers in fishing   % After all your monthly expenses, about what percentage of your revenue {park_name} goes to   % savings/profits Has your fishing catch been negatively impacted by the wildlife from the park?   text What is your estimate of the share of your monthly revenue that has been lost?   % 80 Measuring the Local Economic Impacts of Nature-Based Tourism in Uganda Households1 HOUSEHOLDS   UNIT About what percentage of your monthly income comes from:     Salaries and wages earned by male unskilled workers in the household   % Salaries and wages earned by female unskilled workers in the household   % Salaries and wages earned by male skilled workers in the household   % Salaries and wages earned by female skilled workers in the household   % Profits from household-owned farms or businesses or renting property the household owns   % Migrant remittances (domestic and foreign)   % Amount you receive monthly from the government (pensions, social cash transfers, other)   % Other income source   % Please specify the other household spending   text Total percentage household income   100% About what percentage of your monthly spending goes to:     Buying food from local farmers or animals or animal products (eggs, milk, and so on) from local   % ranchers Value of fish and fish parts purchased from local fishermen   % Tourism products and third-party tour operators   % Things besides food that are sold by people or businesses in your community, including services   % Things you buy from businesses/households, and so on, in places outside your community   % Household local income tax payments   % Household other income tax payments   % Rent (if household rents its house; don’t include business rentals here)   % Other household spending   % Please specify the other household spending   text Household total expenditure percentage   100% Other information     Share of wages earned locally   % What is your household’s average monthly income in Ugandan Shilling   amount What is your household’s size?   number 1 For this LEWIE-LITE pilot in Uganda, household surveys were not implemented. Data was collected from the Living Standards Measurement Survey (LSMS). However, the questions contained in this survey were used as the basis to extract relevant information from the LSMS. 81 Measuring the Local Economic Impacts of Nature-Based Tourism in Uganda APPENDIX B. Questions and Answers about the LEWIE-LITE Model and Analysis This appendix outlines common questions and answers about the use of the LEWIE-LITE model and potential limitations of the model for analysis. Q: Can the model tell us how different tourist groups (for example, foreign versus domestic) affect the local economy? A: The current model does not differentiate between tourist groups. It focuses on an average park visitor. For example, if half of all visitors are foreign and half domestic, and foreigners spend an average of $1,000 while domestic visitors spend an average of $600, the average tourist in the model spends ½x$1000 + ½x$600 = $800. The model is used to quantify the likely impacts of this spending. If one wanted to know how impacts between the two groups differ, it would be necessary to extend the model to have both tourist groups in it. It is feasible to do this using the LEWIE-LITE framework, but it would require some additional work. It would also require collecting data on a large enough sample of visitors in each group to reliably estimate these spending differences. Q: Park fees can vary considerably from one protected area to another. How does this affect the local economy impacts of tourism? A: This depends crucially on where the park fees go. If they go to the central government treasury, they represent a leakage from the local economy. Leakages reduce local income, production, and employment multipliers by shifting benefits to other parts of the country. On the other hand, if the park authority spends entry fees to hire workers and purchase goods and services in the local economy, it creates local linkages and can create local income multipliers. Q: How does the current LEWIE-LITE model’s multiplier estimates reflect these differences in park fees? A: When the model calculates the multiplier effect of an additional dollar of tourist spending, it assumes that a share goes to park entry fees. This share is equal to the share of park fees in total spending by the average tourist. In the case of BINP, most park entry fees are sent to the central government (the Uganda Wildlife Authority). This results in smaller local multiplier effects. The tourist spending multipliers make sense if the goal of the project is to increase tourist spending by bringing in more tourists who pay park fees or encouraging tourists to stay longer (if park fees are collected daily). Q: Can the current LEWIE-LITE model be used to calculate multiplier effects of tourist spending net of park fees? A: Yes. The dashboard provides an adjustment factor to calculate all multipliers net of park fees simply by multiplying them by the adjustment factor. It might make sense to do this if the goal of a project is to encourage tourists to spend more money per day while at 82 Measuring the Local Economic Impacts of Nature-Based Tourism in Uganda the protected area, but without any change in park entry fees. The adjustment factor is 1/ (1-pfs), where pfs is the share of park fees in average tourist spending. Q: Apart from park entry fees, what can make local multipliers higher or lower? A: Multipliers are created as cash cycles through local economies. They tend to be higher in relatively isolated economies in which more of what households and businesses spend money on is supplied locally. In places that are integrated with outside markets, multipliers tend to be smaller because cash leaks out of the local economy through trade. Multipliers also depend critically on local production capacity. If production is limited, for example, in rugged mountainous areas like BINP, businesses and households, by necessity, rely more on trade with outside markets. At BINP, more of the food tourists consume and the workers that lodges hire are brought in from other parts of the country. In Uganda, local production is greater at QENP and money thus circulates more at QENP than BINP, creating larger multipliers at QENP. Q: Can the model be used to come up with practical policy options? A: Yes and no. It is important to distinguish what the model does from what policy makers do with it. In its current form, the dashboard offers some examples of potential impacts of changes in tourism patterns which could arise from certain policy options. It can inform but not develop practical policies. Q: Can the model tell us what skills would be needed to raise wage earnings for local workers, who should be targeted, and whether this would help poor households move into more skilled and better-paying jobs? Can it tell us whether there are structural reasons why local workers are unable to access higher-paying jobs? A: These are questions of policy and program design, and they go beyond what this LEWIE-LITE model can provide. The dashboard presents simulations of local economic impacts of changes in wage earnings of different worker groups. How one increases wages for local workers is a matter of program design. The model does not tell us how to design a job training program, who to target, or why some labor groups are unable to access higher-paying jobs. However, it does tell us how higher wages are likely to benefit the local economy and its different sectors, workers, and social groups. Q: Is it possible to simulate local economy impacts of specific employment programs? A: Yes and no. The key to modeling impacts of specific policies is figuring out what is simulated in each case. For example, if an employment program provides local workers with skills to get jobs in tourism so that their wages increase, the model will predict the local economy impacts of this program. If the choice is between two employment programs, one must specify how each program would affect employment and earnings of the worker group in question so that the model can simulate the local economy impacts of each. This might require modifying the existing model or using a more comprehensive LEWIE modeling approach. 83 Measuring the Local Economic Impacts of Nature-Based Tourism in Uganda Q: Can the model be used to estimate impacts of “buy local” policies that create more benefits for local farmers? A: Yes and no. The dashboard shows us the local economy benefits of increasing local farmers’ sales. This could be the result of connecting tourism facilities (for example, lodges and restaurants) with farmers, making farmers more productive, changing crop choices, changing the quality of what they produce, or—as is most certainly the case—all of these things. The model does not tell us how to design and implement programs to change the demand and supply of local farm goods, like connecting farmers with lodges and restaurants or providing them with access to new technologies. Again, these are policy design questions that would require additional work to model. This is similar for the local economy benefits of increasing sales of nonfarm goods and services. 84 Measuring the Local Economic Impacts of Nature-Based Tourism in Uganda APPENDIX C. Calculation of Multipliers The various income and production multipliers are derived from the social accounting matrix (SAM) and SAM multiplier matrix generated from the data collected around QENP and BINP. As an example, tables C.1 and C.2 show the SAM and resulting SAM multiplier matrix from QENP. In table C.1, the columns show expenditures of farm and nonfarm businesses, male and female skilled and unskilled labor, household groups, park expenditures and community revenue sharing spending, and spending outside the local economy. The rows show income from the same local economy actors. 85 TABLE C.1 Social Accounting Matrix of the Economy in and Surrounding QENP AGRICU TOURISM NONAGR. FISH LMUSK LMSK LFUSK LFSK K POOR NONPOOR RESTAURANTS LODGES TOURISTS PROT.​AREA COMREVSH G ROW TOTAL EXPENDITURES Agricultural 28212836 936194 119644250 460476 0 0 0 0 0 8446183 134466981 3138610 3266632 0 87226 10438 0 0 298669826 Tourism 0 0 0 0 0 0 0 0 0 0 0 0 602747 10552691 0 0 0 0 11155438 Nonagricultural 39040465 1930322 155827637 1359289 0 0 0 0 0 5905493 196768294 1469000 4439747 3232163 1326537 322851 0 0 411621797 Fish 2287527 62413 15166173 75270 0 0 0 0 0 0 2760845 505640 526264 0 0 0 0 0 21384132 LMUSK 29280348 219426 14201632 717280 0 0 0 0 0 0 0 97632 354899 0 3536 178491 0 0 45053244 LMSK 31720377 219426 10586671 53132 0 0 0 0 0 0 0 256284 981636 0 393 20876 0 0 43838795 LFUSK 0 303624 8520979 8855 0 0 0 0 0 0 0 79326 241633 0 498529 92899 0 0 9745846 LFSK 0 63787 0 0 0 0 0 0 0 0 0 67122 558777 0 100702 0 0 0 790388 K 65967928 2392529 50097214 1322545 0 0 0 0 0 140156 4877271 2118750 5009280 0 0 0 0 0 131925673 Poor 0 0 0 0 1678151 831952 699958 49161 3272357 0 0 0 0 0 0 0 0 1290353 7821933 Nonpoor 0 0 0 0 43375093 43006844 9045889 741226 128653316 0 0 0 0 0 0 0 0 17909299 242731667 Restaurants 0 0 0 0 0 0 0 0 0 0 7982528 0 0 1434138 0 0 0 0 9416667 Lodges 0 0 0 0 0 0 0 0 0 0 0 0 0 29946784 0 0 0 0 29946784 Tourists 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 48293551 48293551 Protected area 0 0 0 0 0 0 0 0 0 0 0 0 0 3127775 0 0 0 0 3127775 ComRevSh 0 0 0 0 0 0 0 0 0 0 0 0 0 0 625555 0 0 0 625555 G 1525018 200677 8707132 146113 0 0 0 0 0 3120 3096331 155375 896770 0 0 0 0 0 14730536 ROW 21858593 2527575 73560253 1607238 0 0 0 0 0 155549 11183547 635625 2278678 0 245128 0 14730536 0 128782723 Total 219893093 8855973 456311942 5750197 45053244 43838795 9745846 790388 131925673 14650502 361135798 8523364 19157064 48293551 2887606 625555 14730536 67493203 0 expenditures Source: World Bank data (Dashboard for LEWIE-LITE model of Queen Elizabeth National Park). Note: LMUSK = Labor male unskilled workers; LMSK = Labor male skilled workers; LFUSK = Labor female unskilled workers; LFSK = Labor female skilled workers; K = Capital; ComRevSh = Community revenue sharing; LocalG = Local government; G = National government; ROW = Rest of the world (outside of the local economy) TABLE C.2 Social Accounting Matrix Multiplier Model of the Economy in and Surrounding QENP AGRICU TOURISM NONAGR. FISH LMUSK LMSK LFUSK LFSK K POOR NONPOOR RESTAURANTS LODGES TOURISTS PROT.​ COMREVSH AREA Agricultural 3.11 1.61 1.97 1.59 2.33 2.33 2.34 2.34 2.33 2.61 2.32 2.30 1.95 1.89 1.95 2.16 Tourism 0.00 1.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.03 0.24 0.00 0.00 Nonagricultural 2.90 2.29 3.68 2.31 3.21 3.21 3.21 3.21 3.21 3.19 3.21 2.83 2.68 2.70 3.20 3.45 Fish 0.15 0.12 0.16 1.12 0.16 0.16 0.16 0.16 0.16 0.15 0.16 0.20 0.16 0.15 0.15 0.16 LMUSK 0.52 0.33 0.40 0.42 1.43 0.43 0.43 0.43 0.43 0.47 0.43 0.43 0.38 0.38 0.44 0.70 LMSK 0.52 0.31 0.37 0.29 0.41 1.41 0.42 0.42 0.41 0.45 0.41 0.43 0.40 0.38 0.37 0.43 LFUSK 0.05 0.08 0.07 0.05 0.06 0.06 1.06 0.06 0.06 0.06 0.06 0.06 0.06 0.08 0.27 0.21 LFSK 0.00 0.01 0.00 0.00 0.00 0.00 0.00 1.00 0.00 0.00 0.00 0.01 0.03 0.02 0.04 0.00 K 1.33 1.06 1.07 1.02 1.15 1.15 1.15 1.15 2.15 1.22 1.14 1.34 1.23 1.17 1.01 1.11 Poor 0.07 0.05 0.05 0.05 0.09 0.08 0.13 0.12 0.08 1.06 0.06 0.06 0.06 0.06 0.07 0.08 Nonpoor 2.37 1.74 1.85 1.73 2.96 2.98 2.93 2.94 2.97 2.14 2.99 2.21 2.04 1.97 2.05 2.37 Restaurants 0.05 0.04 0.04 0.04 0.07 0.07 0.06 0.06 0.07 0.05 0.07 1.05 0.05 0.07 0.05 0.05 Lodges 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 1.00 0.62 0.00 0.00 Tourists 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 1.00 0.00 0.00 Protected area 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.06 1.00 0.00 ComRevSh 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.01 0.22 1.00 Source: World Bank data (Dashboard for LEWIE-LITE model of Queen Elizabeth National Park). Note: LMUSK = Labor male unskilled workers; LMSK = Labor male skilled workers; LFUSK = Labor female unskilled workers; LFSK = Labor female skilled workers; K = Capital; ComRevSh = Community revenue sharing; LocalG = Local government; G = National government; ROW = Rest of the world (outside of the local economy). Measuring the Local Economic Impacts of Nature-Based Tourism in Uganda From the SAM matrix (table C.1), you can calculate the SAM multiplier model (table C.2) by balancing the SAM (in this case using the RAS method), converting it into a coefficient matrix, then subtracting it from the identity matrix and inverting the result. The SAM multiplier model captures the links among revenue, income, and expenditure flows of households and firms in the protected area. As an example, the income multiplier of $2.03 of income generated in the local economy for every tourist dollar spent is calculated from the SAM multiplier model (table C.2) by looking at the column for “Tourists” (third from the right) and adding the multipliers for “Poor” households (0.06) and “Nonpoor” households (1.97). For a more detailed explanation of this process, see the open access training guide on social accounting matrices and multiplier analysis published by the International Food Policy Research Institute (IFPRI 2009). However, there are limitations to the model. As a member of the family of fixed-coefficient linear multiplier models, the SAM model assumes that the supply response is perfectly elastic. This assumption describes an economic environment without scarcity surrounding the protected area. That is, there are always unused resources such as labor and capital sufficient to meet the new demands projected by our simulations. This model is also static and represents a single snapshot in time. If there is a dramatic change in the economy of the protected area, one would need to redo the model with new data after the shock. Finally, a last assumption of the model is that prices do not change in response to an exogenous shock. The fixed-price assumption does not invalidate simulations if the shock is small relative to the size of the local economy (most changes in tourism demand occur gradually and at a less than 10 percent growth rate, leading to relatively small shocks to the local economy). For larger shocks, we would anticipate larger price changes, creating larger effects in output and factor markets that could not be captured in this framework. One would have to update data after a large price shock to estimate its impacts. 88