Measuring the Local Economic Impacts of Nature-Based Tourism in Madagascar 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 photo: © michelepautasso / Shutterstock Design: Sergio Moreno Tellez / World Bank Group 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 were conducted by CAETIC Developpement. The authors are grateful for inputs and technical review from Sachiko Kondo (Senior Natural Resources Management Specialist, WBG), Amy Chamberlain (Consultant, WBG), and Laza Rakotondrasoa (Natural Resources Management Specialist, WBG). The team thanks Cristian Quijada Torres (Senior Private Sector Specialist, WBG), Prisca Mamitiana (Private Sector Development Specialist, WBG), and Mirenty Raveloson (Consultant, WBG) from the Madagascar Economic Transformation for Inclusive Growth Project team, and the project implementation unit for overall collaboration and facilitation of field work. Peer review was provided by Nicholas Zmijewski (Senior Environmental Engineer, WBG) and Cristian Quijada Torres. The team thanks Idah Pswarayi-Riddihough (Country Director, WBG), Africa Olojoba (Practice Manager, WBG), Atou Seck (Country Manager, WBG), Marie-Chantal Uwanyiligira (Country Director, WBG), and Diana Styvanley (External Affairs Officer, WBG) for their support. Many other World Bank Group colleagues participated in LEWIE-LITE workshops and offered insights to improve the methodology and analysis. The team is also grateful for the support of the many ministries, agencies, and stakeholders within the government of Madagascar that enabled this work. This report was funded by PROBLUE, PROGREEN, and the Global Environment Facility- financed Global Wildlife Program. Table of Contents Abbreviations and Acronyms / 9 Glossary of Terms / 9 Executive Summary / 10 How Was the Study Done? / 11 What Did the Study Find? / 13 What Recommendations and Lessons Can Be Drawn from the Study? / 15 SECTION 1. Introduction / 16 Direct and Indirect Impacts of Tourism / 17 Nature-based Tourism in Madagascar and the LEWIE-LITE Pilot / 19 Structure of this Report / 19 SECTION 2. Methodology / 20 Sampling Design / 23 SECTION 3. Data Collection and Analysis / 27 Data Analysis and the LEWIE-LITE Dashboard / 30 SECTION 4. Descriptive Statistics of Tourist Numbers and Spending at Ranomafana and Nosy Tanikely National Parks / 33 SECTION 5. Using the Model to Simulate Impacts of Tourism / 45 Local Economy Impacts of Tourist Spending ($) / 47 Impacts of Future Growth in Tourism / 54 SECTION 6. Testing the Robustness of Results and Analyzing the Differences in Multipliers Between the Two Protected Areas / 56 SECTION 7. Using the Model to Simulate Changes in Park and Community Projects and Local Government / 60 Impacts of Changes in Park and Community Projects / 61 Impacts of Changes in Local Government Spending / 65 SECTION 8. Using the Model to Simulate Complementary Interventions / 67 Impacts of Increased Demand for Agricultural and Nonagricultural Goods / 68 Impacts of Increased Wage Earnings for Local Workers / 71 Conclusions and Recommendations / 76 Main Results and Findings / 77 Study Limitations / 79 Recommendations / 81 References / 83 Appendixes Appendix A Questionnaires / 84 Appendix B Questions and Answers about the LEWIE-LITE Model and Analysis / 95 Appendix C. Calculation of Multipliers / 98 Figures, Maps, and Tables FIGURE ES.1 Direct and Indirect Impacts of Tourist Spending in Protected Areas / 12 FIGURE ES.2 Effects of a $100 Increase in Tourist Spending on the Local Economy Around Ranomafana National Park / 13 FIGURE ES.3 Effects of a $100 Increase in Tourism Spending on the Local Economy Around Nosy Tanikely National Park / 14 FIGURE 1.1 Direct and Indirect Impacts of Tourist Spending in Protected Areas / 18 FIGURE 3.1 Social Accounting Matrix for Ranomafana National Park as shown on the LEWIE-LITE Dashboard / 31 FIGURE 3.2 LEWIE-LITE Dashboard for Ranomafana National Park—Multipliers for $100 of Tourist Spending under the Simulations Tab / 32 FIGURE 5.1 LEWIE-LITE Dashboard Visual of Tourist Spending Multipliers for Ranomafana National Park / 48 FIGURE 5.2 Effects of a $100 Increase in Tourist Spending on the Local Economy Around Ranomafana National Park / 49 FIGURE 5.3 LEWIE-LITE Dashboard of Tourist Spending Multipliers for Nosy Tanikely National Park / 50 FIGURE 5.4 Effects of a $100 Increase in Tourism Spending on the Local Economy Around Nosy Tanikely National Park / 52 FIGURE 5.5 Effects of a $1.27 Million Increase in Tourist Spending on the Local Economy Around Ranomafana National Park / 55 FIGURE 5.6 Effects of a $1.04 Million Increase in Tourist Spending on the Local Economy Around Nosy Tanikely National Park / 55 FIGURE 7.1 Effects of a $100 Increase in Park Spending on the Local Economy Around Ranomafana National Park / 62 FIGURE 7.2 Effects of a $100 Increase in Community Project Spending on the Local Economy Around Ranomafana National Park / 62 FIGURE 7.3 Effects of a $100 Increase in Park Spending on the Local Economy Around Nosy Tanikely National Park / 63 FIGURE 7.4 Effects of a $100 Increase in Community Project Spending on the Local Economy Around Nosy Tanikely National Park / 63 FIGURE 7.5 Effects of a $100 Increase in Local Government Spending on the Local Economy Around Ranomafana National Park / 66 FIGURE 7.6 Effects of a $100 Increase in Local Government Spending on the Local Economy Around Nosy Tanikely National Park / 66 FIGURE 8.1 Effects of a $100 Increase in Local Agricultural Production on the Local Economy Around Ranomafana National Park / 69 FIGURE 8.2 Effects of a $100 Increase in Local Nonagricultural Production on the Local Economy Around Ranomafana National Park / 69 FIGURE 8.3 Effects of a $100 Increase in Local Agricultural Production on the Local Economy Around Nosy Tanikely National Park / 70 FIGURE 8.4 Effects of a $100 Increase in Local Nonagricultural Production on the Local Economy Around Nosy Tanikely National Park / 70 FIGURE 8.5 Effects of a $100 Increase in Earnings for Unskilled Female Workers on the Local Economy Around Ranomafana National Park / 72 FIGURE 8.6 Effects of a $100 Increase in Earnings for Skilled Female Workers on the Local Economy Around Ranomafana National Park / 72 FIGURE 8.7 Effects of a $100 Increase in Earnings for Unskilled Male Workers on the Local Economy Around Ranomafana National Park / 73 FIGURE 8.8 Effects of a $100 Increase in Earnings for Skilled Male Workers on the Local Economy Around Ranomafana National Park / 73 FIGURE 8.9 Effects of a $100 Increase in Earnings for Unskilled Female Workers on the Local Economy Around Nosy Tanikely National Park  / 74 FIGURE 8.10 Effects of a $100 Increase in Earnings for Skilled Female Workers on the Local Economy Around Nosy Tanikely National Park / 74 FIGURE 8.11 Effects of a $100 Increase in Earnings for Unskilled Male Workers on the Local Economy Around Nosy Tanikely National Park  / 75 FIGURE 8.12 Effects of a $100 Increase in Earnings for Skilled Male Workers on the Local Economy Around Nosy Tanikely National Park / 75 MAP 2.1 Map of Madagascar Highlighting Ranomafana National Park in the South and Nosy Tanikely National Marine Park in the North / 22 TABLE 2.1 Summary of Sample Size by Types of Actor / 23 TABLE 2.2 Definitions of Skilled and Unskilled Workers / 26 TABLE 3.1 Summary of Information Collected from Visitors, Local Businesses, Protected Area Authorities, and Households (from Data Collection Instruments) / 29 TABLE 4.1 Number of Visitors and Their Expenditures at Ranomafana and Nosy Tanikely National Parks / 35 TABLE 4.2 Expenditure Shares in Tourism Activities, Restaurants, and Hotels or Lodges Surrounding Ranomafana and Nosy Tanikely National Parks / 37 TABLE 4.3 Expenditure Shares for Agriculture, Fishing, Retail, Services, and Production Businesses Surrounding Ranomafana and Nosy Tanikely National Parks / 38 TABLE 4.4 Population, Income, and Expenditures of Poor and Nonpoor Households Surrounding Ranomafana and Nosy Tanikely National Parks / 40 TABLE 4.5 Expected Park Budgets and Community Project Spending in 2023 / 42 TABLE 4.6 Revenue and Spending by the Local Governments of Ranomafana and Nosy Be / 44 TABLE 5.1 Summary of the Impact of a $100 Increase in Tourist Spending and the Impact of an Additional Tourist / 53 TABLE 6.1 Incomes in the Local Economies Around Each Park / 57 TABLE 6.2 Leakage Share from Production Sectors and Households ($) / 59 TABLE C.1 Social Accounting Matrix of the Economy In and Surrounding Ranomafana National Park / 99 TABLE C.2 Social Accounting Matrix Multiplier Model of the Economy In and Surrounding Ranomafana National Park / 100 Measuring the Local Economic Impacts of Nature-Based Tourism in Madagascar Abbreviations and Acronyms 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 MGA Ariary, the Malagasy currency MNP Madagascar National Parks ROW Rest of the world (outside of the local economy) SAM Social Accounting Matrix WBG World Bank Group *All dollar amounts are US dollars unless otherwise indicated. Glossary of Terms 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 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 A tool to estimate the direct and indirect economic impacts of Local Economy-Wide Impact tourism in and around protected areas Evaluation “Lite” (LEWIE-LITE) Spillover effects Indirect effects that occur after direct economic impacts 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 9 Measuring the Local Economic Impacts of Nature-Based Tourism in Madagascar Executive Summary This study addresses the critical connection between Madagascar’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 ministries of tourism and environment, protected area management authorities, local authorities, and task teams supporting nature-based tourism. In Madagascar, where many 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 help 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. 10 How Was the Study Done? LEWIE-LITE was piloted in two of Madagascar’s most visited protected areas: Ranomafana National Park and Nosy Tanikely National Marine Park. 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. In addition, there can also be negative impacts such as park rules which can limit some human activities or human-wildlife conflict around protected areas which can result in losses for communities. 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. Ranomafana. Photo credit: CEATIC 11 Measuring the Local Economic Impacts of Nature-Based Tourism in Madagascar 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 2023. 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. 12 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 $0.71 in the local economy surrounding Ranomafana National Park and $2.48 in the local economy of Nosy Be surrounding Nosy Tanikely. Local production increases by $1.75 in Ranomafana and $4.36 in Nosy Tanikely. The impacts of tourism are lower in Ranomafana because there is higher spending of businesses buying goods and services from larger urban centers further away–close to 68 percent compared to 28 percent in Nosy Tanikely. This is considered a leakage from the local economy. 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 Ranomafana in figure ES.2 and Nosy Tanikely in figure ES.3. FIGURE ES.2 Effects of a $100 Increase in Tourist Spending on the Local Economy Around Ranomafana National Park E ects of this tourism spending... A. ...ON PRODUCTION B. ...ON INCOMES C. ...ON LABOR INCOME 48.62 12.4 94.86 Additional Production Value ($) Additional Labor Income ($) Additional Income ($) 45.27 22.66 5.2 3.42 23.94 2 5.32 3.87 1.59 Lodges Restaurants Tourism 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 2023. 13 Measuring the Local Economic Impacts of Nature-Based Tourism in Madagascar FIGURE ES.3 Effects of a $100 Increase in Tourism Spending on the Local Economy Around Nosy Tanikely National Park E ects of this tourism spending... A. ...ON PRODUCTION B. ...ON INCOMES C. ...ON LABOR INCOME 49.24 217.95 137.51 110.54 Additional Production Value ($) Additional Labor Income ($) 33.13 Additional Income ($) 19.83 74.06 77.95 12.7 32.63 29.33 4.12 Lodges Restaurants Tourism 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 2023. Poor households benefit from the direct and indirect impacts of tourism in the local economies surrounding the two parks, receiving 32 percent of tourism-generated income in Ranomafana and 56 percent in Nosy Be. Per dollar of tourist spending, poor households receive $0.23 at Ranomafana and $1.38 at Nosy Be. These relatively large impacts on poor household incomes suggest the importance of protected area tourism for poverty alleviation. Investing in park management generates economic benefits for local communities. In 2022, tourism to Ranomafana generated $3.2 million in benefits against a park budget of $234,000, while tourism to Nosy Tanikely generated benefits of $23.2 million against a budget of $352,000. Of this, poor households received $1 million at Ranomafana and $12.9 million at Nosy Be. This shows the high return on government investment in protected areas. Complementary interventions around protected areas–such as to increase local sourcing of goods and services–can magnify these impacts. 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. 14 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 Ranomafana and Nosy Tanikely National Parks 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. Introducing additional programs to share 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. And complementary interventions such as skills training for workers from poor households could increase the amount of tourism benefits that poor households capture. These kinds of interventions could also address how to increase the employment of women in nontourism activities. 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. 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. Nosy Be. Photo credit: CEATIC 15 SECTION 1 Introduction 16 Photo credit: © jordieasy / Adobe Stock Measuring the Local Economic Impacts of Nature-Based Tourism in Madagascar A country’s protected areas and tourism are intimately connected and yet, the economic impacts of tourism on the businesses and households in and around protected areas are largely unknown. Protected areas in Madagascar are visited by many thousands of tourists every year. Unfortunately, there are overwhelming data gaps on the economic impact of these many tourists visiting protected areas. Without the information they need, tourism ministries, park services, communities, and the central and local government in general are largely in the dark about the economic value of protected areas and the costs and benefits of investing in protected areas and tourism-related projects. Studies of the impact of tourist dollars on specific protected areas in Madagascar 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 look 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 used. 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 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. 17 Measuring the Local Economic Impacts of Nature-Based Tourism in Madagascar 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. Source: World Bank 2023. 18 Measuring the Local Economic Impacts of Nature-Based Tourism in Madagascar Nature-based Tourism in Madagascar and the LEWIE-LITE Pilot Endowed with dense forests, rich biodiversity, and nearly 5,000 kilometers of coastline, Madagascar has immense potential for nature-based tourism. At the same time, it faces a persistently high poverty rate and is one of the poorest countries in the world. The government of Madagascar is committed to sustainable tourism development that maximizes economic and social benefits and contributes to poverty reduction. Additionally, the government aims to increase foreign visitation to its national parks and protected areas to generate sustainable financing for biodiversity conservation (World Bank 2022). For these reasons, Madagascar was selected as a pilot country for the LEWIE-LITE methodology. Two Madagascar parks, Ranomafana National Park and Nosy Tanikely National Marine Park, were selected with the help of Madagascar National Parks (MNP) to complete prototype models. They were chosen because they are two of Madagascar’s most visited MNP-managed parks and because they represent a terrestrial (Ranomafana) and a marine (Nosy Tanikely) protected area. 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 nature-based tourism 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 Ranomafana and Nosy Tanikely National Parks (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). 19 SECTION 2 Methodology Photo credit: © kris-la-Reunion / Adobe Stock Measuring the Local Economic Impacts of Nature-Based Tourism in Madagascar The two protected areas selected for this project were surveyed separately but with similar DCIs and methods. They are: Nosy Tanikely National Marine Park, a marine island park known for snorkeling, diving, and rich biodiversity, including lemurs, sea turtles, dolphins, and multiple other species. It is located off the northern coast of Madagascar (map 2.1). It receives approximately 31,000 tourists a year who pay an average park entry fee of $6.84. Nosy Tanikely was chosen for this study because it is a marine park managed by MNP. There are no lodgings or other businesses on Nosy Tanikely. Visitors to the park can find places to stay, restaurants, tour guides, and other services on Nosy Be, the large nearby island that is a popular tourist destination. Nosy Tanikely is one of several attractions around Nosy Be, which includes Nosy Sakatia, with many sea turtles; Nosy Komba, where one can see lemurs; and Andilana Beach, a nice sandy beach with blue water and resort-style hotels. The local population of Nosy Be is 109,465 (2018 population census) and the main economic activities on the island are tourism and agriculture. Ranomafana National Park is a terrestrial park in the south-central part of Madagascar and known for abundant biodiversity. It receives approximately 15,200 tourists a year who pay an average park entry fee of $7. It was chosen because it is a popular terrestrial park managed by MNP. Ranomafana Commune is the major community that surrounds the park and is more rural and smaller than Nosy Be. As of 2018, it had a local population of 9,705 and the main economic activities are predominantly agriculture, followed by some retail and services. Ranomafana caters to less high-end tourists than those that visit Nosy Be. It does not have resorts, for example, but is still popular among visitors for its biodiversity, being the only park in the surrounding area where visitors can see chameleons, frogs, lemurs, and other rare species. The two parks differ in terms of the type of tourists that they attract. Since Nosy Be is popular for its marine activities and has a selection of high-end resorts, it attracts wealthier visitors who can explore coral reefs with marine life, enjoy beautiful beaches, and take boat trips. Ranomafana National Park, on the other hand, offers opportunities for diverse wildlife viewing via guided treks through the rainforest. Visitors are not traveling for rest but rather for hikes and to see rare animal species. They will spend less money in the local economy and spend far less time at the park and the surrounding area, which also reduces the relative amount of spending per visitor. These differences complicate comparison of the impacts of tourism across the two protected areas. Ideally, one would want to compare Ranomafana with similar terrestrial parks (such as Andasibe-Mantadia National Park near Antananarivo) and Nosy Tanikely with a similar coastal destination (such as Tamatave Coral Reef in eastern Madagascar). However, for the purposes of this pilot, it was agreed with the government to study one terrestrial and one marine park.1 1 This could be alleviated with application of the model at other protected areas in Madagascar. 21 Measuring the Local Economic Impacts of Nature-Based Tourism in Madagascar MAP 2.1 Map of Madagascar Highlighting Ranomafana National Park in the South and Nosy Tanikely National Marine Park in the North Source: World Bank. 22 Measuring the Local Economic Impacts of Nature-Based Tourism in Madagascar 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. This study uses Nosy Be as the local economy for Nosy Tanikely National Park and Ranomafana Commune (Ranomafana Municipality) as the local economy surrounding Ranomafana National Park. This definition of the local economy was determined in consultation with MNP. 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 actors. TABLE 2.1 Summary of Sample Size by Types of Actor ACTOR OR ENTITY SAMPLE Park manager Budget and entrance fees to Ranomafana National Park and Nosy Tanikely National Park Community projects for each of the parks Tourists 200 tourists randomly selected from different locations (hotel or lodge, park entrance, port or airport boarding, and so on) 100 questionnaires per site Hotels and lodges 30 hotels or lodges selected randomly from a list of properties 15 hotels per site, including five units per hotel category (3 to 5 starts; 1 to 2 stars, and others) Restaurants 20 restaurants selected randomly based on a list of restaurants 10 restaurants per site, including five units per category (small or budget and large or upscale) Other tourism-related 30 units selected randomly based on a list of enterprises businesses 15 units per site, including five 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, including five by type of business (retail trade, other services, and nonservices businesses) Commercial farmers 20 commercial farmers selected randomly based on a list of farmers 10 farmers per site, including five per type of farm (crop farmers and livestock breeders) Commercial fishers 10 Nosy Be fishers, including five traditional fishers and five artisanal fishers Households 60 households surveyed 20 households for Ranomafana and 40 for Nosy Be, including 10 for each focus group (poor and nonpoor households) Local government One commune surveyed, per site Tour operators in the capital 10 tour operators surveyed in the capital, Antananarivo Source: World Bank 2023. 23 Measuring the Local Economic Impacts of Nature-Based Tourism in Madagascar The information was collected as follows: 1. Protected areas Ranomafana National Park and Nosy Tanikely National Park are managed by MNP. Data were collected through face-to-face interviews at each site in which park staff were asked about park budgets. 2. Community projects Communities surrounding the two protected areas benefit from some funds sent to them by the parks. This is not an official revenue sharing program like those that exist in some other countries (for example, Uganda), but it is still significant enough to create meaningful benefits for local communities. Data were obtained by interviewing park authorities for each protected area and asking them about the amounts sent to local communities and how that money was spent.2 3. Tourists Tourists were interviewed at the two sites. The study considered three types of tourists: nationals, foreigners with residency in Madagascar, and foreigners visiting Madagascar for tourism. In total, 200 tourists (100 at each site) were randomly surveyed in person at various locations. In the case of a group,3 only one member was interviewed. 4. Hotels and lodges Information was gathered from 30 hotels/lodges, 15 from each site, that were randomly selected from a list in each area. Data were collected through interviews according to three categories: three-to five-star hotels, one-or two-star hotels, and others (bed and breakfasts, residences, ecolodges, and so on). Five properties were sampled from each of the three categories. 5. Restaurants Interviews were conducted with 20 restaurants, 10 from each site, randomly selected from a list in each area. Data were collected through face-to-face interviews according to two types of restaurants: large or upscale (average price of the main course above MGA4 15,000, or $3.375) and small or budget (average price of the main course below MGA 15,000). Five establishments were sampled for each type of restaurant. 6. Other tourism-related businesses Besides lodges and restaurants, 30 tourism businesses, 15 from each site, were randomly selected. Data were collected from three types of tourism businesses: tour operators or guides such as excursions, souvenir shops, and equipment rental stores. Five interviews were conducted for each type of business per park. 2 Park authorities joined local leaders to determine which community projects would be supported and how spending was allocated. 3 Either an individual, family, or tour group. 4 Abbreviation for the ariary, Madagascar’s currency. 5 The exchange rate used in this analysis was the official exchange rate published by the Madagascar Central Bank at the beginning of the data collection (as of July 28, 2023): €1 = MGA 4876.27 and $1 = MGA 4451.35. 24 Measuring the Local Economic Impacts of Nature-Based Tourism in Madagascar 7. Other nonagricultural businesses Thirty other nonagricultural businesses not related to tourism, 15 from each site, were randomly selected from a list at each site. Data were collected through 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 (carpentry shops, food processing companies, and so on). Five interviews were carried out for each type of business per park. 8. Commercial farmers Twenty farmers, 10 from each site, were randomly selected from a list at each site. Data collection was carried out through in-person interviews of two types of farmers: agricultural farmers or crop producers and livestock breeders. Five of each type of farmer were interviewed. 9. Commercial fishers As natural resource users, 10 fishers were randomly selected from a list at the marine site, the local economy of Nosy Be surrounding Nosy Tanikely National Park. Data collection was carried out through face-to-face interviews of fishers engaged in two types of activities: artisanal fishing and fishing using motorized boats. Five fishers were sampled for each type of activity. 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 companies and operations. A total of 60 households, 20 at Ranomafana and 40 at Nosy Be, were randomly sampled using the itinerary method in two fokontany6 at each site. Data collection was carried out through a focus group of 10 participants representing two types of households, poor and nonpoor. Two household focus groups were convened for this purpose at Ranomafana and four at Nosy Be.7 The national poverty line in Madagascar was used to distinguish between poor and nonpoor households. For 2022, the national poverty line was MGA 1,400,000 ($314.51) per person per year. Household income was estimated based on total consumption, comprised of food expenditures (including consumption of home-produced food) and nonfood expenditures (education, health, transport, rental, energy, and so on). 11. Local governments Local governments (municipalities) collect taxes and spend tax revenue in ways that can create fiscal impacts on local economies. Information on tax receipts and spending was gathered from the communes of Ranomafana and Nosy Be. Data collection was carried out through face-to-face interviews with the mayor and the person responsible for finances in each municipality. 6 A fokontany is the smallest administrative unit in Madagascar, similar to a community or village. 7 According to the 2018 census, the total number of households in Ranomafana is 2,284 and 33,636 in Nosy Be. 25 Measuring the Local Economic Impacts of Nature-Based Tourism in Madagascar Linking to this, each of the businesses surveyed was also asked questions about spending on local taxes. Since tax collection and reporting challenging in Madagascar, the information collected from businesses may contain some biases. This is somewhat mitigated by the small percentages reported by them. 12. Tour operators Many protected area visitors purchase travel packages from tour operators outside the local economy, typically in Antananarivo, the capital. Ten tour operators were surveyed around Antananarivo. These data are important because tourists who buy package deals generally do not know how the costs are allocated to local spending by sector (lodges, tour operators, meals, and so on). To get this information, tour operators in Antananarivo 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 2023. 26 SECTION 3 Data Collection and Analysis Photo credit: © ArtushFoto / Adobe Stock Measuring the Local Economic Impacts of Nature-Based Tourism in Madagascar Data were collected for the LEWIE-LITE survey from July 28 to August 13, 2023, including travel. Two teams were mobilized to carry out the field survey. One team of three field supervisors and two enumerators oversaw the Ranomafana site while another team of three field supervisors and three enumerators was responsible for data collection for Nosy Tanikely and Nosy Be. 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 protected area (national park) authorities and households (focus groups surrounding the two parks). 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) are 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.8 The spending categories for community revenue sharing and park spending were the same, as given in the column. The household data surveys (d) are used to calculate average per capita income, remittances, government transfer incomes, and other variables for each household group, as well as household budget shares and spending on goods and services outside the local economy. 8 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. 28 TABLE 3.1 Summary of Information Collected from Visitors, Local Businesses, Protected Area Authorities, and Households (from Data Collection Instruments) A. VISITORS B. BUSINESSES C. PROTECTED AREA D. HOUSEHOLDS (FROM VISITOR SURVEY) (FROM BUSINESS SURVEY) (FROM NATIONAL PARK ACCOUNTS) (FROM HOUSEHOLD SURVEY) Number of multi-day tourists (adults and About what percentage of monthly spending goes to Total expected annual park budget 2023 What is the population of the communities constituting this children >5) (in the whole zone of impact) each of the following: local economy (number of people) Average stay (days) Salaries and wages for male unskilled workers Total expected park entry fees 2023 How many households are in this local economy (number of households) Average stay (nights) 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 operators, supervisors, receptionists, accountants, etc.) to community sharing households in this region ($) Average nightly price per room (total, Salaries and wages for female unskilled workers Total community sharing budget About what is the average annual government transfers including taxes, double occupancy, (e.g., social cash transfers) to households in this region including resort and other fees) Expected number of single-day tourists Salaries and wages for female skilled workers (machine How much of this budget goes to: About what percentage of household income comes from: (adults and children >5, no lodging) operators, supervisors, receptionists, accountants, etc.) Expected spending per person per day Crop purchases from local farmers or animal products Salaries and wages for male unskilled workers Wages earned by male unskilled workers in the household while visiting this protected area, on: from local ranchers Park entry Purchases from tourism Salaries and wages for male skilled workers Wages earned by male skilled workers in the household Local restaurants (food and drink) Local fish Salaries and wages for female unskilled workers Wages earned by female unskilled workers in the household Guides and tours Services (machine maintenance, construction, repairs) Salaries and wages for female skilled workers Wages eamed by female skilled workers in the household from local providers Souvenirs/handicrafts Purchases from local stores and other businesses Payment of rents on land, buildings, etc. Profits from household-owned farms or businesses or renting property the household owns Retail shops, local markets Purchases outside the local economy, like merchandise Locally produced agricultural products (fruits, Migrant remittances (domestic and foreign) (for stores) or supplies vegetables, meats) Other Farm tax/fishing business tax rate (%) Locally produced fish or other natural resources About what percentage of household spending each month is on: Services (laundry, maintenance, construction, Food bought from local grocery stores Other variables repairs) from local providers Share of businesses locally owned Purchases from local stores and other businesses Buying food direct from local farmers or your own farm Share of wages paid to local workers Purchases made outside the local economy Buying local fish Average profit margin Other variables Buying food and drink at local restaurants Percentage of salaries and wages paid to local Things besides food that are sold by people or businesses in workers your community, including services Things you buy from businesses, etc., in places outside your community Rental income Income tax payments Source: World Bank 2023. Measuring the Local Economic Impacts of Nature-Based Tourism in Madagascar 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 Ranomafana National Park. 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 an additional $100, using the “Simulations” tab, for Ranomafana National Park. 30 FIGURE 3.1 Social Accounting Matrix for Ranomafana National Park as shown on the LEWIE-LITE Dashboard AGRICU TOURISM NONAGR. FISH LMUSK LMSK LFUSK LFSK K POOR NONPOOR RESTAURANTS LODGES TOURISTS PROT. COMREVSH LOCALG G ROW TOTAL AREA EXPENDIT. Agricultural 2086 2101 0 0 0 0 0 0 0 36628 10436 938 525 0 0 0 0 0 0 52714 Tourism 0 0 0 0 0 0 0 0 0 10465 5218 0 34 489914 0 0 0 0 0 505632 Nonagricultural 13728 89931 115181 0 0 0 0 0 0 308722 365259 35555 4063 507735 72365 528 12228 0 0 1525295 Fish 0 0 0 0 0 0 0 0 0 36628 0 0 0 0 0 0 0 0 0 36628 LMUSK 10223 23113 115477 0 0 0 0 0 0 0 0 3615 1498 0 7929 0 3373 0 0 165228 LMSK 0 21712 0 0 0 0 0 0 0 0 0 0 473 0 109182 1162 20240 0 0 152770 LFUSK 2211 2101 6793 0 0 0 0 0 0 0 0 2933 1392 0 0 0 843 0 0 16274 LFSK 0 3222 0 0 0 0 0 0 0 0 0 477 680 0 18071 106 3795 0 0 26350 K 24716 173804 391827 0 0 0 0 0 0 62791 20872 28422 3901 0 0 0 0 0 0 706333 Poor 0 0 0 0 132450 63281 12911 21123 57807 0 0 0 0 0 0 0 0 0 16018 303590 Nonpoor 0 0 0 0 32778 89489 3363 5227 648526 0 0 0 0 0 0 0 0 0 0 779384 Restaurants 0 0 0 0 0 0 0 0 0 0 0 0 0 113687 0 0 0 0 0 113687 Lodges 0 0 0 0 0 0 0 0 0 0 0 0 0 24872 0 0 0 0 0 24872 Tourists 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1248835 1248835 Protected area 0 0 0 0 0 0 0 0 0 0 0 0 0 112627 0 0 0 120934 0 233561 ComRevSh 0 0 0 0 0 0 0 0 0 0 0 0 0 0 10562 0 0 0 0 10562 LocalG 1252 6934 31010 0 0 0 0 0 0 0 0 2046 925 0 0 0 0 0 0 42167 G 0 11977 49469 0 0 0 0 0 0 5233 1459 3922 1327 0 0 0 0 0 0 73387 ROW 12226 129013 789582 36628 0 0 0 0 0 62791 120014 35777 10054 0 26014 8766 1687 0 0 1232551 Total 66442 463908 1499339 36628 165228 152770 16274 26350 706333 523257 523257 113687 24872 1248835 244123 10562 42167 120934 1264853 0 expenditures Source: World Bank 2023. 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 Madagascar FIGURE 3.2 LEWIE-LITE Dashboard for Ranomafana National Park—Multipliers for $100 of Tourist Spending under the Simulations Tab LOCAL ECONOMY WIDE IMPACTS OF TOURIST SPENDING $ You may wish to evaluate di erent 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 ($) do you want to simulate? 100 E ects of this tourism spending on... A. ...ON PRODUCTION B. ...ON INCOMES C. ...ON LABOR INCOME 115.68 56.55 12.32 Additional Production Value ($) Additional Labor Income ($) Additional Income ($) 7.57 40.18 19.43 1.21 1.34 9.1 1.99 2.88 1.36 Lodges Restaurants Tourism 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 2023. 32 SECTION 4 Descriptive Statistics of Tourist Numbers and Spending at Ranomafana and Nosy Tanikely National Parks Nosy Be. Photo credit: CEATIC Measuring the Local Economic Impacts of Nature-Based Tourism in Madagascar Tables 4.1 to 4.6 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 patterns across the local economies surrounding Ranomafana and Nosy Tanikely National Parks. Table 4.1 shows the number of visitors projected by the parks for the full year 20239 and their spending on goods and services during their visit to each site.10 A total of 15,200 visitors were expected to visit Ranomafana National Park in 2023. Nosy Tanikely expected 31,000 visitors by year end. The visitor surveys showed that 60 percent of visitors to Nosy Be and 79 percent of visitors to Ranomafana were foreigners. The average stay at Ranomafana was 2.2 days (1.9 nights) while Nosy Be had an average of 12 days (11 nights).11 Rooms were, on average, more expensive at Nosy Be at $112 per room, compared to $64 at Ranomafana. Entry fees averaged $7.41 for Ranomafana and $6.84 for Nosy Tanikely.12 Visitors must also pay for a guide to enter the parks. This mandatory fee is included in the spending data on guides and tours and impacts the local economy directly through guide earnings and indirectly as they spend the money earned at nearby businesses.13 In summary, the average visitor to Ranomafana spent $144 per day, of which 44.7 percent went to accommodation, 5.1 percent to park fees, 5.1 percent to local restaurants, 19.9 percent to guides and tours, 2.2 percent to souvenirs and handicrafts, 21.5 percent to retail shops and local markets, and 1.4 percent to other goods and services. The average visitor to Nosy Tanikely spent $190 per day, of which 58.8 percent went to accommodation, 3.6 percent to park fees, 6.5 percent to local restaurants, 22 percent to guides and tours, 3.8 percent to souvenirs and handicrafts, 4.7 percent to retail shops and local markets, and 0.6 percent to other goods and services. The data point to some interesting differences between Ranomafana and Nosy Tanikely. For instance, the great difference in the average stays near each protected area could indicate different types of tourism, such as longer vacations in Nosy Be due to the variety of popular attractions and beaches, compared to the short visits for Ranomafana. There is also a wide disparity in spending. The total expected spending per person per day is significantly higher in Nosy Be ($190.07) compared to Ranomafana ($144.03). This suggests that visitors to Nosy Tanikely National Park tend to spend more during their stay due to higher-end accommodations (there are more resort style and upscale hotels in Nosy Be than near Ranomafana National Park), a greater number of tours, and increased expenditure on souvenirs. The distribution of visitor spending at Ranomafana and Nosy Be is similar for most categories except accommodation and retail. Tourists to Ranomafana allocate a larger 9 Projections on 2023 visitors were collected at the end of Q3 2023, thus reflecting most of the year’s total. 10 In the sample of visitors, 47 percent answered that they visited or planned on visiting Nosy Tanikely during their stay in Nosy Be. That subsample was used estimate the averages in table 4.1. However, these tourists are similar to those coming to Nosy Be, and using the full sample of tourists to Nosy Be does not create any statistical difference in the average estimates. 11 See subsequent discussion on attribution of the days (and nights) spent in Nosy Be to Nosy Tanikely. For the reasons mentioned, this study includes only one day and two nights attributed to Nosy Tanikely. 12 Park entry fees are averages of fees for foreign or domestic visitors. They also depend on the group composition averages as there are discounted prices for children. See the Madagascar National Parks website for more details. 13 There is an additional small community fee of approximately $1, not captured here. 34 Measuring the Local Economic Impacts of Nature-Based Tourism in Madagascar portion of their budget to retail shops, local markets, and services (21 percent of daily spending), while those visiting Nosy Tanikely spend only 5 percent in this category. A larger portion goes to more expensive accommodations at Nosy Be (59 percent compared to 45 percent for lodging around Ranomafana). Understanding these spending patterns and other differences can be valuable for local businesses and authorities to tailor their offerings and marketing strategies and increase local spillover effects. Attribution can be challenging at some protected area sites. Can one attribute tourist spending in the local economy to the existence of the park itself? In the case of Ranomafana National Park, attribution is straightforward, because most tourists only come to the area Ranomafana Commune to visit the park. Because of this, all or almost all tourist spending in the local economy surrounding Ranomafana National Park can be attributed to the park. Attribution is more complex for Nosy Tanikely National Park, which is one of several tourist sites around Nosy Be. A tourist could visit Nosy Be without ever TABLE 4.1 Number of Visitors and Their Expenditures at Ranomafana and Nosy Tanikely National Parks VISITOR INFORMATION GATHERED RANOMAFANA NOSY TANIKELY NATIONAL PARK NATIONAL PARK Park records and Number of visitors (adults and children > 5) 15,200 31,000 visitor surveys Visitor surveys Percent foreign visitors 79% 60% Average stay (days) 2.2 1* Average stay (nights) 1.9 2* Average nightly price per room (total, including taxes, double occupancy, $64.44 $112.09 including resort and other fees, $) Expected spending in $ per person per day on: Expenditure category Park entry $7.41 $6.30 Local restaurants (food and drink) $7.38 $12.46 Guides and tours $28.65 $41.84 Souvenirs and handicrafts $3.17 $7.30 Retail shops, local markets and local $30.94 $8.94 transportation Other $2.03 $1.14 Total $144.03 $190.07 Source: World Bank 2023. Note: *The average length of stay reported by tourists was 12 days and 11 nights (medians of 10 days and nine nights) but this was for stays at Nosy Be and not necessarily attributable to Nosy Tanikely. 35 Measuring the Local Economic Impacts of Nature-Based Tourism in Madagascar going to Nosy Tanikely. In fact, more tourists surveyed at Nosy Be (53 percent) did not and were not planning to visit Nosy Tanikely National Park during their stay. Moreover, all tourists who said they visited or will visit Nosy Tanikely said they would have come to Nosy Be even if the park did not exist or they had not heard of it. This suggests that all tourist spending at Nosy Be cannot be attributed to Nosy Tanikely National Park. The simulations therefore only consider spending by visitors who visited Nosy Tanikely, and they only attribute one day and two nights, the minimum amount of time necessary to go to and from Nosy Be to visit the park.14 This complication highlights the importance of thinking about attribution when there is more than one reason to visit a protected area site. One way to address it is by including contingent behavior questions on tourist questionnaires to provide more insight into their choices. 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.15 Visitor demands for local goods and services direct more money to 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 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 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) for Ranomafana and Nosy Tanikely National Parks. These were calculated from the interviews with souvenir stores, tour operators and tour equipment rental shops, restaurants, and hotels and lodges surrounding the parks. 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. 14 A different sample of tourists may yield different results, as hypothetical questions might be difficult for visitors to answer and, lacking the counterfactual, there is no way of knowing for sure how the number of visitors to Nosy Be and their length of stay there might change if Nosy Tanikely did not exist. Therefore, it was decided to establish an additional upper-bound estimate of the impact of tourism attributable to Nosy Tanikely that uses all days and nights spent by tourists in Nosy Be who visited Nosy Tanikely. Assumptions about how many days and nights to attribute to Nosy Tanikely do not make much of a difference if one is interested in estimating impacts per dollar of tourist (or other) spending. The multipliers are very similar other than a small difference in the impact on lodges. However, the attribution assumption does make a difference if one wishes to simulate the impacts of increases in the number of tourists, because the absolute amount generated by tourists increases much more if one attributes all days and nights spent at Nosy Be to the existence of Nosy Tanikely. 15 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. 36 Measuring the Local Economic Impacts of Nature-Based Tourism in Madagascar TABLE 4.2 Expenditure Shares in Tourism Activities, Restaurants, and Hotels or Lodges Surrounding Ranomafana and Nosy Tanikely National Parks RANOMAFANA (AVG. %) NOSY BE (AVG. %) About what percentage of monthly TOURIST RESTAURANTS HOTELS/ TOURIST RESTAURANTS HOTELS/ spending goes to each of the following: BUSINESSES LODGES BUSINESSES LODGES Salaries and wages for male unskilled 11% 5% 10% 4% 5% 7% workers Salaries and wages for male skilled 10% 0% 3% 6% 2% 8% workers (machine operators, supervisors, receptionists, accountants, etc.) Salaries and wages for female unskilled 1% 4% 9% 5% 14% 9% workers Salaries and wages for female skilled 2% 1% 4% 2% 2% 3% workers (machine operators, supervisors, receptionists, accountants, etc.) Crop purchases from local farmers or animal 1% 1% 3% 1% 10% 4% products from local ranchers Purchases from tourism activities 0% 0% 0% 0% 0% 1% Local fish 0% 0% 0% 1% 7% 3% Services (machine maintenance, 13% 11% 8% 28% 7% 13% construction, repairs) from local providers Purchases from local stores and other 16% 30% 13% 11% 27% 15% businesses Purchases outside the local economy like 41% 40% 37% 36% 20% 30% merchandise (for stores) or supplies Local nonfarm tax/meal tax/lodge tax rate 2% 2% 5% 4% 4% 7% (%) Nonlocal nonfarm tax/meal tax/lodge tax 4% 5% 7% 2% 2% 2% rate Other variables Share of businesses locally owned 99% 100% 66% 100% 100% 82% Share of wages paid to local workers 67% 80% 81% 100% 100% 98% Average profit margin 36% 25% 24% 32% 22% 21% Number of observations 15 10 15 15 10 15 Source: World Bank 2023. 37 Measuring the Local Economic Impacts of Nature-Based Tourism in Madagascar 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 Ranomafana and Nosy Tanikely National Parks RANOMAFANA (AVG. %) NOSY BE (AVG. %) ABOUT WHAT PERCENTAGE OF MONTHLY SPENDING AGRICUL- FISHING RETAIL/ AGRICUL- FISHING RETAIL/ GOES TO EACH OF THE FOLLOWING: TURE (NOT SERVICES/ TURE SERVICES / AVAILABLE PRODUCTION PRODUC- LOCALLY) TION Salaries and wages for male unskilled workers 25% - 14% 21% 33% 13% Salaries and wages for male skilled workers 0% - 0% 1% 3% 5% Salaries and wages for female unskilled workers 5% - 1% 2% 0% 2% Salaries and wages for female skilled workers 0% - 0% 0% 0% 15% Crop purchases from local farmers, animal products 5% - 0% 10% 2% 3% from local ranchers, or ag inputs to farmers/herders Purchases from tourism activities 0% - 0% 0% 0% 0% Local fish 0% - 0% 0% 15% 0% Services (machine maintenance, construction, 2% - 3% 21% 21% 10% repairs) from local providers Purchases from local stores and other businesses 31% - 7% 34% 22% 17% Purchases outside the local economy like 29% - 68% 8% 4% 28% merchandise (for stores) or supplies Local farm tax/fishing tax/business tax rate (%) 3% - 3% 2% 1% 3% Nonlocal farm tax/fishing tax/business tax rate (%) 0% - 4% 1% 0% 3% Other variables Share of businesses locally owned 100% - 100% 100% 100% 100% Share of wages paid to local workers 100% - 77% 100% 100% 100% Average profit margin 37% - 26% 25% 30% 19% Number of observations 10 0 15 10 10 15 Source: World Bank 2023. 38 Measuring the Local Economic Impacts of Nature-Based Tourism in Madagascar Tables 4.2 and 4.3 are snapshots of the data input for the model, gathered from 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 15 percent. However, the percentages going to male-worker wages in these activities are usually much higher—13 percent in local businesses to 33 percent in the case of fishing. Female-worker wages are higher in tourism-related activities, especially in restaurants and lodges (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 as 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. It is also more common in Ranomafana than in Nosy Be. 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.16 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 some services and production activities; for example, carpentry shops and local taxis purchase many of their inputs in outside markets. At Ranomafana, which is more isolated and has less local business activity than Nosy Be, purchases from outside markets make up 68 percent of total local, nonfarm, nontourism business spending, compared with 28 percent at Nosy Be. 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. Households in Ranomafana spend a larger share of their income in local stores (46 percent for poor households and 50 percent for nonpoor households) than in Nosy Be (22 percent for poor households and 24 percent for nonpoor households). Because stores at Nosy Be source more of what they sell locally and households either tend to buy from local farmers or grow their own produce, they circulate money in communities around Nosy Tanikely, and create larger multipliers. 16 It is important to reiterate here that “local” pertains to Ranomafana Commune for Ranomafana National Park and Nosy Be for Nosy Tanikely National Park. It could be that benefits (or purchased inputs to businesses), especially from Ranomafana, come from areas outside of the commune but still closer than Antananarivo, which would also be positive for rural development. However, to capture those, the study would have needed to draw a larger net and it was not clear what would be included in the “local” category outside of the commune. Future LEWIE- LITE surveys could be designed with this larger local economy. 39 Measuring the Local Economic Impacts of Nature-Based Tourism in Madagascar TABLE 4.4 Population, Income, and Expenditures of Poor and Nonpoor Households Surrounding Ranomafana and Nosy Tanikely National Parks HOUSEHOLD POPULATION, RANOMAFANA NOSY BE INCOME AND EXPENDITURES POOR NONPOOR POOR NONPOOR What is the population of the communities constituting this 7,764 1,941 86,613 21,653 local economy (number of people) How many households are in this local economy (number of 1,827 457 26,909 6,727 households) About what is the average annual per-capita income of $216 $647 $162 $620 households in this region ($) About what is the average annual government transfers $4 $6 $4 $16 (e.g., 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 40% 10% 23% 36% Wages earned by male skilled workers in the household 7% 10% 10% 10% Wages earned by female unskilled workers in the 38% 10% 21% 23% household Wages earned by female skilled workers in the household 4% 1% 7% 9% Profits from household-owned farms or businesses or 6% 68% 23% 14% renting property the household owns Migrant remittances (domestic and foreign) 3% 0% 9% 4% About what percentage of household spending each month is on: Food bought from local grocery stores 46% 50% 22% 24% Buying food direct from local farmers or your own farm 7% 2% 30% 21% Buying local fish 7% 0% 11% 8% Buying food and drink at local restaurants 0% 0% 0% 0% Things besides food that are sold by people or businesses 13% 20% 11% 13% in your community, including services Things you buy from businesses, etc., in places outside 12% 23% 3% 3% your community Rental income 12% 4% 14% 20% Tourism products 2% 1% 5% 8% Household local income tax payments 0% 0% 6% 6% Household federal income tax payments 1% 0% 0% 0% Source: World Bank 2023. 40 Measuring the Local Economic Impacts of Nature-Based Tourism in Madagascar Generally, households surrounding Ranomafana National Park tend to be slightly better off than those surrounding Nosy Tanikely (living in Nosy Be). The average annual per capita income is $216 for poor households and $647 for nonpoor households surrounding Ranomafana, $54 and $27 higher, respectively, than in Nosy Be.17 In Ranomafana, poor households receive more income from unskilled male and female workers than nonpoor households—40 percent and 38 percent compared to 10 percent and 10 percent. This is not the case at Nosy Be, however, where the two groups’ shares are similar. Less skilled female labor makes up a larger percentage of income at Ranomafana than at Nosy Be.18 In Ranomafana, nonpoor households have more income from renting property or having their own businesses than poor households. Nonpoor households earn 68 percent of income from farms, businesses, or renting property while poor households earn only 6 percent. Households near Nosy Tanikely spend from 21 percent to 30 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. In contrast, households surrounding Ranomafana purchase a higher percentage of outside goods and services—12 percent to 23 percent. Similar to business inputs purchased from outside, this result shows that there are more leakages from the local economy surrounding Ranomafana, leading to lower multiplier impacts. Table 4.5 shows the park budgets and park revenue from entry fees for Ranomafana and Nosy Tanikely National Parks. 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. Park budget is determined by MNP and other ministries of the Madagascar government. At both protected areas, park revenue is less than the park budget. A portion of park revenue is transferred to the national government (via MNP). In addition, the parks share 9 percent to 14 percent of their entry fees to support local communities and create incentives for conservation.19 17 The share of poor households was applied to household population data for Nosy Be and Ranomafana from the Madagascar Census 2018 (RGPH-3) using the World Bank statistics for poverty headcount ratio. 18 One income source that is not broken out but noted as “other income” from the household focus groups at Nosy Be is the presence of prostitution as an income source for some poor and nonpoor households. 19 In Madagascar, there is no formal agreement for protected areas to share their budget with local communities, but since COVID-19, MNP has sought to give a portion of their funds for community projects. This contrasts with other countries that have a formal revenue sharing agreement such as Uganda, which sends 20 percent of park entry fees to local communities surrounding protected areas. 41 Measuring the Local Economic Impacts of Nature-Based Tourism in Madagascar TABLE 4.5 Expected Park Budgets and Community Project Spending in 2023 RANOMAFANA NOSY TANIKELY PARK BUDGET COMMUNITY PARK BUDGET COMMUNITY PROJECTS PROJECTS Total expected park budget 2023 $233,561 $351,807 Total expected park entry fees 2023 $112,627 $211,964 What share of park entry fees are being assigned to 9% 14% community projects Amount sent to the national government ($120,934) ($139,842) Total community budget $10,562 $29,092 Share of park entry fees being assigned to community projects Salaries and wages for male unskilled workers 3.4% 0.0% 0.0% 1.0% Salaries and wages for male skilled workers 46.7% 11.0% 16.9% 1.0% Salaries and wages for female unskilled workers 0.0% 0.0% 0.0% 1.0% Salaries and wages for female skilled workers 7.7% 1.0% 12.9% 1.0% Payment of rents on land, buildings, etc. 0.0% 0.0% 0.0% 0.0% Locally produced agricultural products (fruits, 0.0% 0.0% 0.0% 0.0% vegetables, meats) Locally produced fish or other natural resources 0.0% 0.0% 0.0% 0.0% Services (laundry, maintenance, construction, repairs) 26.2% 0.0% 41.4% 27.0% from local providers Purchases from local stores and other businesses 4.8% 5.0% 23.7% 8.0% Purchases made outside the local economy 11% 83% 5.2% 61.0% % of salaries and wages paid to local workers 100% 100% 100% 100% Source: World Bank 2023. 42 Measuring the Local Economic Impacts of Nature-Based Tourism in Madagascar The lower half of the table shows how the parks and communities allocate their budgets across wages and other goods and services. The numbers shown are the percentages of park and community budgets spent on each item. These are the same as percentages of revenue in the case of community and park budgets, which the government must supplement. Parks contract local labor and a high percentage of their goods and services locally. This increases the multiplier effect of tourist dollars on the local economy. While using 100 percent local labor, community projects purchase 61 percent and 83 percent of goods and services outside of Nosy Be and Ranomafana, respectively. These community projects create local social assets and generate economic benefits by hiring local workers but could do more by purchasing a higher percentage of their materials from local businesses. However, local purchases at both parks impact the local economy in ways beyond just the community projects. They make up 89 percent of the Ranomafana National Park budget and 94.8 percent of the Nosy Tanikely National Park budget.20 The local governments or communes of Ranomafana and Nosy Be contribute to the local economy through investments in infrastructure, wages to local workers, purchases from local businesses, and so on. Table 4.6 shows the government revenue from taxes and how local governments spend this money in communities around the parks.21 Nosy Be generates more than three times higher revenue from taxes than Ranomafana, demonstrating an overall larger local economy. However, most of the money Ranomafana earns is spent locally (96 percent) with a large proportion to male skilled workers (48 percent). Nosy Be spends a higher portion of revenue on female skilled workers than Ranomafana—16 percent compared to 9 percent—but only 70 percent of purchases and wages stay local and the local government spends a higher percentage on local stores and other businesses which, in turn, have higher leakages. 20 The percentage of the budget spent on local purchases is calculated as 100 percent minus the percentage of “purchases made outside the local economy.” 21 The calculation of local tax revenue is made from summing up the reported percentages of local taxes paid by farm and nonfarm businesses. 43 Measuring the Local Economic Impacts of Nature-Based Tourism in Madagascar TABLE 4.6 Revenue and Spending by the Local Governments of Ranomafana and Nosy Be TOTAL EXPECTED LOCAL GOVERNMENT REVENUE 2023* RANOMAFANA NOSY BE $170,547 $684,920 How much of the local government revenue goes to: Salaries and wages for male unskilled workers 8.0% 2.0% Salaries and wages for male skilled workers 48.0% 11.0% Salaries and wages for female unskilled workers 2.0% 1.0% Salaries and wages for female skilled workers 9.0% 16.0% Payment of rents on land, buildings, etc. 0.0% 1.0% Locally produced agricultural products (fruits, vegetables, meats) 0.0% 0.0% Locally produced fish or other natural resources 0.0% 0.0% Services (laundry, maintenance, construction, repairs) from local providers 25.0% 20.0% Purchases from local stores and other businesses 4.0% 19.0% Purchases made outside the local economy 4.0% 30.0% Percentage of salaries and wages paid to local workers 100% 100% Source: World Bank 2023. Note: * Estimated based on reported local taxes paid in the other farm and nonfarm sectors of the local economy. 44 SECTION 5 Using the Model to Simulate Impacts of Tourism Photo credit: © dblumenberg / Adobe Stock Measuring the Local Economic Impacts of Nature-Based Tourism in Madagascar To illustrate uses of the model, the two LEWIE-LITE dashboards were used to simulate impacts of changes in tourist spending surrounding Ranomafana and Nosy Tanikely National Parks (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 nonskilled 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 $100 could come from more tourists or the same number of tourists spending more. Other simulations can be undertaken (for example, increasing the number of tourists or increasing the spending of tourists per day; see appendix B for more information on how LEWIE-LITE can be used). Nosy Be. Photo credit: CEATIC 46 Measuring the Local Economic Impacts of Nature-Based Tourism in Madagascar Local Economy Impacts of Tourist Spending ($) 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. Ranomafana National Park The tourist spending multipliers for Ranomafana National Park are shown in figure 5.1. An additional dollar of tourist spending increases total local production by $1.75 and local income or GDP by $0.71. Most of this income gain, $0.48, accrues to nonpoor households although there is significant impact on poor household income rising by $0.23 per dollar spent by park visitors.22 Of the $0.71 per dollar increase in income, $0.23 is worker wages and $0.48 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 Ranomafana in figure 5.2. This was done through multiplying the previously mentioned multipliers and others produced by the model by $100 in extra tourist dollars. For example, the increase in poor- household income shown in figure 5.2b is the $100 increase in tourist spending times the multiplier on poor income, 0.23, giving a total impact of almost $23. Should tourist spending increase by $100, a total increase of $71 in the local GDP can be expected (that is, $48.62 in nonpoor plus $22.66 in poor household income). 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. 22 Due to rounding, adding up poor and nonpoor multipliers does not exactly equal the total income multiplier shown here (the total is off by $0.01). If they wish, users of the model can add precision by including more digits to the right of the decimal. point. However, since LEWIE-LITE is a quick assessment tool, the authors decided not to do this to simplify the presentation of simulation results. 47 Measuring the Local Economic Impacts of Nature-Based Tourism in Madagascar The large impacts on nonagricultural activities 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 all six production bars in figure 5.2a, the total impact on all production sectors, $175, exceeds the simulated $100 increase in tourist spending. FIGURE 5.1 LEWIE-LITE Dashboard Visual of Tourist Spending Multipliers for Ranomafana National Park Local-economy impacts of tourist spending ($): For every dollar of tourist spending, the total production multiplier is: WHICH CAN BE SPLIT INTO: $1.75 $0.75 Total production multiplier Tourism activities $1.00 Nontourism activities For every dollar of tourist spending, the total income multiplier is: WHICH CAN BE SPLIT INTO: OR. ALTERNATIVELY, CAN BE SPLIT INTO: $0.71 $0.23 $0.23 Total income multiplier Accruing to labor Accruing to poor households $0.48 $0.48 Accruing to capital Accruing to nonpoor households Source: World Bank 2023. 48 Measuring the Local Economic Impacts of Nature-Based Tourism in Madagascar FIGURE 5.2 Effects of a $100 Increase in Tourist Spending on the Local Economy Around Ranomafana National Park E ects of this tourism spending... A. ...ON PRODUCTION B. ...ON INCOMES C. ...ON LABOR INCOME 48.62 12.4 94.86 Additional Production Value ($) Additional Labor Income ($) Additional Income ($) 45.27 22.66 5.2 3.42 23.94 2 5.32 3.87 1.59 Lodges Restaurants Tourism 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 2023. Ranomafana. Photo credit: CEATIC 49 Measuring the Local Economic Impacts of Nature-Based Tourism in Madagascar Nosy Tanikely National Park Production, income, and wage and profit multipliers are much larger at Nosy Tanikely National Park, as shown in figure 5.3. An additional dollar of spending by Nosy Tanikely park visitors in Nosy Be increases total local production by $4.36 and local income or GDP by $2.48. These impacts are much larger than at Ranomafana National Park due to the larger share of local spending among farm and nonfarm businesses and the smaller share of household spending on local stores, services, and production, which have high leakages. FIGURE 5.3 LEWIE-LITE Dashboard of Tourist Spending Multipliers for Nosy Tanikely National Park Local-economy impacts of tourist spending ($): For every dollar of tourist spending, the total production multiplier is: WHICH CAN BE SPLIT INTO: $4.36 $1.11 Total production multiplier Tourism activities $3.25 Nontourism activities For every dollar of tourist spending, the total income multiplier is: WHICH CAN BE SPLIT INTO: OR. ALTERNATIVELY, CAN BE SPLIT INTO: $2.48 $1.15 $1.38 Total income multiplier Accruing to labor Accruing to poor households $1.33 $1.11 Accruing to capital Accruing to nonpoor households Source: World Bank 2023. The actual income multipliers accruing to poor and nonpoor households in Nosy Tanikely are $1.375 and $1.105, respectively, which is equivalent to the total income multiplier of $2.48. However, for purposes of this report, they have been rounded off to $1.38 and $1.11. 50 Measuring the Local Economic Impacts of Nature-Based Tourism in Madagascar Both Ranomafana and Nosy Tanikely generate large benefits to poor households per dollar of tourist spending—$0.23 and $1.38, respectively. Poor households receive 32 percent of the income multiplier in Ranomafana and 56 percent in Nosy Be. In Nosy Be, poor households receive slightly more income per tourist dollar than nonpoor households—poor households’ income rises by $1.38 per tourist dollar spent, while nonpoor households’ income rises by $1.11. Relatively large impacts on poor household incomes suggest the importance of protected area tourism for poverty alleviation. These results do not mean that poor households receive more income per household on average from tourism than nonpoor households. Per-household impacts depend on how many households there are. For a given group (for example, poor households), the per- household impact would be calculated as the total group impact divided by the number of households in the group. Although poor households receive 32 percent and 56 percent of every dollar of income generated by tourism at Ranomafana and Nosy Be, respectively, there is a considerably higher number of poor households. Poor households make up over 80 percent of the population at both sites. Because of this, on average, a nonpoor household in Ranomafana and Nosy Be receives more income and income spillover from visitors than a poor household. Nonpoor households have a larger share of employment and ownership of capital (that is, businesses) in the local economies surrounding the two parks. The simulated production impacts of an additional $100 of tourist spending in the local economy around Nosy Tanikely National park are shown in figure 5.4, disaggregated by sector, income impacts by household group, and wage income impacts by worker group. Similarly to Ranomafana, the largest production impacts at Nosy Be are on nonagricultural activities (figure 5.4a). Of particular interest here is that female skilled workers (figure 5.4c) benefit more than at Ranomafana because they receive a larger share of income from wages in retail, service, and local production activities as well as from park employment. Table 5.1 summarizes these multiplier impacts of the extra $100 in tourist spending on the economies surrounding Ranomafana and Nosy Tanikely National Parks. 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 Nosy Be, agriculture. Also notable is the fact that a large portion of the incomes goes to female unskilled and skilled labor, especially in Nosy Be. Ranomafana. Photo credit: CEATIC 51 Measuring the Local Economic Impacts of Nature-Based Tourism in Madagascar FIGURE 5.4 Effects of a $100 Increase in Tourism Spending on the Local Economy Around Nosy Tanikely National Park E ects of this tourism spending... A. ...ON PRODUCTION B. ...ON INCOMES C. ...ON LABOR INCOME 49.24 217.95 137.51 110.54 Additional Production Value ($) Additional Labor Income ($) 33.13 Additional Income ($) 19.83 74.06 77.95 12.7 32.63 29.33 4.12 Lodges Restaurants Tourism 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 2023. multipliers, this translates into an additional $209 GDP gain per visitor at Ranomafana and $749 per visitor at Nosy Be (attributable to Nosy Tanikely). When applying the number of tourists expected to visit each protected area in 2023—15,200 and 31,000, respectively—the total amount generated by these protected area tourists is $3.17 million at Ranomafana and $23.23 million at Nosy Tanikely. Each amount far exceeds the annual park budget to operate these sites ($0.23 million and $0.35 million, respectively). Tourism benefits poor households at both sites. It provides $1.01 million to poor households surrounding Ranomafana and $12.88 million to poor households in Nosy Be. 52 Measuring the Local Economic Impacts of Nature-Based Tourism in Madagascar TABLE 5.1 Summary of the Impact of a $100 Increase in Tourist Spending and the Impact of an Additional Tourist IMPACTS OF TOURIST SPENDING OUTCOME $ % RANOMAFANA NOSY BE RANOMAFANA NOSY BE Impacts per $100 of tourist spending Gross revenue from local production Agriculture $4 $78 2% 18% Fishing $2 $29 1% 7% Tourism businesses $24 $33 14% 7% Retail/services/production $95 $218 54% 50% Restaurants $5 $4 3% 1% Lodges $45 $74 26% 17% Total production multiplier $175 $436 100% 100% Payments to: Labor (wages) $23 $114 100% 100% Male unskilled labor (wages) $12 $49 55% 43% Female unskilled labor (wages) $3 $13 14% 11% Male skilled labor (wages) $5 $20 23% 17% Female skilled labor (wages) $2 $32 8% 28% Capital (profits) $49 $133 100% 100% Income to: Poor households $23 $138 32% 55% Nonpoor households $49 $111 68% 45% Total income (GDP) multiplier $71 $248 100% 100% Park revenue $5.10 $3.60 Community revenue $0.48 $0.54 Impacts of an additional tourist Average spending per tourist $293 $302 Total local GDP impact per tourist $209 $749 100% 100% To poor households $66 $415 32% 56% To nonpoor households $142 $334 68% 45% Source: World Bank 2023. The following table also shows total impacts per visitor. A tourist spends $293 on average around Ranomafana and $302 around Nosy Tanikely. Using the income (local GDP) 53 Measuring the Local Economic Impacts of Nature-Based Tourism in Madagascar Impacts of Future Growth in Tourism The LEWIE-LITE tool was used to estimate the impact of projected growth in protected area tourism for Ranomafana and Nosy Tanikely National Parks. COVID-19 was a major shock to Madagascar’s tourism, causing an 89 percent decline in visitors to national parks—from 229,970 visitors in 2019 to 25,151 in 2020.23 Tourism has only partially recovered. From 2020 to 2021, the number of park visitors increased by 22 percent, from 25,151 to 30,891. Surveys of park authorities found that the number of visitors to Ranomafana was expected to increase 28.4 percent—from 11,837 in 2022 to 15,200 expected visitors in 2023. The number at Nosy Tanikely was expected to increase 11.1 percent—from 27,906 in 2022 to 31,000 in 2023. These percentages are consistent with the 22 percent national increase between 2020 and 2021.24 At these same rates of increase, one could expect 4,318 more visitors to Ranomafana and 3,437 to Nosy Tanikely in 2024.25 The model was used to estimate the likely impacts of these increases in tourism on the local economies around each park. At the average spending per tourist estimated from the survey data, these increases would add $1.27 million and $1.04 million in visitor spending at Ranomafana and Nosy Be, respectively. Entering this average additional tourist spending onto the dashboard for tourism impacts, the model gives the local economy impacts shown in figures 5.5 and 5.6.26 At Ranomafana (figure 5.5), the overall impact on the local economy from this increase in tourism would be a $2.21 million increase in local production and a $901,000 increase in income to households. For Nosy Tanikely (figure 5.6), the impact on the local economy of Nosy Be would be a $4.53 million increase in local production and a $2.58 million increase in income to households.27 23 MNP statistics 2023. 24 Assuming that the protected areas can absorb the new tourists without a cost to biodiversity or other park constraints. This is a reasonable assumption for the near future, given park authorities’ responses to questions on visitor numbers in 2022 and expected visitor numbers for 2023. Also, Madagascar has not yet achieved prepandemic numbers of total visitors. 25 These numbers are slightly different from what one would get by multiplying the growth rates by the 2023 visitor numbers due to rounding error. 26 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. 27 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. 54 Measuring the Local Economic Impacts of Nature-Based Tourism in Madagascar FIGURE 5.5 Effects of a $1.27 Million Increase in Tourist Spending on the Local Economy Around Ranomafana National Park How much tourist spending ($) do you want to simulate? 1265 E ects of this tourism spending... A. ...ON PRODUCTION B. ...ON INCOMES C. ...ON LABOR INCOME 1199.99 615.04 156.84 Additional Production Value ($) Additional Labor Income ($) Additional Income ($) 572.7 286.61 65.81 302.81 43.26 25.33 67.33 47.85 20.06 Lodges Restaurants Tourism 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 2023. FIGURE 5.6 Effects of a $1.04 Million Increase in Tourist Spending on the Local Economy Around Nosy Tanikely National Park How much tourist spending ($) do you want to simulate? 1039 E ects of this tourism spending... A. ...ON PRODUCTION B. ...ON INCOMES C. ...ON LABOR INCOME 2264.46 1428.76 511.64 1148.46 Additional Production Value ($) Additional Labor Income ($) 344.27 Additional Income ($) 206.08 769.46 809.87 131.93 339.04 304.72 42.76 Lodges Restaurants Tourism 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 2023. 55 SECTION 6 Testing the Robustness of Results and Analyzing the Differences in Multipliers Between the Two Protected Areas Nosy Be. Photo credit: CEATIC Measuring the Local Economic Impacts of Nature-Based Tourism in Madagascar 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. TABLE 6.1 Incomes in the Local Economies Around Each Park RANOMAFANA NATIONAL PARK HOUSEHOLD GROUP POPULATION TOTAL INCOME PER CAPITA INCOME $ PER DAY Poor 7,764 $1,027,407 $132 $0.36 Nonpoor 1,941 $2,064,050 $1,063 $2.91 NOSY TANIKELY NATIONAL PARK HOUSEHOLD GROUP POPULATION TOTAL INCOME PER CAPITA INCOME $ PER DAY Poor 86,613 $79,294,278 $916 $2.51 Nonpoor 21,653 $63,372,273 $2,927 $8.02 Source: World Bank 2023. At Ranomafana, poor households have per capita incomes substantially less than $1 per day, whereas at Nosy Be, poor households have per capita income a little over $2.50 per day. Nonpoor households also have more money at Nosy Be than at Ranomafana. The incomes of both household groups are lower, on average, near Ranomafana National Park. One explanation for this is that, in contrast to the more remote and landlocked area of Ranomafana, there are more economic opportunities in communities around Nosy Tanikely National Park. Nosy Tanikely National Park also attracts a more affluent category of visitors drawn by resort hotels, nice beaches, and higher-spend tourism products such as diving. 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 57 Measuring the Local Economic Impacts of Nature-Based Tourism in Madagascar rarely purchase food directly from farmers, yet at Nosy Be, an additional dollar of tourist spending has a large, positive effect on local farm and fisher 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). 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. The total income and production multipliers are larger for Nosy Tanikely than Ranomafana. It is particularly striking that an additional dollar of tourist spending increases local household income by more than one dollar at Nosy Be. Several factors shape the size of these multipliers. Foremost among these is that the local economy surrounding Ranomafana is less developed, likely due to its remote setting and terrestrial nature, and this creates more reliance on goods and services obtained from outside markets. Local purchases contribute to multipliers by keeping more cash circulating 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. Connections with outside markets through trade can offer many advantages for local economies. For example, it can provide 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 much larger at Ranomafana, as shown in table 6.2, and this explains the smaller multipliers there. 58 Measuring the Local Economic Impacts of Nature-Based Tourism in Madagascar TABLE 6.2 Leakage Share from Production Sectors and Households ($) Ranomafana National Park Nosy Tanikely National Park (Nosy Be) PARK AND EXTERNAL ACCOUNT Expenditures Goods and Total Leakage Expenditures Goods and Total expenditures Leakage from the sector or services expenditures share from the sector services share household group purchased or household purchased from to the national from outside group to outside of the government of the local the national local economy economy government by sector or by sector or household household group group PRODUCTION AGRICULTURE 29,993 163,001 18.40% 236,305 2,835,659 44,712,375 6.90% SECTOR/ ACTIVITY TOURISM 25,066 270,004 970,887 30.40% 454,148 7,058,761 24,713,569 30.40% NONAGRICULTURE 132,775 2,119,244 4,024,238 56.00% 3,557,523 29,144,323 126,276,687 25.90% FISH - 71,917 71,917 100.00% - 4,924,246 16,802,434 29.30% RESTAURANTS 7,423 67,713 215,167 34.90% 48,748 621,541 3,896,127 17.20% LODGES 97,659 739,775 1,830,164 45.80% 563,405 9,317,291 33,709,625 29.30% HOUSEHOLD POOR 10,274 123,289 1,027,407 13.00% - 1,982,357 79,294,278 2.5% GROUP NONPOOR 5,755 473,408 2,064,050 23.20% 75 1,584,307 63,372,273 2.5% PROTECTED 24,612 230,962 10.70% - 11,437 239,905 4.80% AREAS COMMUNITY 8,294 9,992 83.00% . 11,177 18,323 61.00% PROJECT LOCAL - 6,910 172,739 4.00% . 4,566,378 15,069,046 30.30% GOVERNMENT Source: Authors based on balanced SAM . 59 SECTION 7 Using the Model to Simulate Changes in Park and Community Projects and Local Government Photo credit: © ggfoto / Adobe Stock Measuring the Local Economic Impacts of Nature-Based Tourism in Madagascar Impacts of Changes in Park and Community Projects Governments generally decide on protected area budgets and the share (if any) of park revenue going to community projects. 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. 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 project spending. To illustrate the use of this feature, the dashboard was used to simulate the impacts of $100 of park and community project spending on the local economy around each park, and summarize the impacts on production, household incomes, and wages. At Ranomafana, a $100 increase in park spending (or park budget) leads to a $106 increase in local GDP, or an income multiplier of 1.06 for the local economy (with impacts disaggregated in figure 7.1). At Nosy Tanikely, a $100 increase in park spending leads to a $281 increase in local GDP, or a local income multiplier of 2.81 (figure 7.3). In comparison, a $100 increase in community project spending leads to an increase of $22 in local GDP, or an income multiplier of 0.22 in Ranomafana (figure 7.2) while a similar increase of $100 in community project spending in Nosy Tanikely leads to a GDP gain of $110 or a local income multiplier of 1.10 (figure 7.4). Again, the multiplier for Ranomafana is much less than one—a result of its high reliance on imported goods and services for community projects. It is possible that tenders for local community projects could create more local benefits by encouraging the purchase of local materials, if available. However, this might not be sufficient to substantially increase the local income multiplier given the large leakages from the Ranomafana economy. In contrast, at both parks the impact of park revenue on local incomes has a greater than $1 impact on the local economy for every additional dollar in park spending. The local GDP multiplier from park spending is 1.06 at Ranomafana and 2.81 at Nosy Be. Park spending favors local labor and local goods and services, and that is why there is a more than 1-for-1 increase in local income. This is on top 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. 61 Measuring the Local Economic Impacts of Nature-Based Tourism in Madagascar FIGURE 7.1 Effects of a $100 Increase in Park Spending on the Local Economy Around Ranomafana National Park E ects of this park budget... A. ...ON PRODUCTION B. ...ON INCOMES C. ...ON LABOR INCOME 109.36 46.41 61.66 Additional Production Value ($) Additional Labor Income ($) Additional Income ($) 44.32 12.62 7.67 4.48 3.1 0.7 0 0 1.5 Lodges Restaurants Tourism 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 2023. FIGURE 7.2 Effects of a $100 Increase in Community Project Spending on the Local Economy Around Ranomafana National Park E ects of this community spending... A. ...ON PRODUCTION B. ...ON INCOMES C. ...ON LABOR INCOME 21.32 13.05 11.24 Additional Production Value ($) Additional Labor Income ($) Additional Income ($) 8.8 1.83 1.04 0.91 0.62 0.14 0 0 0.31 Lodges Restaurants Tourism 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 2023. 62 Measuring the Local Economic Impacts of Nature-Based Tourism in Madagascar FIGURE 7.3 Effects of a $100 Increase in Park Spending on the Local Economy Around Nosy Tanikely National Park E ects of this park budget... A. ...ON PRODUCTION B. ...ON INCOMES C. ...ON LABOR INCOME 283.57 153.88 53.32 Additional Production Value ($) 50.5 Additional Labor Income ($) Additional Income ($) 127.54 32.98 17.9 85.71 7.96 30.85 0 0 Lodges Restaurants Tourism 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 2023. FIGURE 7.4 Effects of a $100 Increase in Community Project Spending on the Local Economy Around Nosy Tanikely National Park E ects of this community spending... A. ...ON PRODUCTION B. ...ON INCOMES C. ...ON LABOR INCOME 124.32 60.03 23.22 Additional Production Value ($) Additional Labor Income ($) Additional Income ($) 49.57 17.74 8.38 33.76 4.35 6.97 12.02 0 0 Lodges Restaurants Tourism 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 2023. 63 Measuring the Local Economic Impacts of Nature-Based Tourism in Madagascar 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 site. 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). To take a hypothetical example, suppose historical experience suggests that if the government invests an additional $1 million in Nosy Tanikely park amenities and staff, it will attract a minimum of 10 percent more tourists. The dashboard can be used to simulate a $1 million increase in park spending and 10 percent increase in the number of park visitors. The results from this simulation show a $2.81 million increase in Nosy Be income from park spending plus an additional $2.01 million increase from increased tourist spending. In this example, the benefit-cost ratio would be $4.8 ($2.81 + $2.01) per $1 increase in park investment. It would be even greater if the increase in tourism due to higher park spending became permanent. Park purchases outside the local economy are a leakage that reduces local income multipliers. The model could also be used to explore the impacts of each dollar the park spends inside the local economy— that is, the park spending multipliers net of local imports. The percentage of park budget spent on outside goods is 0.11 at Ranomafana and 0.05 at Nosy Tanikely (table 4.5). To obtain the multiplier effects of park spending inside the local economy, one would multiply the GDP multiplier by 1/(1-.11) or 1.12 for Ranomafana and 1/(1- .05) or 1.05 for Nosy Tanikely. This would give a total household income multiplier of 1.19 at Ranomafana and 2.95 at Nosy Tanikely—larger than the multipliers of 1.06 and 2.81 reported earlier. (The other park spending multipliers previously presented can be converted to local spending multipliers by multiplying them by the same conversion coefficient). Photo credit: © Artush / Shutterstock 64 Measuring the Local Economic Impacts of Nature-Based Tourism in Madagascar Impacts of Changes in Local Government Spending Local governments generally decide how their tax revenue will be spent on local services, infrastructure development, and wages for government employees. They are important actors in the local economy surrounding the two national parks. Although tourists do not directly give the local government dollars, governments can also benefit from an increase in visitors through the additional tax revenue they collect from lodges, restaurants, and other tourism and nontourism businesses. Additional spending by the local government helps stimulate the local economy. The model can also be used to simulate impacts of changes in local government spending. To illustrate the use of this feature, the dashboard was used to simulate the impacts of $100 in local government spending on the local economy around each park, and assess impacts on production, household incomes, and wages. At Ranomafana, a $100 increase in local government revenue leads to a $123 increase in local GDP, or an income multiplier of 1.23 for the local economy (figure 7.5). At Nosy Be, a similar $100 increase in local government revenue leads to a $230 increase in local GDP, or a local income multiplier of 2.3 (figure 7.6). Both parks show a greater than $1 increase in GDP for every additional $1 spent by the local government. The local governments spend in such a way as to increase, directly and indirectly, the nontourism production sector, especially local retailers, services, and other production. They even have strong impacts on agriculture, especially at Nosy Be, where demand linkages are strong between households and local farms. Among worker groups at Nosy Be, the largest impact is on skilled female workers. Sixteen percent of local government revenue is used to hire skilled female workers (as shown in table 4.6). Adding in indirect impacts, this contributes to a $46 increase in earnings for skilled female workers for every $100 in local government spending. The second largest impact is on unskilled male workers ($44), followed by skilled male workers ($25) and unskilled female workers ($7). A limitation to the model is that if community revenue sharing or community investment by the parks increase, this may cause the local government to reduce its spending. The model does not take this into account when simulating increases in community investment. This may be an area of future research where local government spending priorities and processes are incorporated into the model or an alternative method is used to complement LEWIE-LITE. 65 Measuring the Local Economic Impacts of Nature-Based Tourism in Madagascar FIGURE 7.5 Effects of a $100 Increase in Local Government Spending on the Local Economy Around Ranomafana National Park E ects of this increase in local government spending... A. ...ON PRODUCTION B. ...ON INCOMES C. ...ON LABOR INCOME 120.64 69.01 49.34 Additional Production Value ($) Additional Labor Income ($) Additional Income ($) 54.22 18.41 9.25 5.35 3.8 2.78 0 0 1.77 Lodges Restaurants Tourism 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 2023. FIGURE 7.6 Effects of a $100 Increase in Local Government Spending on the Local Economy Around Nosy Tanikely National Park E ects of this increase in local government spending... A. ...ON PRODUCTION B. ...ON INCOMES C. ...ON LABOR INCOME 215.71 124.71 45.66 43.76 Additional Production Value ($) Additional Labor Income ($) 104.76 Additional Income ($) 24.49 69.39 7.12 25.12 14.62 0 0 Lodges Restaurants Tourism 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 2023. 66 SECTION 8 Using the Model to Simulate Complementary Interventions Photo credit: © dennisvdwater / Adobe Stock Measuring the Local Economic Impacts of Nature-Based Tourism in Madagascar 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.28 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 Ranomafana, a $100 increase in demand for local agricultural production leads to a local GDP multiplier of 1.05 (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 only 0.6 (figure 8.2). For the economy surrounding Nosy Tanikely, the $100 increase in demand for local agricultural production leads to a local GDP multiplier of 3.13 (figure 8.3) and a $100 increase in demand for local nonagricultural production leads to a local GDP multiplier of 2.66 (figure 8.4). The multipliers from increases in agricultural production are larger than those from tourism for a variety of reasons. The most important is a higher share of local spending by agriculture. Table 6.2 showed that leakages from the agricultural sector’s spending are only 6.9 percent at Nosy Be, whereas for lodges they are 29.3 percent, 17.2 percent for restaurants, and 30.4 percent for other tourism activities. Leakages from agriculture are higher at Ranomafana (18.4 percent), but they are nonetheless lower than from businesses engaged in tourism. 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 in Ranomafana. The larger leakages in the local economy surrounding Ranomafana National Park, specifically in the nonagricultural sector, lead to this smaller multiplier effect. Even in this case, local GDP rises by 60 cents for every $1 increase in nonagricultural production. 28 There are no impacts on the park and community projects because there are no feedback effects from production activities on their budgets. 68 Measuring the Local Economic Impacts of Nature-Based Tourism in Madagascar The ministries of agriculture and fisheries may be interested in using LEWIE-LITE to simulate changes in local agricultural production. Since the model is inclusive of the local economy surrounding the two parks, considering the full range of industries and households, one could use this model for purposes other than tourism impact analysis. FIGURE 8.1 Effects of a $100 Increase in Local Agricultural Production on the Local Economy Around Ranomafana National Park E ects of this increase in local agricultural production... A. ...ON PRODUCTION B. ...ON INCOMES C. ...ON LABOR INCOME 107.13 71.41 24.65 Additional Production Value ($) 100.88 Additional Labor Income ($) Additional Income ($) 33.36 4.11 2.04 1.38 2.34 0.38 0 0 Lodges Restaurants Tourism 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 2023. FIGURE 8.2 Effects of a $100 Increase in Local Nonagricultural Production on the Local Economy Around Ranomafana National Park E ects of this increase in local nonagricultural production... A. ...ON PRODUCTION B. ...ON INCOMES C. ...ON LABOR INCOME 153.41 43.1 12.44 Additional Production Value ($) Additional Labor Income ($) Additional Income ($) 16.98 1.58 0.83 0.77 2.12 1.19 0.3 0 0 Lodges Restaurants Tourism 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 2023. 69 Measuring the Local Economic Impacts of Nature-Based Tourism in Madagascar FIGURE 8.3 Effects of a $100 Increase in Local Agricultural Production on the Local Economy Around Nosy Tanikely National Park E ects of this increase in local agricultural production... A. ...ON PRODUCTION B. ...ON INCOMES C. ...ON LABOR INCOME 282.8 173.06 73.17 Additional Production Value ($) Additional Labor Income ($) Additional Income ($) 140.03 202.91 39.21 18.6 34.4 9.93 19.86 0 0 Lodges Restaurants Tourism 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 2023. FIGURE 8.4 Effects of a $100 Increase in Local Nonagricultural Production on the Local Economy Around Nosy Tanikely National Park E ects of this increase in local nonagricultural production... A. ...ON PRODUCTION B. ...ON INCOMES C. ...ON LABOR INCOME 321.63 146.32 56.08 Additional Production Value ($) Additional Labor Income ($) Additional Income ($) 119.81 43.1 18.83 82.57 8.53 29.21 16.9 0 0 Lodges Restaurants Tourism 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 2023. 70 Measuring the Local Economic Impacts of Nature-Based Tourism in Madagascar 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 were simulated, summarizing impacts on production, household incomes, and wages.29 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 poor 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 Ranomafana National Park, a $100 increase in wage earnings for either unskilled female workers or skilled female workers leads to a GDP multiplier of 1.61 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. For unskilled male workers, a $100 increase in wage earnings leads to a GDP multiplier of 1.61 for the local economy (figure 8.7), reporting the same result as for female workers. A $100 increase in wage earnings for skilled male workers leads to a slightly lower local GDP multiplier of 1.57 (figure 8.8). Results from Nosy Tanikely report much higher 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 4.11. 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 leads to the same local GDP multiplier of 4.11 (figure 8.11). Finally, a $100 increase in wage earnings for skilled male workers leads to a local GDP multiplier of 4.12, with disaggregated impacts shown in figure 8.12. 29 Impacts on the park and community projects are nil because there are no feedback effects from local wage earnings to their budgets. 71 Measuring the Local Economic Impacts of Nature-Based Tourism in Madagascar Ranomafana National Park FIGURE 8.5 Effects of a $100 Increase in Earnings for Unskilled Female Workers on the Local Economy Around Ranomafana National Park E ects of this increase in earnings to low-skilled female workers... A. ...ON PRODUCTION B. ...ON INCOMES C. ...ON LABOR INCOME 113.28 99.07 100.86 Additional Production Value ($) Additional Labor Income ($) Additional Income ($) 61.8 10.36 8.44 6.94 2.6 0.25 1.34 0 0 Lodges Restaurants Tourism 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 2023. FIGURE 8.6 Effects of a $100 Increase in Earnings for Skilled Female Workers on the Local Economy Around Ranomafana National Park E ects of this increase in earnings to skilled female workers... A. ...ON PRODUCTION B. ...ON INCOMES C. ...ON LABOR INCOME 113.27 99.78 100.25 Additional Production Value ($) Additional Labor Income ($) Additional Income ($) 61.8 10.36 8.48 6.98 2.61 0.86 1.34 0 0 Lodges Restaurants Tourism 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 2023. 72 Measuring the Local Economic Impacts of Nature-Based Tourism in Madagascar FIGURE 8.7 Effects of a $100 Increase in Earnings for Unskilled Male Workers on the Local Economy Around Ranomafana National Park E ects of this increase in earnings to low-skilled male workers... A. ...ON PRODUCTION B. ...ON INCOMES C. ...ON LABOR INCOME 113.27 99.78 110.36 Additional Production Value ($) Additional Labor Income ($) Additional Income ($) 61.8 8.48 6.98 2.61 0.86 1.34 0 0 0.25 Lodges Restaurants Tourism 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 2023. FIGURE 8.8 Effects of a $100 Increase in Earnings for Skilled Male Workers on the Local Economy Around Ranomafana National Park E ects of this increase in earnings to skilled male workers... A. ...ON PRODUCTION B. ...ON INCOMES C. ...ON LABOR INCOME 113.8 93.38 101.31 Additional Production Value ($) Additional Labor Income ($) Additional Income ($) 63.23 10.08 6.51 2.2 4.43 0.79 0 0 0.24 Lodges Restaurants Tourism 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 2023. 73 Measuring the Local Economic Impacts of Nature-Based Tourism in Madagascar Nosy Tanikely National Park FIGURE 8.9 Effects of a $100 Increase in Earnings for Unskilled Female Workers on the Local Economy Around Nosy Tanikely National Park E ects of this increase in earnings to low-skilled female workers... A. ...ON PRODUCTION B. ...ON INCOMES C. ...ON LABOR INCOME 293.77 222.97 109.08 Additional Production Value ($) Additional Labor Income ($) Additional Income ($) 188.39 64.89 121.84 41.32 45.02 19.68 26.22 0 0 Lodges Restaurants Tourism 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 2023. FIGURE 8.10 Effects of a $100 Increase in Earnings for Skilled Female Workers on the Local Economy Around Nosy Tanikely National Park E ects of this increase in earnings to skilled female workers... A. ...ON PRODUCTION B. ...ON INCOMES C. ...ON LABOR INCOME 293.61 218.53 141.29 Additional Production Value ($) Additional Labor Income ($) 192.64 Additional Income ($) 64.74 121.35 44.82 19.66 26.34 9.07 0 0 Lodges Restaurants Tourism 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 2023. 74 Measuring the Local Economic Impacts of Nature-Based Tourism in Madagascar FIGURE 8.11 Effects of a $100 Increase in Earnings for Unskilled Male Workers on the Local Economy Around Nosy Tanikely National Park E ects of this increase in earnings to low-skilled male workers... A. ...ON PRODUCTION B. ...ON INCOMES C. ...ON LABOR INCOME 293.42 213.51 164.57 Additional Production Value ($) Additional Labor Income ($) 197.44 Additional Income ($) 120.81 41.26 44.6 26.47 19.65 9.06 0 0 Lodges Restaurants Tourism 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 2023. FIGURE 8.12 Effects of a $100 Increase in Earnings for Skilled Male Workers on the Local Economy Around Nosy Tanikely National Park E ects of this increase in earnings to skilled male workers... A. ...ON PRODUCTION B. ...ON INCOMES C. ...ON LABOR INCOME 293.91 226.65 119.69 Additional Production Value ($) Additional Labor Income ($) Additional Income ($) 184.87 65.02 122.24 41.34 45.18 26.12 9.06 0 0 Lodges Restaurants Tourism 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 2023. 75 Conclusions and Recommendations Photo credit: © aylerein / Adobe Stock Measuring the Local Economic Impacts of Nature-Based Tourism in Madagascar 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 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 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 14 percent of employees across all tourism-related businesses (hotels, restaurants, and other tourism businesses) are women; however, in nontourism-related businesses, they are only employed in retail and not in agriculture or fishing. 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 Ranomafana and Nosy Tanikely National Parks are greater than 1 (1.75 and 4.36, respectively). 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, especially at Nosy Be, 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, including the income of poor households Local income rises by more than one dollar per dollar spent by tourists at Nosy Tanikely National Park and by 71 cents at Ranomafana National Park. The income multiplier at Nosy Tanikely is 2.48, of which 1.10 goes to nonpoor households and 1.38 goes to poor households. The smaller GDP multiplier at Ranomafana is due to leakages from the purchase of goods and services outside the local economy. The 71 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 Nosy Be. 77 Measuring the Local Economic Impacts of Nature-Based Tourism in Madagascar Poor households benefit from the direct and indirect impacts of tourism in the local economies surrounding the two parks. In Ranomafana, 32 percent of the income generated from tourist spending goes to poor households. The impact is greater at Nosy Be with 56 percent of income gains accruing to poor households.30 These findings suggest the potential for nature-based tourism to contribute towards poverty alleviation in these communities. 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 $209 during their stay at Ranomafana and $749 at Nosy Tanikely. 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 expected to visit these protected areas in 2023 (15,200 at Ranomafana and 31,000 at Nosy Tanikely), the total income generated by tourists is approximately $3.2 million at Ranomafana and $23.2 million at Nosy Be. Of this, poor households receive $1 million at Ranomafana and $12.9 million at Nosy Be. These amounts far exceed the current park budgets at these sites, which are $234,000 and $352,000, respectively. If the number of future tourists for 2023–24 increases at the same annual growth rates,31 the protected areas will generate an additional production value to businesses of $2.21 million and $4.53 million and income to households of $901 thousand and $2.58 million, respectively, for the local economies of Ranomafana Commune and Nosy Be. Parks support the local economies through their spending on local labor and local goods and services. Every $100 increase in park spending adds $106 to the GDP of communities around Ranomafana and $281 to income around Nosy Tanikely. Income multipliers from park spending are greater than one: every additional dollar spent by the parks increases local incomes 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 $161 in local income to households in Ranomafana and $411 in Nosy Be. If a program was to generate additional female employment in the local economy, the multiplier impact in local incomes would be large, 30 It is important to note that over 80 percent of these two economies are poor; therefore, even capturing 56 percent of local economy benefits means each household receives less income than the equivalent nonpoor household. 31 Annual growth rates for 2022–23 were 28 percent for Ranomafana and 11 percent for Nosy Tanikely. These growth rates, given by park authorities, are similar to that of national tourist numbers from 2020–21 of 22.8 percent and are conservative, given that Madagascar’s national tourism decreased by 89 percent due to COVID-19 from 2019–20 and has not yet achieved prepandemic levels. These estimates also assume that the protected areas can absorb the new tourists without a cost to biodiversity or other park constraints. This is a reasonable assumption, given the lower than historical achieved visitations. 78 Measuring the Local Economic Impacts of Nature-Based Tourism in Madagascar 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. 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 Photo credit: © dennisvdwater / Adobe Stock 79 Measuring the Local Economic Impacts of Nature-Based Tourism in Madagascar 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 depends 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. Photo credit: © Bertrand Godfroid / Adobe Stock 80 Measuring the Local Economic Impacts of Nature-Based Tourism in Madagascar Recommendations Based on the above analysis, this report offers several recommendations for policy makers to promote Madagascar’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. Promote local sourcing The economic benefits of tourism are reduced in rural communities with less developed local economies in which goods and service are purchased in urban centers farther away. This means less money is retained in the local economy and economic impacts are generated. Local linkages to sectors such as agriculture, fishing, and manufacturing could be strengthened so that multipliers increase. Using Ranomafana as an example, if local businesses sourced more inputs locally, the production and income multipliers from tourism could be boosted. It may be necessary to address common challenges that affect local sourcing for tourism, such as the quality and quantity of local goods, prices, infrastructure and pricing, and communications with producers and vendors. The pilot’s findings prompt further studies to explore how to strengthen local market linkages in Ranomafana and to grow the local economy to reduce leakages and maximize the impacts of tourism. 2. Scale up benefit sharing Sharing the benefits is the means by which both tangible and intangible returns from nature-based tourism are channeled to the local communities living in and around the protected areas. It is a mechanism through which local development can be financed directly from revenue generated by tourism. This can be through policies which establish a formal mechanism for benefit sharing, small investments in infrastructure (schools, clinics, water supply, etc.) and, addressing human-wildlife conflict. 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. National tourism revenue sharing programs such as those in Rwanda and Uganda may provide interesting examples of how to formalize and scale up benefit sharing from protected area tourism with surrounding communities. Local economies around protected areas in other countries or at other sites in Madagascar can benefit from community projects agreements like those at the two sites studied here. 3. Strengthen employment and training 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 Ranomafana and Nosy Be, particularly for women. Tourism requires skills and training, which can increase the types of jobs accessible to workers and promote entrepreneurship. Skills training for workers from poor households could increase the tourism benefits that poor households capture. The World Bank is currently involved in such programs in Madagascar. The private sector, such as tour operators and hotels, can play a role by providing and expanding employment programs, targeting women and workers from poorer communities or households, and offering attractive jobs and fair remuneration. 81 Measuring the Local Economic Impacts of Nature-Based Tourism in Madagascar 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 Madagascar, 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 making 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 local 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 Madagascar. 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 both terrestrial and marine parks, applying the tool to other protected areas in Madagascar would allow comparison across sites and help obtain a national average of local production, employment, and income impacts. 82 References CEATIC (Center for Advanced Studies in Information and Communication Technologies). 2023. “PAT-SAM LEWIE Multiplier Data Collection in Madagascar.” Final report prepared for the World Bank, Jaén. Stone, Richard and Alan Brown. 1962. “A Computable Model for Economic Growth.” In A Programme for Growth, Volume 1, edited by Richard Stone. Cambridge: Chapman and Hall. 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. World Bank. 2021. Banking on Protected Areas: Promoting Sustainable Protected Area Tourism to Benefit Local Economies. https://www.worldbank.org/en/topic/ environment/publication/banking-on-protected-areas-promoting-sustainable- protected-area-tourism-to-benefit-local-communities.  World Bank. 2022. Madagascar Country Environmental Analysis: Promoting Green, Resilient, and Inclusive Development. https://documents1.worldbank.org/curated/ en/099635010242211316/pdf/P17701803653a407f0bbd80a9da77fa6f51.pdf. Zhu, Heng, Anubhab Gupta, Mateusz Filipski, Jaakko Valli, Ernesto Gonzalez- Estrada, and J. Edward 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. Measuring the Local Economic Impacts of Nature-Based Tourism in Madagascar APPENDIX A Questionnaires Park Entry PARK RECORDS 2022 2023 UNIT Tourists number       What was/is the number of multiday tourists (adults and children > 5) in     number the park for all of 2022 (2023)? What was/is the number of single-day tourists (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 (for example, maintenance     amount workers) Salaries and wages for female unskilled workers (for example,     amount maintenance workers) Salaries and wages for male skilled workers (for example, some guides,     amount wardens, ticket sales, admin) Salaries and wages for female skilled workers (for example, some guides,     amount wardens, ticket sales, admin) Payment of rents on land, buildings, and so on     amount Locally produced agricultural products (for example, fruits, vegetables,     amount 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 84 Community Projects COMMUNITY PROJECTS SURVEY 2022 2023 UNIT What was/is the total annual community projects budget for 2022 (2023)?     amount Community projects 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     % Salaries and wages for female skilled workers     % Payment of rents on land, buildings, and so on     % Locally produced agricultural products (for example, fruits, vegetables, and meats)     % Services (for example, laundry, maintenance, construction, and 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 projects budget for 2022     100% Government Budget (Commune) GOVERNMENT BUDGET 2022 2023 UNIT What was the total budget of the commune for 2022 (2023)?     amount What was the total amount received from local taxes for 2022 (2023)?     amount Gov budget for 2022 (2023)       How much of the 2022 (2023) budget 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     % Salaries and wages for female skilled workers     % Payment of rents on land, buildings, and so on     % Locally produced agricultural products (for example, fruits, vegetables, and meats)     % Services (for example, laundry, maintenance, construction, and repairs) from local providers     % Purchases from local stores and other businesses     % Purchases made outside the local economy     % Other percent of the expected 2022 (2023) budget     % Specify other percent of the expected 2022 (2023) budget     text Total     100% Other information       Share of salaries and wages paid to local workers (budget 2022 (2023)     % 85 Measuring the Local Economic Impacts of Nature-Based Tourism in Madagascar 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 or foreign resident of Madagascar?   text Have you visited or do you plan on visiting (park_name)?   text If Nosy Tanikely Park was not here, would you still have come to Nosy Be?   text If you had not heard of Nosy Tanikely Park, would you have known about   text Nosy Be? Expected spending     Are you paying for this trip to {park_name} as part of a package tour   text (hotel, tour, food, and so on)? 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   amount {park_name} (lodging, food, park fees, and other local activities)? Number of days and 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 Ranomafana   amount Park entry in Nosy Tanikely   amount Park entry in Lokobe   amount Hotel shops or other hotel amenities   amount Local restaurants (food and drink)   amount 86 Measuring the Local Economic Impacts of Nature-Based Tourism in Madagascar TOURISTS SURVEY   UNIT Guides and tours   amount Tourist equipment rental and purchases   amount Souvenirs and handicrafts   amount Retail shops, local markets, and so on   amount Local transportation (taxi, bajaj, and so on)   amount Other   amount Please specify the other spending   text Tour Operators (based in Antananarivo) 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   amount park (lodging, food, park fees, and 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   % Souvenirs and handicrafts   % Retail shops, local markets   % Hotel shops or 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 or profits? Tour operator total percentage   100% 87 Measuring the Local Economic Impacts of Nature-Based Tourism in Madagascar 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 Nosy Be) 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 or hotel is locally owned?   % Share of wages paid to local workers   % After all your monthly expenses, about what percentage of your revenue   % from {park name} goes to savings or profits? 88 Measuring the Local Economic Impacts of Nature-Based Tourism in Madagascar 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, and others) Salaries and wages for female unskilled workers (for example, bussers,   % dishwashers, and others) Salaries and wages for male skilled workers (for example, cooks, servers,   % and admin) Salaries and wages for female skilled workers (for example, cooks, servers,   % and 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 Nosy Be) 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 from   % {park name} goes to savings or profits? 89 Measuring the Local Economic Impacts of Nature-Based Tourism in Madagascar 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, and supervisors) Salaries and wages for female skilled workers (for example, machine   % operators, clerks, and 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 Nosy Be) 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 nonagricultural, 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 expenses, about what percentage of your revenue from   % {park name} goes to savings or profits? 90 Measuring the Local Economic Impacts of Nature-Based Tourism in Madagascar 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, and supervisors) Salaries and wages for female skilled workers (for example, machine   % operators, clerks, and 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 Nosy Be) 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 local tax   % Nonfarm tax other   % Other nonagricultural producers’ nontourism-related spending (specify)   % Please specify the other nonagricultural nontourism-related producers   text spending 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   % from {park name} goes to savings or profits? 91 Measuring the Local Economic Impacts of Nature-Based Tourism in Madagascar 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 and supervisors) Salaries and wages for female skilled workers (for example, machine   % operators and supervisors) Land rate   % Value from own harvest, livestock, or neighbors (for example,   % seedlings and 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,   % and 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)   text spending Commercial farmers (crop producers) total percentage   100% Other information     After all your monthly expenses, about what percentage of your   % revenue from {park name} goes to savings or profits? Has your crop harvest been negatively impacted by the wildlife from   text the park? 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   text from the park? What is your estimate of the share of your monthly revenue that has   % been lost? 92 Measuring the Local Economic Impacts of Nature-Based Tourism in Madagascar 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   % and supervisors) Salaries and wages for female skilled workers (for example, machine   % operators and 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, and repairs) from local   % providers Purchases from local stores and other businesses (for example, nets, lines,   % hooks, and bait) About what percentage of your monthly costs are purchases of inputs from   % 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 from   % {park name} goes to savings or 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? 93 Measuring the Local Economic Impacts of Nature-Based Tourism in Madagascar Households 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 (for example, pensions,   % social cash transfers, or 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 Ariary?   amount What is your household’s size?   number Source: World Bank 2023 94 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$1,000 + 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 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. 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 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. 95 Measuring the Local Economic Impacts of Nature-Based Tourism in Madagascar 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 isolated areas like Madagascar’s Ranomafana National Park, businesses and households, by necessity, rely more on trade with outside markets. For example, more of the food tourists consume and the workers lodges hire are brought in from other parts of the country. In Madagascar, local production is greater at Nosy Be and money thus circulates more at Nosy Be than Ranomafana, creating larger multipliers at Nosy Be. 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. 96 Measuring the Local Economic Impacts of Nature-Based Tourism in Madagascar 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. 97 Measuring the Local Economic Impacts of Nature-Based Tourism in Madagascar 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 Ranomafana and Nosy Tanikely. As an example, tables C.1 and C.2 shows the SAM and resulting SAM multiplier matrix from Ranomafana National Park. 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 projects, and spending outside the local economy. The rows show income from the same local economy actors. 98 TABLE C.1 Social Accounting Matrix of the Economy In and Surrounding Ranomafana National Park AGRICULTURE TOURISM NONAGRI. FISH LMUSK LMSK LFUSK LFSK K POOR NONPOOR RESTAURANTS LODGES TOURISTS PROTEC. AR COMREVSH LOCALG G ROW TOTALEXP Agriculture 7751 4107 0 0 0 0 0 0 0 117210 25046 1833 39897 0 0 0 0 0 0 195844 Tourism 0 0 0 0 0 0 0 0 0 33488 12523 0 2613 957681 0 0 0 0 0 1006305 Nonagriculture 51001 175796 289919 0 0 0 0 0 0 987909 876621 69504 308780 992518 72365 528 49459 0 0 3874400 Fish 0 0 0 0 0 0 0 0 0 117210 0 0 0 0 0 0 0 0 0 117210 LMUSK 37980 45181 290662 0 0 0 0 0 0 0 0 7067 113860 0 7929 0 13644 0 0 516323 LMSK 0 42443 0 0 0 0 0 0 0 0 0 0 35953 0 109182 1162 81863 0 0 270603 LFUSK 8216 4107 17098 0 0 0 0 0 0 0 0 5734 105827 0 0 0 3411 0 0 144393 LFSK 0 6298 0 0 0 0 0 0 0 0 0 933 51660 0 18071 106 15349 0 0 92417 K 91827 339751 986256 0 0 0 0 0 0 200931 50093 55558 296457 0 0 0 0 0 0 2020873 Poor 0 0 0 0 435493 131322 120793 77949 214664 0 0 0 0 0 0 0 0 0 51258 1031479 Nonpoor 0 0 0 0 80830 139281 23600 14468 1806210 0 0 0 0 0 0 0 0 0 0 2064389 Restaurants 0 0 0 0 0 0 0 0 0 0 0 0 0 222234 0 0 0 0 0 222234 Lodges 0 0 0 0 0 0 0 0 0 0 0 0 0 1890275 0 0 0 0 0 1890275 Tourists 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 4175335 4175335 Protected area 0 0 0 0 0 0 0 0 0 0 0 0 0 112627 0 0 0 120934 0 233561 ComRevSh 0 0 0 0 0 0 0 0 0 0 0 0 0 0 10562 0 0 0 0 10562 LocalG 4651 13554 78055 0 0 0 0 0 0 0 0 4000 70287 0 0 0 0 0 0 170547 G 0 23412 124517 0 0 0 0 0 0 16744 3501 7667 100867 0 0 0 0 0 0 276708 ROW 45421 252194 1987432 117210 0 0 0 0 0 200931 288033 69937 764073 0 26014 8766 6822 155775 0 3922608 Total 246846 906845 3773939 117210 516323 270603 144393 92417 2020873 1674422 1255817 222234 1890275 4175335 244123 10562 170547 276709 4226593 0 expenditures Source: World Bank 2023. 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 Ranomafana National Park AGRICULTURE TOURISM NONAGRI. FISH LMUSK LMSK LFUSK LFSK K POOR NONPOOR RESTAURANTS LODGES TOURISTS PROT. AR. COMREVSH LOCALG Agriculture 1,07 0,03 0,02 0 0,08 0,07 0,08 0,08 0,04 0,09 0,04 0,03 0,05 0,04 0,04 0,01 0,05 Tourism 0,01 1,01 0,01 0 0,03 0,02 0,03 0,03 0,02 0,03 0,02 0,01 0,01 0,24 0,02 0 0,02 Nonagriculture 1,01 0,87 1,53 0 1,13 1,14 1,13 1,13 1,14 1,13 1,15 0,87 0,68 0,95 1,09 0,21 1,21 Fish 0,02 0,02 0,01 1 0,07 0,04 0,07 0,07 0,02 0,08 0,01 0,01 0,02 0,02 0,03 0,01 0,04 LMUSK 0,25 0,13 0,12 0 1,1 0,1 0,1 0,1 0,1 0,1 0,1 0,11 0,12 0,12 0,13 0,02 0,18 LMSK 0,02 0,06 0,02 0 0,01 1,01 0,01 0,01 0,01 0,01 0,01 0,02 0,04 0,05 0,46 0,11 0,49 LFUSK 0,04 0,01 0,01 0 0,01 0,01 1,01 0,01 0,01 0,01 0,01 0,03 0,06 0,03 0,01 0 0,03 LFSK 0 0,01 0 0 0 0 0 1 0 0 0 0,01 0,03 0,02 0,08 0,01 0,09 K 0,74 0,67 0,45 0 0,48 0,44 0,48 0,48 1,4 0,5 0,39 0,53 0,4 0,48 0,39 0,08 0,43 Poor 0,33 0,23 0,17 0 1 0,63 0,99 1 0,25 1,16 0,14 0,19 0,25 0,23 0,44 0,09 0,54 Nonpoor 0,71 0,66 0,43 0 0,61 0,93 0,62 0,61 1,28 0,47 1,37 0,51 0,42 0,49 0,62 0,13 0,69 Restaurants 0 0 0 0 0 0 0 0 0 0 0 1 0 0,05 0 0 0 Lodges 0 0 0 0 0 0 0 0 0 0 0 0 1 0,45 0 0 0,00 Tourists 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 Protected areas 0 0 0 0 0 0 0 0 0 0 0 0 0 0,03 1 0 0 ComRevSh 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0,04 1 0 LOCALG 0,04 0,03 0,03 0,00 0,03 0,03 0,03 0,03 0,02 0,03 0,02 0,04 0,05 0,04 0,02 0,00 1,03 Source: World Bank 2023. 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 Madagascar 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 $0.72 dollars 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” (fourth from the right) and adding the multipliers for “Poor” households (0.23) and “Nonpoor” households (0.49). 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. 101