Assessing the Economic Impact of Tourism in Protected Areas on Local Economies in Zambia Heng Zhu, Anubhab Gupta, Edward Whitney, Urvashi Narain, Iretomiwa Olatunji, Hasita Bhammar, Ngao Mubanga, Phoebe Spencer and J. Edward Taylor © 2021 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. designer Sergio Andres Moreno Tellez cover photo Nico Adriaan Kelder / Shutterstock.com SUPPORTED BY: acknowledgements We are grateful for the valuable assistance to this project provided by Donald Banda, Chief Administrative Officer, Chipata Town; Moses Saul Kaoma, Area Warden, Lower Zambezi National Park; Iain Shuker, Nathalie Johnson, and Hellen Mungaila, World Bank; Chiwala Matesamwa, CRB executive officer, Chiawa GMA; Petros Muyunda and Choizya Mbewe, CRB officers, South Luangwa; Ian Stevenson, CEO of Conservation Lower Zambezi; Keira Langford-Johnson, Commercial Director of PROFLIGHT Zambia; Adrian Coley, Flatdogs Camp; Paul Barnes, Pioneer Camp; and Grant Cumings, Chiawa Camp. Thanks are also due to stakeholders who provided invaluable feedback during a two-week virtual consultation process. This project would not have been possible without the dedicated and enthusiastic work of our Zambian survey team: Alick Bruce Makondo, Kenneth Mulenga, Sarai Sinyolo, Nozyenji Mwale, Janet Mulla, Chilufya Chisanga, Memory Bwalya, Liseli Moira Banda, Mwila Lunda, Margret Mbewe, Chipo Shimoomba, Christopher Chibwe, Keren Chakaba, and Vincent Katowa. Contents Acknowledgements ........................................................................................................................ 4 Executive Summary ...................................................................................................................... 9 How was the study done?..............................................................................................................10 What did the study find? ............................................................................................................... 11 What lessons can policy makers draw from the study? ........................................................ 12 1. Introduction ................................................................................................................................. 14 2. Background .................................................................................................................................20 2.1 Policy and Institutional Context ............................................................................................. 21 2.2 Study Sites ................................................................................................................................ 22 2.3 Government Revenues and Expenditures ....................................................................... 24 3. Methodology................................................................................................................................26 3.1 Avenues for Economic Impacts of Protected Areas ........................................................ 27 3.2 LEWIE Model.............................................................................................................................29 3.3 Data Collection......................................................................................................................... 30 4. Data Summary............................................................................................................................32 4.1 Tourists and Tourism Businesses.......................................................................................... 33 4.2 Households............................................................................................................................... 35 4.3 Local Business and Entrepreneurial Activities.................................................................. 37 5. LEWIE Model Findings.............................................................................................................38 5.1 Impact of Additional Tourist on the Local Economy......................................................... 40 5.2 Total Impacts of Nature Tourism on the Local Economy .............................................. 43 5.3 Impacts of Complementary Investments and Outside Shocks.................................... 45 Conclusions and Policy Recommendations........................................................................ 48 Household and Local Business Survey ......................................................................... 57 Tourist Surveys...................................................................................................................... 58 Tourism Businesses Survey............................................................................................... 58 Crops and Livestock............................................................................................................59 References........................................................................................................................................ 64 List of figures, maps, boxes and tables Figures Figure ES- 1. Economic Impact Pathways of Protected Area Tourism. . . . . . . . . . . . . . . . . . . . . . . . . 10 Figure ES-2. Income Multipliers for an Additional Kwacha of Tourist Spending. . . . . . . . . . . . . . . . . 11 Figure ES-3. Distribution of Multiplier across Poor and Non-Poor Populations. . . . . . . . . . . . . . . . . . 11 Figure 1. Number of Visitor Entries to Lower Zambezi (LZ) and South Luangwa (SL) National Parks, 2015-2018 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24 Figure 2. Non-consumptive fees. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24 Figure 3. Non-consumptive fees. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25 Figure 4. Economic Pathways of Tourism in Protected Areas. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27 Figure 5. Tourists by Origin. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33 Figure 6. Real-Income Multipliers for an Additional Kwacha of Tourist Spending . . . . . . . . . . . . . 41 Figure 7. Which Households Benefit from Tourist Spending Multipliers? . . . . . . . . . . . . . . . . . . . . 42 Figure 8. Distribution of Multiplier across Poor and Non-Poor Populations . . . . . . . . . . . . . . . . . . 42 Figure 9. Which Households Benefit from Government Spending Multipliers?. . . . . . . . . . . . . . . 44 Figure 10. Zambia Travel and Tourism Competitiveness Index Profile . . . . . . . . . . . . . . . . . . . . . . . 52 Figure A3.1. Crop Types. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59 Figure A3.2. Livestock at Lower Zambezi (LZ) and South Luangwa (SL) . . . . . . . . . . . . . . . . . . . . . . 61 Figure A3.3. New Business Formation. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62 Maps Map 1. Zambia’s Protected Area Network . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15 Map 2. Zambia has one of the highest proportions of protected land in Africa . . . . . . . . . . . . . . 17 Map 3. Lower Zambezi and South Luangwa National Parks on the national map of Zambia. . 23 Map 4. Lower Zambezi National Parks and related GMAs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23 Map 5. South Luangwa National Park and related GMAs. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23 Tables Table 1. GoZ revenues and expenditures at the two study sites (LZ and SL National Parks) in 2018*. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25 Table 2. Sample Sizes and Local Populations. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30 Table 3. Trip Characteristics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33 Table 4. Local Tourist Expenditures in Kwacha. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34 Table 5. Poverty Headcount . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35 Table 6. Household Demographics and Activities. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35 Table 7. Wage Income and Employment. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36 Table 8. Household Business Types by Site. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37 Table 9. Avenues of Impact Captured by LEWIE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39 Table 10. Local Income Impacts of an Additional Tourist . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40 Table 11. Production Impacts of an Additional Tourist . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41 Table 12. Estimated Impact of Tourism. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43 Table 13. Government Hiring an Additional Local Laborer. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44 Table 14. Estimated Losses from Human-Wildlife Conflict. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45 Table 15. Impacts of 5 Percent Increase in Local Purchases by Businesses. . . . . . . . . . . . . . . . . . 46 Table 16. Monthly Income Loss from No Tourism. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46 Table 17. Monthly Production Loss from No Tourism . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47 Table 18. Opportunities to Strengthen the Income Multiplier for Local Households. . . . . . . . . . . 55 Table A3.1. Employment by Sector . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59 Table A3.2. Crop Production and Inputs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60 Table A3.3. Crop Use and Sales. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61 Table A3.4. Livestock and Inputs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62 Table A3.5. Business Operations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63 Table A3.6. Business Costs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63 Table A3.7. Share of Inputs Purchased Outside the Local Economy . . . . . . . . . . . . . . . . . . . . . . . . 63 Boxes Box 1. Definition of protected area categories. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16 Box 2. What is a Local Economy?. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22 Box 3. Building Capacity While Doing Research. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30 Box 4. Assessment of Factors Needed for a Strong Concessions Program. . . . . . . . . . . . . . . . 53 8 ASSESSING THE ECONOMI C IMPAC T OF TOURISM IN PROTE CTED AREAS ON LO CA L E CONOMIES IN Z AMBIA List of figures, maps, boxesand tables Executive Summary Zambia is a country richly endowed with natural inflicted by COVID-19. Awareness is growing resources, and home to a substantial protected that these two challenges – precipitous declines area network. Approximately 40 percent of in global biodiversity, and the imperative for the country’s land area enjoys some form of a green recovery from the pandemic – must protection, with globally significant biodiversity. be addressed as one: neither problem can be This combination of protected areas and rich solved without solving the other. biodiversity is equally a major tourism asset, in Additionally, these challenges must be met in an industry which world over attracts eight bil- poor and often isolated rural areas in which lion visitors a year to protected areas, provides many of Zambia’s biodiversity-rich protected ar- one-in-ten jobs globally, and contributes up to eas are located. Through the economic benefits 10 percent of global GDP. it generates, protected area tourism is moreover But the potential of Zambia’s protected area one of the few avenues through which govern- network, and its contribution to economic ments can help support livelihoods, simulate development in the country, is yet to be fully economic development, while cultivating local realized. This situation mirrors that of many community support for conservation in these ru- countries in which governments see protected ral communities. In this context, the importance areas as key to addressing biodiversity conser- of protected area tourism cannot be overstated, vation but often overlook these natural assets in because of its potential to address losses to economic development plans. This oversight is economies, promoting recovery, addressing of great concern, as countries, globally, struggle longstanding development challenges, all the to contain unprecedented biodiversity losses while supporting biodiversity conservation. while trying to address development setbacks This study therefore set out to strengthen the economic case for Government of Zambia to promote sustainable and inclusive tourism in its protected areas by estimating the direct and indirect benefits to local economies from protected area tourism. 10 ASSESSING THE ECONOMI C IMPAC T OF TOURISM IN PROTE CTED AREAS ON LO CA L E CONOMIES IN Z AMBIA How was the study done? Working in South Luangwa and Lower Zambezi actors (businesses and households) within a National Parks, and using survey data collected local economy. Direct impacts refer to monies from communities living in the vicinity of the spent directly by tourists in protected areas, Parks, from tourism and other local businesses, while indirect impacts describe the knock-on and from tourists to the two Parks, the study effects of this spending, via production linkag- traced and quantified the economic pathways es which grow to support expanding tourism through which protected area tourism stimulates markets, and consumption linkages, through local economies. A general equilibrium mod- which wages and profits trigger fresh rounds of el for local economy-wide impact evaluation spending which ripple through the local econo- (LEWIE) was used to describe direct and indirect mies (Figure ES-1). impacts of tourism by integrating models of Figure ES-1. Economic Impact Pathways of Protected Area Tourism Revenue sharing, community projects Environmental Impact b PARK AUTHORITY, Businesses pay GOVERNMENT taxes and fees b a Pay non-consumptive a Pay a Park and consumptive fees Entrance Fee and taxes b Park hires guards or employs households for PA activities a Protected Areas a Purchase goods Terrestrial / Marine Spend money on and services lodging, tourist TOURISM, HOUSEHOLDS LODGES AND OU activities G N BUSINESSES RIST S VISITI T Wages paid to Purchase food, a workers employed b,c goods and in tourism activities Local incomes services c Ex ecu tiv e Su mma ry increase; b Source goods households spend and services their income to source goods LOCAL FARMS AND BUSINESSES a Trade with outside/ non-local markets Direct Impact Legend of Pathways of Influence a. Direct impacts Indirect impact through b. Production linkages production linkages c. Income and consumption linkages ASSESSING T H E E C ONOMI C IMPACT OF TOURISM IN PROTE CTED AREAS ON LO CA L E CONOMIES IN Z AMBIA 11 What did the study find? Public investment in protected areas pays off, generating a net of 16.17 million kwacha, and generating per-kwacha economic returns on establishing these protected areas as sources government spending of approximately 28.2 of revenue, rather than financial burdens. kwacha in South Luangwa National Park and Tourism in Zambia’s two protected areas gen- about 16.7 kwacha in Lower Zambezi National erates significant income multipliers, defined Park. Tourist spending strongly infiltrates local as the change in local household incomes per economies, while government revenues from kwacha of fresh infusion of cash into the econo- park fees exceed investments in the two parks, my through tourist spending. These multipliers apply to households directly tied to the tourism Figure ES-2 Income Multipliers for an Additional Kwacha of Tourist sector, but also those which are not, and in both Spending poor and non-poor households alike, as indicat- ed by the estimates shown in Figure ES-2, which 2,5 show the distribution of income gains for poor and non-poor households. Each kwacha spent by visitors at Lower Zambezi National Park rais- 2,0 es household incomes around the park by 1.82 kwacha and around South Luangwa National 1,5 Park by 1.53 kwacha, reflecting the penetra- tion of tourist spending into local economies. Additionally, when normalized by population, 1,0 1,82 as shown in Figure ES-3, multiplier shares per 1,53 resident are revealed as comparable between 0,5 poor and non-poor populations both within and outside of protected areas, showing that the 0,0 impacts of protected area tourism are equitably Lower Zambezi South Luangwa beneficial to communities. Tourism in protected areas also creates signifi- Figure ES-3. Distribution of Multiplier across Poor and Non-Poor cant job opportunities. The study estimates that Populations protected area tourism supports approximately 7,463 full-time equivalent jobs around Lower Zambezi National Park and 28,210 jobs around South Luangwa National Park, equivalent to 14 and 30 percent of the populations around the two parks, respectively. Jobs are created di- Lower rectly through tourism activities. Additional jobs 47% Zambezi 53% are supported when businesses such as tourism operators and tourism employees purchase supplies and services from other local busi- nesses, thus creating indirect effects of visitor E xe cu t iv e Su mma ry spending surrounding the park. While the economic benefits of protected areas are strong, the costs to local communities must be managed. Human-wildlife interactions around protected areas occur mostly in the form of crop losses, and have negative impacts on South household incomes, with wild animal move- 48% Luangwa 52% ments on farms reducing crop yields by 11–14 percent according to household surveys. These direct impacts, and indirect impacts through production and income linkages amount to income losses in the local economy of approx- imately 23.7 million kwacha (US$1.8 million) at Nonpoor households Poor households Lower Zambezi and 16.3 million kwacha (US$1.2 12 ASSESSING THE ECONOMI C IMPAC T OF TOURISM IN PROTE CTED AREAS ON LO CA L E CONOMIES IN Z AMBIA million) at South Luangwa. These figures are reduces local real GDP by 17.6 million kwacha important in that they underpin arguments in (US$ 1.26 million) per month at Lower Zambezi favor of compensation, which both mitigates and 53 million kwacha (US$ 3.78 million) per these losses, and retains the needed support of month at South Luangwa, and these losses also local communities. negatively affected retail, livestock, agricultural production, and various service sector outputs. The study also points to the need to address These impacts indicate the extent to which losses suffered by the sector following the support for protected areas will be needed to COVID-19 pandemic. The study shows that offset these losses and to realize the potential the pandemic has led to substantial losses of these areas to support a green economic in tourism and tourism incomes. A complete recovery. loss of tourist revenue around the two parks What lessons can policy makers draw from the study? With over 40 percent of its land area under 3. Share benefits with local communities. some form of protection, including 20 national Zambia’s protected area regulations require parks, there is great potential for protected sharing of revenues with local communities, areas in Zambia to contribute to development but experience indicates that mechanisms to goals and to maintain the country’s rich biodiver- this effect need to be more timely, equi- sity. In order to realize this potential, the report table, and transparent. Additionally, while recommends enhanced protection of Zambia’s tourist-spend income multipliers for local natural assets, growing and diversifying the households are significant, opportunities ex- tourism sector, and sharing benefits with local ist for governments to raise these multipliers communities. These approaches form the three through their policies, and these opportuni- pillars of a strategy to jointly address biodiversi- ties need to be explored. ty loss, development challenges, and a green, In conclusion, and in the wake of the COVID-19 post-COVID recovery. pandemic, Zambia needs to address losses 1. Protect the natural asset base. To support to its protected area tourism sector in order to conservation and secure the natural assets regain benefits to park-adjacent communities that draw visitors to Zambia, the protected and to secure the conservation status of its area network needs to be better managed. significant natural assets. To do this, Zambia To achieve this, specific recommendations should champion sustainable and inclusive from the study are to (i) increase public tourism in protected areas. It should increase investment in protected area management; public and private investment in protected areas (ii) build capacity of protected area managers; on the growing evidential basis for attractive and to (iii) assess and monitor the impacts of and far-reaching returns which support both visitor spending. conservation and sustainable development 2. Diversify and grow the tourism sector. The strategies. Finally, in response to a pandem- Zambian tourism sector needs to expand ic which has caused development setbacks, and diversify beyond the five parks current- Zambia’s protected area tourism sector should Ex ecu tiv e Su mma ry ly visited by tourists. This requires that the enact mechanisms to distribute its benefits fairly country’s protected areas be assessed, and in the face of poverty and losses incurred by ranked by their tourism potential in order to local communities. select priority sites for development. A strong commercial services/concessions program will be needed to develop the new sites, draw tourists and generate revenue. 14 ASSESSING THE ECONOMI C IMPAC T OF TOURISM IN PROTE CTED AREAS ON LO CA L E CONOMIES IN Z AMBIA 1 Introduction In tro d uc tio n ASSESSING T H E E C ONOMI C IMPACT OF TOURISM IN PROTE CTED AREAS ON LO CA L E CONOMIES IN Z AMBIA 15 Zambia is a low-income country with an Zambia’s biodiversity is managed within a net- economy that is driven largely by subsistence work of protected areas that include 20 national agriculture and mining. Zambia is also endowed parks, 39 game management areas (GMAs), with rich natural resources including forests, 432 forest reserves, 59 botanical reserves, 42 wildlife, and rivers. Forests cover 61 percent of important bird areas, and 2 bird sanctuaries, as Zambia’s land area (World Bank 2019), and the illustrated in Map 1 below. Different categories country is home to globally significant biodiver- of protected areas (see Box 1 for definitions of sity and about 40 percent of southern Africa’s categories of protected areas) have varying lev- freshwater resources. els of protection for biodiversity and wildlife: no human settlements are allowed inside national parks, for example, while GMAs include resident communities and permit multiple uses of wildlife. Map 1. Zambia’s Protected Area Network Source: Adapted from Zambia’s Second National Biodiversity Strategy and Action Plan (NBSAP) 2015-2025. In t ro du c tio n 16 ASSESSING THE ECONOMI C IMPAC T OF TOURISM IN PROTE CTED AREAS ON LO CA L E CONOMIES IN Z AMBIA box 1 Definition Protected Areas are clearly defined areas, recognized, dedicated, and managed, through legal or other of protected area effective means, to achieve the long-term conservation of nature with associated ecosystem services and categories cultural values (Dudley, N. 2008). They range from Category I to Category VI on a decreasing scale of level of regulation. National Parks are classified under Category II of protected areas. They are defined as large natural or near-natural areas set aside to protect large-scale ecological processes, along with complementing species and ecosystems characteristic of the area, which also provide a foundation for environmentally and culturally compatible spiritual, scientific, educational, recreational and visitor opportunities. In Zambia, national parks cover ~20 percent of the country’s land area. Game Management Areas (GMA) are a category of protected areas in Zambia that are mostly customarily owned lands designated as buffer zones between national parks and open areas under Section 28 of the Zambia Wildlife Act, 2015. They currently cover ~22 percent of the country’s total land area. Human settle- ment is allowed in designated areas, as are agriculture, forestry and mining, as defined by the GMA’s General Management Plan (a document that sets forth the basic management and development philoso- phy for a protected area and provides land use strategies to address problems and achieve management objectives over a set time period). Three types of hunting are permitted within GMAs: (i) safari hunting; (ii) resident hunting; and (iii) bona-fide hunting*. Additionally, photographic tourism is permitted in a few GMAs. Community Resource Boards (CRB) are identified by the Zambia Wildlife Act of 2015 as institutions legally mandated to co-manage and benefit from wildlife in GMAs. The Chief of an area is regarded as the patron of the CRB. * According to the Zambia Wildlife Act, “bona fide client” means a licensed non-Zambian hunter who is a client of a hunting outfitter that has a hunting concession or owns an unfenced private wildlife estate; Zambia created its first protected area in 1920. Despite this economic value, Lindsey et al. 2014 Since then, Zambia has increasingly dedicated found that Zambia’s protected areas are ecologi- land to conservation. Currently, approximate- cally, economically, and socially underperforming, ly 40 percent of the land area is under some and face challenges including poaching, ecolog- form of protection, giving Zambia one of the ical threats from development, and ineffective highest proportions of land under protection in management1 (Lindsey et al. 2014). Africa (see Map 2). Conservation efforts by the Poor regulation, open access, population Government of Zambia (GoZ) aim to maintain growth, poverty, fuelwood harvesting and agri- species populations and promote the provision culture encroach on protected areas, advancing of ecosystem services such as water, food, and deforestation around major roads at a rate of carbon storage. up to 2 kms² per year toward national parks, Protected areas in Zambia also attract tourists and habitat fragmentation which threatens the who visit parks for wildlife tourism and con- integrity of protected areas. tribute to the country’s economy. Moreover, Communities have had little incentive to prevent tourists visit Zambia predominantly for nature encroachment, poaching, and other threats to and wilderness (MoTA 2016). Contributions to nearby protected areas as they have received the economy are direct in the form of visitor little benefit from conservation in the past. spending on park fees, hotels, lodges, transport, Rather than benefitting from conservation and leisure, and recreation services. This results tourism, Zambian protected area neighbors in local job creation, with additional jobs and have suffered from human‐wildlife conflict. One In tro d uc tio n economic activity supported when tourism of the few ways in which communities have ben- operators and employees purchase supplies efited is game hunting, where the law allows for and services from other local businesses, thus 45 percent of hunting revenues to flow directly creating indirect effects of visitor spending sur- to communities via CRBs in GMAs. However, rounding the park. As per estimates provided hunting revenues intended to be shared with by the World Travel and Tourism Council, travel communities have fallen short of their promise and tourism contributed 7 percent of the GDP in in the past, and used instead to bridge budget 2019, supporting 7.2 percent of employment in shortfalls in the Department of National Parks the country (WTTC 2019). 1 Unless otherwise stated, this section on factors constraining effective protected area management draws on two World Bank project appraisal documents - the Transforming Landscapes for Resilience and Development Project (P164764), Zambia Integrated Forest Landscapes Project (P161490), and Lindsey et al. (2014). ASSESSING T H E E C ONOMI C IMPACT OF TOURISM IN PROTE CTED AREAS ON LO CA L E CONOMIES IN Z AMBIA 17 Map 2. Zambia has one of the highest proportions of protected land in Africa In t ro du c tio n Source: World Database on Protected Areas (UNEP-WCMC and IUCN 2020) and Wildlife (DNPW), previously known as the Too few enforcement staff, inadequate housing, Zambia Wildlife Authority (ZAWA). Benefits to and lack of basic infrastructure, all stemming local communities from tourism are also limited from lack of funding, further constrain park by the size and concentration of the sector; management and reduce visitor numbers. 95 percent of tourism is clustered around the For example, Lukusuzi National Park is one of five most popular national parks - Kafue, South Zambia’s largest parks, and has high potential Luangwa, Mosi-oa-Tunya, Lower Zambezi and for biodiversity conservation and ecotourism, Lochinvar (MoTA 2018). but lacks management and receives little government funding. In fact, protected areas 18 ASSESSING THE ECONOMI C IMPAC T OF TOURISM IN PROTE CTED AREAS ON LO CA L E CONOMIES IN Z AMBIA all over Africa are under-funded, with estimated impacts may strengthen the economic case for deficits of over US$1 billion annually. Zambia’s public investment in protected areas, much like funding deficit is estimated at 91 percent public investments in roads and other forms of (Lindsey et al. 2018)we compiled a dataset of public infrastructure and assets. The study also funding in Africa’s PAs with lions and estimated estimates the benefits to local communities, for a minimum target for conserving the species poor and non-poor households, to understand and managing PAs effectively. PAs with lions how protected area tourism incentivizes sup- require $1.2 to $2.4 billion or $1,000 to $2,000/ port for conservation, and how it may improve km2 annually, yet receive just $381 million or household incomes. $200/km2 (median with inadequate funding This study builds on two previous studies on the from the central treasury for wildlife conserva- economic impacts of protected areas in Zambia tion (MoTA 2018). – World Bank 2007 and Chidakel, Child, and Globally, governments do not prioritize invest- Muyengwa 2021. World Bank (2007) estimated ments in protected areas, in part because these the economic impact of tourist spending on investments are seen to support conservation National Government revenues, and the overall but not to further development. Scarce pub- effects of GMAs on local consumption, but did lic resources are instead allocated to other, not estimate the impacts of tourism in protected competing development needs. But protected areas on local economies and poor households. areas can provide development opportunities, Chidakel et al., (2021) included the impact of as noted above, and may generate returns on tourist spending on tourist and other businesses public investments that far exceed the amounts near South Luangwa National Park, together governments spend. In the United States, in with reported local consumption spending by 2019, an annual investment of US$3 billion of businesses and employees. While this approach public resources in the National Parks System progressed towards capturing direct and indi- resulted in a contribution to GDP of US$41.7 rect impacts, it used a 2007 national income billion through visitor spending (US NPS 2019). multiplier for Zambia to estimate national Similarly, in 2018, Parks Canada generated a impacts and did not consider price effects (while contribution to GDP of US$3.1 billion and tax this study focuses specifically on local economic revenues of almost US$0.4 billion for a public and community impacts). investment of approximately US$1 billion (Parks This study assesses the full range of impacts on Canada 2019). Moreover, investments in pro- the local economy, and includes expenditures tected areas can generate significant benefits made in the local economy by households for local economies through job creation and and local businesses who benefit from tourism income generation, lifting households out of through employment or through tourism-relat- poverty and providing them with incentives ed local businesses. The study also estimates to support conservation goals. US Parks are the additional income created around parks estimated to support 329,000 jobs in gateway per dollar spent by visitors, and therefore communities, and Parks Canada 40,469 jobs. the income multiplier from tourism in protect- Governments often lack evidence for the ed areas, and the rate of return per kwacha economic impact of protected area tourism on invested by the government in protected areas local and national economies, and fail to see the like national parks. Additionally, the study development gains from public expenditure on provides estimates on the economic impacts conservation. The objective of this study is to of human-wildlife conflict and the COVID-19 make the economic case for public investment pandemic, and quantifies possible effects of In tro d uc tio n in protected areas in Zambia by estimating government policies to increase local benefits the direct and indirect benefits of tourism on from protected area tourism. local economies. This estimate of economic ASSESSING T H E E C ONOMI C IMPACT OF TOURISM IN PROTE CTED AREAS ON LO CA L E CONOMIES IN Z AMBIA 19 In t ro du c tio n 20 ASSESSING THE ECONOMI C IMPAC T OF TOURISM IN PROTE CTED AREAS ON LO CA L E CONOMIES IN Z AMBIA 2 Background Backg ro u n d ASSESSING T H E E C ONOMI C IMPACT OF TOURISM IN PROTE CTED AREAS ON LO CA L E CONOMIES IN Z AMBIA 21 2.1 Policy and Institutional Context The GoZ has established the requisite legal CRBs are responsible for community instruments, policies, and institutional frame- projects such as boreholes, maintenance works to conserve biodiversity in its network of community infrastructure including local of protected areas. The framework behind the markets, and hiring of community scouts. There are sometimes several Chiefdoms in current institutions and management systems for a GMA, each with its own CRB. protected areas and wildlife was first enacted • Lodge owners, hunting outfitters in 1998 through the Policy for National Parks (individuals responsible for managing and Wildlife. This policy led to the Wildlife Act of licensed hunting in the GMA’s consumptive 1998 which provided for the establishment of a zones), and other tourism service corporate body, ZAWA, as the lead agency for providers. wildlife management and wildlife estates2 (MoTA About 40 percent of tourism in Zambia is based 2018). The Wildlife Act of 1998 was replaced on consumptive use of wildlife (licensed hunting by the Zambia Wildlife Act (ZWA) No.14 of 2015 and other extractive activities), and 60 percent to improve implementation and linkages with is based on non-consumptive use (mainly pho- related economic sectors. At the same time, the tographic tourism). The government receives ZAWA was replaced with the DNPW, operating revenues from tourism in protected areas under the Ministry of Tourism and Art (MoTA). through various mechanisms: through park entry Placing DNPW under MoTA reflects GoZ’s vision fees paid by visitors, non-consumptive fees paid to create an environment supportive of conser- by lodges3 located inside national parks, safari vation and the emergence of protected areas hunting concession fees4 also known as the which function as economic assets and contrib- “right to hunt fee” paid by businesses operating ute to development. in hunting blocks, outfitter license fees paid All wildlife is protected within the boundaries by hunting operators/outfitters5, animal fees of national parks. On the other hand, GMAs are or trophy fees (varies by species) and hunting separated into consumptive and non-consump- fees which include application and basic fees. tive use zones, with land closer to protected Furthermore, based on stakeholder consulta- areas designated for non-consumptive use. tions in July 2019, it was decided that lodges in Parks typically contain privately-run lodges that GMAs6 would in the near future be expected to offer tourists accommodation, food, viewing pay a land user fee to the government. safaris (photo-tourism) within the park, and other Park entry fees are different for domestic and amenities, but no settlements. Villages and international visitors: domestic visitors pay 41.7 lodges exist within the GMAs surrounding parks. kwacha (US$2.98), and international visitors pay There are three primary stakeholders involved US$25 each. Killing or capturing a game animal7 in the management of protected areas, sur- generally requires payment of the four different rounding areas, and tourism-related activities: fees noted above - safari hunting concession fee, outfitter license fee, animal fee, and hunting • DNPW, which monitors the protected area, fee. The GoZ establishes a quota for each spe- deploys park rangers, and represents cies that may be hunted within the consumptive national government. zone of a GMA each year. Animal and hunting • CRBs, which represent the interests of fees are per-animal and vary by species. They local communities living inside GMAs, apply to GMA members as well as visitors from with each CRB representing a Chiefdom. outside the GMA. B ac kgrou n d 2 Includes national parks, wildlife sanctuaries, GMAs and any other area devoted to wildlife and managed by public institutions. 3 Non-consumptive fees include leasing and park entry, bed levies, fees for commercial filming and photography, game drives, walking safaris and water-based activities. 4 DNPW in consultation with the local community is mandated to grant hunting concessions to a business in a specified hunting block. The business enterprise follows the process laid out under section 48 of the Zambia Wildlife Act No.14 and is awarded a contract after a successful bid. The concessionaire has obligations to fulfil during the term of the hunting conces- sion agreement. A hunting block means a Game Management Area or an area within it that is set aside for hunting. 5 A hunting outfitter is a company which offers safari hunting, and which holds a tourism enterprise license and a hunting concession; a hunting license is a license issued under section 40 of the ZWA. 6 For a lodge to be built in a Chiefdom, there must be agreement among three separate parties: first, consent from the local Chief who owns the land must be obtained, followed by approval of the Chiefdom’s CRB board; afterwards, the plan is sub- mitted to DNPW and the lodge operator for final approval. 7 Wild animals defined as such under legislation for the purposes of game management, hunted for food and/or particular products, and/or sports, including trophy hunting. 22 ASSESSING THE ECONOMI C IMPAC T OF TOURISM IN PROTE CTED AREAS ON LO CA L E CONOMIES IN Z AMBIA There is no law governing how the GoZ and maintaining scouts for patrols. Once the Treasury uses revenues from park entry fees land user fee is operational, the law stipulates and lodges, and Treasury decisions about park that 45 percent of the fee will be given to the funding are formally independent of the fees CRB, 5 percent to the chief, and the remainder collected. The ZWA, however, decides how retained by DNPW. fees from killing or capturing game animals In addition, lodges inside the GMA are obliged are to be used. Fifty percent of revenues from to participate in social responsibility programs, outfitter licenses, animal fees, and hunting fees; usually involving reinvestment in the com- and 20 percent from safari hunting conces- munity in the form of schools, hospitals, and sions is shared with the CRB, as per the ZWA. conservation. This is another mechanism for Moreover, five percent from these fees is shared communities to benefit from tourism in protect- with the local Chieftains, and the remainder ed areas.8 allocated to the CRB. In situations in which one GMA has several chiefdoms (thus several CRBs), In theory, these revenue-sharing arrangements the money is split equally among the CRBs. The channel economic benefits from resource con- funds to CRBs are further split three ways: 20 servation to local communities, create incentives percent for CRB administrative costs; 35 percent for communities to protect the wildlife within the for community development projects (bore- GMA, and compensate communities for losses holes, toilets and schools); and the remaining 45 from human-wildlife conflict. percent for resource protection, primarily hiring 2.2 Study Sites The study focuses on two national parks - the conservation system and borders Zimbabwe’s Lower Zambezi and South Luangwa National Mana Pools National Park, with the Zambezi Parks (see Map 3). Both parks have high biodiver- River separating the two countries and parks. sity values, particularly wildlife, and both attract The Lower Zambezi and Mana Pools National large numbers of visitors from around the world.9 Parks form a contiguous 628,800 hectare (2,428 square mile) protected area, and wildlife moves Lower Zambezi National Park, established in freely between the two countries by crossing the 1983, runs along the north bank of the Zambezi Zambezi River. Surrounding the Lower Zambezi River in southeastern Zambia. Prior to 1983, it National Park to the west is Chiawa GMA, with was the private game reserve of the President Rufunsa GMA to the north (see Map 4). Chiawa of Zambia. It is part of a unique binational GMA is also a de facto buffer zone for the Zimbabwe park, which extends farther along the Zambezi River than the Zambia park (see Map 4). Box 2. What is a Local Economy? South Luangwa National Park borders Chisomo and Sandwe GMAs to the south, Lupande and A local economy could be a village, a collection of villages, a town, region, or even country. The wider the demarcation, the more economic activity Lumimba GMAs to the east, and Munyamadzi and economic benefits that will likely be captured, so the definition chosen GMA to the north and west (see Map 5). will depend on the goals of the study. To be effective, conservation poli- cies which underpin protected areas rely on surrounding communities to According to the DNPW, there are six lodges in act as stewards of biodiversity. In Zambia, people living around protected Lower Zambezi and 21 in South Luangwa. Road areas need to see the benefits—including economic benefits—of preserving access is difficult, particularly during the rainy Backg ro u n d wildlife. For the purpose of this study, therefore, the local economy is defined season. Most visitors, therefore, arrive by plane. as the villages within the GMAs surrounding each national park. Moreover, A landing strip is located outside each of the because village households and businesses are linked to nearby market towns just outside the GMA through the purchase of goods and services, the two parks, and in the case of South Luangwa market towns nearest to each park were also included as part of the local National Park, the Mfuwe International Airport is economy for this study. the main point of entry, and there are two strips inside Lower Zambezi Park. 8 In South Luangwa, some study sites i.e., lodges in Chiefdoms, are required to hire 80 percent of their staff from the Chiefdom in which the lodge operates. In practice, this rule is not enforced, and can be satisfied by people from the CRB/ Chiefdom’s catchment area. 9 The study sites were selected because they are important tourism destinations, and on the recommendation of the Zambian Government. Map 3: Lower Zambezi and South Luangwa National Parks on the national map of Zambia Map 4: Lower Zambezi National Parks and related GMAs Map 5: South Luangwa National Park and related GMAs 24 ASSESSING THE ECONOMI C IMPAC T OF TOURISM IN PROTE CTED AREAS ON LO CA L E CONOMIES IN Z AMBIA At both sites, a small regional commercial center constitute what is called the “local economy” for relatively close to the park supplies goods and the purposes of this study. Specifically, Chiawa services to the households and businesses in GMA and the market town of Chirundu con- the GMA and to park lodges: Chirundu, in the stitute the local economy for Lower Zambezi case of Lower Zambezi, and Chipata in the National Park, and Upper and Lower Lupande case of South Luangwa. The villages within GMA and the market town of Chipata for South the GMAs around each park, along with these Luangwa National Park. nearby market towns with which they interact, 2.3 Government Revenues and Expenditures Figure 1. Number of Visitor Entries to Lower Zambezi (LZ) and South Figure 1 shows the number of domestic and Luangwa (SL) National Parks, 2015-2018 international visitor entries to each of the two parks between 2015 and 2018. 60,842 visits International Local were made to these two parks in 2018: 11,161 50,000 at Lower Zambezi (LZ) and 43,469 at South Luangwa (SL). Most—49,858, or 82 percent— 40,000 were by international visitors. Over these four years, the number of visits has not varied signifi- cantly, averaging about 53,907per year. Number of Visitors 30,000 The top panel of Table 1 summarizes GoZ revenues from the two national parks in 2018. 20,000 Expenditures are summarized in the bottom panel.10 Revenues totaled 64.5 million kwacha 10,000 (US$5.4 million), and expenditures totaled 50.9 million kwacha (US$4.2 million). 0 Non-consumptive fees or fixed leases paid by LZ SL LZ SL LZ SL LZ SL lodges inside the park are a significant source 2015 2016 2017 2018 of revenue for DNPW, with a yearly average Source: DNPW, Government of Zambia of approximately US$520,000 (see Figure 2). South Luangwa generates more money through this revenue stream due to its high number of Figure 2. Non-consumptive fees lodges. South Luangwa Lower Zambezi Revenue from hunting constitutes the largest $600.000 share of revenues from tourism in protected areas to the GoZ Treasury. In 2013, Zambia $500.000 banned hunting for two years, which resulted in a loss of revenue for communities and DNPW. However, since the ban was lifted in 2015, $400.000 Total amount in US$ there have been 44 outfitters paying license fees totaling US$107,000 per year. Since 2016, Backg ro u n d $300.000 revenues to communities from hunting have fluctuated, but amounted to 13.92 million kwa- $200.000 cha (US$1.2 million) in 2019 (see Figure 3). $100.000 Notably, expenditure by the GoZ on the two parks is less than the revenue earned from $- them. The bottom panel of Table 1 presents GoZ 2015 2016 2017 2018 2019 expenditures to maintain the two parks. They in- clude a payment of 20.7 million kwacha (US$1.73 Source: DNPW, Government of Zambia 10 Further details on revenues and expenditures are provided in Annex 1. ASSESSING T H E E C ONOMI C IMPACT OF TOURISM IN PROTE CTED AREAS ON LO CA L E CONOMIES IN Z AMBIA 25 Figure 3. Non-consumptive fees million) to CRBs in the GMAs around the two parks from consumptive (hunting) fees, and ex- 25 penditures on game wardens, maintenance and office workers, and other workers. Subtracting these expenditures from revenues, the Treasury 20 appears to have netted approximately 16.17 Total Amount in Millions ZMK million kwacha (US$1.34 million) from Lower 15 Zambezi and South Luangwa National Parks. This revenue from two of Zambia’s major na- tional parks may be used to support biodiversity 10 conservation at other sites, or other national priorities. However, these revenues are only part of the overall economic impact of these 5 parks on their local economies. The next section describes the methodology used to estimate the total economic impacts – direct and indirect 0 – on local economies and communities. 2016 2017 2018 2019 Source: DNPW, Government of Zambia Table 1. GoZ revenues and expenditures at the two study sites (LZ and SL National Parks) in 2018* REVENUES  ZMK USD Park visitor fees 13,727,178 $1,143,330* Non-consumptive Fees 6,858,243 $571,220 Safari Hunting Concession Fees 8,653,694 $720,766 Outfitter License Fees 1,284,675 $107,000 Animal Fees 30,996,458 $2,581,680 Hunting Fees 2,993,532 $249,331 Total Revenues to GoZ 64,513,780 $5,373,327 EXPENDITURES   GoZ Payment to CRBs 20,724,533 $1,726,146 Wage Expenditures 27,613,450 $2,299,924 Non-wage Expenditures 2,595,230 $216,157 Total Expenditures 50,933,213 $4,242,227 GoZ Revenues Minus Expenditures 13,580,567 $1,131,100 B ac kgrou n d Note: This is an estimate gained by multiplying the number of visitors (provided by DNWP) by the fee per international and national visitor. Complete wage expenditure information was only available for South Luangwa in 2018. The last year in which it was available for both parks was 2015. Because of this, 2015 data was used to calculate total GoZ wage expenditures on the two parks. These are conservative estimates of what these expenditures were likely to have been in 2018. * Information in this table was reported for the two parks together and is given in sum here. Revenues reported by GoZ are conservative compared with local estimates gathered through consultation with experts. Informal estimates for SL National Park range between US$2–3 million, concentrated mainly in non-consumptive tourism. 26 ASSESSING THE ECONOMI C IMPAC T OF TOURISM IN PROTE CTED AREAS ON LO CA L E CONOMIES IN Z AMBIA 3 Methodology Me th od o lo gy ASSESSING T H E E C ONOMI C IMPACT OF TOURISM IN PROTE CTED AREAS ON LO CA L E CONOMIES IN Z AMBIA 27 3.1 Avenues for Economic Impacts of Protected Areas As noted, tourism in protected areas can impact local economies through direct (shown by arrows a in Figure 4) and indirect channels. Indirect channels can, in turn, be broadly clas- sified as: production linkages (shown by arrows b in Figure 4) and income and consumption linkages (shown by arrows c in Figure 4). Figure 4. Economic Pathways of Tourism in Protected Areas Revenue sharing, community projects Environmental Impact b PARK AUTHORITY, Businesses pay GOVERNMENT taxes and fees b a Pay non-consumptive a Pay a Park and consumptive fees Entrance Fee and taxes b Park hires guards or employs households for PA activities a Protected Areas a Purchase goods Terrestrial / Marine Spend money on and services lodging, tourist TOURISM, HOUSEHOLDS LODGES AND OU activities G N BUSINESSES RIST S VISITI T Wages paid to Purchase food, a workers employed b,c goods and in tourism activities Local incomes services c increase; b Source goods households spend and services their income to source goods LOCAL FARMS AND BUSINESSES Trade with outside/ M e tho do logy a non-local markets Direct Impact Legend of Pathways of Influence a. Direct impacts Indirect impact through b. Production linkages production linkages c. Income and consumption linkages 28 ASSESSING THE ECONOMI C IMPAC T OF TOURISM IN PROTE CTED AREAS ON LO CA L E CONOMIES IN Z AMBIA 3.1.1 Direct impacts first round of indirect effects in a local economy through production linkages. For example, more Protected areas attract tourists who spend tourists mean increased demand for lodging money on various services. Tourists spend mon- and restaurant meals, and therefore greater de- ey at lodges inside national parks and GMAs mand for everything from ingredients (meat, fish, (often via tour packages they purchase outside fruits, vegetables, etc.) to beverages and nap- the country which in turn channel money into kins and workers. To the extent that lodges and the lodges). Tourists also partake in tourism tour operators hire workers from local house- activities such as game drives, walking safaris, holds and purchase goods and services from wildlife photographic tourism, hunting, etc. ei- local farms and businesses, the greater demand ther through the lodges or hunting outfitters and for lodging and meals will have positive linkage other tourism service providers. Tourists some- effects on the local economy. Inputs purchased times purchase goods and services directly from outside the local economy will create from local businesses and households. Finally, positive linkages for other parts of the country, tourists also pay park entrance fees that accrue or potentially in other countries and not for the to the GoZ Treasury along with consumptive local economy. Similar impacts are realized and non-consumptive fees and taxes from when a park hires local guards or employs local tourism lodges and businesses. Among these households, when hunting and other revenues channels of tourist spending, the only one that are shared with CRBs and used to source local contributes directly to local economies is when goods or to employ local people, and when the tourists buy directly from local businesses and corporate social responsibility (CSR) programs households, and these opportunities are limited of lodges and other businesses generate local or nonexistent when visitors fly directly to parks economic activity. When tourist services, pro- or airstrips close to them. A tourism impact tected area management activities, and those analysis based on tourist expenditures would promoted by CRB and CSR programs expand, stop here and would only capture a fraction of they create positive indirect impacts on local the impact of tourism in protected areas on the economies. On the other hand, a contraction local economy. in resource-extraction may have an opposite Protected areas also affect local economies effect if extraction relies on local inputs. An in- directly by affecting resource extraction—in put-output (IO) analysis would stop here, and the case of Zambia’s national parks, through only capture the direct impacts and the indirect restrictions on hunting and fishing. By regulating impacts through production linkages.  these activities, protected areas can have an ad- A critical issue when analyzing these production verse effect on the incomes of households that linkages is whether the local supply of goods would otherwise hunt or fish. On the other hand, and services can expand to meet the new de- by promoting the recovery of over-exploited mand. If it does not, growth in demand around common property resources (forests, animals, protected areas may inflate prices. This reduces and fish), protected areas may also increase the real or inflation-adjusted income gains from sustainable resource extraction. Wildlife in parks protected areas. Estimation of indirect impacts is protected but recovering populations may must take these potential inflationary effects into spill over into GMAs or buffer zones. The ben- account. efits of larger wild animal populations include hunting opportunities for local people in the consumptive zones inside GMAs. Of course, 3.1.3 Indirect Impacts Through larger wild animal populations also increase the Income and Consumption Linkages Me th od o lo gy likelihood of human-wildlife conflicts, as when elephants raid farmers’ fields or predators attack In addition, all production activities in the local livestock. The overall impacts of wildlife on local economy triggered by tourism in protected incomes are therefore unclear, and need to be areas generate income in the form of wages both quantified, and better understood. and profits. Wages paid to workers in tourism potentially also have a positive indirect effect on the local economy as they trigger fresh rounds 3.1.2 Indirect Impacts Through of spending. Wages and profits from locally Production Linkages owned tourist businesses, and local businesses which supply them, flow into local households, As tourism activities expand and resource which in turn spend income in the local econo- extraction contracts, these activities’ demand my. Of course, if resource extraction contracts for intermediate inputs will change, producing a ASSESSING T H E E C ONOMI C IMPACT OF TOURISM IN PROTE CTED AREAS ON LO CA L E CONOMIES IN Z AMBIA 29 household incomes, then indirect income more than a dollar. Local income multipliers are effects from the protected area via that activity not necessarily greater than one, because the may be negative instead of positive. new demand created by tourist spending may be met by purchases from other parts of Zambia As local activities expand to supply new house- or abroad. In this case, the income “leaks out” hold demands, new rounds of increased input from the local economy to other places, creating demand, income, and household expenditures benefits there instead. If the supply of goods follow, creating additional increases in income and services in the local economy is elastic, and demand in the local economy. Successive prices will not change much as local demand rounds of impacts become smaller and smaller, increases. Otherwise, rising local demand could and the overall (direct and indirect) effect of the place upward pressure on prices, causing real or expansion in tourism converges to an income price-adjusted multipliers to diverge from nominal multiplier, defined as the change in local house- (cash income) ones. The general equilibrium (GE) hold incomes per unit of fresh infusion of cash model will capture all of the effects, the direct into the economy through tourist spending. If impacts and both channels of indirect impacts. local market linkages are strong, each dollar of tourist spending may increase local income by 3.2 LEWIE Model Quantifying the direct and indirect impacts of intermediate inputs (fertilizer, seed, and a variety tourism in protected areas on local economies of purchased inputs) to produce an output therefore requires an applied GE approach. For (corn, prepared meals, a service), which may be this study, a GE method called “local econo- consumed locally or sold. The household and my-wide impact evaluation” (LEWIE) was used.11 household-farm models describe productive activities, income sources, and consumption/ LEWIE uses simulation methods to estimate expenditure patterns. In a typical model, house- the direct and indirect (or “spillover”) effects of holds participate in activities such as crop and protected area-induced tourism. LEWIE uses a livestock production, resource extraction (e.g., structural approach that integrates models of fishing), retail, other business activities, and in actors (businesses and households) within a GE the labor market. Production functions for each model of the local economy. Businesses include activity are the recipes that turn inputs into locally owned firms and businesses not owned outputs. by locals but typically employing some local workers and purchasing some locally supplied Micro survey data are required to populate the inputs. There is a rich tradition in economics LEWIE model, and play two main roles in its con- of using micro survey data to construct mod- struction. They provide initial values for variables els of agricultural households that are both in the model i.e., inputs and outputs of each producers and consumers of food (Singh, production activity, and household expenditures Squire, and Strauss 1986). LEWIE initially uses on goods and services. The data are also used micro-survey data and econometric methods to econometrically estimate model parameters to construct models of firms, households, and for each household group and sector, together household-farms within local economies. Then, with standard errors on these estimates. These M e tho do logy these micro-models are “nested” within a GE initial values and parameter estimates are orga- model of the local economy, drawing from the nized into a data input spreadsheet designed to literature on GE modeling in economics (Dixon interface with the GAMS (Generalized Algebraic and Jorgenson 2012). The models of firms de- Modeling System) software used to program the scribe how businesses combine various factors LEWIE model. (e.g., hired labor, family labor, land, capital) and 11 A basic reference for this methodology and examples of recent studies using the LEWIE methodology can be found at http://beyondexperiments.org/ (Taylor and Filipski 2014). 30 ASSESSING THE ECONOMI C IMPAC T OF TOURISM IN PROTE CTED AREAS ON LO CA L E CONOMIES IN Z AMBIA 3.3 Data Collection12 To build the LEWIE model, data were gathered through surveys of tourists, lodges and resorts, local businesses, and local households. Surveys gathered information on production, income, expenditures, and the locations of transactions (i.e., whether they are inside or outside the local economy). The household and local business surveys were entered onto tablets using the Open Data Kit (ODK) platform for Android. A team of Zambian enumerators were trained to carry out the business and household surveys (see Box 3). Table 2 shows the total number of households, the sample size, and the percentage of house- Box 3: Building Capacity While Doing Research holds surveyed at each site. A team of 15 Zambian university students and recent graduates were trained Data on visitor expenditures were gathered to carry out the fieldwork for this study. This included a one-week face-to- through hard-copy survey forms distributed by face course on the LEWIE methodology, and on how to conduct detailed Proflight, the air carrier that services the majority household and business surveys with questionnaires using the ODK plat- of tourists, on its return flights to Lusaka from form. At the end of the week, the team visited a village near Lusaka for field each of the two parks. In total, 226 visitors testing. This was followed by two weeks of data collection around Lower Zambezi National Park and two weeks around South Luangwa National Park. returned complete and usable questionnaires Enumerators were awarded certificates of completion of the LEWIE survey for our survey. Of these, most—all but 12—were training course and fieldwork. international visitors. The students (8 men and 7 women) were extremely appreciative of the Reliable data on lodges’ income and expendi- hands-on experience through which they gained research, survey, and team- work skills, and felt privileged to view their country’s natural assets up close. tures is proprietary and difficult to obtain. For They reported seeing lots of elephants and other animals on the way to this study, these data were gathered from three fieldwork sites, and appreciated learning about protected area management sources: data collected by the survey team and the tourism sector. Some quotes from their feedback report include: from six lodges at both parks, a recent survey My experience in the field in lower Zambezi was an eye opener to the of 13 lodges in South Luangwa conducted by adverse effects of drought and climate change and how it affects ordinary Chidakel, Child, and Muyengwa (2021), and people who heavily depend on agriculture in this region for their livelihoods information about lodges, including nightly rates, -Mr. Kenneth Mulenga. available online.13 The data were broadly consis- With this experience, I gained more knowledge on protected areas, gained tent across sources. more skills with field work, my group working skills have improved and my communication skills have really been enriched. It was an extremely, over- whelming experience working in the field – Miss Nozyenji Mwale. Table 2. Sample Sizes and Local Populations LZ SL Me th od o lo gy GMA Market Town GMA* Market Town Estimated number of households 6100 3000 8000 10000 Sample size for survey 311 164 330 124 Percent surveyed 5.1% 5.5% 4.1% 1.2% *Note: Number of households for SL only includes the two surveyed Chiefdoms. 12 Further details on the data collection methods are provided in Annex 2. 13 For example, see https://www.zambiatourism.com/accommodation/lower-zambezi/national-parks/zambia/38/ ASSESSING T H E E C ONOMI C IMPACT OF TOURISM IN PROTE CTED AREAS ON LO CA L E CONOMIES IN Z AMBIA 31 M e tho do logy 32 ASSESSING THE ECONOMI C IMPAC T OF TOURISM IN PROTE CTED AREAS ON LO CA L E CONOMIES IN Z AMBIA 4 Data Summary Data S um m a ry ASSESSING T H E E C ONOMI C IMPACT OF TOURISM IN PROTE CTED AREAS ON LO CA L E CONOMIES IN Z AMBIA 33 4.1 Tourists and Tourism Businesses Figure 5 shows the shares of visitors by region to local economic impacts),16 and dividing by of origin. The largest share came from the the average stay at each park, gives the local United Kingdom, followed by the rest of Europe expenditure per tourist per night. These expen- and the United States and Canada. Similar ditures include spending on lodging, meals, shares were from Africa and Oceania. park entry, tours, and out-of-pocket items. They equal 3,097 kwacha (US$221) at Lower Zambezi Less than one-fifth (18 percent) of surveyed and 1,909 (US$136) at South Luangwa (Table tourists reported visiting Lower Zambezi during 4). Most of this money flows into lodges, which their trip, while many more (93 percent) reported offer rooms, meals, and park tours. Spending visiting South Luangwa. These percentages add outside of tourism packages was higher at up to more than 100 percent, because many Lower Zambezi (1,430 kwacha, or US$130) than people visited both parks.14 The majority of South Luangwa (881 kwacha, or US$68). visitors came for tourist activities (greater than 90 percent). On average, the duration of stay was 3.4 nights and 5.0 nights for Lower Zambezi Figure 5. Tourists by Origin and South Luangwa, respectively; the sizes of tourist groups were comparable between sites Asia 4% (see Table 3). UK 30% Oceania By far the largest expenditure was on tourism 12% packages that included accommodation, meals, and park entry fees and tours. In most cases, packages also included international airfares, Africa transport to and from the park (usually by plane), 13% and commissions to booking agents. On aver- age, tourists spent 40,661 kwacha (US$2,90415) per person to visit Lower Zambezi and 34,359 US/Canada Europe 22% kwacha (US$2,454) to visit South Luangwa 19% National Park. Subtracting booking agent commissions, airfares, and other costs of getting Source: Tabulations from visitor survey to and from parks (as these do not contribute Table 3. Trip Characteristics  Share of Average Average Number in Party Surveyed Number of Tourists Nights Spent Total Adults Lower Zambezi Mean 0.18 3.37 5.9 5.66 (N=41 visits) SD (0.39) (1.3) (6.34) (5.92)             Data S um m a ry South Luangwa Mean 0.93 5.01 5.7 5.39 (N=211 visits) SD (0.25) (0.39) (6.18) (5.66) Source: Tabulations from visitor survey 14 The percentages from the survey data roughly align with the most recently available official data on park visits: in 2018 there were 11,161 visits to Lower Zambezi and 43,469 to South Luangwa (DNPW estimates). The percentage of visits to Lower Zambezi in the official data (20 percent of the total for the two parks) is close to that of our survey. In our survey, the percentage of respondents who reported visiting South Luangwa is somewhat higher than that from the official data for 2015 (80 percent). 15 The exchange rate for survey and model outputs is 1 ZMK = 14 USD unless otherwise noted. This rate reflects an average exchange across the time of the survey. 16 It is estimated that about 60.1 percent of the cost of tour packages is spent on lodges. This is a rough estimate based on the costs of direct bookings as advertised on websites, and is used in the absence of such information from hotels themselves. Hotels that do not advertise their prices, including some luxury lodges, were not included in this estimate, and therefore, accommodation costs are conservatively estimated. 34 ASSESSING THE ECONOMI C IMPAC T OF TOURISM IN PROTE CTED AREAS ON LO CA L E CONOMIES IN Z AMBIA Taking the average for each expenditure catego- retailers. On average, lodges purchase approx- ry, it was estimated that lodges spent an average imately 16 percent of their daily inputs (in value of 6,314 kwacha per visitor on wages, 249 terms) from non-local sources.17 These expen- kwacha on locally-grown crops, 202 kwacha on ditures are the indirect pathways through which local livestock products, 1,213 kwacha on local local communities benefit from tourist spending. services, and 740 kwacha on goods from local Table 4. Local Tourist Expenditures in Kwacha Lower Zambezi South Luangwa 0.68 0.79 % Purchasing Packages (0.47) (0.41) Mean Local Expenditures Per Tourist/Night (lodging, meals, park entry, tours, out-of-pocket) All Visitors 3097 1909 SD (3523) (2170) International Visitors 3,600 1,998 SD (3617) (2191) Domestic Visitors 331 242 SD (295) (198)  Mean Local Expenditure Per Tourist/Night, by Type* (Only Available for International Tourists) Lodges (includes room, meals, tours, park entry) 1,668 1,028 SD (1897) (1168) Mean Out-of-Pocket Spending Total 1430 881 SD (1626) (1001) At Lodge (e.g., curios, bar) 336 207 SD (382) (235) Local Retail Shops† 720 444 SD (819) (504) Local Services† 107 66 SD (122) (75) Local Transport 267 164 SD (304) (187) Sample (Max) 33 177 Data S um m a ry Sample Size 226 *Note: Breakdown by expenditure type is only possible for international visitors, due to small sample size of domestic visitors. Airfare and Hotel costs are derived using average flight costs when costs are included in tour packages. Out-of-Pocket Spending not included in package cost. †Local retail shops include all manner of convenience shops, liquor shops, souvenir shops and other retail goods. Local Services include bars and clubs, massages, and other services. **Sample size varies due to availability of information. 17 As previously noted, the definition of local economy includes the nearby market towns (Chirundu for Lower Zambezi and Chipata for South Luangwa), where most hotels purchase inputs in bulk. ASSESSING T H E E C ONOMI C IMPACT OF TOURISM IN PROTE CTED AREAS ON LO CA L E CONOMIES IN Z AMBIA 35 4.2 Households The household surveys provide rich data on Table 6 reports households’ demographic char- household demographics, economic activities, acteristics and activities. Households at Lower and spending, which determine economic im- Zambezi and South Luangwa have 4.5–6.2 pacts within local economies around each park. members on average, with poorer households having more members on average (6.2 vs 5.5 Average household expenditure per capita for Lower Zambezi and 5.6 vs 4.5 for South is 5,067 kwacha (US$362) at Lower Zambezi Luangwa poor and non-poor, respectively). The and 3,108 kwacha (US$222) at South Luangwa most common source of income for households (see Table 5). At Lower Zambezi, 56 percent of at both locations is agriculture, followed by live- households have average per-capita expen- stock, and wage employment. Forty percent of ditures below the poverty line of US$1.90/day poor households in Lower Zambezi and 24 per- (using PPP adjusted exchange rates).18 The cent in South Luangwa have some form of wage poverty rate is higher at South Luangwa at 83 employment; these numbers are higher, at 59 percent. The World Bank reports that Zambia percent and 47 percent for their non-poor coun- as a whole had a poverty rate of 58 percent in terparts in Lower Zambezi and South Luangwa, 2015, the latest year for which this information is respectively. Wage employment includes work available.19 in non-tourism and tourism activities. Table 5. Poverty Headcount Average Per Capita Expenditure Poverty Headcount* (Annual) Mean 5,067 Lower Zambezi sample: 329 0.56 SD (6,478) Mean 3,108 South Luangwa sample: 233 0.83 SD (3,377) Note*: Poverty Headcount calculated as the proportion of households with under $1.90/day per capita income (PPP adjusted using PPP exchange rate of 4.475) Table 6. Household Demographics and Activities Share of Households Participating in: Head Head HH Size Age Educ Wage Agri Livestock Fishing Business Employment Lower Zambezi Mean 6.15 47.2 5.52 0.76 0.50 0.15 0.27 0.40 Poor SD (2.57) (14.0) (3.76) (0.43) (0.50) (0.36) (0.44) (0.49) Lower Zambezi Mean 5.52 43.1 7.86 0.77 0.59 0.23 0.34 0.59 Non-poor SD (2.44) (12.9) (3.50) (0.42) (0.49) (0.42) (0.48) (0.49) Data S um m a ry South Luangwa Mean 5.60 46.1 5.77 0.94 0.60 0.11 0.23 0.24 Poor SD (2.33) (15.3) (3.42) (0.24) (0.49) (0.32) (0.42) (0.43) South Luangwa Mean 4.54 46.3 6.78 0.85 0.44 0.13 0.33 0.47 Non-poor SD (2.29) (15.8) (3.53) (0.36) (0.50) (0.33) (0.47) (0.50) Source: World Bank survey 18 World Bank 2018 PPP exchange rates were used in lieu of 2019 figures, which were not available at the time of this study. 19 The World Bank in Zambia: Overview. https://www.worldbank.org/en/country/zambia/overview 36 ASSESSING THE ECONOMI C IMPAC T OF TOURISM IN PROTE CTED AREAS ON LO CA L E CONOMIES IN Z AMBIA The importance of local tourism is evident and tour businesses, tend to pay higher wages in household wage activities (see Table 7). than other activities: 58 kwacha per day for poor Twenty-four percent of poor households and workers and 84 kwacha per day for non-poor 34 percent of non-poor households around workers at Lower Zambezi, and 58 kwacha per Lower Zambezi National Park and 17 percent of day for poor workers and 66 kwacha per day poor households and 42 percent of non-poor for non-poor workers at South Luangwa.20 Types households around South Luangwa had wage of tourism-related jobs listed on the household income from tourism employment during the survey include visitor services (restaurant work, year prior to the survey. Non-poor households employment at hotels/lodges and tour agen- have a higher percentage of members working cies), maintenance (repairs, ground keeping) in tourism-related enterprises. Tourism-related and crafts (handicraft manufacturing). activities, including jobs in lodges, restaurants Table 7. Wage Income and Employment Average Share Average Share Average Share with Days Wages with Tourism Working > Wage Tourism Worked per day Second Employment 150 Days Income Employment (kwacha) Job Wage Lower Zambezi Mean 152.2 0.47 7325 52.2 0.08 0.24 58.4 Poor N=117 SD (113.1) (0.50) (8,208) (66.3) (0.27) (0.43) (31.6) Lower Zambezi Mean 191.0 0.67 13710 71.4 0.01 0.34 83.7 Non-poor N=136 SD (103.9) (0.47) (11,898) (57.6) (0.12) (0.47) (57.7) South Luangwa Mean 120.9 0.31 4065 33.2 0.04 0.17 58.2 Poor N=95 SD (90.7) (0.46) (5,176) (32.1) (0.20) (0.38) (30.7) South Luangwa Mean 200.3 0.64 11525 56.4 0.00 0.42 66.0 Non-poor N=36 SD (89.8) (0.49) (8,264) (30.4) N/A (0.50) (20.3) Source: World Bank survey Data S um m a ry 20 Variation in wages between poor and non-poor workers in the same sector is due to differing positions held by workers within these groups (e.g. a poor employee in the tourism sector likely works in a job that pays lower wages than the job of a non-poor employee). ASSESSING T H E E C ONOMI C IMPACT OF TOURISM IN PROTE CTED AREAS ON LO CA L E CONOMIES IN Z AMBIA 37 4.3 Local Business and Entrepreneurial Activities Thirty percent of households in Lower Zambezi varies between the two survey sites (Table 8). and 25 percent of households in South More service-type businesses are found in Luangwa own and operate some form of South Luangwa.21 This may be because visitors business. Businesses are defined as entrepre- pass through communities on their way from neurial activities, including small business types the airport to South Luangwa, while at Lower such as hawkers, small grocery stalls, and other Zambezi, there are no communities or business- roadside vendors. Distribution of business types es between the airport and the park. Table 8. Household Business Types by Site Business Type Lower Zambezi South Luangwa Agriculture 8.8% 17.7% Livestock 0.6% 0.6% Retail 68.5% 43.7% Services 16.6% 33.7% Hotels and Lodges* 1.7% 0.0% Tour Operators 3.9% 4.4% Total 100% 100% Sample (number of businesses) 172 181 *No households in the South Luangwa sample operated a hotel or lodge. Source: World Bank Survey Data S um m a ry 21 A number of restaurants, coffee shops, barbers, and other services line the road leading up to the main gate of South Luangwa National Park. 38 ASSESSING THE ECONOMI C IMPAC T OF TOURISM IN PROTE CTED AREAS ON LO CA L E CONOMIES IN Z AMBIA 5 LEWIE Model Findings LE WIE M o de l F in di n g s ASSESSING T H E E C ONOMI C IMPACT OF TOURISM IN PROTE CTED AREAS ON LO CA L E CONOMIES IN Z AMBIA 39 As noted above, the LEWIE model can be used Simulations require making judgements, based to estimate the direct and indirect impacts of on the survey data, about where and how prices tourism in protected areas on a local economy. are determined (that is, market closure, which There are many avenues through which these is not known with certainty). Sensitivity analyses impacts manifest. Data availability determines were performed, and combined with the Monte in large part the extent to which these avenues Carlo method described above, to test the ro- can be captured through the LEWIE model. A bustness of simulated impacts to market-closure summary of the avenues and how they are mod- assumptions. eled by LEWIE is provided in Table 9. The impact of tourism in protected areas on the Once built, the LEWIE model can be used to local economy is estimated in two steps. Step quantify impacts on a local economy. Because one entails simulating the impact of an addition- the model parameters have been estimated al tourist on the local economy. This step also econometrically, Monte Carlo methods are used provides an estimate of the income multiplier to perform significance tests and construct of an additional dollar of tourist spending. The confidence intervals around the simulated total impact is estimated in the second step by impact results as shown by Taylor and Filipski multiplying the per-tourist estimate by the total (Taylor and Filipski 2014). For this study, 500 number of tourists who visit the national park. iterations of the simulations for each park were Comparing the total impact with public invest- conducted. Additionally, the LEWIE model ment in the park provides an estimate of the considers nonlinearities and local price effects. rate of return on the public investment. Table 9. Avenues of Impact Captured by LEWIE Included Impact Avenue Comment in LEWIE? Direct Tourist spending at local Yes businesses Restrictions on resource use Yes These impacts are built into the base run of the model. It is important to and positive spillovers from note that this version of LEWIE is static and therefore does not account Park to GMA for changes in the resource base and its use. Impact of human-wildlife Yes As per the information provided in the household surveys, crop conflict damage caused by animals (primarily elephants, and in Lower Zambezi hippos as well) was between 11–14 percent of total output. This impact is included in the base run as households currently receive no compensation for this damage. Indirect – Hiring and local sourcing Yes These linkages are included for lodges but not for other tourism production of goods by tourism service providers due to data limitations. Only four percent of farmers linkages establishments reported selling produce directly to lodges. Most crop sales are through traders and intermediaries who collect goods from farms and sell them on to hotels and other businesses. Hotels also source goods and services from businesses in local towns. Hiring and local sourcing of No CRBs receive money from hunting licenses and use it to provide goods by CRBs services such as maintenance of public facilities and hiring of scouts L E WIE M o de l F in di n g s for the park. The LEWIE model does not account for CRB spending due to lack of information on their budget and expenditure. Hiring of local labor, however, is reflected in the household surveys. Hiring and local sourcing of Partially Hiring staff for park operations is captured in the household section goods by park managers of the surveys. Scouts and rangers are largely locals. However, operational costs of park management (DNPW) are not included due to lack of information on their expenditures. Activities supported by CRS No Because breakdowns of CSR expenditures into different categories, programs as required for the LEWIE estimation, were not available, CSR expenditures are not included in the analysis. Input use spillover effects of Yes resource use restriction Indirect – Household expenditures Yes consumption based on wages and profits linkages earned through tourism sector linkages * This question was only added towards the end of the survey period. 40 ASSESSING THE ECONOMI C IMPAC T OF TOURISM IN PROTE CTED AREAS ON LO CA L E CONOMIES IN Z AMBIA 5.1 Impact of Additional Tourist on the Local Economy Table 10 shows average tourist spending and are 17,959–20,075 (US$1,283–1,434) at Lower the impacts of an additional average park Zambezi and 14,106–15,007 (US$1,008–1,072) tourist on household incomes around the two at South Luangwa. These income impacts are parks. Simulations find that an additional tourist particularly striking when one considers that adds 18,968 kwacha (US$1,355) to local real data limitations did not allow for the inclusion of (inflation-adjusted) income to the economy some key avenues of impact, such as expendi- surrounding Lower Zambezi National Park and tures by the CRBs. 14,625 kwacha (US$1,045) at South Luangwa Most of the local income gain goes to house- National Park. We estimate that each additional holds within the GMAs surrounding the parks. visitor to South Luangwa and Lower Zambezi Poor households, which are more numerous, spends an average of 10,437 kwacha (US$745) tend to receive more benefits. Poor households and 9,564 kwacha (US$683) in their local econ- located in the GMA at Lower Zambezi receive omies, respectively. Total spending per tourist is 6,429 kwacha (US$459) per additional tour- higher: 40,661 kwacha (US$2,904) and 34,359 ist; non-poor households gain 5,984 kwacha kwacha (US$2,454), respectively. The difference (US$427). The gains are 4,865 kwacha (US$347) between total and local spending includes and 1,139 kwacha (US$81) for poor and non-poor international and local airfares, commission GMA households at South Luangwa, respective- fees on packages, and costs of visiting other ly. Poor and non-poor households in the market parks in Zambia, which do not enter into the town near Lower Zambezi gain 3,876 kwacha local economies of the study sites. The impacts (US$277) and 2,679 kwacha (US$191), respec- on local incomes are larger than the amount of tively. Poor households in the market town near money tourists spend in the local economy be- South Luangwa gain 7,916 kwacha (US$565), cause of the income and production spillovers and non-poor households gain 706 kwacha this spending generates. The bottom of the (US$50). top panel of Table 10 presents 95% confidence intervals around these local GDP impacts. They Table 10. Local Income Impacts of an Additional Tourist Lower Zambezi South Luangwa Income effects of one additional tourist Kwacha US$ Kwacha US$ Average total spent by an additional tourist 40,661 2,904 34,359 2,454 Average total spent by an additional tourist 10,437 745 9,564 683 in LZ/SL Changes in local economy incomes LE WIE M o de l F in di n g s Real (inflation-adjusted) Income 18,968 1,355 14,625 1,045 95% CIs [17,959; [1,283; [14,106; [1,008; 20,075] 1,434] 15,007] 1,072] Changes in household incomes, by location       Poor households in GMA 6,429 459 4,865 347 Non-poor households in GMA 5,984 427 1,139 81 Poor households in local market town 3,876 277 7,916 565 Non-poor households in local market town 2,679 191 706 50 ASSESSING T H E E C ONOMI C IMPACT OF TOURISM IN PROTE CTED AREAS ON LO CA L E CONOMIES IN Z AMBIA 41 Figure 6. Real-Income Multipliers for an Additional Kwacha of Tourist at these two sites, respectively. There are also Spending positive impacts on crop (1,591 kwacha at Lower 2,5 Zambezi and 3,227 kwacha at South Luangwa) and livestock (1,050 kwacha at Lower Zambezi 2,0 and 882 kwacha at South Luangwa) production. Households pursuing these activities, as their revenues increase, hire labor, purchase inputs, 1,5 and generate profits that add further to local incomes. 1,0 1,82 Figure 6 shows the income multipliers, that 1,53 is, impacts on local income or GDP for each 0,5 additional kwacha that visitors spend locally. Note that the additional kwacha that visitors 0,0 spend is primarily at lodges, and does not add Lower Zambezi South Luangwa to local income until it creates income gains in local households. Thus, these multipliers Tourist spending creates these income impacts represent mainly indirect effects of tourist by stimulating the local demand for goods and spending on local incomes. The multipliers are services, either directly (as when tourists or adjusted for price inflation and thus represent lodges buy goods and services from local busi- real-income effects. An additional kwacha nesses and households), or indirectly (as when spent by visitors at Lower Zambezi raises the lodges pay wages to local households, which total income of households around the park by in turn spend their income on locally-supplied 1.82 kwacha. This is higher than the multiplier goods and services). Table 11 summarizes the at South Luangwa: 1.53 kwacha. The vertical impacts of an additional park visitor on produc- line at the top of each bar gives the 95-percent tion (in value) by local farms and businesses. confidence interval around the income multipli- The largest impact is on business activities er, obtained by running 500 iterations of each (retail and service), mostly the small fami- simulation. The two lines are short compared ly-owned stores at which households around with the corresponding bars, indicating high parks spend the largest share of their incomes. confidence in the estimates. Both multipliers are The value of retail sales increases by 2,995 positive and large, and the confidence intervals (US$214) kwacha at Lower Zambezi and 8,085 lie well above 1.0. This indicates that, in most kwacha (US$578) at South Luangwa, while gross cases, each kwacha that tourists spend creates revenue in service and other activities rises by significantly more than one additional kwacha 5,442 and 3,788 kwacha (US$389 and US$271) of income in communities around the parks. Table 11. Production Impacts of an Additional Tourist Production effects of one additional tourist Lower Zambezi South Luangwa L E WIE M o de l F in di n g s Kwacha US$ Kwacha US$ Agricultural Crops 1591 114 3227 230 Livestock 1050 75 882 63 Retail 2995 214 8085 578 Services and Other Production 5442 389 3788 271 42 ASSESSING THE ECONOMI C IMPAC T OF TOURISM IN PROTE CTED AREAS ON LO CA L E CONOMIES IN Z AMBIA Figure 7 shows how much of the multiplier town. Poor and non-poor GMA households in benefits households within the GMAs versus Lower Zambezi gain 0.62 kwacha and 0.57 households in the market towns. At Lower kwacha, respectively, of new income for every Zambezi, where roads are poor, more of the kwacha that tourists spend there. Poor and non- income benefits stay inside the GMA than at poor households in the market town near Lower South Luangwa, where there is better market Zambezi gain 0.37 kwacha and 0.26 kwacha, integration between the park, GMA, and nearby respectively. In South Luangwa, households in the GMA receive 0.51 (poor) and 0.12 (non-poor) Figure 7. Which Households Benefit from Tourist Spending kwacha of income per kwacha spent by tourists. Multipliers? Market town households near South Luangwa benefit substantially, with poor and non-poor 2 households gaining 0.83 and 0.07 kwacha, re- 1,8 spectively, per kwacha spent locally by tourists. 0,26 Nonpoor households 1,6 0,07 in local market town The discrepancy in multiplier shares between 1,4 0,37 poor and non-poor households partly reflects 1,2 Poor households in local market town the size of the population in each category. 1 0,83 Normalizing multiplier shares by populations (i.e., 0,57 0,8 Nonpoor households dividing the multiplier share by the population 0,6 in GMA share for each group; see Figure 8) reveals that 0,12 0,4 the multiplier share per resident is largely compa- 0,62 Poor households in GMA 0,2 0,51 rable between poor and non-poor populations. In the pie charts in the figure, a number greater 0 Lower Zambezi South Luangwa than 1.0 indicates that a household group’s share of benefits is larger than its share of population, and a number less than 1.0 indicates the oppo- site. At Lower Zambezi, the relative shares are slightly less than 1 (0.83) for poor households in Figure 8: Distribution of Multiplier across Poor and Non-Poor the GMA and slightly greater than 1 for the other Populations three household groups. At South Luangwa, the Poor households relative share is slightly higher for the market Nonpoor households in GMA town poor and GMA non-poor, and it is slightly in local market town 0,87 lower for the other two groups. 1,05 There are several reasons why the total multipli- er effects of an additional tourist who spends an Lower additional kwacha are higher at Lower Zambezi Zambezi than South Luangwa. At Lower Zambezi, a higher percentage of workers are employed by the tourism industry, and wages are higher than Nonpoor at South Luangwa. Wages are the main avenue Poor households in households through which tourism directly affects household Local market town In GMA incomes. Anecdotally, people explained that 1,05 1,13 they have sustained themselves through the LE WIE M o de l F in di n g s recent drought in part from the wages of lodge workers, and the movement of this income Nonpoor households through their tightly knit communities. Villages Poor households in the GMAs around South Luangwa are more in local market town in GMA 0,58 integrated with outside markets, and thus, more 0,87 tourism-derived monies escape from the South Luangwa economy. The larger impact on house- holds in the market town near South Luangwa South reflects this greater market integration between Luangwa the GMA and the town. Poor households in Local market Nonpoor town households 1,14 In GMA 1,24 ASSESSING T H E E C ONOMI C IMPACT OF TOURISM IN PROTE CTED AREAS ON LO CA L E CONOMIES IN Z AMBIA 43 5.2 Total Impacts of Nature Tourism on the Local Economy It is impossible to know what the local econo- and 28.2 kwacha per 1 kwacha of government mies around these two parks would look like in spending on South Luangwa National Park. the absence of tourism. Nevertheless, the total The impact of tourism on employment around impact of nature-based tourism on incomes the two parks includes employment by tourist around the two parks can be approximated by businesses and indirect employment impacts multiplying the impact per additional tourist by from tourism. These employment effects can be the number of tourists visiting the parks.22 We estimated by dividing the total labor value-add- estimate that tourism adds 212 million kwa- ed by the average local wage.23 Based on this cha (US$15.1 million) to total income or GDP method, we estimate that national park tourism around Lower Zambezi and 635 million kwacha generates 7,463 full-time equivalent jobs around (US$45.4 million) to the GDP around South Lower Zambezi National Park and 28,210 jobs Luangwa National Park (Column A of Table 12). around South Luangwa National Park.24 To It is important to note that even though the ef- put these employment impacts into perspec- fect of an additional visitor and kwacha spent is tive, they are equivalent to 14 percent and 30 higher at Lower Zambezi, impacts of nature tour- percent of the total populations around the two ism are much larger in South Luangwa because parks, respectively.25 of its higher tourist numbers. Our results are comparable to findings from a Dividing these economic impacts by govern- recent study of the economic impact of tourism ment expenditures (wage and non-wage) on the around South Luangwa National Park (Chidakel, two parks provides estimates of the returns on Child and Myengwa, 2021).26 This study found government spending. Based on this calcula- that, even though only 25 percent of tourist tion, there are substantial economic returns on spending is captured locally, tourism contributes government spending at both parks: 16.7 kwa- around 40 percent of local income, and 70 per- cha of income gain per 1 kwacha of government cent of local businesses are highly dependent spending on Lower Zambezi National Park, on tourism expenditures. The study concludes that multiplier effects are an order of magnitude Table 12. Estimated Impact of Tourism (>10 times) higher than park management costs. Our study differs from Chidakel et al. (2021) in A B C D both its modeling approach and geographic Estimated Expenditure Expenditure Rate of scope. Besides surveying households in the Economic on Park on Wages Return GMA area, our study includes businesses and Impact of Maintenance (US$) households in and around Chipata town, which Tourism (US$) (US$) is a major commercial hub in the South Luangwa area. We also include the nearby market town Lower Zambezi in the Lower Zambezi survey. By casting our net 15,121,783 43,990 861,352 16.7 NP more widely in this way at both sites, we detect- South Luangwa ed more of the economic benefits created by 45,410,793 172,167 1,438,573 28.2 NP L E WIE M o de l F in di n g s nature-based tourism. * Expenditures are reported by GoZ for 2018 and therefore use the exchange rate noted for Table 1. 22 In reality, marginal impacts, as approximated by the impact per additional tourist, may differ at different levels of tourism. Over time, local economies would have to adjust to a complete loss of tourism revenue, and to new access to currently protected lands and activities (e.g., hunting), and these in turn would change the economic impact of the protected area tourism on the local economy. We do not attempt to address these complex, dynamic questions in our modeling, nor the sustainability and resource-management questions that inevitably would arise. These are left for future research. 23 The effect of labor value-added is estimated in the LEWIE model as the returns to labor, a productive asset, and represents the wage income gains to the local economy. Dividing by wages allows us to estimate the extra employment generated through tourist spending. 24 For these calculations, we used average local tourism industry (lodge, restaurant and tour operator) daily wages of 74.4 and 62.0 kwacha/day and average full-time equivalents of 173 and 142 days/year at Lower Zambezi and South Luangwa National Parks, respectively. Average wages are higher in the tourism industry than in other economic activities. Thus, the full-time equivalent job gains reported here understate actual impacts on local employment. 25 For South Luangwa the GMA study area consisted of the two chiefdoms surrounding SLNP. Employment impacts include the two market towns, Chirundu and Chipata. 26 Chidakel, Child and Myengwa estimate that including local contributions, tourism in SLNP in 2015 was responsible for about US$38 million in national GDP. 44 ASSESSING THE ECONOMI C IMPAC T OF TOURISM IN PROTE CTED AREAS ON LO CA L E CONOMIES IN Z AMBIA Governments can create additional benefits This park-hiring impact can also be expressed in for local populations by hiring local people to terms of an income multiplier. An additional kwa- work in parks as guards, guides, game wardens, cha spent by the government on park wages etc. This brings income directly to households creates a local economy real (inflation-adjusted) around the parks. The LEWIE model estimates multiplier of 3.02 kwacha at Lower Zambezi impacts beyond the benefits of park-related and 3.1 kwacha at South Luangwa (Figure 9). jobs, by describing indirect income effects. Households in the GMAs receive 2.67 kwacha The model estimates that an additional worker of this government spending multiplier at Lower hired by the park generates an increase in local Zambezi and 2.41 kwacha at South Luangwa, real income of 19,467 kwacha (US$ 1,479) at and households in the nearby market towns Lower Zambezi and 13,655 kwacha (US$ 1,038) get 0.35 and 0.69 kwacha, respectively. These at South Luangwa. The cost to government of park employment multipliers are higher than hiring an additional worker is 12,871 kwacha tourist spending multipliers because wages paid (US$978) at Lower Zambezi and 8,804 kwacha to locally hired park personnel go directly to (US$669) at South Luangwa per annum, which local households, whereas a fraction of tourist is considerably less than the local income gains spending does. from hiring the additional worker (see Table 13 below). Table 13. Government Hiring an Additional Local Laborer Figure 9. Which Households Benefit from Government Spending Multipliers? 3,5 Income effects Lower Zambezi South Luangwa Kwacha US$ Kwacha US$ 0,05 0,06 3 Changes in local economy incomes 0,29 0,64 Real (inflation-adjusted) 19,467 1,479 13,655 1,038 2,5 Income Changes in household incomes, by location 1,10 Poor households in 2 10,136 770 5,958 453 1,06 GMA Non-poor households 1,5 7,083 538 4,660 354 in GMA Poor households in 1,861 141 2,825 215 1 local market town Non-poor households 1,57 387 29 211 16 1,35 in local market town 0,5 Cost of   12,871 978 8,804 669 Implementation 0 Loss in crop production Not applicable Lower Zambezi South Luangwa LE WIE M o de l F in di n g s Nonpoor households Nonpoor households in local market town in GMA Poor households Poor households in local market town in GMA ASSESSING T H E E C ONOMI C IMPACT OF TOURISM IN PROTE CTED AREAS ON LO CA L E CONOMIES IN Z AMBIA 45 5.3 Impacts of Complementary Investments and Outside Shocks Besides estimating the economic impacts of conflict is included in the base model. A hu- tourism and park hiring on the local economy, man-wildlife conflict simulation which returns lost the LEWIE model can be used to simulate the crops to households was conducted to estimate local economic impacts of other government the loss to the local economy (i.e., it estimates interventions and economic shocks. the counterfactual of no human-wildlife conflict, the negative of which is the local-economy impact of the human-wildlife conflict that actually 5.3.1 Local Economy-Wide Costs of occurred). Human-Wildlife Conflicts Crop losses can have major implications for the Animal incursions from each of the two parks households suffering them, and send negative cause crop losses for nearby households, and ripple effects through local economies. Table 14 information about these losses was gathered presents the impact of crop losses from animal during the surveys. Human-wildlife conflicts re- incursions on income around the two parks. The duce crop output by almost 14 percent at Lower total real (inflation-adjusted) income losses are Zambezi and 11 percent at South Luangwa. This around 23.7 million kwacha (US$ 1.8 million) at is equivalent to 10.4 million kwacha (US$0.8 Lower Zambezi and 16.3 million kwacha (US$ million) in crop losses at Lower Zambezi and 14.1 1.2 million) at South Luangwa. These numbers million kwacha (US$1.1 million) at South Luangwa. represent the total economic cost of crop losses The base LEWIE model uses harvest data re- due to human-wildlife conflicts, and they signifi- ported at the time of the survey. Thus, the 11–14 cantly exceed the direct negative impacts on percent loss of crop value from human-wildlife households that suffer these losses. Table 14: Estimated Losses from Human-Wildlife Conflict Human-wildlife conflict Income effects Lower Zambezi South Luangwa Kwacha US$ Kwacha US$ Changes in local economy incomes Real (inflation-adjusted) Income -23,665,409 -1,798,571 -16,302,224 -1,238,969 Changes in household incomes, by location Poor households in GMA -11,084,700 -842,437 -9,302,568 -706,995 L E WIE M o de l F in di n g s Non-poor households in GMA -7,774,781 -590,883 -2,576,346 -195,802 Poor households in local market -3,684,955 -280,057 -4,182,148 -317,843 town Non-poor households in local market -1,120,973 -85,194 -241,162 -18,328 town Cost of Implementation Not Applicable Loss in crop production -14,728,733 -1,119,384 -12,883,629 -979,156 46 ASSESSING THE ECONOMI C IMPAC T OF TOURISM IN PROTE CTED AREAS ON LO CA L E CONOMIES IN Z AMBIA 5.3.2 Local Economy-Wide Impact of businesses. This was done by increasing the a 5 Percent Increase in Local Input volume of local purchases by businesses (both services and retail) by 5 percent while holding Purchases by Businesses purchases from outside the local economy con- Governments can increase local benefits from stant. The results are shown in Table 15. tourism by encouraging businesses to source A 5 percent increase in local purchases boosts more inputs locally. The LEWIE model was used local incomes by 2.73 million kwacha (US$0.21 to simulate the impact of a 5 percent increase million) in Lower Zambezi and 4.64 million in the amount of goods sourced locally by kwacha (US$0.35 million) in South Luangwa. The largest share of benefits in Lower Zambezi Table 15: Impacts of 5 Percent Increase in Local Purchases by go to non-poor households in the GMA (1.83 Businesses million kwacha, or US$0.14 million), and non- poor households in market towns follow with an Income effects Lower Zambezi South Luangwa estimated impact of 763,958 kwacha (US$0.06 million). Poverty alleviation effects of increased Kwacha US$ Kwacha US$ local purchasing are larger for poorer commu- Changes in local economy incomes nities in South Luangwa. Poor GMA and town Real (inflation- households in the South Luangwa region re- 2,731,311 207,573 4,643,502 352,894 adjusted) Income ceive an estimated 1.63 million kwacha (US$0.12 Changes in household incomes, by location million) and 1.79 million kwacha (US$0.14 million), Poor households in respectively, from the increased local purchas- 61,586 4,680 1,634,221 124,197 GMA ing. This is at least partially driven by the larger Non-poor proportion of poor households at the South 1,826,932 138,842 896,254 68,113 households in GMA Luangwa study site. Poor households in 78,835 5,991 1,789,238 135,978 local market town 5.3.3 Local Economy-Wide Losses Due Non-poor households in local 763,958 58,059 323,789 24,607 to COVID-19 market town Just as increases in tourism and tourist spend- ing have positive multiplier effects, negative shocks produce negative income multipliers in Table 16. Monthly Income Loss from No Tourism local economies. The COVID-19 pandemic has resulted in substantial losses in tourism and Income Loss Per Lower Zambezi South Luangwa tourism income. The LEWIE model was used Month of Lost to simulate the impact of a complete loss of Tourism Million Million Million Million tourism for one month on the local economies Kwacha US$ Kwacha US$ around Lower Zambezi and South Luangwa Loss in local economy incomes parks. Tables 16 and 17 present the estimated Real (inflation- impacts on income and production, respectively. 17.6 1.26 53.0 3.78 adjusted) Income The simulations showed that a complete 95% CIs [17.9;17.4] [1.28;1.24] [54.0;51.9] [3.86;3.71] loss of tourist revenue around the two parks Loss in Household Real Income, by location LE WIE M o de l F in di n g s reduces local real GDP by 17.6 million kwacha Poor households in (US$1.26 million) per month without tourists 6.0 0.43 17.6 1.26 GMA at Lower Zambezi and 53.0 million kwacha Non-poor households (US$3.78 million) per month without tourists at 5.6 0.40 4.1 0.29 in GMA South Luangwa (Table 16). Each month without Poor households in tourists at Lower Zambezi reduces the income 3.6 0.26 28.7 2.05 local market town of GMA poor households by 6.0 million kwacha Non-poor (US$0.43 million) and GMA non-poor house- households in local 2.5 0.18 2.6 0.18 holds by 5.6 million kwacha (US$0.40 million). market town Poor market town households at Lower Zambezi lose 3.6 million kwacha (US$0.26 million), and non-poor households lose 2.5 million kwacha (US$0.18 million). Losses at South Luangwa are substantially larger: Poor (non-poor) GMA households lose 17.6 (4.1) million kwacha, or ASSESSING T H E E C ONOMI C IMPACT OF TOURISM IN PROTE CTED AREAS ON LO CA L E CONOMIES IN Z AMBIA 47 Table 17. Monthly Production Loss from No Tourism US$1.26 (US$0.29) million per month. Market town households are also heavily impacted: Monthly Production Lower Zambezi South Luangwa poor market town households lose 28.7 million Loss kwacha (US$2.05 million), and non-poor house- Kwacha US$ Kwacha US$ holds lose 2.6 million kwacha (US$0.18 million). Agricultural Crops 811,983 57,999 1,479,344 105,667 All production activities lose, with sales losses Livestock 173,876 12,420 976,532 69,752 ranging from 0.17 million kwacha (US$12,420) in Retail 872,033 62,288 2,785,957 198,997 livestock to 0.87 million kwacha (US$62,288) in retail activities at Lower Zambezi, and from Services and other 622,123 44,437 5,061,314 361,522 0.98 million kwacha (US$69,752) in livestock production to 5.1 million kwacha (US$361,522) in services Hotels 69,902 4,993 4,838,106 345,579 and other production at South Luangwa (see Table 17). L E WIE M o de l F in di n g s 48 ASSESSING THE ECONOMI C IMPAC T OF TOURISM IN PROTE CTED AREAS ON LO CA L E CONOMIES IN Z AMBIA 6 ConclusionsandPolicyRecommendations Conclusions and Policy Recommendations ASSESSING T H E E C ONOMI C IMPACT OF TOURISM IN PROTE CTED AREAS ON LO CA L E CONOMIES IN Z AMBIA 49 The study set out to make the case for greater local economy. Both these channels add to the investment of public resources in protected area economic return per kwacha of government management by estimating the local econom- spending. Secondly, because of data limitations, ic impact – direct and indirect – of tourism at some pathways through which tourist spend- two biodiversity-rich areas, through the appli- ing benefits the local economy have not been cation of the LEWIE model. Its focus is on the accounted for, and baseline figures e.g., for park local economy, defined as the households and fees and hotels, are conservatively estimated. businesses in the GMA, and the CRBs in the Thirdly, impacts on the local economy when vicinity of the protected areas and in the main CRBs use revenues from hunting fees to hire market town. This focus was chosen in order local households or source local goods were to understand the potential of protected areas not considered. A limitation of this approach, to benefit local households. These households like other ex-post economic impact evaluations, often suffer from restrictions placed on natural is that we do not know what local economies resource use by conservation authorities, and looked like before the advent of national parks from human-wildlife conflict; and their support and tourism. As tourism expands, economies is critical to the integrity of parks as they can around protected areas evolve. Private and discourage encroachment, poaching, and other public investments stimulate and transform the threats. Economic development of the local structure of the economy in ways that the model economy is also a goal in-and-of itself, and an is not able to capture. Because of this, it is additional reason to have a local focus. possible that this study understates the full eco- nomic impact of nature-based tourism around One of the key findings of the study is that the the two parks. economic return per kwacha of government spending in protected areas is significantly Interestingly, government revenues from tourism greater than 1 to 1: about 16.7 kwacha at South in the two protected areas – gathered through Luangwa National Park and about 28.2 kwa- park visitor fees, non-consumptive fees, safari cha at Lower Zambezi National Park. Public hunting fees, outfitter license fees, animal fees, investment in protected areas not only helps and hunting fees – exceed current investments to conserve biodiversity, it also helps to make in the park, generating a net of 16.17 million protected areas more attractive to tourists kwacha (US$1.34 million) for the GoZ Treasury. – for example, by securing wildlife through Biodiversity in these two parks is therefore a investments in anti-poaching measures or by source of revenue for the government and not providing well maintained safari trails. When a financial burden. This may not be the case for tourists visit protected areas, they not only pay every protected area, as not all protected areas park entry fees, but also for lodging, meals, will attract tourists, even when the supporting transport, souvenirs, and other tourism services. infrastructure – roads, air strips, lodging etc., Co n clusi on s a n d Pol ic y Re co mm e n dati on s These expenditures directly benefit the tourism –are available. Surplus revenues from protect- sector, but the benefits do not stop there. ed areas that do attract tourists can be used Tourism service providers hire labor and source to subsidize investments in other parts of the goods and services from the local economy, protected area network. and trigger a chain of benefits for local busi- Another key finding of the study is that ex- nesses and households that are not directly penditures by tourists visiting protected areas connected with the tourism sector. It is the sum generate significant income multipliers for of these direct and indirect benefits that result households in the local economy, benefiting in the high economic returns per kwacha of households directly involved in the tourism investment by the government. Investment in sector and those not, and benefitting both poor protected areas is therefore good for biodiver- and non-poor households. The study estimates sity conservation and for the development of that an additional kwacha spent by visitors at the local economy. Lower Zambezi National Park raises household It is important to note that these estimates incomes around the park by 1.82 kwacha and of economic return are conservative. Firstly, around South Luangwa National Park by 1.53 only benefits to the local economy have been kwacha. Tourists spend money at local retail estimated. Tourists who visit protected areas stores, on local services, and on local transport, also spend money outside the local economy generating incomes for households in the local – for example, while traveling to the protected economy. These transactions establish a direct area – and tourism businesses are likely to link between tourists and the local econo- source goods and services from outside the my but are only a part of the local economic 50 ASSESSING THE ECONOMI C IMPAC T OF TOURISM IN PROTE CTED AREAS ON LO CA L E CONOMIES IN Z AMBIA benefits generated. Furthermore, the bulk of Zambezi and 635 million kwacha (US$45.4 tourist spending goes towards airfares, commis- million) around South Luangwa National Park. sions to booking agents, and expenditures at Because the base run of the LEWIE model lodges, further limiting the benefit transmission includes damages from human-wildlife con- through the direct link. Households addition- flict, the estimated economic impact of tourism ally benefit indirectly through production and presented here already is net of these losses. income linkages, when tourism operators hire Households that incur losses from wild animals local households and source local goods, and may or may not be the same households which when households spend wages and business- benefit from tourism in the protected area. es spend profits earned through the tourism Particularly for households that do not benefit sector. This implies that not only households directly or indirectly from tourism, there may directly engaged in the tourism sector, but also be a need to compensate for losses incurred other households benefit. Moreover, both poor from animal encroachments. Currently, no such and non-poor households benefit. An addi- mechanism is in place. tional kwacha of spending by visitors at Lower In summary, the analysis finds that the two study Zambezi raises the real income of poor house- sites are important tourist attractions which holds within the GMA and local market town by protect biodiversity and support their local 0.99 kwacha, and those of non-poor house- economies, providing jobs for poor and non- holds by 0.83 kwacha. At South Luangwa, the poor households, and for those directly involved real income multipliers for poor and non-poor in the tourism sector, and those not. households are 1.34 and 0.19, respectively. With over 40 percent of Zambia’s land area Studies that look only at tourism expenditures to under some form of protection, including 20 estimate impacts will underestimate impacts on national parks, there is even greater potential the local economy, and over emphasize leakage for protected areas to contribute to develop- from tourism activities outside the local economy. ment goals while maintaining the country’s rich It is critical to consider both direct and indirect biodiversity asset base. For this vision to be mechanisms to assess the economic impact of realized, however, protected area manage- tourism in protected areas on the local economy. ment challenges must be addressed, tourism in Tourism generates a significant number of jobs, protected areas promoted and diversified, and directly and indirectly. The study estimates that benefits shared with local communities fairly. national park tourism generates 7,463 full-time Protecting natural assets, growing and diversi- equivalent jobs around Lower Zambezi National fying the tourism business, and sharing benefits Park and 28,210 jobs around South Luangwa with local communities form the basis of pursu- National Park, equivalent to 14 percent and ing development and biodiversity conservation 30 percent of the populations around the two goals together. parks, respectively. PROTECT NATURAL ASSETS The study also provides an estimate of the significant negative impact of human-wildlife To promote biodiversity conservation and ConclusionsandPolicyRecommendations conflict on the local economy through crop secure the natural assets which attract visitors, losses. Animal incursions on to farms reduced it is critical that protected areas be protected, crop output by 11–14 percent around the two enhanced to reverse degradation, and generally parks, as per the information gathered through well managed. This requires addressing the the household surveys. The direct impact from underlying factors that are contributing to poor crop damage, together with the indirect impacts performance of Zambia’s protected areas. The through production and income linkages, following are identified in this report: amount to income losses in the local economy Increase public investment in protected area of around 23.7 million kwacha (US$1.8 million) at management: As indicated in this study, public Lower Zambezi and 16.3 million kwacha (US$1.2 funding of protected areas results in a high re- million) at South Luangwa. These are significant turn on investment. Using public funds for park losses, and they do not consider losses from management is especially important, as well human injury, including mortality, to which this managed parks attracts tourists, underpinning study does not attempt to assign economic a sustainable tourism industry and maintaining value. The magnitude of economic losses from livelihoods and benefits for local communi- crop loss is much less than the total impact of ties. Existing investments in natural resource tourism on income around the two parks – 212 management, such as those supported by the million kwacha (US$15.1 million) around Lower World Bank projects “Transforming Landscapes ASSESSING T H E E C ONOMI C IMPACT OF TOURISM IN PROTE CTED AREAS ON LO CA L E CONOMIES IN Z AMBIA 51 for Resilience and Development Project” location. Some relevant competencies include: (P164764) and “Zambia Integrated Forest understanding the legal framework that applies Landscapes Project” (P161490), aim to improve to operators, how to develop contracts or other livelihoods, land rights, ecosystem services, and authorizing instruments and solicit bids if appli- sustainability. While these projects and other cable, how to monitor and evaluate operators, government investments lay the groundwork skills related to data collection and analysis, for a well-supported nature-based tourism business acumen, negotiation skills, and asset sector, further resources are needed to attract management training, if government facilities tourists, for example, through improved access are to be assigned to an operator. As recom- and connection with local communities. Greater mended by the World Bank (2007), standards investment is also needed to increase the num- and curricula are needed to grow the skills ber of wildlife rangers and scouts, to strengthen required by a nature-based tourism industry, ranger capacity, and to upgrade infrastructure and on-the-job training in commercial services and equipment. DNPW is currently working at programs support this goal. one-fourth the capacity needed to properly Undertake regular Visitor Spending Effects manage parks, and the technology being used Assessments at the national level: This study for research and monitoring is outdated. Public presented a methodology to assess the eco- investment in protected area management can nomic impacts of tourism in protected areas on ensure that areas continue to attract tourists local economies at two study sites. To make the without degrading Zambia’s rich biodiversity, case for regular allocation of public resources, and support its continuation, so that protected and to support planning and program design area management and tourism are mutually to identify, for example, where tourism ser- beneficial, and self-reinforcing. Retaining reve- vices can be improved, it is important that such nue for resource protection and operations can assessments be conducted by the government also ensure the sustainability of tourism, so that regularly, and at the national level. This will protected area management benefits from the require first and foremost that data on tourists, industry it supports. tourism businesses, the local economy, and Build capacity of protected area managers: park management be collected systematically. Protected area managers require the right Therefore, a complementary recommenda- blend of education, training and expertise. In tion is to: Implement regular visitor surveys for particular, to manage commercial and business monitoring and evaluation. A key challenge for operations within and around protected areas this study was the lack of available tourist infor- requires that staff are well versed in the laws mation. Carrying out visitor surveys is crucial to and policies of the protected area, understand understand the impacts of tourism and how they the business needs of tourism operators, and may change over time. The number of visitors Co n clusi on s a n d Pol ic y Re co mm e n dati on s can manage commercial entities to reflect pro- to each park, and their spending habits are im- tected area values. The Zambia Tourism Master portant for informing policy plans. Strengthening Plan 2018–2038 identifies lack of tourism and data collection has also been previously identi- business acumen among managers tasked with fied as a critical input for national tourism plans tourism development in protected areas as a in Zambia (World Bank, 2007). Visitor surveys cause for concern. While required skills will vary would ideally be deployed on a rolling basis to depending on the protected area, education, capture seasonal trends in tourism activities, experience, and training in certain fields are and administered at the end of a visitor’s trip. needed for commercial services regardless of 52 ASSESSING THE ECONOMI C IMPAC T OF TOURISM IN PROTE CTED AREAS ON LO CA L E CONOMIES IN Z AMBIA GROW AND DIVERSIFY THE TOURISM BUSINESS additionally support the industry’s growth while Zambia has been attracting increasing num- building local ownership of protected areas. bers of tourists to its protected areas; however, Promoting local tourism improves the social the country lags behind some of its neighbors equity of nature-based tourism, and encour- in terms of numbers of visitors. Based on the ages Zambians to enjoy the natural beauty of World Economic Forum (WEF) 2019 Travel and their country, while preferential pricing supports Tourism Competitiveness ranking, Zambia, with such local tourism and keeps leisure spending a score of 3.2 out of a maximum of 7, ranked low by Zambian professionals within the country. overall: 113 out of 140 countries (WEF 2019); see This strategy can also buffer the industry from Figure 10 below. Categories in which Zambia future shocks—a clear lesson learned from the scored poorly, and significantly below the global Covid-19 pandemic. average included infrastructure (air and land transport, and tourist services), health and hy- Growing tourism beyond the five parks that are giene services, and international openness. On currently visited by tourists will require planners the other hand, Zambia ranked 41st among 140 to assess the tourism potential of Zambia’s countries for its natural resource assets. protected area network and to prioritize sites that could be developed to diversify Zambia’s Many of Zambia’s natural assets are inac- tourism offering. cessible to tourists because of poor road infrastructure. Diversifying tourism through Another intervention to promote tourism in pro- increased connectivity not only allows the tected areas relates to the concessions policy. sector to develop in new protected areas, but Four factors are critical for the development of also combats over-reliance on the few natural a strong commercial services/concessions pro- assets that currently attract tourists. The need gram in any country: strong protected area laws for investment in critical infrastructure, echoed and regulations, public support for commercial from World Bank (2007) and (Sichilongo et activity in parks, demonstrated economic bene- al. 2012), goes beyond roads, and applies to fits, and systems to evaluate the implementation air, telecommunications, and power supply. of the laws and regulations on a continuous Updating policies related to the tax regime and basis and modify them when necessary. As visa requirements are also needed. Policies per information gathered from government that increase domestic investment in tourism officials, private sector tourism operators, and Figure 10. Zambia Travel and Tourism Competitiveness Index Profile Performance Overview Key International Openness Price competitiveness 2.9 5.1 93rd 90th Prioritization of Travel & Tourism Environmental sustainability ConclusionsandPolicyRecommendations 3.9 4.4 108th 53rd ICT readiness Air transport infrastructure 3.2 1.8 117th 124th Human resources Ground & port & labour market infrastructure 3.8 2.4 119th 117th Tourist service Health & hygiene infrastructure 2.6 2.5 132nd 117th Safety & Natural security resources 5.3 3.6 3.2 81st 41st Cultural Business resources & environment business travel 4.4 69th Overall Score 110th 1.3 7 6 5 4 3 2 1 Score 1-7 (best) 1 2 3 4 5 6 7 ASSESSING T H E E C ONOMI C IMPACT OF TOURISM IN PROTE CTED AREAS ON LO CA L E CONOMIES IN Z AMBIA 53 non-governmental organizations, some of these 1. Establish completion dates for all actions in conditions are met in Zambia, but not all (see the Tourism Master Plan. Priority should be Box 4 for detailed analysis). given to drafting regulations, policies and then procedures. The Tourism Master Plan provides three main 2. Ensure tourism concession regulations, recommendations to reform tourism conces- policies, and procedures are drafted in sions in Zambia’s protected areas: establish accordance with the law and global best a tourism specialization in DNPW and cultural practices including from neighbouring agencies, prioritize tourism development, nations. Pursue a streamlined business incentivize DNPW to take a more commercial environment for these, the absence of which approach to managing parks and protected has stifled previous tourism business initia- areas, plough back revenues directly into its tives in Zambia (Sichilongo et al. 2012). Seek core conservation mandate, and establish a input into the regulations and policies from transparent and consistent concessions policy community groups, the private sector, tourism for nature tourism areas. Additional actions to businesses, citizens, and other stakeholders. further strengthen Zambia’s nature-based tour- A needed policy adjustment is to lengthen ism sector include: concessions agreements to improve the sus- tainability of private sector business models. Box 4: Assessment of Factors Needed for a Strong Concessions 3. DNPW should develop a strategic plan Program specifically for tourism concessions. This plan Strong protected area foundational laws and regulations: ZWA serves should enable the organization to identify as a foundational law; however, NGOs and private actors were unfamiliar goals, objectives, and specific tasks, then set with the regulations and processes for soliciting concessions. According to priorities for action. It will also open dialogue those interviewed, there are no procedures to award commercial service/ on partnerships between the government, concession contracts in a competitive, transparent, fair and easily under- community organizations, private entities, and stood manner, criteria to evaluate proposals are not publicly available, contracts are too short for a fair return on investment, and new concessions NGOs to attract further capital.27 are awarded on an ad hoc basis. Interviewees also stated that contracts are “generally renewable” but that poor park management is affecting feasibility, SHARE BENEFITS WITH LOCAL COMMUNITIES and that reasonable opportunities for a profit by the concessioner are not As noted, development of local communities considered. Similar concerns were raised for the Zambia Tourism Master Plan 2018–20381, which noted that there is no clear concessions system in the is a goal in itself. Ensuring that benefits from country and that land allocations in national parks and GMAs are managed protected area tourism are shared with local on an ad hoc basis. communities helps further this goal. Moreover, Public support for commercial activity in park areas: Interviewees believe when local communities benefit from tourism to that there is much public support for protected areas, and for commercial protected areas, they are incentivized to support activities in these areas if such activities are compatible with the protection conservation efforts and discourage encroach- of wildlife; and regulations which require operators/concessioners to hire ment, poaching, and other activities that lead to Co n clusi on s a n d Pol ic y Re co mm e n dati on s local people bolster this public support. It is not clear, however, whether any provisions governing the award of contracts give preference to local the degradation of protected areas. Furthermore, entrepreneurs. engagement with local communities can provide Demonstrated economic benefit: There was general concern that if protect- a unique tourist experience beyond typical ed areas are not better managed, that the economic stability of surrounding protected area-focused activities. Strengthening communities will falter and that this would lead to more poaching, thus tourist interactions with communities through reducing tourism. Interviewees expressed a high level of concern over the homestays, traditional cultural exhibitions, and need for resource protection, and to ensure that protected area revenues other learning experiences further develops eco- benefit local communities. nomic opportunities and buy-in from communities Evaluate the implementation of the law and regulations on a continuous living near protected areas. basis and modify the regulations when necessary: The DNPW has not evaluated, or updated the laws and regulations governing concessions, even Zambia’s protected area regulations mandate though the Tourism Master Plan suggests steps to improve the regulatory sharing of revenues with CRB in GMAs. These process. revenues can be substantial. Communities Note: The information on Zambian concession operations comes from brief conver- sations with DNPW officials, interviews with private individuals and NGOs operating in have, however, expressed concern that they do Zambian parks, and internet research. not always receive their share of the revenues in a timely manner. Previous studies of CRBs in Zambia’s GMAs have also found that non- 1 Zambia Tourism Master Plan, 2018–2038, Final Report, March 2018, accessed via internet August 18, 2020: https://www.mota.gov.zm/wp-content/uploads/2020/03/ poor households benefit more from revenue Zambia-Tourism-Master-Plan.pdf sharing than poor households (World Bank, 27 Enhancing PPPs as a strategy for Zambia’s protected areas is also recommended by World Bank (2007), Simasiku et al. (2008), and Sichilongo et al. (2012). 54 ASSESSING THE ECONOMI C IMPAC T OF TOURISM IN PROTE CTED AREAS ON LO CA L E CONOMIES IN Z AMBIA 2007). Greater transparency in the transfer Labor Department, and the US Army. During of funds would help to alleviate these con- the CCC’s nine-year operation, approximately cerns. Revenue sharing with communities can 5 percent of the US male workforce (about be expanded where it is currently limited, for three million people) was employed through example, in situations in which hunting revenue the program. Beyond the temporary creation is shared, but not photography or lodge-relat- of jobs, the CCC is credited with increasing ed revenues. Furthermore, while the income visitors to national and state parks from 3.2 multiplier for local households from visitor to 20.4 million over its nine years. Today, the spending in protected areas is significant, there parks system attracts over 320 million visitors are opportunities to raise this multiplier through each year, benefitting local gateway regions government policies and programs. Table 18 with estimated spending of over US$21 billion, summarizes these opportunities. supporting more than 340,500 jobs, and gener- ating US$41.7 billion in economic output (World Finally, it is critical to reflect on the costs of Bank, Environment, Natural Resources & Blue the COVID-19 pandemic for local households Economy 2020). and businesses, and the role that protected areas can play in an economic recovery. The In the wake of the COVID-19 pandemic, Zambia study shows that the pandemic has resulted could benefit from a CCC-like scheme to renew in substantial losses in tourism and tourism the nature-based tourism sector in a way that income: 17.6 million kwacha (US$1.26 million) of maximizes benefits to park-adjacent communi- real GDP loss per month without tourist revenue ties. As the global economy re-opens, Zambia’s at Lower Zambezi, and 53.0 million kwacha tourism industry can empower small- and me- (US$3.78 million) of income loss per month at dium-sized firms through concessions policies South Luangwa. At the same time, investment in in and around natural areas, and encourage protected areas can serve as a mechanism for spending in local communities. It can empow- green economic recovery. er citizens to gain entry to the nature-based tourism industry through training and support, Large-scale investments in protected areas as also recommended by Sichilongo et al. can create jobs and boost economic recov- (2012). Additionally, supporting local businesses ery and resilience. In the United States, the through loans, fast-track financing, or technical Civilian Conservation Corps (CCC) was estab- assistance to diversify and use digital technolo- lished during the Great Depression, creating gies can contribute towards business continuity. jobs, infrastructure, and businesses which Jobs created, for example, to improve accessi- exist to this day. The initiative was created bility (e.g., road network improvements), patrol through a government-wide partnership which protected areas, and improve park infrastructure brought together the Forest Service (under the can grow tourism and create a sustainable Department of Agriculture), the National Parks incomes for households in surrounding areas. Service (under the Department of Interior), the ConclusionsandPolicyRecommendations ASSESSING T H E E C ONOMI C IMPACT OF TOURISM IN PROTE CTED AREAS ON LO CA L E CONOMIES IN Z AMBIA 55 Table 18: Opportunities to Strengthen the Income Multiplier for Local Households Impact Avenue Opportunities to deepen the linkage Direct Tourist spending at local Explore opportunities for tourists to interact directly businesses with local communities. This will involve a combination of investment in transport infrastructure so that tourists pass by local towns and villages on their way to parks, and strengthening the capacity of local communities to provide goods and services to tourists. As recommended by World Bank (2007), establishing market niches and diversifying tourism offerings can strengthen the tourism industry and its supply chains. Restrictions on resource While restrictions on resource use by communities extraction and positive within protected areas will have to be maintained, spillovers from Parks to GMAs positive spillovers from protected areas to GMAs – e.g., when wild animals move from a protected area into a GMA due to growing herd size – will increase the benefits to the local economy. Improved protected area management with a focus on enforcement will be critical to increase positive spillovers. Impact of human-wildlife As noted, animal incursions lead to significant loss of conflict income for households near protected areas and within GMAs. Mitigating losses through investment in local level strategies (seasonal fences, livestock corrals, etc.) or establishing mechanisms to compensate households for their losses can promote coexistence. Indirect – Hiring of local labor by tourism Employment in the tourism industry is an important production establishments, CRBs, and source of income for households in the local economy. linkages park managers Opportunities to further increase employment occur both through increasing demand by growing the tourism industry, and increasing supply by building human capital. This study shows that when local people are hired to work in the park as guards, guides, game wardens etc., that the impact on the local economy is greater than the cost to the state. In fact, an additional kwacha spent by the government on park wages creates higher income multipliers (3.02 in Lower Zambezi and 3.1 in South Luangwa) than those for tourist Co n clusi on s a n d Pol ic y Re co mm e n dati on s spending (1.57 in Lower Zambezi and 1.42 in South Luangwa). Sourcing of local goods by Greater sourcing of goods and services by tourism tourism establishments, CRBs, establishments is another potential avenue for higher and park managers impact on the local economy. World Bank (2007) highlights that national-level leakages from tourism occurring through imports and profit repatriations will not decrease unless local participation in the sector is improved. 56 BANKING ON PROTE C TED AREAS ANNEX 1 Detailed Expenditures from DNPW Estimates for South Luangwa and Lower Zambezi Expenditures South Luangwa Lower Zambezi Total   ZMK US$ ZMK US$ ZMK US$ GoZ Payment to CRBs         20,724,533 1,726,146 Wildlife Police and Game 12,970,184 1,080,287 8,416,489 701,009 21,386,673 1,781,296 Wardens Maintenance Workers 2,797,241 232,982 689,324 57,414 3,486,565 290,396 Office Workers 1,504,427 125,304 1,142,182 95,132 2,646,609 220,436 Other Workers     93,603 7,796 93,603 7,796 Wage Expenditures 17,271,852 1,438,573 10,341,598 861,352 27,613,450 2,299,924 General Administration 200,180 16,673 108,690 9,053 308,870 25,726 Financial Management and 150,000 12,494 50,000 4,165 200,000 16,658 Accounting Tourism Sector Development 70,000 5,830 40,000 3,332 110,000 9,162 Programs Community Based Wildlife 276,250 23,009 239,290 19,930 515,540 42,939 Management Wildlife Conservation and 954,490 79,499 90,170 7,510 1,044,660 87,010 Management Road Maintenance 416,160 34,662   0 416,160 34,662 Non-wage expenditures 2,067,080 172,167 528,150 43,990 2,595,230 216,157 Total Expenditure         50,933,213 4,242,227 Source: DNPW, Government of Zambia A N N EX ES 57 ANNEX 2 Data Collection Methodology Household and Local Business Survey the number of households in the village). At this time, the headman and guides were request- The Lower Zambezi survey covered all of the ed to convey the following information to the Chiawa GMA under the Chiawa Chiefdom which selected households: 1) the purpose of the study surrounds the main entrance to the park. At and length of the survey; 2) that participation in South Luangwa National Park, the large area the study was voluntary; 3) that not all members covered by the combined Upper Lupande and of the household were needed at the time of Lower Lupande sites made it unfeasible to the survey; and 4) that the data are confiden- collect data from all chiefdoms. Thus, two chief- tial and to be used solely for the purposes of doms were selected to comprise the survey site this study. Once households consented to the in South Luangwa: Senior Chief Nsefu’s chief- interview, they were confirmed on the list; if they dom in Upper Lupande, and Chief Kakumbi’s declined, a nearest neighbor household was chiefdom in Lower Lupande. approached as a replacement. An estimated 85 Four visits were made to seek permission, to percent of households approached agreed to identify villages, to randomly select house- be surveyed. In a few cases in which the head- holds, and to conduct the survey. In the initial man had a roster of households in the village, visit, members of the team approached chief- the second and third visits were combined. tains from each chiefdom/township to brief On the final visit, enumerators, assisted by local them on the purpose of the study, to gain their guides, approached households for interviews. blessing for the research, and to request a list One village (or one village cluster) was visited of villages and the number of households in each day of the survey. The order in which each village under their administration.28 This villages were surveyed depended largely on the list was subsequently collected by the team. availability of the headman or his deputy. An aver- Villages consisting of fewer than 40 households age of 45 households were surveyed each day. were either combined with nearby villages, or dropped if that option was unavailable due to lo- The household survey included a module de- gistical constraints. Villages were then randomly signed to gather information about businesses, selected from each site (Chiawa GMA, Nsefu and this was administered to households with Chiefdom and Kakumbi Chiefdom). businesses. Other businesses in the villages and nearby market towns were surveyed to If a village was randomly selected for survey, the supplement the household business sample, next visit was to the headman (leader) for further and gathered the same information as for briefing, at which point a list of households in households. Lacking a master list of businesses, the selected villages was constructed with help all small businesses in each surveyed village from the headman and 2–3 locally-hired guides. were approached (villages typically had only A N N E XE S Permission to interview the villagers was also a few businesses). In market towns, an ev- obtained from the headman during this visit. ery-other-business approach was adopted for On the third visit, 30–50 households were surveying. As in the household surveys, own- randomly chosen for interviews (depending on er-operator participation in the business surveys was voluntary. 28 In practice chiefs did not know the exact number of households in each village and thus gave an estimate. 58 BANKING ON PROTE C TED AREAS Tourist Surveys Tourism Businesses Survey Information on tourists was collected through The key tourism activities are lodges inside questionnaires that Proflight Zambia graciously and outside the national parks, which provide made available to its passengers returning from visitors with accommodation, meals, game view- the two parks. This is the most comprehensive ing and hunting (in the consumptive use areas way to gather such data because most visitors of GMAs). Lodges, in turn, spend money on (i) fly to the two parks and over 90 percent of them taxes and concession fees to the GoZ, (ii) wages use Proflight as their carrier. It was not possible to workers from communities near the park, to gather information from visitors who drove other parts of Zambia, or abroad, (iii) food, crafts, to the parks and thus we do not know whether and services purchased locally, in other parts of their behavior and expenditures are similar to Zambia, or abroad, and (iv) maintenance, utili- those in the survey. However, it is unlikely that ties, and other costs of managing and running a their inclusion would alter the findings because lodge. Some also support local communities by they constitute such a small share of park visi- investing in specific “corporate responsibility” tors. The survey was conducted during the peak projects. season in October and November, and is thus representative of the majority of visitors, who visit parks during this period. A N N EX ES 59 ANNEX 3 Additional Data Summary Statistics Table A3.1 provides a breakdown of employment by sector Crops and Livestock at each site. The largest percentage of workers at Lower Zambezi are employed by hotels/restaurants/tour operators Owing to large difference in climate and growing conditions, (32 percent), followed by domestic work (21 percent), agri- households in Lower Zambezi and South Luangwa grow culture (15 percent), and construction (12 percent). At South different types of crops. Figure A3.1 displays the percentage Luangwa the largest shares are construction (29 percent), of households engaged in various types of crop cultivation. hotels/restaurants/tour operators (24 percent), and domestic Households at both sites are largely subsistence farm- work (22 percent). ers, with some farmers producing a surplus to sell in local markets. Maize is the main staple crop grown at both sites, with rice only produced in South Luangwa. More vegeta- bles, fruits and tubers are grown in Lower Zambezi; the two primary cash crops, cotton and tobacco, are grown in South Luangwa. Table A3.1. Employment by Sector Figure A3.1. Crop Types Peanuts Cotton/Tobacco Lower South 33 9 Zambezi Luangwa Domestic Work 21% 22% Fruits/Veg 82 Agriculture 15% 4% Lower Maize Store/Factory/Food Processing 4% 5% Zambezi 324 Construction 12% 29% Cereals/Pulses /Tubers Beauty/Transportation 3% 2% 35 School 0% 2% Government 7% 2% Cotton/Tobacco 129 NGOs 4% 2% Hotels/Restaurants/Tour Operators 32% 24% Peanuts A N N E XE S 97 Other Services 2% 10% South Maize Fruits/Veg Luangwa Sample Size (Households with Wages) 253 131 15 408 Source: World Bank survey Cereals /Pulses /Tubers 8 Rice 81 60 BANKING ON PROTE C TED AREAS Table A3.2 summarizes crop production at Lower Zambezi; in South Luangwa this figure is the two sites at the plot level. Harvested crop 15 percent (18 percent) for plots owned by poor values are low at both sites but substantially (non-poor) households. lower in Lower Zambezi, especially for poorer Table A3.3 uses data on South Luangwa, and households. A severe drought in 2019 caused a subset of Lower Zambezi farmers who were widespread crop failure at both sites, especially able to obtain some harvest despite the drought Lower Zambezi. The majority of farmers in Lower to summarize harvest use. Over half of farmers Zambezi experienced large-scale crop failure; who managed some harvest in Lower Zambezi poor households were able to harvest only 8 sold a portion of their crops (58 percent for percent of their plots, and non-poor households poor and 65 percent for non-poor), compared were only able to harvest 15 percent of their to around 40 percent of both poor and non- plots (compared with 50 and 55 percent in poor households in South Luangwa.30 In Lower South Luangwa, respectively).29 Zambezi, just under half of all crops produced Average plot sizes are larger in Lower Zambezi, (by value) were consumed by households (46 2.71 and 2.62 acres compared to 0.62 and percent and 42 percent, for the poor and non- 0.40 acres in South Luangwa for the poor and poor, respectively), while households in South non-poor, respectively. Labor for agricultural Luangwa consumed a smaller percentage of activities is supplied primarily by the household their harvest at 38 percent and 36 percent for itself, though some households (primarily the poor and non-poor, respectively. Spoilage is non-poor) do hire outside labor. Overall inputs higher for households at Lower Zambezi (8 and are low, 21 percent (32 percent) of the plots 10 percent, for poor and non-poor households, owned by poor (non-poor) households in Lower respectively) when compared to South Luangwa Zambezi used pesticides, while this number is 14 (3–4 percent, for poor and non-poor house- percent (15 percent) for poor (non-poor) house- holds, respectively), while households in South holds in South Luangwa. Similarly, fertilizer use Luangwa have a higher percentage of their is low, 13 percent (28 percent) of plots owned by harvest being stored or given away. poor (non-poor) households applied fertilizer in Table A3.2. Crop Production and Inputs Average Average Family Hired Inputs % Plot Size Harvest Labor Labor Harvested (acres) Value days days Pesticides Fertilizer Lower Zambezi Mean 2.71 0.08 240.65 155.47 0.83 0.21 0.13 Poor N=208 SD (2.30) (0.27) (1089.8) (177.5) (5.0) (0.41) (0.34) Lower Zambezi Mean 2.62 0.15 1131.90 180.85 28.82 0.32 0.28 Non-poor N=171 SD (2.45) (0.35) (4133.2) (189.6) (115.7) (0.47) (0.45) South Luangwa Mean 0.62 0.55 1001.47 235.41 3.92 0.14 0.15 Poor N=575 SD (1.10) (0.50) (2,138.0) (198.9) (37.8) (0.34) (0.36) South Luangwa Mean 0.40 0.50 1153.65 196.91 15.82 0.15 0.18 Non-poor A N N EX ES N=117 SD (1.13) (0.50) (2460.6) (165.2) (74.1) (0.35) (0.39) Source: World Bank Survey Note: Information presented at the plot level 29 In the survey we asked households how much they actually harvested and how much they had expected to harvest from each plot under ideal conditions. A plot was considered to have experienced crop failure if the farmer failed to harvest more than a third of what was expected. 30 Crop sales happen primarily at local markets; 72 percent of crops sold in South Luangwa and 96 percent of crops sold in Lower Zambezi are sold in nearby town markets. 61 Table A3.3. Crop Use and Sales   Share of Share Share Share to Households Share of Share Damaged Sold to Spoilage Gifts and Selling Crop Sold Consumed by a Lodge Storage Crops Animals Lower Zambezi Poor Mean 0.58 0.11 N/A 0.46 0.08 0.03 0.18 N=129 SD (0.50) (0.28) - (0.44) (0.24) (0.11) (0.29) Lower Zambezi Mean 0.65 0.23 N/A 0.42 0.10 0.04 0.11 Non-poor N=139 SD (0.48) (0.33) - (0.47) (0.24) (0.11) (0.17) South Luangwa Poor Mean 0.40 0.21 0.04 0.38 0.03 0.19 0.11 N=342 SD (0.49) (0.35) (0.20) (0.35) (0.13) (0.25) (0.23) South Luangwa Mean 0.39 0.21 0.06 0.36 0.04 0.21 0.09 Non-poor N=120 SD (0.49) (0.36) (0.24) (0.35) (0.16) (0.27) (0.22) Source: World Bank Survey Figure A3.2. Livestock at Lower Zambezi (LZ) and South Luangwa The composition of livestock varies between (SL) the two sites (see Figure A3.2). Chickens are common at both sites; however South Luangwa lacks goats. Locals reflected that livestock Other Poultry predation was severe in South Luangwa, and Cattle 13 that this discouraged goat farming. Only a small 20 percentage of livestock-rearing households Ducks had cattle, due to Tsetse flies and the risk of 49 Trypanosomiasis to unvaccinated animals. Table A3.4 summarizes livestock values, sales Lower Goat/Sheep and purchases, and input use. Total livestock 163 value is substantially higher at Lower Zambezi, Zambezi a more livestock intensive region. Households in Lower Zambezi consumed 9 and 12 percent (in value, for poor and non-poor households, Chickens respectively) of their herds over the 12 month 138 period prior to the survey, compared with 14 and 22 percent (for poor and non-poor house- holds, respectively) in South Luangwa. A third of all households sold or purchased livestock Pigs during the year prior to the survey, and nearly all 24 Goat/Sheep livestock transactions took place within the GMA Other Poultry 22 16 or nearby market towns. Inputs for livestock production are low as most animals are free ranging. Cattle A N N E XE S 9 Ducks South Chickens 60 Luangwa 196 62 BANKING ON PROTE C TED AREAS Table A3.4. Livestock and Inputs Sales Purchase Input values (kwacha) Total Share Value Consumed Share Local Share Local Pens Vet Feed Selling % Buying % Lower Zambezi Poor Mean 2,809 0.09 0.50 0.80 0.31 0.98 58.42 26.78 3.55 N=115 SD (4,431) (0.16) (0.50) (0.40) (0.47) (0.1) (164.4) (118.5) (18.5) Lower Zambezi Mean 3,797 0.12 0.38 0.80 0.32 0.95 105.49 46.54 34.88 Non-poor N=116 SD (5,408) (0.19) (0.49) (0.40) (0.47) (0.23) (253.1) (138.2) (140.5) South Luangwa Poor Mean 954 0.14 0.36 0.99 0.28 1.00 21.75 14.66 37.78 N=194 SD (2,778) (0.24) (0.48) (0.06) (0.45) - (217.3) (75.1) (244.5) South Luangwa Mean 821 0.22 0.17 1.00 0.31 1.00 1.43 1.43 11.86 Non-poor N=35 SD (1,478) (0.32) (0.38) - (0.47) - (6.0) (8.5) (50.9) Source: World Bank Survey The business surveys (and business modules Luangwa.32 Businesses in South Luangwa tend in household surveys) asked the year in which to hold more assets (and inventory), with an businesses formed. Figure A3.3 provides a average asset value for South Luangwa busi- frequency distribution of business formation, nesses of 38,155 kwacha (US$2,935) compared which shows that most businesses started up to 19,369 kwacha (US$1,489) in Lower Zambezi. recently.31 Businesses in South Luangwa make a higher profit: 5,427 kwacha (US$417)/month, compared Most businesses in communities around the with 4,145 kwacha (US$318) in Lower Zambezi. two parks are family operated. Table A3.5 presents information about businesses around Table A3.6 summarizes local business costs the two parks. They pay wages: hired, as op- and whether they are met inside or outside of posed to family, laborers are paid an average the local economy. Business costs inside the of 445.8 kwacha (US$35) per month in Lower local economy potentially create income growth Zambezi and 570.8 (US$43) per month in South linkages for other businesses or households. A large percentage of businesses do not pay rent, either because they are street vendors Figure A3.3. New Business Formation or because they own the land on which their businesses operate. Among those who do pay 0.2 rent, the average monthly rental is 204 kwa- cha (US$17) at Lower Zambezi and 130 kwacha (US$11) at South Luangwa. Transport costs are 0.15 substantially higher in Lower Zambezi due to poor road conditions. The majority of input purchases are local. Most smaller businesses Frequency buy their inputs locally from larger traders who 0.1 source goods from outside the local economy. Using the total value of input purchases from all businesses in the sample, Table A3.7 presents 0.05 the share of goods and services purchased A N N EX ES from outside the local economy by business at the two sites. For retail type businesses (corner shops, traders, etc.), almost half of their inputs 0 1980 1990 2000 2010 2020 are sourced from outside the local economy, while the percentage of outside procurement is lower for service-type businesses (hotels, restaurants, repair shops, etc.). 31 An increasing number of recently-formed businesses may indicate high levels of business failure. Information on failed busi- nesses is not available. 32 As is often the case with household and business data, there are some differences in average wages paid from the two data sources. 63 Table A3.5. Business Operations Labor Asset Months Revenue Profit Value Operated Hired Monthly Family (kwacha) (kwacha) Workers Wage Workers (kwacha) Mean 9.5 0.58 445.8 1.96 19,369 14,035 4,145 Lower Zambezi (N=172 Businesses) SD (3.7) (1.8) (781.6) (2.14) (52,139) (43,748) (11,995) Mean 9.1 0.65 570.8 1.55 38,155 16,584 5,427 South Luangwa (N=181 Businesses) SD (3.9) (1.7) (662.5) (1.1) (168,540) (89,873) (27,546) Table A3.6. Business Costs Monthly Crop Monthly Livestock Monthly Services Monthly Retail Purchases Purchases Hired Goods Purchased Rent Transp. % % % % Kwacha Purchased Kwacha Purchased Kwacha Purchased Kwacha Purchased Outside Outside Outside Outside Lower Zambezi Mean 204 725 112 12 142 12 127 17 632 20 (N=172 Businesses) SD (797) (2,379) (417) (0.28) (1,042) (0.31) (468) (0.36) (4,176) (0.39) South Luangwa Mean 130 285 223 0.03 75 0.25 80 0.03 653 0.12 (N=181 Businesses) SD (421) (1,085) (1,705) (0.09) (521) (0.4) (225) (0.14) (3,608) (0.31) Table A3.7. Share of Inputs Purchased Outside the Local Economy Retail Businesses Service Businesses Lower Zambezi 46% 12% South Luangwa 47% 1% A N N E XE S References Chidakel, Alexander, Brian Child, and Shylock Muyengwa. 2021. “Evaluating the Economics of Park- Tourism from the Ground-up: Leakage, Multiplier Effects, and the Enabling Environment at South Luangwa National Park, Zambia.” Ecological Economics 182 (April): 106960. https:// doi.org/10.1016/j.ecolecon.2021.106960. Dixon, Peter, and Dale Jorgenson. 2012. “Handbook of Computable General Equilibrium Modeling.” Handbook of Computable General Equilibrium Modeling. Elsevier. https://econpapers. repec.org/bookchap/eeehacgem/1.htm. Dudley, N. 2008. “Guidelines for Applying Protected Area Management Categories.” Gland, Switzerland: IUCN. https://www.iucn.org/theme/protected-areas/about. Lindsey, Peter A., Jennifer R. B. Miller, Lisanne S. Petracca, Lauren Coad, Amy J. Dickman, Kathleen H. Fitzgerald, Michael V. Flyman, et al. 2018. “More than $1 Billion Needed Annually to Secure Africa’s Protected Areas with Lions.” Proceedings of the National Academy of Sciences 115 (45): E10788. https://doi.org/10.1073/pnas.1805048115. Lindsey, Peter A., Vincent R. Nyirenda, Jonathan I. Barnes, Matthew S. Becker, Rachel McRobb, Craig J. Tambling, W. Andrew Taylor, Frederick G. Watson, and Michael t’Sas-Rolfes. 2014. “Underperformance of African Protected Area Networks and the Case for New Conservation Models: Insights from Zambia.” PloS One 9 (5): e94109. https://doi.org/10.1371/ journal.pone.0094109. MoTA. 2016. “National Tourism Policy.” Ministry of Tourism and Arts, Republic of Zambia. https://www. mota.gov.zm/?wpfb_dl=70. ———. 2018. “Zambia Tourism Master Plan 2018-2038.” Government of the Republic of Zambia, Ministry of Tourism and Arts. https://www.mota.gov.zm/wp-content/uploads/2020/03/ Zambia-Tourism-Master-Plan.pdf. Parks Canada. 2019. “Parks Canada Agency - Development Plan.” Washington D.C. Sichilongo, Mwape, Pamela Mulozi, Biston Mbewe, Chomba Machala, and Jean-Michel Pavy. 2012. “Zambia Wildlife Sector Policy: Situation Analysis and Recommendations for a Future Policy.” Simasiku, Phyllis, Hopeson I. Simwanza, Gelson Tembo, Sushenjit Bandyopadhyay, and Jean- Michel Pavy. 2008. “The Impact of Wildlife Management Policies on Communities and Conservation in Game Management Areas in Zambia: Message to Policy Makers.” Natural Resources Consultative Forum. Singh, Inderjit, Lyn Squire, and John Strauss. 1986. “A Survey of Agricultural Household Models: Recent Findings and Policy Implications.” The World Bank Economic Review 1 (1): 149–79. Taylor, J. Edward, and Mateusz J. Filipski. 2014. Beyond Experiments in Development Economics: Local Economy-Wide Impact Evaluation. Oxford, New York: Oxford University Press. UNEP-WCMC and IUCN. 2020. “Protected Areas (WDPA).” Protected Planet. 2020. https://www. protectedplanet.net/en/thematic-areas/wdpa. US NPS. 2019. “National Park Visitor Spending Contributed $40 Billion to U.S. Economy - Office of Communications (U.S. National Park Service).” 2019. https://www.nps.gov/orgs/1207/nation- al-park-visitor-spending-contributed-40-billion-to-u-s-economy.htm. WEF. 2019. “Country Profiles: Travel & Tourism Competitiveness Index2019 Edition.” Travel and Tourism Competitiveness Report 2019 (blog). 2019. https://reports.weforum.org/ travel-and-tourism-competitiveness-report-2019/country-profiles/#economy=ZMB. World Bank. 2007. “Zambia: Economic and Poverty Impact of Nature-Based Tourism.” Washington, D.C.: World Bank. https://openknowledge.worldbank.org/handle/10986/7553. ———. 2019. “Country Forest Note: Zambia. Towards a Sustainable Way of Managing Forest.” Lusaka: World Bank. http://documents1.worldbank.org/curated/en/571651580133910005/ pdf/Zambia-Country-Forest-Note-Towards-a-Sustainable-Way-of-Managing-Forest.pdf. World Bank Environment, Natural Resources & Blue Economy. 2020. “Transforming Landscapes, Creating Jobs and Laying the Foundation for Economic Growth in the Long Run - Learning from the Civilian Conservation Corps: A Green Economic Recovery Initiative.” Washington, D.C.: World Bank. WTTC. 2019. “Zambia 2020 Annual Research: Key Highlights.” World Travel and Tourism council. https://wttc.org/Research/Economic-Impact.