This is an Accepted Manuscript of the article de Lima, L. S., Barón, P. A. R., Villamayor-Tomas, S., & Krueger, T. (2019). Will PES Schemes Survive in the Long-term Without Evidence of Their Effectiveness? Exploring Four Water-related Cases in Colombia. Ecological Economics, 156, 211-223, available online at https://doi.org/10.1016/j.ecolecon.2018.09.005 ©2018. This manuscript version is made available under the CC-BY-NC-ND 4.0 license. 1 Will PES Schemes Survive in the Long-term Without Evidence of Their Effectiveness? 2 Exploring Four Water-related Cases in Colombia 3 Letícia Santos de Lima; Pablo Andres Ramos; Sergio Villamayor-Tomas; Tobias Krueger 4 Accepted version for Ecological Economics 5 6 ABSTRACT 7 This paper explores the reasons why payers, intermediaries, and providers engage in PES 8 even under uncertainty about outcomes, and how they relate to the long term durability of 9 the scheme. In theory, it is expected that, in face of high uncertainty, payers would leave the 10 projects if effectiveness cannot be demonstrated and providers would not keep their 11 conservation practices if no money remains available. Consequently, it is also expected that 12 PES proponents would do their best to demonstrate ES improvement/maintenance. To 13 explore these hypotheses we use field data collected from PES schemes in Colombia. Our 14 results show that payers have additional motivations for engaging beyond ES improvement 15 (e.g., CSR, green image). These motivations may explain their permanence in the scheme 16 even without evidence of effectiveness. Water quantity/quality concerns were the main 17 driver for participation of providers, evidencing that they do not see themselves only as 18 providers but also as users. Therefore, the lack of evidence of effectiveness could discourage 19 their permanence. Intermediaries are the ones mostly concerned about presenting evidence 20 of PES effectiveness for many reasons (e.g. reputation, engaging stakeholders). PES may 21 survive in the long term due to additional motivations from stakeholders, however, evidence 22 of effectiveness is still expected. 23 1. INTRODUCTION 24 The Ecosystem Services (ES) concept has gained strong rhetorical power (Kull et al., 25 2015) and is shaping practices among conservationists (Fisher and Brown, 2014). Non- 26 government organizations and development agencies are increasingly working with 27 instruments derived from the ES approach such as Payments for Ecosystem Services (PES) 28 schemes (Ezzine-De-Blas et al., 2016; Grima et al., 2016). 29 Funding from private and public companies has been strategic for several 30 conservation schemes in Latin America. In order to communicate better with the business 31 sector so as to attract more funding conservation organizations have been adapting their 32 discourses towards a more business-like language with the utilitarian arguments of the ES 33 approach (Fisher and Brown, 2014). In doing so, several of them have ended up using 34 business jargons, for example, investment portfolio, business case, performance indicator, 35 return-on-investment (e.g. Boyd et al., 2012; Hanson et al., 2012, 2011). The growing use of 36 business-like language on ES approaches is frequently supported by the use of predictive 37 numbers derived from ES modelling (e.g. Crossman et al., 2013; Quintero et al., 2009; 38 Ruckelshaus et al., 2015; Tallis and Polasky, 2009). Practitioners have also placed strong 39 emphasis on cost comparison of conservation strategies versus conventional solutions for 40 the same ends (e.g. ‘green infrastructure’ versus ‘gray infrastructure’, see Bennett and Ruef, 41 2016; Calvache et al., 2012; Postel and Thompson, 2005). Finally, they have given priority to 42 PES over other policy tools expecting to achieve more effective ES conservation. Along these 43 lines, there has been a recent increase in the interest of some PES payers to adopt 44 performance-based payments and thus the need for PES impact evidence (Gammie, 2016). 45 The idea of securing ES provision through payment schemes has been strongly 46 promoted by some scholars (e.g. Wunder, 2005; Wunder et al. 2008; Ferraro and Kiss, 2002; 47 Ferraro 2011) and propagated in practical discourses towards potential payers. However, 48 this approach to conservation has been deeply criticized in the literature because of the 49 implied simplification of social-ecological systems (Kosoy and Corbera, 2010; Muradian et al., 50 2010; Norgaard, 2010) and the practical difficulty to prove ‘service delivery’ (Carpenter et al., 51 2009; Lele, 2009; Palmer and Filoso, 2009; Ponette-González et al., 2014). This way of 52 approaching and practicing conservation implies a control over a service being sold which 53 may not actually be secured, especially in the case of water-related PES schemes (Barnaud 54 and Antona, 2014; Kosoy and Corbera, 2010). Examples of targeted services in these cases 55 are water flow regulation, water quality maintenance, and water provision. So far, 56 incomplete knowledge about the processes behind water-related services together with a 57 lack of monitoring best practices in most cases have hindered the evidence of PES impacts 58 (Bohensky and Lynam, 2005; Carpenter et al., 2009; Lele, 2009; Norgaard, 2010; Palmer and 59 Filoso, 2009). 60 Discourses based strongly on economic aspects disregarding scientific uncertainty 61 underlying ES provision can lead to unrealistic expectations from actors involved in these 62 schemes. If expectations are not fulfilled those schemes may not endure. Under this 63 perspective, a question remains empirically under-explored: will PES schemes survive in the 64 long-term if no evidence of their impact on the target ES is achieved? Which motivations 65 would maintain the long-term participation of providers, payers and intermediaries in the 66 schemes, even under uncertainty? If uncertainty is large, practitioners cannot prove PES 67 effectiveness through impact evidence, and payers are only looking for returns in terms of 68 ES, then they would leave the PES scheme. This hypothesis would align with conventional 69 views of PES (e.g. Ferraro and Kiss, 2002). As pointed by Wunder (2005, p. 3), “the less 70 realistic the scientific basis of a PES scheme, the more exposed it is to the risk of buyers 71 questioning its rationale and abandoning payments”. Consequently, assuming that 72 landowners are acting solely as “providers” of a service based on a rational and utility- 73 maximizer perspective, they would abandon the scheme too as no money is left to pay 74 them. By the same token, PES intermediaries, for instance, non-government organizations 75 (NGOs), would lose credibility due to their “unfulfilled promises” of ES delivery and would 76 have their reputation at risk (Fisher and Brown, 2014). This hypothesis implies that ES 77 monitoring, evaluation and reporting are strongly needed to produce the right evidence to 78 keep payers and providers on board (e.g. Naeem et al., 2015). 79 However, if additional motivations, perceptions and values are in place, then the 80 assumptions made in theory regarding expectations from participants and the importance of 81 evidence of ES improvement may be questioned. Preliminary evidence suggests that 82 providers and payers may indeed be motivated by a plurality of reasons. Recent studies have 83 proposed that intrinsic motivations, for example, a desire to “care for the land” (Méndez- 84 López et al., 2015, p. 695), or the “warm-glow effect” (Andreoni, 1990), can play a strong 85 role on conservation schemes engagement by land owners (Ezzine-de-blas et al., 2015; Kits 86 et al., 2014; Kosoy et al., 2008; Zanella et al., 2014). In the same way, a mixture of 87 motivations, for example, green marketing or to maintain reputational capital, may be in 88 place when corporate leaders decide to invest in environmental programs (Babiak and 89 Trendafilova, 2011; Chin et al., 2013; Ditlev-simonsen and Midttun, 2011; Hemingway and 90 Maclagan, 2004). Further disentangling the motivations of providers and payers, as well as 91 intermediaries can not only shed light on alignments and misalignments across each group 92 of actors, but most importantly contribute to a better understanding of the long-term 93 sustainability of PES schemes, even in the presence of high uncertainty about their impact. 94 In this paper, we explore motivations and expectations of actors involved in PES 95 schemes in order to assess the importance of impact evidence in the scheme long-term 96 durability. We compare the perspectives of three groups involved in four water-related PES 97 schemes in Colombia: intermediaries, landowners (providers), and major donors (payers). 98 We focus our questions on whether additional motivations from payers and providers would 99 play a role in their engagement and permanence in the scheme; if payers demand evidence 100 of PES impacts on ES provision/maintenance; and if intermediaries feel in the need to 101 provide evidence of achievements of ES goals through monitoring and reporting. 102 103 104 2. METHODS 105 Study sites 106 This study was conducted in four water-related PES schemes (here also called payments for 107 watershed services - PWS) in Colombia: 108 1) “Agua Somos” (Chisacá and Mugroso river watersheds - tributaries of Tunjuelo river; 109 located in Usme, Bogotá D.C.); 110 2) Asociación de Usuarios de Agua del Río Bolo – “ASOBOLO” (Aguaclara river 111 watershed - tributary of Bolo River; located in Pradera and Palmira municipalities, 112 Valle del Cauca Department); 113 3) Compensación por Servicios Ambientales Hidricos Cuenca del Río Cali - “CSAH Cali” 114 (Felidia and Pichindé rivers watersheds – tributaries of Cali River; located in Cali 115 municipality, Valle del Cauca Department); 116 4) “Cuenca Verde” (Chico river watershed - tributary of Riogrande River; located in 117 Belmira municipality, Antioquia) (Figure 1). 118 The selection of schemes was mainly based on data and literature accessibility, 119 availability and openness of key informants, and the safety in the field. We conducted a 120 preliminary fieldwork of two weeks in January 2015, followed by a 6-months fieldwork from 121 January to June 2016. The study was based on semi-structured interviews, questionnaires, 122 participant observation, and literature review of reports and related documents from the 123 organizations managing the schemes. We followed intermediaries in their daily work in the 124 field and office, and attended a major conference with the presence of PWS managers and 125 payers from three of the studied schemes (June 13th to 17th, 2016, Bogotá D.C.). Interviews 126 and questionnaires were applied to three groups: providers, intermediaries, and major 127 payers. Here we consider PWS intermediaries as “actors who take on roles that connect and 128 facilitate transactions between buyers and sellers” (Huber-Stearns et al., 2013, p. 105). In 129 the studied cases they are civil society organizations (e.g. water user associations, 130 environmental non-government organizations). Providers in these schemes are considered 131 those who live in areas that are strategic for the provision of water resources and have 132 voluntarily agreed to participate. They are usually small landowners living in the upstream 133 area of the watersheds. Payers in the four schemes are mostly private companies, public 134 companies and environmental authorities, although individuals also pay a small contribution 135 as water users in one of the schemes (ASOBOLO case). 136 Among the four schemes, two of them (Agua Somos and Cuenca Verde) were created 137 as “water funds” within the frame of the Latin-American Alliance for Water Funds, an 138 initiative led mainly by the international non-governmental organization (NGO) The Nature 139 Conservancy (TNC) (Calvache et al., 2012; Goldman-Benner et al., 2012). ASOBOLO, in turn, 140 is a water user association created in the early-90’s to manage the water resources shared 141 by different users in the Rio Bolo watershed, mainly from the sugarcane production sector 142 (Munoz Escobar et al., 2013). In the late-2000’s ASOBOLO joined an initiative also framed as 143 a water fund together with several other water user associations from the same region, 144 forming what is now called Fondo Agua por la Vida y la Sostenibilidad (FAVS) (Moreno- 145 Padilla, 2016). ASOBOLO receives funds from FAVS and other programs such as the 146 “Conservation Incentives”, a finished 5-years project funded by The Netherlands and 147 managed by the NGO Fondo Patrimonio Natural (Table 1). The same project incentivized and 148 funded the creation of CSAH Cali putting together local government agencies and private 149 sector to fund conservation practices in upstream catchments that provide water to the city 150 of Cali (See references in Table 1 for additional details). 151 The PWS conservation activities undertaken in the watersheds through the schemes 152 under study started quite recently; from 2010 onwards (Table 1). In all four schemes, 153 conservation practices are mostly concentrated in river and spring fencing for protection 154 against the cattle in upstream lands (Table 2). Some of the initiatives (e.g. ASOBOLO) also 155 include implementation of live fences and agrosilvopastoral systems, cattle rotational 156 grazing, and other farm practices to reduce the impact of cattle on the soil of mountainous 157 areas. In order to set those practices, the intermediaries establish voluntary conservation 158 agreements with each upstream landowner. The agreements include payments to the 159 landowners, usually in-kind, including technical assistance to improve productivity in the 160 farm, materials for the conservation practices, and farm infrastructure improvement. Among 161 the studied cases, CSAH Cali is the only program that pays also in cash. 162 163 Table 164 1: Description of the four PWS schemes: literature references, watershed location, intermediaries involved, major payers, year in which 165 conservation interventions started, and number of contracts signed up to June, 2016 (only in the referred watershed). # of Scheme Watershed, Partnership Intermediary Intermediary Major payers1 Year contracts in References (Department) Programs (Catalysts) (Operator) this watershed Chisacá and TNC, EAAB, Agua Somos 35 Mugroso Rivers, Latin-American Fundación SDA, Bavaria/ (Goldman- (as of tributary of Alliance TNC Patrimonio SabMiller§, 2013 Benner February Tunjuelo River for Water Funds Natural Coca-Cola/ et al., 2012) 2016) (Cundinamarca) FEMSA§ ASOBOLO Latin-American Aguaclara iver, ASOCAÑA, CVC, (Moreno-Padilla, Alliance 56 tributary of ASOBOLO Syngenta§, RioPaila§ 2016; Munoz for Water Funds; ASOBOLO 2010* (as of June Bolo River ASOCAÑA y Castilla, Escobar Conservation 2016) (Valle del Cauca) Mayagüez§ et al., 2013) Incentives Chico River, Cuenca Verde tributary of Latin-American EPM§, Postobon§, 22 Cuenca (Gómez-Ochoa, Riogrande Alliance TNC, EPM Nutresa§, Argos§, 2015 (as of June Verde 2016) Reservoir for Water Funds AMVA, CORNARE 2016) (Antioquia) CSAH Cali Felidia and The Netherlands (Fondo Pinchindé Rivers, Fundación 46 Conservation government, Patrimonio Tributaries of Patrimonio AcuaCali.co 2013 (as of June Incentives CVC, EMCali, Natural Cali River, Natural 2016) ASOCAÑA et al., 2014) (Valle del Cauca) 166 Abbreviations: 1Major payers as mentioned by interviewed intermediaries. In addition to these, there are several others that contribute with smaller funds. AMVA – Area Metropolitana del Valle 167 de Aburrá; ASOCAÑA - Asociación de Cultivadores de Caña de Azúcar de Colombia; ASOBOLO – Asociación de Usuarios de las Aguas Superficiales y Subterraneas de la Cuenca del Río Bolo; 168 CORNARE – Corporación Autónoma Regional de las Cuencas de los Ríos Negro y Nare; CVC - Corporación Autónoma Regional del Valle del Cauca; EAAB – Empresa de Acueducto, Alcantarillado y 169 Aseo de Bogotá; EMCali – Empresas Municipales de Cali; EPM – Empresas Publicas de Medellín; FEMSA - Fomento Económico Mexicano, S.A.B. de C.V.; PN - Fundación Patrimonio Natural; SDA - 170 Secretaria Distrital de Ambiente; TNC - The Nature Conservancy; *Refers to the year in which activities started in Aguaclara watershed only. ASOBOLO itself was launched in 1993 and started 171 working in other regions first. §Those companies are among the 100 biggest industry companies in Colombia. 172 173 Table 2: Target ES, activities performed in the field, type of payment and payments for each PWS scheme. Watershed Conservation practices Type of Payments Scheme Target ES area Undertaken Payment1 already performed Energy saving wood stove, cattle drinking Chisacá and Water quality Agua River fencing, tree planting, fountains, materials for fences, seedlings, small-scale Mugroso and provision, In-kind Somos cattle rotational grazing systems poultry systems, improvement of pastures, (8,245 ha)2 sediment retention technical assistance River fencing, springs protection, tree planting, Material for fences, seedlings, septic tanks, financial support Water flow Aguaclara forest protection, farm septic tanks, for implementing forest reserves, community vegetable ASOBOLO regulation, In-kind (11,141 ha)3 agrosilvopastoral systems, cattle gardens, fruit trees, environmental education in schools, sediment retention rotational grazing systems, capacity building with local associations, technical assistance farm live fences. River fencing, forest protection, Cattle drinking fountains, materials for fences, Cuenca Chico springs protection, bridges for cattle, improvement of paddocks, forage for Water quality In-kind Verde (17,172 ha)4 forest restoration, cattle, compost systems, septic tank, tree planting seedlings, technical assistance Felidia Water quality Cash, materials for fences, seedlings, (6,635 ha)5 River fencing, forest protection, In-kind CSAH Cali and provision, technical assistance, environmental education projects, Pichindé springs protection and Cash sediment retention vegetable gardens, sewage treatment system (5,272 ha)5 174 Sources: 1It refers to the type of benefit that has been provided to the land owners who are implementing conservation practices in their properties. 2Calculated from digital elevation map, 175 upstream Chisacá reservoir; 3CENICAÑA (Centro de Investigación de la Caña de Azúcar de Colombia) technical staff – personal communication; 4Calculated from digital elevation map, upstream 176 confluence with El Hato stream. 5Fondo Patrimonio Natural et al., 2014. 177 178 179 180 Figure 1. Schematic location of the studied water-related PES Schemes 181 182 Questionnaires and Interviews 183 Initial interviews with key informants were used to understand the history and 184 context of each scheme and to support the design of the questionnaires. Questionnaires 185 were structured in questions that combined selection of options, ranking, and degree of 186 agreement with some statements. For questions regarding degree of agreement a Likert 187 scale of 0 (totally disagree) to 5 (totally agree) was applied. Providers answered their 188 questionnaires in the field. Intermediaries answered their questionnaires either directly in 189 the field or online through an internet link sent to their email address after permission 190 requested in person or by phone. Questionnaires for major payers were applied in the field 191 during the interviews, through teleconference meetings or via email after contact by phone, 192 depending on the availability of the respondents. Semi-structured interviews followed the 193 same topics of the questionnaires in a more flexible approach in order to capture each 194 group's perspectives, differences, opinions, and explanations for their choices and to cross- 195 check the answers from questionnaires. 196 The structure of the questionnaires started with a session with general information 197 about the respondent and followed topics such as: motivations to participate in the scheme, 198 general expectations based on their own roles and perceptions about the functioning of the 199 scheme and the environmental changes and/or improvements. In addition, the following 200 topics were added according to each respondent group: (1) providers: role of the payment 201 as an incentive for entering the scheme; (2) intermediaries: importance of environmental 202 monitoring, importance of reporting impact evidence, perception of expectations and 203 demands from payers; (3) payers: water-related environmental services of interest, 204 importance of evidence of the achievement of the PES goals for the maintenance of the 205 payment, expectations on impact monitoring and reporting. The initial interviews with the 206 three groups provided the statements used for the agreement questions and the list of 207 motivations of providers and payers. After selecting the motivations from the list with the 208 option of adding new ones, these two groups ranked up to three of their most important 209 motivations. We then analyzed all answers in a lumped way. 210 211 Characteristics of questionnaires' respondents 212 Intermediaries group: 25 individuals answered the intermediaries' questionnaire. 213 The questionnaire included intermediaries working not only in the study sites reported here 214 but also in additional PES schemes in Colombia. Three respondents were involved in 215 schemes in Colombia and in other Latin-American countries as part of an international 216 program. Most of the respondents were from NGOs. Other respondents included: 217 representatives of local water users’ associations, representatives of other civil associations, 218 researchers and technicians belonging to local research centers involved in monitoring PES 219 schemes, and one representative from a regional environmental authority. Table 3 220 summarizes the characteristics of the respondents and their organizations in terms of the 221 roles played in the scheme in which they are involved as intermediaries. 222 223 Table 3: Individual role and organizational role of intermediaries’ respondents. Individual Role in the Scheme n (%) Organizational Roles in the Scheme n (%)* Supporting Technician 9 (36) Design 15 (60) Manager 6 (24) Support 14 (56) Scientist in charge of research 5 (20) Implementation 10 (40) Mixed role: scientist/technician 2 (8) Management 4 (16) Mixed role: manager/technician 2 (8) Evaluation 1 (4) Other 1 (4) Professional Background n (%) Organizational Roles related to Monitoring n (%)* Biologist 7 (28) Funding 15 (60) Economist 4 (16) Getting secondary data from third parts 15 (60) Agricultural engineer 4 (16) Direct monitoring of env. indicators 14 (56) Forest engineer 2 (8) Logistics for monitoring in the field 13 (52) Agronomy engineer 2 (8) Processing monitoring data and report 13 (52) Sanitary engineer 1 (4) No role on monitoring 1 (4) Environmental engineer 1 (4) Social worker 1 (4) Environmental manager 1 (4) Oceanographer 1 (4) Environmental technician 1 (4) 224 Total of respondents: 25. * Non-exclusive roles, i.e. the organization may perform different 225 roles in the same scheme. 226 227 Payer’s group: 15 individuals answered the payers' questionnaire. Respondents belonging to 228 the payers group were representatives of several types of organizations/firms, including: 229 two public water supply companies; private companies from the beverage, sugarcane and 230 dairy production sectors, a restaurant, a health clinic, a lawyers firm, an infrastructure 231 company and one service company; and one development organization acting as payer. 232 Respondents reported both in kind and cash investments to the scheme (Table 4). The 233 following types of financial participation were reported: annual payments (n=6); every 3-6 234 months (n=3); a one-time payment (n=2); punctual payments through projects in the form of 235 contracts with the intermediaries (n=3); no payments in cash (n=2). Most of the payers are 236 users of the water from the watershed that they are paying for conservation (Table 4). 237 238 Table 4: Use of water from the watershed by the 15 paying organizations and type of 239 payment provided to PES schemes (non-exclusive categories). Use of water from the watershed n (%) Type of payment provided for the scheme n (%) Incorporated in products 7 (46.7) Cash* 13 (86.7) Public water supply 4 (26.7) Materials (e.g. construction, technology) 3 (20) Private water supply 3 (20) Logistic Services (e.g. transportation) 2 (13.3) Use of water for industrial processes 4 (26.7) Technical assistance (e.g. laboratory, GIS support) 4 (26.7) No use of water 2 (13.3) Other types 3 (20) Other 1 (6.7) 240 * Note that cash could be used by the intermediaries to cover implementation costs (materials, staff…) and/or converted into in-kind 241 payments. The only case where cash was entirely disbursed as such to the providers is the CSAH Cali. 242 243 Providers group: 72 individuals from a total of 159 providers (45%) from the four PES 244 schemes answered the providers' questionnaire. The distribution of respondents in the four 245 schemes was: 12 respondents out of 35 providers from Agua Somos (34%); 30 respondents 246 out of 56 providers from ASOBOLO (54%); 18 respondents out of 22 providers from Cuenca 247 Verde (82%); 12 out of 46 respondents from CSAH Cali (26%) (Table 5). 248 Landowners living in farms inherited from relatives are the majority of respondents. 249 A small fraction is not owners but long-term tenants. Most of the respondents have lived 250 their entire life in the region, some of them in the nearby city. They are usually farmers who 251 depend on agricultural production as their main income. Cattle ranching is the most 252 common use of land. Table 5 summarizes the main characteristics of the respondents among 253 providers. 254 255 Table 5: Main characteristics of questionnaire respondents among providers and their land 256 properties. Characteristics of providers’ respondents Avg. number Number of Average Most Scheme of years with Respondents farm size Common Main land use signed n (%) (median) [ha] type of property PES agreements Cattle ranching, Agua Somos 12 (48) 18.7 (7.1) Inherited 2.2 potato crops Cattle ranching, ASOBOLO 30 (54) 8.8 (3.3) Acquired 4.5 fruit production Fruits and CSAH Cali 12 (26) 19.4 (4.5) Acquired herbs production, 3.8 leisure, tourism Cuenca Verde 18 (82) 41 (8.5) Inherited Cattle ranching 1.1 Fruits and herbs CSAH Cali 12 (26) 19.4 (4.5) Acquired production, 3.8 leisure, tourism 257 3. RESULTS 258 This section presents the results grouped in four core topics: (1) motivations to engage in a 259 PES scheme from payers and providers’ perspective; (2) importance of reporting PES impacts 260 from the intermediaries and payers’ perspectives; (3) payment upon evidence of PES 261 impacts; (4) environmental indicators of interest. 262 1. Motivations to engage in a PES scheme 263 Payers: during the interviews, payers reported different motivations to engage in PES 264 schemes. The most frequently mentioned was the protection of water resources from which 265 they depend directly or indirectly. In addition, several interviewees have mentioned that 266 their companies have environmental and social responsibility policies and that the schemes 267 fit into their goals. A representative of a multinational company emphasized that there is a 268 new generation of businessmen with a newer mentality towards environmental issues such 269 as climate change and water resources. According to him, with potential water shortages in 270 the future, companies cannot count only on efficient use anymore - there must be combined 271 strategies. 272 Interviewees from two different water supply companies had different perspectives 273 regarding the conservation schemes. The first mentioned that the scheme was a way to 274 compensate the municipalities from which they have been withdrawing water for decades. 275 Now there was a need to invest back in these communities and, therefore, the focus of their 276 participation was the social aspect more than the environmental one. The representative 277 from the second company, on the other hand, emphasized the protection of water resources 278 as the fundamental reason as the company is in charge of the water treatment and 279 distribution service as well as electricity supply from hydropower. Legal requirements were 280 also mentioned: Colombian Law n. 99/1993 requires that energy supply companies pay a tax 281 to environmental authorities and municipalities. 282 Several schemes in Colombia have been set in regions with long term environmental 283 conflicts of different levels, for example, the Valle del Cauca where sugarcane producers are 284 great water users. In other regions, small landowners are in conflict with environmental 285 authorities and water supply companies over land use in upstream areas due to impacts on 286 springs and creeks. Therefore, from the perspective of many companies acting as payers, 287 their participation is also a way to reduce conflicts by investing back in those communities. 288 The questionnaires confirmed several aspects mentioned during the interviews. ES 289 improvement (e.g. better water quality, better regulation of flows) was selected as one of 290 the motivations by all respondents from the payers group (n=15) (Table 6). The second most 291 selected motivation was corporate socio-environmental policy (n=12). Following, corporate 292 image/green marketing and improvement of private-public relations (e.g. with local 293 communities) were selected by 7 respondents each. 294 Table 6: Motivations selection and ranking by payers (absolute values followed by 295 percentage relative to total of answers). First value column refers to the number of times 296 the motivation was selected in the list, including when they did not count among the most 297 important ones. The following value columns refer to the number of times the motivation 298 was ranked as first, second and third most important. Total number of answers is shown in 299 the headings. # selections # selected 1st # selected 2nd # selected 3rd id Motivation total=15 total=15 total=12 total=11 n (%) n (%) n (%) n (%) A ES improvement 15 (100) 10 (66.7) 3 (25) 0 B Corporate social-environmental responsibility (CSR)* 12 (80) 4 (26.7) 4 (33.3) 1 (9.1) C Corporate image, i.e. green marketing 7 (46.7) 0 0 5 (45.5) D Improvement of private-public relations 7 (46.7) 0 0 2 (18.2) E Part of the fundamental goals of the organization 3 (20) 1 (6.7) 0 0 F Part of a mitigation process of our production chain 3 (20) 0 2 (16.7) 1 (9.1) G Restoration of areas affected by the organization. 2 (13.3) 0 1 (8.3) 0 H Sustainability index in stock market 3 (20) 0 0 0 I Taxes deduction 2 (13.3) 0 1 (8.3) 0 J Demanded by shareholders 2 (13.3) 0 1 (8.3) 1 (9.1) K Environmental compensation required by law 1 (6.7) 0 0 1 (9.1) L Environmental investment required by law 1 (6.7) 0 0 0 M Part of the organization' duties according to the law 1 (6.7) 0 0 0 N ES was already part of their business tradition 1 (6.7) 0 0 0 O Facilitate environmental certification 0 0 0 0 P ES is incorporated in business 0 0 0 0 300 *Corporate Social Responsibility here is understood as “context-specific organizational actions and policies that 301 take into account stakeholders’ expectations and the triple bottom line of economic, social, and environmental 302 performance (Aguinis and Glavas, 2012, p. 933).” 303 304 In sequence, payers’ respondents ranked the first, second, and third most important 305 motivations for them (Table 6). The first most important motivations were: ES improvement 306 (n=10), corporate social responsibility (n=4), and ES as part of the goals of the organization 307 (n=1). The second most important motivations were: corporate social responsibility (n=4), ES 308 improvement (n=3), and impact mitigation (n=2). Restoration of areas affected by the 309 organization, taxes deduction and shareholders’ demands were selected as second most 310 important motivation by one respondent each. The third most important motivations were: 311 corporate image/green marketing (n=5), improvement of private-public relations (n=2), 312 corporate social responsibility (n=1), mitigation (n=1), compensation required by law (n=1), 313 and shareholders’ demand (n=1). 314 Although not identified as initial motivations by all respondents, the following items 315 were indicated by some of the respondents during interviews as non-expected benefits 316 gained after their engagement in PES schemes: better private-public relations, better 317 corporate image, tax deductions, internal sustainability indicators, sustainability index in 318 stock markets, environmental awareness. 319 Providers: during the interviews, providers mentioned several motivations for 320 engaging in the conservation scheme. Among the mentioned ones, they have said that the 321 scheme represented progress to them and that it provided learning experience. Many have 322 mentioned the environmental consciousness as a motivation for engaging, relating it to the 323 protection of water resources and some concerns with future dry seasons. Some 324 interviewees mentioned that they want to protect water resources that are strategic for 325 their farm for the sake of those who will inherit it. Personal conviction and economic 326 incentives (even in the in-kind form) were also mentioned as motivations. Interviews and 327 participant observation led to the reflection that the main motivations can be interpreted as 328 an intention to stay in the territory maintaining or improving their livelihoods. 329 Interviews with Agua Somos providers showed that similar initiatives had already 330 been undertaken in the region some years ago and that helped them to feel motivated to 331 participate in Agua Somos. It was reported that when some landowners first started 332 engaging with Agua Somos, others observed the benefits gained by them through the 333 program and decided to participate, a behavior also observed by Goldman-Benner et al. 334 (2012) and described as a social spillover effect. Participants that engaged later also had the 335 chance to observe the commitment of the intermediary actor with the first participants 336 allowing them to feel more confident in the process. 337 Additional benefits later perceived by some of the providers were the farm and 338 landscape aesthetics. Several interviewees mentioned that the farm looks better managed 339 and not abandoned when receiving the interventions proposed by the intermediaries, which 340 could be interpreted as an expression of perceived added value to the property. River fences 341 were also pointed as beneficial for avoiding accidents with the cattle such as when the cows 342 fall from steep areas. Other interviewees mentioned that the planted trees would favor 343 avoiding high temperatures during the day by providing shadow for the cattle. Several 344 interviewees mentioned the importance of protecting the springs in their farms for their 345 own sake. One interviewee translated that into a clear statement: “a farm without water has 346 no value”. Extra income provided by the fruit trees planted in one of the schemes was also 347 mentioned as perceived benefit. From participant observation, it was possible to realize that 348 several reasons reported by providers were being disseminated by intermediaries during 349 negotiations as a way to create environmental awareness. 350 Providers were asked what they know about the reasons why some landowners 351 decided not to engage in the schemes. Several interviewees reported that those who are not 352 participating believe that the fenced land (e.g. for riparian forest recovery) would be later 353 expropriated and taken from their control. Others pointed that resistant landowners were 354 worried about not being able to use the water for their cattle anymore. 355 The questionnaire phase started with a list of statements that were taken from the 356 interviews. Providers were asked whether they agree with statements connected to their 357 motivations and expectations in participating in the scheme (Table 7). Answers supported 358 the notion of an existing interest in water resources conservation (item B) but also to a 359 desire to improve their livelihoods (item A and C). Results from items D and E showed that 360 the majority of providers would have participated in similar conservation schemes even 361 without direct economic benefits for themselves (in kind or cash payments), confirming 362 other statements in which it becomes clear that providers understand the protection of 363 water resources as a beneficial for themselves anyway. Most providers also recognize that 364 the effects of the schemes will require some years to be observed (item F). 365 From a list of potential motivations extracted from interviews, respondents were 366 asked to select those that they had. More than half of respondents pointed to the 367 expectation of a beneficial effect of the scheme on the water quality/quantity (Table 8, item 368 C) and half of respondents indicated a concern with water resources conservation (item A), 369 which may be explained by the importance attributed to water as a factor of production. 370 Almost half of the respondents reported the potential for improvement of farm productivity 371 as one of their motivations (item B). 37% of respondents also reported to have been 372 concerned with conservation even before the proposition of the scheme (item D). Only 6 373 among 71 providers (8.6%) reported that one of the motivations to engage was that the 374 economic incentives proposed by the intermediaries caught their attention (item K). More 375 common motivations than economic incentive were: enjoying the idea of having a protected 376 forest in the farm (item E); shadow provided by trees for the cattle (item F); and farm’s 377 aesthetics (item G) (Table 8). 378 Concerns with water resources conservation figured out as the first most important 379 motivation for 30% of respondents, while improvement of farmers productivity assumed the 380 same position for 25.7%. Expectations with improvement of water quality/quantity assumed 381 the second place as most important motivation for 20% and third place for 27.5% of 382 respondents (Table 8). 383 384 Table 7: Summary of providers’ answers to agreement questions. Neither # Totally Partially Partially Totally id Agreement Statement Agree nor Answers Disagree Disagree Agree Agree Disagree n (%) This conservation project will improve A 71 3 (4.3) 0 11 (15.7) 25 (35.7) 31 (44.3) a lot the productivity conditions of my farm. I have always been interested in water resources conservation, B 72 2 (2.8) 0 6 (8.3) 12 (16.7) 52 (72.2) even before this project. Environmental concerns are important, but it is more important C 72 12 (16.7) 8 (11.1) 25 (34.7) 12 (16.7) 15 (20.8) to obtain the benefits of the scheme for my farm. I would participate in any environmental project, D 71 1 (1.4) 2 (2.8) 5 (7.0) 21 (29.6) 42 (59.2) even without economic benefits. I would have participated in this scheme, E 72 1 (1.4) 3 (4.2) 5 (6.9) 25 (34.7) 38 (52.8) even without having received any economic incentive for that. Detection of the desired effects of the scheme on F 73 1 (1.4) 1 (1.4) 4 (5.5) 11 (15.1) 56 (76.7) water quality/quantity will still require some years. 385 386 387 388 Table 8: Ranking of providers’ motivations. First number column shows the total number of 389 times the motivation was included among the three most important ones and percentage 390 among respondents. Next number columns report the number of times the motivation was 391 ranked as first, second and third most important. # ranking # selected # selected # selected # Motivation (“I engaged in the scheme because...”) selections 1st 2nd 3rd (N=70) (N=70) (N=70) (N=69) n (%) A ... I began to feel concerned with water resources conservation 35 (50.0) 21 (30.0) 11 (15.7) 3 (4.3) B ... I thought the scheme would improve my farm’s productivity 30 (42.9) 18 (25.7) 6 (8.6) 6 (8.7) C ... I thought the scheme would improve water quality/quantity 44 (62.9) 11 (15.7) 14 (20.0) 19 (27.5) D ... I was already concerned with conservation before this scheme 26 (37.1) 8 (11.8) 7 (10.0) 11 (15.9) E ... I like the forest and I wanted to have a bit in my farm 20 (28.6) 4 (5.7) 9 (12.9) 7 (10.1) F ... With more trees my cattle would have more shadow 14 (20.0) 2 (2.9) 6 (8.6) 6 (8.7) ... I believed the farm would appear more organized/beautiful with the 11 (15.7) 4 (5.7) 4 (5.7) 3 (4.3) G interventions proposed H ... I like that others are concerned with our community 13 (18.6) 1 (1.4) 7 (10.0) 5 (7.2) ... In our community we all are interested in participating in 3 (4.3) 1 (1.4) 0 (0) 2 (2.9) I environmental projects J ... My neighbors were already in the scheme 4 (5.7) 0 (0) 4 (5.7) 0 (0) K ... The economic incentives caught my attention 6 (8,6) 0 (0) 1 (1.4) 5 (7.2) L ... Of other non-mentioned motivations 3 (4,3) 0 (0) 1 (1.4) 2 (2.9) 392 393 2. Importance of evidence of PES impacts 394 Intermediaries: During the questionnaire phase, intermediaries were asked whether they 395 feel it is important to report/communicate actual PES impacts on water-related ES (e.g. 396 water turbidity reduction, river discharge increment, etc.) to payers (Table 9). 96% of 397 respondents (n=24) totally agreed that it is important. Following, they were asked whether 398 they feel that payers expect them to demonstrate PES impacts on water-related ES. The 399 question did not include the expectation over indicators of activities performed (e.g. number 400 of planted trees), but only indicators regarding water-related ES. 68% of respondents (n=17) 401 totally agreed, while 24% (n=6) mostly agreed. Following, intermediaries were asked 402 whether they agree that demonstrating PES impacts on the target water-related ES was 403 important for the reputation of their organizations (Table 9). Among respondents, 68% 404 (n=17) totally agreed and 20% (n=5) mostly agreed. These answers highlighted the 405 importance attributed by the intermediaries to the effective communication processes 406 among stakeholders. 407 Interviews confirmed the intermediaries’ concerns on proving the effectiveness of 408 their schemes in terms of water-related ES indicators. Interviewees expressed that they are 409 concerned about demonstrating PES impacts on ES not only for the sake of their reputation, 410 but also because: (1) they want to keep payers on board, (2) they want to attract more 411 potential payers by demonstrating the effectiveness of their schemes, (3) they want to keep 412 good relations with the providers who entered the scheme motivated by environmental 413 awareness, (4) they feel personally engaged with the cause and want the schemes to be 414 successful. It was also observed in the field that NGOs and civil associations employ several 415 technicians and managers to work in the PES scheme and lack of funding would mean 416 unemployment for many of them. Therefore, evidence of an effective conservation scheme 417 also supports more stable working conditions. Another reason mentioned by interviewees 418 from one intermediary organization is that their organization portraits itself as a “science- 419 based” NGO, therefore, evidence of ES outcomes is important for them to keep their 420 tradition and image. 421 422 Table 9: Summary of intermediaries’ respondents answers to agreement questions. Totally Partially Neither Agree Partially Totally id Agreement Statement # Answers Disagree Disagree nor Disagree Agree Agree n (%) It is important to communicate PES impacts/outcomes to those who A 25 0 0 0 1 (4.0) 24 (96.0) are supporting and paying this scheme Those who are supporting and paying this B scheme expect us to 25 0 1 (4.0) 1 (4.0) 6 (24.0) 17 (68.0) demonstrate PES impacts Those who are supporting and paying this scheme only pay under the C 25 6 (24.0) 2 (8.0) 9 (36.0) 4 (16.0) 4 (16.0) condition of PES impacts being demonstrated I think that demonstrating PES impacts is D important for the reputation of 25 0 1 (4.0) 2 (8.0) 5 (20.0) 17 (68.0) those in charge of the schemes PES schemes are too new to be able to E generate impacts that could be 25 1 (4.0) 1 (4.0) 7 (28.0) 1 (4.0) 15 (60.0) detected on water-related ES 423 424 Payers: payers’ respondents were asked how much they agree that their organizations 425 require regular reports of environmental indicators monitoring from PES intermediaries 426 (Table 10, item E). Most respondents totally agreed (n=9, 60%). Respondents were then 427 asked how much they agree that they only require reports about the activities performed by 428 the PES intermediaries, but not about the impacts of these activities on the environmental 429 services of interest (item F). Most of the respondents disagreed (66.7%), while 26,6% agreed. 430 These answers were similar to those obtained in several interviews with payers and 431 intermediaries. However, although payers are interested in checking out the actual PES 432 outcomes in terms of ES, several of them were aware that it would take longer to see the 433 effects of the scheme and that monitoring could be expensive, counterbalancing their 434 expectation of receiving full reports about the impacts of these activities on ES. 435 436 Table 10: Summary of payers’ answers to agreement questions. Neither Totally Partially Agree Partially Totally Id Agreement Statement # Answers Disagree Disagree nor Agree Agree Disagree n (%) We believe that ES will improve with A 15 0 0 1 (6.7) 5 (33.3) 9 (60) the PES scheme The most important motivation for engaging was the B 15 0 0 2 (13.3) 2 (13.3) 11 (73.3) improvement of environmental service We only pay under the condition of C PES impacts being 15 6 (40) 2 (13.3) 1 (6.7) 5 (33.3) 1 (6.7) demonstrated If PES impacts on environmental services are not proven D 15 4 (26.7) 2 (13.3) 2 (13.3) 4 (26.7) 3 (20) with the time, we would stop paying for the scheme We require reports with monitoring results of PES impacts on E 15 1 (6.7) 1 (6.7) 1 (6.7) 3 (20) 9 (60) environmental services regularly We only require reports of activities performed, not of PES F 15 7 (46.7) 3 (20) 1 (6.7) 2 (13.3) 2 (13.3) impacts on environmental services 437 438 3. Payment upon evidence of PES impacts 439 Intermediaries: intermediaries were asked whether they agree that payers pay (or will keep 440 paying) under the condition that PES managers provide proofs of PES impacts on ES (Table 441 9). This question raised no clear consensus: 36% of respondents (n=9) neither agreed nor 442 disagreed, while those who mostly or totally disagreed summed 32% (n=8), and the same 443 was obtained for those who mostly or totally agreed (32%, n=8). During the interviews, 444 questions regarding this point also raised uncertainty among intermediaries. However, most 445 of them expressed that although reporting progress in terms of PES impacts on water- 446 related ES was important and required by payers, payers do not appear to be willing to leave 447 the scheme in case no proof was delivered. When questioned about potential reasons, 448 several interviewees pointed that although payers were interested in the ES, they also 449 engaged in the scheme because of other motivations. Table S1 presents their opinion about 450 potential motivations from payers (separated in two categories only, private and public 451 companies). 452 Interviews revealed that intermediaries understand that large companies acting as 453 payers, in general, are aware that several factors are not under control on a conservation 454 scheme and that the outcomes may not be the expected ones. Even so, they would be able 455 to handle it through other investments in case they depend on these water resources. For 456 example, in case reduction of turbidity is not achieved for a certain water source they could 457 pay for new water treatment facilities to overcome the issue. According to one 458 representative from an intermediary, not achieving the ES goals would not be a major issue 459 for big companies; for instance, he mentioned, if a beverage company does not obtain 460 better water quality, any additional cost on water treatment required for production would 461 be passed to consumers through an increase in price. The respondent did not mention any 462 effect that it could have on the competition with other beverage companies. 463 Therefore, according to the intermediaries, other benefits that big payers have while participating in 464 PES projects would explain their permanence in the schemes, such as stating their 465 commitment with environmental projects each year in a corporate sustainability report. 466 Intermediaries did not express the same perception for small payers. 467 Some intermediary representatives also expressed that their responsibility over 468 providing clear evidence of ES benefits out of the conservation practices are limited by the 469 fact that payers usually pay for conservation but they do not pay for research or monitoring. 470 Therefore, payers would not be in the position to demand clear evidence from 471 intermediaries. 472 Payers: Payer’s representatives were then asked how much they agree that their 473 payments were conditional on evidence of ES improvement to be provided by intermediaries 474 (Table 10, item C). The question divided respondents; 8 out of 15 respondents disagreed 475 (53.3%) and 6 of them agreed (40%), while 1 neither agreed nor disagreed. Average value for 476 this question was 2.53, indicating a trend towards disagreeing. Following, payers were asked 477 how much they agree that their organizations would leave the scheme in case no ES 478 improvement is proven over time (item D). This question also divided respondents. 4 479 respondents totally disagreed, 2 mostly disagreed, 1 neither agreed nor disagreed, 4 mostly 480 agreed, and 3 totally agreed. Average value reached for these answers was 3.0, indicating 481 that answers were slightly balanced towards agreeing with the statement. 482 During interviews, payers indicated in their answers that ES improvement is 483 important and is the main driver for their engagement. However, most of them have 484 entered the scheme for additional reasons, confirming the results presented in Table 6 and 485 interviews with intermediaries. Several interviewees expressed that they would not stop 486 paying in case ES improvement is not proven. Reasons for that included: (a) some 487 understanding that water flows in a watershed is a result of complex processes and 488 monitoring is costly; (b) payers invest part of their money for environmental compensation 489 and other legal requirements and, therefore, engaging in a PES scheme fits those 490 requirements; (c) additional motivations pay-off; (d) big companies are somehow used to 491 fund risky investments and PES costs are not large enough to compromise their budget. One 492 water supply company representative stated that evidence of ES outcomes is highly 493 important, but they are aware, from their own experience, that a good hydrologic 494 monitoring system is too expensive. Interviewees from another water supply company 495 attributed less importance to evidence of ES outcomes, in part because their participation in 496 the scheme was much more a response to a social demand than to the need of improving 497 watershed conditions. 498 For one water supply company, the PES scheme was part of a whole strategy on 499 conservation that was already established in the company’s core functioning. Therefore, 500 even if the PES scheme does not achieve the expected outcomes, other internal 501 conservation projects would play a role on satisfying environmental concerns. Another 502 respondent from a beverage company expressed that although the outcomes of such 503 schemes were highly uncertain their company already had environmental concerns as part 504 of their core values. Moreover, as in any other investment, companies are used to take risks 505 and deal with uncertain situations. 506 507 4. Environmental and social indicators of interest 508 Payers: payers representatives received a list of potential environmental and social 509 indicators and were asked to choose the ones that their organization had required reports 510 from intermediaries. In addition, they had the option of adding new ones to the list. In 511 sequence, they were asked to choose the first, second and third most important indicators 512 (Table S2). All 15 respondents selected the indicators, however just 12 indicated their first, 513 and 9 chose their second and third most important ones. The most selected indicators were 514 A to E. While indicators A to D refer to the conservation practices performed in the field (e.g. 515 area under vegetation protection, number of planted trees, number of protected springs), E 516 refers to a social indicator (number of families receiving payments). None of these five 517 indicators is related to the ES outcomes, i.e. the ES maintenance or improvement obtained 518 from the conservation practices. However, when asked about their first most important 519 indicator, 33.3% of respondents (n=4) said river discharge increment (item G), would be their 520 choice. Area under vegetation protection (item A) follows as first choice for 25% of 521 respondents (n=3). Area under vegetation protection and sediment load reduction (item I) 522 were chosen as second most important motivation by 2 respondents each. Number of 523 planted trees (item B) was chosen by 4 respondents (44.4%) as the third most important 524 motivation. 525 The fact that the most important indicators selected by payers are not about ES 526 outcomes but are rather proxies associated to farmer´s land use practices and their impact 527 on the land/vegetation could be due to three potential explanations drawn from interviews: 528 (i) Payers strongly believe in a positive interaction between vegetation protection 529 and water-related ES maintenance; 530 (ii) payers are more inclined to use those indicators as numbers to their annual 531 sustainability reports for their clients, government, shareholders or local community; for 532 that matter indirect indicators of conservation, i.e. proxies, would be sufficient; 533 (iii) at least part of the payers are aware that ES monitoring is not easy to perform 534 and is expensive, therefore, they would rely on proxy indicators as metrics for PES 535 effectiveness. 536 As one of the payers from a water supply company reported on an interview: “Impact 537 indicators are more important, i.e. those who reflect the benefit of the activities performed, 538 however it is not always possible to get them due to information availability issues or the 539 cost of getting information. Therefore, we should seek a balance in those terms (Bogotá, 540 June 2016). ” These results apparently contradict a comment made by one PES manager that 541 said, “business men who are investing in those payments schemes expect that scientists 542 could come with an equation that would tell them for every dollar spent how much water 543 quality improvement they will gain in return (Medellín, February 2015).” 544 545 4. DISCUSSION 546 5. Who is interested in evidence of PES environmental impacts? 547 Intermediaries are concerned about getting impact evidence of their schemes on ES 548 for several reasons, most of them linked to their wish to keep their modus operandi. 549 Evidence of ES improvement or maintenance would support keeping payers on board, 550 attract more payers in the long term, sustain conservation jobs, and maintain trust 551 relationships with both payers and providers. These concerns are seen on interviews, 552 questionnaires and are similar to those found in the practitioners’ literature (e.g. Bremer et 553 al., 2016; Higgins and Zimmerling, 2013). Intermediaries are also clearly concerned about 554 their reputation as PES managers and are aware that providing evidence of PES impacts on 555 ES is a central issue on this matter. This result is consistent with the one obtained by Fisher 556 and Brown (2014) regarding the use of the ES concept and its derived tools by conservation 557 practitioners. The cases analyzed provide some evidence that personal values also play a 558 role on the intermediaries’ concerns with the evidence of PES impacts. Most of the 559 interviewed representatives feel personally engaged with the local communities in which 560 they work and with the environmental issues, and want the schemes to be effective. 561 Payers do expect evidence of environmental benefits achieved by the scheme in 562 terms of ES and expect intermediaries to perform monitoring and report its results. Answers 563 from both payers and intermediaries’ perceptions about payers’ demands pointed to this 564 direction. However, when it comes to express the most important indicators for them, 565 payers would mostly point to proxy indicators, for example, total area protected or number 566 of planted trees, instead of selecting more indicators linked to the target ES, for instance, 567 the decrease of nitrogen and phosphorus concentration on water. Potential reasons may 568 include the importance of indicators mostly for the sake of corporate sustainability reports 569 in which specific indicators are not needed, or due to an assumption that proxy indicators 570 are directly related to the desired effect in the targeted ES. 571 Providers are also interested in the PES impacts. Questions regarding motivation to 572 engage in the scheme were able to capture their interest in the ES itself, instead of solely the 573 payment, either in cash or in kind, for performing conservation practices in their properties. 574 Most of them pointed water resources conservation as one of the most important 575 motivations and declared that they would engage in the scheme even without economic 576 incentives. This does not necessarily mean that providers are not interested in economic 577 benefits, as at least half of the interviewees reported being motivated by increases in 578 productivity; however, these results contradict the simplistic idea of a “service provider” as 579 proposed in the literature (Wunder, 2005) and give clear indication that the so-called 580 providers understand themselves as ES beneficiaries too and are willing to cooperate for 581 conservation in case they receive technical and material support for that. Other recent 582 studies in Latin-American cases have pointed to similar results. Bremer et al. (2014) showed 583 that in Ecuador, landowners were motivated to participate in the program for a variety of 584 reasons, including a “high value placed on the water provisioning services of the páramo 585 [ecosystem] (Bremer et al. 2014, p. 122).” Intrinsic motivations have been also found in 586 parallel with utilitarian arguments among landowners in other Latin-American cases. For 587 instance, (Kosoy et al., 2008) found that indigenous groups participating in conservation 588 schemes in Mexico had religious reasons in addition to the concerns with the forest 589 provision of benefits. They considered forests as "sacred" places, with caves and other sites 590 being used as temples (Kosoy et al., 2008, p. 2080). In another study, a desire to “care for 591 the land” was found as one of the main motives for participation in PES schemes in Mexico 592 (Méndez-López et al., 2015, p. 695). 593 In part, the results found in the present study could be a consequence of the 594 discourses held in the field by intermediaries while negotiating with landowners. Most of the 595 intermediaries in the studied cases do not frame the proposed conservation practices as 596 something that would only benefit downstream users but most of them emphasize the 597 benefits that conservation will also have for those who live upstream. They argue that 598 protecting water bodies, like springs and creeks, is fundamentally important for the 599 providers themselves. During interviews, it became clear that providers tend to assume that 600 the conservation practices are good for their own sake. In fact, the same water resources are 601 shared and used by all, although distributed unevenly throughout the watershed. 602 6. Why those schemes keep working if there is no evidence of improvement of 603 the target ES yet? 604 Most of respondents from the three groups in all four schemes perceive the lack of 605 evidence as a matter of time, because the schemes under study are relatively new and that 606 the monitoring process is still incipient. They understand that the watershed would require 607 more time to respond to the conservation practices undertaken. Still, most of the 608 respondents from both providers and payers groups showed that they believe the scheme 609 will produce beneficial ES outcomes. This may be explained by the presence of the common 610 belief that “more forest leads to more water quantity” and better water quality (Kosoy et al. 611 2007, p. 451) in the discourses from both providers and payers, according to the interviews. 612 An additional explanation for the perseverance of such schemes relies on the extra 613 benefits it brings to actors. It appears that PES fit well into the agenda of several of them, 614 mainly of the intermediaries and payers. Intermediaries benefit from this approach due to 615 the central role they must play to sustain such a scheme as several tasks involved in the PES 616 design and implementation are required, such as property rights verification, constant visits 617 to the field, and one-by-one negotiation with local farmers (Vatn 2010). As noted by Pham et 618 al. (2010), intermediaries also act as information providers, mediators, watchdogs, 619 arbitrators and bridge builders, among a series of other functions. Vatn (2010, p. 1247) 620 concludes that “this explains why in PES schemes the intermediary is the dominant agent — 621 whether the state, firms or NGOs of various kinds. The intermediary defines the good, 622 establishes the group of ‘sellers’ and ‘buyers’ and even often set a predefined price.” 623 In parallel with that, the strong presence of the ES approach in the global 624 conservation agenda since MEA (2005) means that more funding through international 625 cooperation is available for this type of approach (Fisher and Brown, 2014). Among the 626 studied cases, there was government funding from The Netherlands and US, and 627 development banks. In addition, PES have been recently assuming an important position in 628 the environmental institutional configuration of Colombia through several laws and decrees 629 (Rojas-Sanchez, 2014). It has been promoted as a national conservation strategy that 630 culminated with the announcement (7th July 2017) of the new “Payment for Environmental 631 Services Policy” by the Colombian Ministry of Environment and the National Department of 632 Planning PES schemes are expected to escalate (MINAMBIENTE, 2017). Thus, working with 633 the ES approach could mean more funding available for the conservation business and, 634 therefore, more jobs. 635 From the payers’ side, although ES improvement was the first motivation to enter the 636 scheme for more than half of respondents, additional motivations play a strong role. These 637 include, for example, fulfilling corporate socio-environmental policy requirements, 638 improving corporate image and local relations, and complying with legal environmental 639 requirements. These additional payers’ motivations could explain a potential permanence of 640 payers in the schemes even without evidence of PES impacts on the target ES. 641 Although the literature on why firms engage in PES schemes as payers is almost non- 642 existent, there are several papers discussing the reasons behind the adoption of corporate 643 socio-environmental policy (CSR), which respondents pointed as one of the main reasons for 644 the adoption of PES. Hemingway and Maclagan (2004) discussed the hypothesis of 645 managers’ personal values influencing the adoption of CSR practices by their companies. 646 They proposed that cultural factors, such as religious beliefs and moral values of managers, 647 could play an important role. Vives (2006) presented a survey covering more than a 648 thousand small and medium firms in Latin America that had implemented responsible 649 practices of a variety of types. The major motivations found by Vives were increment profits, 650 but also religion/ethics, motivating workforce, and building relationships. In a survey with 651 corporate leaders, Ditlev-Simonsen & Midttun (2011) found that branding and reputation 652 were the primary CSR drivers among Norway companies. Babiak & Trendafilova (2011) 653 studied the motives behind the adoption of environmental management practices as part of 654 CSR among U.S. sports leagues. They found that strategic and legitimacy motives, connected 655 to the need to address institutional pressures and social expectations, were among the main 656 drivers of the CSR adoption. Executives also associated these choices with increasing chances 657 for financial collaborations with sponsors that were interested in environmental issues. 658 Thus, although economic motives exist, social norms are also important. Indeed, as pointed 659 out by organizational sociology scholars, organizations are particularly attentive to what 660 other organizations do when adopting certain practices (Hougland et al., 1980). Firms may 661 experience pressure from organizations upon which they depend, emulate other 662 organizations because of their success, or just follow the advice of professional associations 663 from their sector (DiMaggio and Powell, 1983). Exploring the extent to which such 664 “institutional isomorphism” applies to PES programs may have implications also for our 665 understanding of their endurance. 666 667 7. Will PES schemes survive in the long term if there is no evidence of the 668 expected environmental outcomes? 669 The results indicate that providers from the studied cases are willing to take part on 670 some costs to adopt conservation practices, for example, by setting aside part of their arable 671 land for vegetation protection, if there is an incentive for that, i.e. if they are not the only 672 ones to bear the costs. As one interviewee expressed, “I have always been concerned with 673 conservation and I expect that with the economic incentives I can dedicate my land to 674 protect water springs without affecting my income for that”. As most providers are engaging 675 motivated mainly by water resources conservation, if the desired environmental impact is 676 not met in the long term, there is a chance that they would drop out from the scheme not 677 because of lack of economic incentives, but mostly because their expected benefits in terms 678 of ES were not achieved. Under this assumption, evidence of PES environmental outcomes is 679 therefore important not only to keep payers on board, but also providers. 680 When intermediaries were asked if payers would only pay under the condition of PES 681 impacts being demonstrated they showed no consensus. This result is consistent with the 682 one obtained from the payers. When asked if they would stop paying in case the ES 683 improvement is not achieved with the time, payers also showed no consensus, with answers 684 relatively balanced along the scale. This lack of consensus could be due to the relative 685 importance attributed by each payer to the additional motivations they have to participate. 686 In addition, it may also reflect the ability of the payer to bear the risk: bigger payers would 687 be less worried about the ES outcomes, while small payers would be more keen to check if 688 the investments in conservation are giving returns. Although the small sample payers in this 689 study do not allow for a clear conclusion regarding this latter possibility, it provides an 690 interesting hypothesis for further research. 691 The question of whether PES schemes will survive or not in the long term remains 692 dependent on additional empirical research, however the present study demonstrates that 693 several motivations are behind the engagement of providers, intermediaries and payers in 694 PES schemes. Therefore, the perceived additional benefits the actors have while 695 participating in such schemes may partly explain the future long-term maintenance of this 696 conservation approach even under lack of evidence of their impacts. 697 8. Limitations of this study 698 We believe the results presented here can shed some light on the issue of the long- 699 term durability of PES schemes; however, it is worth mentioning that the scope of this study 700 is limited to the four cases presented and, therefore, more research would be needed to test 701 similar hypotheses in the field. In addition, the low response rate obtained among payers is 702 of some concern, as in the studied cases they are usually managers of large organizations 703 and have a limited time for interviews or questionnaires. However, several firms that 704 responded the questionnaire are among the largest ones in Colombia and their funding for 705 conservation is expressive. A second concern refers to the accessibility to the field in this 706 type of scheme. As intermediaries are the ones who know exactly who the schemes’ 707 providers are and where they live, it is very hard to perform interviews and questionnaires 708 without the presence of at least one manager or technician belonging to the intermediaries 709 group. We believe that in occasions in which it is not possible to avoid their presence while 710 interviewing, answers from providers may be partially biased as some of them would be 711 concerned with not threatening the trust relationship developed with the intermediaries. 712 Whenever it was possible, questionnaires were conducted without the practitioners 713 presence, and double-check of answers was performed through interviews. An additional 714 point is that interviews and questionnaires are only able to capture declared points of view, 715 however, communicated content is frequently different from actual behavior. Finally, we are 716 aware that the use of Likert Scale can introduce biases to the answers; in order to address 717 this, we built questionnaires based on preliminary interviews and additional interviews were 718 undertaken in order to confirm first answers or to receive explanations on the choices made. 719 720 5. CONCLUSION 721 There is not a straightforward answer for whether PES schemes will survive in the 722 long term in the absence of evidence of their environmental benefits. What it is clear is that 723 the lack of evidence has been already a matter of criticism from the scientific community 724 and a matter of concern among PES intermediaries. In this paper, we explored four water- 725 related PES schemes in which became clear that the evidence of environmental outcomes is 726 important not only for those who pay for the schemes but also for intermediaries and the 727 so-called providers. However, as PES schemes have important indirect benefits, mainly for 728 the intermediaries and payers, there is a chance that those schemes will survive the lack of 729 effectiveness evidence in the long term if that depends solely on these groups’ wishes. Still, 730 as providers understand themselves also as ES beneficiaries in these water-related schemes, 731 and even if most of them understand that ES changes may take some time to be observed, 732 the lack of evidence of PES environmental outcomes in the long run could discourage their 733 permanence in the schemes. 734 735 736 Acknowledgements 737 The authors are thankful for the knowledge exchange and insights provided during debates 738 with colleagues at IRI THESys and with Dr. Esteve Corbera (ICTA-UAB). The authors are also 739 thankful for the fieldwork support, interviews, and data provided by several organizations 740 involved in the Colombian cases. In particular, the authors are thankful for the field 741 assistance provided by ASOBOLO, ASOCAÑA, Cuenca Verde, TNC, Ecoforest-Cali, CIPAV, 742 Patrimonio Natural. This work was funded by the National Counsel of Technological and 743 Scientific Development - CNPq, Science Without Borders Program, Brazil [grant number 744 202203/2014-0], and the THESys Graduate Program, Humboldt-Universität zu Berlin, 745 Germany. IRI THESys is funded by the German Excellence Initiative. 746 747 REFERENCES 748 1. Aguinis, H., 2012. What We Know and Don’t Know About Corporate Social 749 Responsibility : A Review and Research Agenda. J. Manage. 38, 932–968. 750 doi:10.1177/0149206311436079 751 2. Andreoni, J., 1990. Impure Altruism and Donations to Public Goods: A Theory of 752 Warm-Glow Giving. Source Econ. J. 100, 464–477. doi:10.2307/2234133 753 3. Babiak, K., Trendafilova, S., 2011. CSR and environmental responsibility: motives and 754 pressures to adopt green management practices. Corp. Soc. Responsib. Environ. 755 Manag. 18, 11–24. doi:10.1002/csr.229 756 4. Barnaud, C., Antona, M., 2014. Deconstructing ecosystem services: Uncertainties and 757 controversies around a socially constructed concept. Geoforum 56, 113–123. 758 doi:10.1016/j.geoforum.2014.07.003 759 5. Bennett, G., Ruef, F., 2016. Alliances for Green Infrastructure: State of Watershed 760 Investment 2016. Washington, DC. 761 6. Bohensky, E., Lynam, T., 2005. Evaluating Responses in Complex Adaptive Systems: 762 Insights on Water Management from the Southern African Millennium Ecosystem 763 Assessment (SAfMA). Ecol. Soc. 10. 764 7. Boyd, J., Epanchin-Niell, R., Siikamäki, J., 2012. Conservation Return on Investment 765 Analysis: A review of results, methods, and new directions. Washington DC, USA. 766 8. Bremer, L., Vogl, A.L., Bievre, B. de, Petry, P., 2016. Bridging Theory and Practice for 767 Monitoring in Water Funds. 768 9. Calvache, A.S., Benitez, S., Ramos, A., 2012. Water Funds: Conserving green 769 infrastructure. A guide for design, creation and operation. Puntoaparte, Bogotá, 770 Colombia. 771 10. Carpenter, S.R., Mooney, H.A., Agard, J., Capistrano, D., Defries, R.S., Díaz, S., Dietz, 772 T., Duraiappah, A.K., Oteng-Yeboah, A., Pereira, H.M., Perrings, C., Reid, W. V, 773 Sarukhan, J., Scholes, R.J., Whyte, A., 2009. Science for managing ecosystem services: 774 Beyond the Millennium Ecosystem Assessment. Proc. Natl. Acad. Sci. U. S. A. 106, 775 1305–1312. doi:10.1073/pnas.0808772106 776 11. Chin, M.K., Hambrick, D.C., Trevin, L.K., 2013. Political Ideologies of CEOs : The 777 Influence of Executives’ Values on Corporate Social Responsibility. Adm. Sci. Q. 58, 778 197–232. doi:10.1177/0001839213486984 779 12. Crossman, N.D., Burkhard, B., Nedkov, S., Willemen, L., Petz, K., Palomo, I., Drakou, 780 E.G., Martín-Lopez, B., McPhearson, T., Boyanova, K., Alkemade, R., Egoh, B., Dunbar, 781 M.B., Maes, J., 2013. A blueprint for mapping and modelling ecosystem services. 782 Ecosyst. Serv. 4, 4–14. doi:10.1016/j.ecoser.2013.02.001 783 13. DiMaggio, P.J., Powell, W.W., 1983. The Iron Cage Revisited: Institutional 784 Isomorphism and Collective Rationality in Organizational Fields. Am. Sociol. Rev. 48, 785 147. doi:10.2307/2095101 786 14. Ditlev-simonsen, C.D., Midttun, A., 2011. What Motivates Managers to Pursue 787 Corporate Responsibility? A Survey among Key Stakeholders. Corp. Soc. Responsib. 788 Environ. Manag. 18, 25–38. doi:10.1002/csr.237 789 15. Ezzine-de-blas, D., Corbera, E., Lapeyre, R., 2015. Crowding-in or crowding-out? A 790 conceptual framework to understand motivations in payments for ecosystem 791 services, in: Resource Politics 2015. Institute of Development Studies, pp. 1–18. 792 16. Ezzine-De-Blas, D., Wunder, S., Ruiz-Pirez, M., Del Pilar Moreno-Sanchez, R., 2016. 793 Global patterns in the implementation of payments for environmental services. PLoS 794 One 11, 1–16. doi:10.1371/journal.pone.0149847 795 17. Ferraro, P.J., 2011. The Future of Payments for Environmental Services. Conservation 796 Biology 25(6), 1134-1138. doi: 10.1111/j.1523-1739.2011.01791.x. 797 18. Ferraro, P.J., Kiss, A., 2002. Direct payments to conserve biodiversity. Science 298, 798 1718-1719. 799 19. Fisher, J.A., Brown, K., 2014. Ecosystem services concepts and approaches in 800 conservation: Just a rhetorical tool? Ecol. Econ. 108, 257–265. 801 doi:10.1016/j.ecolecon.2014.11.004 802 20. Fondo Patrimonio Natural, CIPAV, CVC, DAGMA, PNN Farallones de Cali, EMCALI, 803 2014. Compensación por servicios ambientales hídricos en la cuenca del río Cali - 804 Valle del Cauca, in: Colección Los Incentivos a La Conservación: Una Mirada Desde La 805 Práctica. Fondo Patrimonio Natural, Bogotá, Colombia, p. 141. 806 21. Gammie, G., 2016. Monitoreo y Evaluación Económica de la Infraestructura Verde, in: 807 Biannual International Conference on Water Funds. Oral Presentation. Bogotá, 808 Colombia. 809 22. Goldman-Benner, R.L., Benitez, S., Boucher, T., Calvache, A., Daily, G., Kareiva, P., 810 Kroeger, T., Ramos, A., 2012. Water funds and payments for ecosystem services: 811 practice learns from theory and theory can learn from practice. Oryx 46, 55–63. 812 doi:10.1017/S0030605311001050 813 23. Gómez-Ochoa, M.I., 2016. Creación de la Corporación CuencaVerde como 814 Mecanismo para la Conservación de las Cuencas Abastecedoras de Agua de los 815 Embalses Riogrande II y La Fe, EPM. Medellín, Colombia. 816 24. Grima, N., Singh, S.J., Smetschka, B., Ringhofer, L., 2016. Payment for Ecosystem 817 Services (PES) in Latin America: Analysing the performance of 40 case studies. 818 Ecosyst. Serv. 17, 24–32. doi:10.1016/j.ecoser.2015.11.010 819 25. Hanson, C., Lugt, C. van der, Ozment, S., 2011. Nature in performance: Initial 820 Recommendations for Integrating Ecosystem Services into Business Performance 821 Systems. Washington DC, USA. doi:978-1-56973-779-8 822 26. Hanson, C., Ranganathan, J., Iceland, C., Finisdore, J., 2012. The Corporate Ecosystem 823 Services Review: Guidelines for indentifying business risks and opportunities arising 824 from ecosystem change, 2nd ed. World Resources Institute, Washington DC, USA. 825 27. Hemingway, C.A., Maclagan, P.W., 2004. Managers ’ Personal Values as Drivers of 826 Corporate Social Responsibility. J. Bus. Ethics 50, 33–44. 827 28. Higgins, J., Zimmerling, A., 2013. A Primer for Monitoring Water Funds. The Nature 828 Conservancy, Arlington, US. 829 29. Hougland, J.G., Shepard, J.M., Douglas, J., 1980. Organizations and Environments., 830 Academy of Management Review. Stanford University Press, Stanford, US. 831 doi:10.5465/AMR.1980.4288948 832 30. Huber-Stearns, H.R., Goldstein, J.H., Duke, E.A., 2013. Intermediary roles and 833 payments for ecosystem services: A typology and program feasibility application in 834 Panama. Ecosyst. Serv. 6. doi:10.1016/j.ecoser.2013.09.006 835 31. Kits, G.J., Adamowicz, W.L., Boxall, P.C., 2014. Do conservation auctions crowd out 836 voluntary environmentally friendly activities? Ecol. Econ. 105, 118–123. 837 doi:10.1016/j.ecolecon.2014.05.014 838 32. Kosoy, N., Corbera, E., 2010. Payments for ecosystem services as commodity 839 fetishism. Ecol. Econ. 69, 1228–1236. doi:10.1016/j.ecolecon.2009.11.002 840 33. Kosoy, N., Corbera, E., Brown, K., 2008. Participation in payments for ecosystem 841 services: Case studies from the Lacandon rainforest, Mexico. Geoforum 39, 2073– 842 2083. doi:10.1016/j.geoforum.2008.08.007 843 34. Kull, C.A., Arnauld de Sartre, X., Castro-Larrañaga, M., 2015. The political ecology of 844 ecosystem services. Geoforum 61, 122–134. doi:10.1016/j.geoforum.2015.03.004 845 35. Lele, S., 2009. Watershed services of tropical forests: from hydrology to economic 846 valuation to integrated analysis. Curr. Opin. Environ. Sustain. 1, 148–155. 847 doi:10.1016/j.cosust.2009.10.007 848 36. Méndez-López, M.E., García-Frapolli, E., Ruiz-Mallén, I., Porter-Bolland, L., Reyes- 849 Garcia, V., 2015. From Paper to Forest: Local Motives for Participation in Different 850 Conservation Initiatives. Case Studies in Southeastern Mexico. Environ. Manage. 56, 851 695–708. doi:10.1007/s00267-015-0522-0 852 37. MINAMBIENTE, 2017. Con la Política de Pago por Servicios Ambientales se da vía libre 853 a los incentivos económicos para la conservación. Available at: 854 http://www.minambiente.gov.co/index.php/noticias/3025-con-la-politica-de-pago- 855 por-servicios-ambientales-se-da-via-libre-a-los-incentivos-economicos-para-la- 856 conservacion (accessed 8.16.17). 857 38. Moreno-Padilla, P.H., 2016. Fondo Agua por la Vida y Sostenibilidad: Manejo integral 858 de cuencas hidrográficas en el valle geográfico alto del Río Cauca. Cali, Colombia. 859 39. Munoz Escobar, M., Hollaender, R., Pineda Weffer, C., 2013. Institutional durability of 860 payments for watershed ecosystem services: Lessons from two case studies from 861 Colombia and Germany. Ecosyst. Serv. 6, 46–53. doi:10.1016/j.ecoser.2013.04.004 862 40. Muradian, R., Corbera, E., Pascual, U., Kosoy, N., May, P.H., 2010. Reconciling theory 863 and practice: An alternative conceptual framework for understanding payments for 864 environmental services. Ecol. Econ. 69, 1202–1208. 865 doi:10.1016/j.ecolecon.2009.11.006 866 41. Naeem, B.S., Ingram, J.C., Varga, A., Agardy, T., Barten, P., Bennett, G., Bloomgarden, 867 E., Bremer, L.L., Burkill, P., Cattau, M., Ching, C., Colby, M., Cook, D.C., Costanza, R., 868 Declerck, F., Freund, C., Gartner, T., Gunderson, J., Jarrett, D., Kinzig, A.P., Kiss, A., 869 Koontz, A., Kumar, P., Lasky, J.R., Masozera, M., Meyers, D., Milano, F., Nichols, E., 870 Olander, L., Olmsted, P., Perge, E., Perrings, C., Polasky, S., Potent, J., Prager, C., 871 Quétier, F., Redford, K., Saterson, K., Thoumi, G., Vargas, M.T., Vickerman, S., 872 Weisser, W., Wilkie, D., Wunder, S., 2015. Get the science right when paying for 873 nature’s services. Science 347, 1206–1207. doi:10.1126/science.aaa1403 874 42. Norgaard, R.B., 2010. Ecosystem services: From eye-opening metaphor to complexity 875 blinder. Ecol. Econ. 69, 1219–1227. doi:10.1016/j.ecolecon.2009.11.009 876 43. Palmer, M.A., Filoso, S., 2009. Restoration of Ecosystem Services for Environmental 877 Markets. Science 325, 575–576. doi:10.1126/science.1172976 878 44. Pham, T.T., Campbell, B.M., Garnett, S., Aslin, H., Hoang, M.H., 2010. Importance and 879 impacts of intermediary boundary organizations in facilitating payment for 880 environmental services in Vietnam. Environ. Conserv. 37, 64–72. 881 doi:10.1017/S037689291000024X 882 45. Ponette-González, A.G., Brauman, K. a., Marín-Spiotta, E., Farley, K. a., Weathers, 883 K.C., Young, K.R., Curran, L.M., 2014. Managing water services in tropical regions: 884 From land cover proxies to hydrologic fluxes. Ambio. doi:10.1007/s13280-014-0578-8 885 46. Postel, S.L., Thompson, B.H., 2005. Watershed protection: Capturing the benefits of 886 nature’s water supply services. Nat. Resour. Forum 29, 98–108. doi:10.1111/j.1477- 887 8947.2005.00119.x 888 47. Quintero, M., Wunder, S., Estrada, R.D., 2009. For services rendered? Modeling 889 hydrology and livelihoods in Andean payments for environmental services schemes. 890 For. Ecol. Manage. 258, 1871–1880. doi:10.1016/j.foreco.2009.04.032 891 48. Rojas-Sanchez, A.M., 2014. The payment for hydrological environmental services. 892 Examining the experiences of Costa Rica, Mexico, Ecuador and Colombia. Ambient. y 893 Desarro. 19, 39–53. 894 49. Ruckelshaus, M., McKenzie, E., Tallis, H., Guerry, A., Daily, G., Kareiva, P., Polasky, S., 895 Ricketts, T., Bhagabati, N., Wood, S.A., Bernhardt, J., 2015. Notes from the field: 896 Lessons learned from using ecosystem service approaches to inform real-world 897 decisions. Ecol. Econ. 115, 11–21. doi:10.1016/j.ecolecon.2013.07.009 898 50. Tallis, H., Polasky, S., 2009. Mapping and Valuing Ecosystem Services as an Approach 899 for Conservation and Natural-Resource Management. Ann. N. Y. Acad. Sci. 1162, 900 265–283. doi:10.1111/j.1749-6632.2009.04152.x 901 51. Wunder, S., 2005. Payments for environmental services: Some nuts and bolts, CIFOR 902 Occasional Paper. Bogor, Indonesia. 903 52. Zanella, M.A., Schleyer, C., Speelman, S., 2014. Why do farmers join Payments for 904 Ecosystem Services (PES) schemes? An Assessment of PES water scheme 905 participation in Brazil. Ecol. Econ. 105, 166–176. doi:10.1016/j.ecolecon.2014.06.004 906 907 908 909 910 911 912 913 914 915 916