Ecological Economics 70 (2011) 2163–2174

Contents lists available at ScienceDirect

Ecological Economics

j ourna l homepage: www.e lsev ie r.com/ locate /eco lecon

Analysis

Towards a unified scheme for environmental and social protection: Learning fromPES and CCT experiences in developing countries☆

Luis C. Rodríguez a,⁎, Unai Pascual b,c, Roldan Muradian d, Nathalie Pazmino e, Stuart Whitten a

a CSIRO Ecosystems Science, Canberra, Australiab Department of Land Economy, University of Cambridge, UKc Basque Centre for Climate Change (BC3) & IKERBASQUE, Basque Foundation for Science, Bilbao, Spaind Centre for International Development Issues (CIDIN) Radboud University of Nijmegen, Netherlandse The World Bank, Sydney, Australia

☆ The findings, interpretations and conclusions exprauthors and do not necessarily reflect the view of the WDirectors or the governments they represent.⁎ Corresponding author.

E-mail address: luis.rodriguez@csiro.au (L.C. Rodrígu1 See three recent special issues by Engel et al. (2008)

and Zilberman (2008) in Environment and Developmen(2010) in Ecological Economics.

0921-8009/$ – see front matter © 2011 Elsevier B.V. Aldoi:10.1016/j.ecolecon.2011.06.019

a b s t r a c t

a r t i c l e i n f o

Article history:Received 3 November 2010Received in revised form 7 June 2011Accepted 17 June 2011Available online 28 July 2011

Keywords:Payments for environmental servicesConditional cash transfersPovertyConservationSocial protection

Environmental protection and poverty alleviation in the developing world are usually heralded as jointobjectives. However, these two goals are often associated with different sectoral policy instruments. Whileso−called payments for environmental services (PES) are increasingly being promoted for environmentalprotection, poverty alleviation is increasingly addressed by conditional cash transfers (CCT) program. Theseinstruments although aimed to achieve distinct objectives have a number of similarities and challenges intheir design and implementation phases. This paper elaborates on these similarities and develops a unifyinggeneric framework that is used to discuss the extent to which both approaches could be unified.

essed herein are those of theorld Bank Group, its Board of

ez).in Ecological Economics, Bulte

t Economics, and Pascual et al.

l rights reserved.

© 2011 Elsevier B.V. All rights reserved.

1. Introduction

Environmental conservation and poverty alleviation are the twokey policy areas that continue to receive increasing attention fromgovernments, donors and NGOs in the developing world. The complexlinks between poverty and environmental degradation continue to bethe focus of much research and debate and it is unlikely to be closed(e.g. Dasgupta, 2003; Duraiappah, 1998;Martínez-Alier, 2002; Ruijs etal., 2008). However, there is increasing consensus about the policyinstruments to be used to tackle both problems, albeit in a ratherpiecemeal approach. The current thrust is mostly on payments forenvironmental services (PES) to correct market failures that lead toexcessive levels of environmental degradation (Pagiola et al., 2005;Wunder, 2006, 2008).1 While on the other hand, conditional cashtransfer (CCT) programs are being implemented to correct marketfailures that lead to underinvestment in social protection in

developing countries (Chapman, 2006; Farrington and Slater, 2006;GTZ/BMZ, 2005).

PES constitute economic transfers (in the form of monetarypayments, or in kind rewards) aiming to compensate a target group(normally land holders) for the opportunity cost of providing positiveenvironmental externalities (environmental services) to a group ofactors that can pay by either creating a market or other informallydevised institutional mechanism between private agents, as well asthrough the State's public funds for the environmental services theyexpect to receive. These transfers when properly designed andimplemented link the payment with a set of environmental conditionsand landmanagement practices that promote the provision of sociallyvaluable environmental services (Bulte and Zilberman, 2008; Engelet al., 2008; Muradian et al., 2010).

While PES programs are primarily cost-effective conservationstrategies (Ferraro and Simpson, 2002), there is increasing interest onthe potential positive effect of PES on the livelihoods of poorlandholders (Grieg-Gran et al., 2005; Niesten and Rice, 2004; Pagiolaet al., 2005; Wunder, 2008; Zilberman et al., 2008). However, PESprograms have also received criticisms due to their potential negativeeffects in terms of maintaining asymmetric power distribution, andgenerating changes in the behavior of participants towards a visionlinking conservation and rent seeking (e.g. Karsenty, 2004; Kosoy andCorbera, 2010; Pascual et al., 2010).

Conditional cash transfers (CCT) in developing countries, on theother hand, are primarily designed as a means to poverty alleviation

2164 L.C. Rodríguez et al. / Ecological Economics 70 (2011) 2163–2174

or more generally as social protection mechanisms. CCT programs arean innovative approach that promotes immediate and intergenera-tional poverty reduction based on the provision of monetary transfersto socially vulnerable households, promoting a minimum level ofconsumption and investments in human capital (Rawlings, 2005;Winters and Davis, 2009). Similarly to PES, they are generally linkedto compliancewith a set of conditions aimed at promoting investmentin human capital. A typical application is offering cash transfers tohouseholds so that children attend school or are brought regularly tohealth centers (Das et al., 2004). However, it has also been argued thatthe direct provision of cash may generate disincentives to participatein the labor market and might not be a satisfactory solution to thepoverty problem if the transfer promotes dependency on external aid(Medeiros et al., 2008). Cash transfers show enough versatility andhave been adapted to different urban and rural contexts andcircumstances (e.g. Schubert and Slater, 2006; Skoufias, 2005) andthere is some emerging evidence of their positive impact on ruralpoverty alleviation (e.g. Rawlings, 2005; Veras Soares et al., 2008) andindirectly on the environment by investing the transferred money inland rehabilitation activities (e.g. Standing, 2007), restoring thecapacity of the agroecosystem to provide goods and services such asfood and pastures and improving livelihoods.

The evidenced effect of some PES schemes on the livelihoods ofpayment receptors has been the starting point of the exploration of itspotential as a poverty reduction tool (Grieg-Gran et al., 2005; Suyanto etal., 2007), attracting the attention of donors not just for their positiveenvironmental impacts but also for the hoped pro-poor effects of thepayments (Wunder, 2008). In a similarway, due to the versatility of CCTprograms, there is a call for revisiting CCT to take advantage of theirpositive effects on poverty reduction and also targeting aspects beyondhuman capital such as livestock and seeds which can have importantenvironmental consequences (Cuesta, 2007; Handa and Davis, 2006).

The nexus between poverty and environmental degradation arecomplex (e.g. Martínez-Alier, 2002) but it is now considered thatenvironmental degradation is a major barrier for poverty reductionand, at the same time, reaching environmental conservation goalsrequires progress in the reduction of poverty. Thus, in order to besuccessful, poverty reduction and environmental initiatives should belinked and implemented together (see Sachs and Reid, 2006).However, the usual policy approach is to deal with poverty andenvironmental degradation in separate ways through either CCT orPES. The CCT and social protection literature shows that linkingprograms might be desirable if they operate efficiently in the samegeographic area and benefit the same population groups, theselinkages usually reduce the transaction and operating costs, improvethe flow of information and the level and type of support for thepayment receptors (see Grosh et al., 2008). Thus, CCT are often linkedto microfinance initiatives, public work programs, access to agricul-tural extension agents, or adult training, improving the overall impactof the intervention (World Bank, 2009).

The purpose of this paper is to build on the similarities of bothapproaches to shed light on the potential for linking CCT and PES. Inthis paper we bridge ideas that are well developed in the still largelyseparate CCT and PES literatures in order to develop a unifyingconceptual framework that based on the high cost effectiveness ofboth programs could serve to design economic transfers to deal withboth environmental protection and poverty alleviation in the samegeographic areas. Of course such unified scheme should not follow a“one size fits all” and would not necessarily be applicable to allsituations and contexts. Under some conditions individualized PES orCCT might be the best choice, while in some others decision makersmight evaluate a unified approach if the potential reduction in costs,including implementation and transaction costs, might compensatethe increase in complexity.

The rest of the paper is organized as follows, in the followingsection we review the conceptual similarities between PES and CCT.

We then present a framework for unifying both approaches. Someimportant considerations for the design of money transfers arediscussed, and some recommendations are proposed for designing anew breed of joint PES-CCT programs in rural areas of developingcountries.

2. PES and CCT: From Concept to Practice

Although both have different objectives, PES and CCT instrumentshave similarities in terms of their respective conceptual designs. Weargue that from a broad perspective both programs can be consideredto be based on i) market based interventions aimed towardsinternalizing an externality in which ii) a number of paymentreceptors are identified in a targeting exercise to iii) receive apayment or reward in kind if iv) they fulfill a set of conditions. Table 1sketches out the main characteristics of PES and CCT in terms of thefour defining aspects identified above. Table 2 also presents acomparison between the transfer amounts, target mechanisms,conditionalities and monitoring of compliance of PES and CCTprograms across countries in Latin America, Asia and Africa.

2.1. PES and CCT as Market Based Interventions

Both policy interventions aim at correctingmarket failures (under-provision of a valuable environmental or social service, respectively)by adjusting individuals' behavior through economic incentives. InCCT programs, interventions targeting poor populations are designedto match individual households' decisions on investment in humancapital with wider social preferences, mainly in the areas of educationand health (World Bank, 2009). It is argued that an increase ineducation or health by an individual generates positive benefits onothers. This positive externalities are not rewarded by market forces,therefore it is expected that poor households under-invest in humancapital from a social optimum perspective (Das et al., 2004; de Janvryand Sadoulet, 2004).

The core justification for the design and implementation of PESschemes is similar. It aims at internalizing environmental externalitiesgenerated by individual landowners due to their under-provision ofenvironmental services (Engel et al., 2008; Muradian et al., 2010;Pagiola et al., 2002). PES aim at aligning the private and socialincentives for investment in environmental assets through a directpayment or reward to individual landholders, thus facilitating thesupply of improved environmental conservation outcomes.

2.2. Targeting the Payment Receptors of the Compensation Payment

Although CCT and PES interventions are based on voluntaryparticipation, both programs are based on a prior targeting exercise oftheir payment receptors as a means of enhancing their cost-effectiveness.

The choice of a particular targetmechanism depends on the type ofinformation available about the potential payment receptors and theobjective of the intervention. CCT and PES rely on both geographic andhousehold/plot targeting approaches. Geographic targeting is rela-tively simple to administer as different areas are ranked by someindicative measure relevant to either CCT intervention, e.g. infantmortality rates, education levels, access to water or electricity, or PES-related goals e.g., biome distribution, biodiversity richness, landscapeconnectivity. Resources are then allocated in proportion to theirexpected impacts. Hence, in terms of CCT regions with poorer humancapital indicators tend to receive higher per capita transfers, and asregards PES, land areas with higher potential to supply environmentalservices or being associated with a higher likelihood of degradationleading to the loss of valuable environmental services are usuallyeligible for PES, although the size of the payment might also vary dueto other factors such as whether the areas are also targeted by other

Table 1Comparison between PES and CCT from a conceptual perspective.

PES CCT

Market basedinstruments

• Designed to internalize an environmental externality whenindividual decisions do not match social preferences because ofmarket failure.

• Interventions designed to match individualdecisions with social preferences internalizingdirect and learning externalities mainly in theareas of education and health.

For a number of selectedpayment receptors

• Target to increase payment efficiency.• Main criteria: environmental quality.• Geographic.• Household/plot target.

• Target to increase payment efficiency.• Main criteria: poverty and vulnerability.• Geographic.• Household target.

A payment amount istransferred

• National programs financed by public funds from country taxesand donors. Private schemes might be also implemented asmarkets for environmental services.

• Payment should compensate for the revenue foregone andopportunity cost of adopting landmanagement activities for theprovision of environmental services.

• Cash or in kind payments.• Payment frequency regular and predictable designed to reduceentry barriers towards environmental protection.

• Payments contribute to safety nets and mitigate risk.

• Financed by public funds from tax revenuesand international lending from donors. Privateparties can participate from the supply side, asproviders of education and health services.

• Payment covers the direct costs of sendingchildren to school or to medical checkups, andthe child income lost due to school attendancerather than working.

• Cash or in kind payments.• Payment frequency regular and predictabledesigned to reduce entry barriers towards theuse of education and health systems.

• Payments contribute to safety nets and miti-gate risk.

Payment receptors mustcomply with a set ofconditionalities

• Conditions designed to adjust individual behavior to matchsocial environmental objectives.

• Design includes mechanisms to verify compliance and definesanctions.

• Conditions designed to adjust individual be-havior to match socio-economic objectives ofthe society.

• Design includes mechanisms to verify compli-ance and define sanctions.

2165L.C. Rodríguez et al. / Ecological Economics 70 (2011) 2163–2174

conservation programs. For example, Honduras and Peru's socialprograms use geo-statistical data and human development indexes toidentify regions where CCT poverty interventions might be bestimplemented (e.g. Glewwe and Olinto, 2004; Schady, 2002a, 2002b).In an analogous way, PES use geo-referenced data and ecologicalindexes to identify priority areas and to define plots for being eligiblefor being part of PES programs (e.g. Asquith et al., 2008; Imbach,2005).

At the household level, CCT programs use different poverty orsocial categorization target approaches; however the use of a proxymeans test is the preferred option to select payment receptors when itis not possible to directly observe the actual income of paymentreceptors (World Bank, 2009). This approach uses household data andeconometric models to estimate the coefficient associated withselected socio-economic variables that are then used to predicthousehold income, which in turn are used to rank the eligibility ofpotential participants, (Perez-Ribas et al., 2008). Likewise, environ-mental metrics are frequently used in many PES and otherconservation payment programs. Such metrics utilize plot-levelbiophysical data and models to predict/estimate ecological outcomes,e.g., biodiversity enhancement or provision of environmental services,and express it as a single unit. The metric value is then used tocompare and rank the expected environmental additionality ofdifferent plots and select those that provide the most benefits perdollar invested in the program (Stoneham et al., 2003).

The Mexican CCT programOportunidades uses the proxy means testto classify households from previously identified geographic areas aseligible for treatment (“poor”) or ineligible (“non-poor”) based oninformation collected from socio-economic surveys (Skoufias andMcClafferty, 2001). In an analogous way, for PES, the VictorianBushtenderprogram in Australia uses a “habitat hectare scoringmethod” as the environmental metric to allocate funds for biodiver-sity conservation (Department of Sustainability and Environment,

2008) and for many PES carbon schemes like the PROFAFOR project inEcuador, the amount of the payment was originally estimated using ametric based on vegetation types and soil criteria and latercomplemented with plot measures and model outputs (Wunder andAlban, 2008).

2.3. The Type of Compensation Payment/Reward

Both payment mechanisms are broadly similar in terms of i) theorigin of the funds, ii) the estimation of the amount to betransferred, iii) the payment vehicle, and iv) the frequency of thepayments.

First, there are similarities with respect to the origin of the funds.Most CCT programs are financed by public funds from country taxrevenues complemented by international lending from donors. Forexample the consolidation and expansion of the Mexican nation-wideProgresaprogram (currently known as Oportunidades) was partiallyfunded by a one billion US dollar loan from the Inter-AmericanDevelopment Bank and $1.3 billion collateral from the MexicanGovernment (IADB, 2002). In fact, the dependence of many CCTprograms on external financial sources is a major concern for theirlong-term viability, particularly as they increasingly become anintegral part of countries' social assistance strategy (Rawlings, 2005).

Something similar occurs for long-term, large-scale publiclyfunded PES programs such as those implemented in Costa Rica,Mexico or China in which the state makes use of fiscal instruments,mainly taxes, and donors' funds to pay providers of environmentalservices that are strategic for those countries. For instance, the CostaRican program, Pago por Servicios Ambientales (PSA) was originallyimplemented using revenues from fuel taxes and is currently co-financed through external funds from World Bank loans and GEFgrants which cover about 45% of the PSA total running costs (Pagiola,2008). Such public PES schemes are seen as less efficient than when

2166 L.C. Rodríguez et al. / Ecological Economics 70 (2011) 2163–2174

the payers are private agents but they are generally larger in scope,have lower transaction costs because of the use of existing institutionsand have the state providing legitimacy, which many private schemesare lacking (Vatn, 2010; Wunder, 2005).2

Secondly, the amount of the payment is critical for the program'scost-effectiveness. Interestingly, despite the different program objec-tives, the transferred amounts across the two of them are similar (c.f.Table 2). Regarding CCT, the amount of the transfer is generallyassociated with the depth of poverty being addressed. An interna-tional rule of thumb is that a poverty motivated cash transfer shouldrepresent between 20% and 40% of the per capita total poverty line inorder to be meaningful to the payment receptors even though themajority of CCT programs transfer less than the lower bound (Handaand Davis, 2006). For example, while in the case of CCT programstargeting investments in education, cash transfers ought to cover theincremental costs of education, what is actually transferred is theactual costs of sending children to schools, which involves school fees,supplies and transport, plus the opportunity cost of their time, i.e.,foregone income by children (PIDS, 2007). A review of the educationand health related CCT literature indicates that this amounts to aboutUSD15 per month and about USD10 for health per child enrolled inthe programs, respectively, albeit payments may increase by up to90% to support specific actions from disadvantaged groups likepromoting female secondary education (e.g. Barrera-Osorio et al.,2008; Gökalp, 2006). Although such amounts are per child, there isnormally a maximum cap on transfers to single households. Forinstance, in Latin American CCT programs, the total transfer per familyrepresents 8–23% of the per capita poverty income line with mostprograms providing transfers of about 15%, or between 10 and 30% ofpoor households' consumption expenditure (Handa and Davis, 2006).

As regards PES, the monetary transfer needs to at least compensatethe landholder for the revenue foregone or opportunity cost of adoptingthe desired land conservation activities. There is however someevidence that shows that this is not achieved in all cases or somebenefits fromPES are nonmonetary andhard toquantify (see e.g. Porraset al., 2008; Rosa et al., 2003). The contribution of PES to householdincome can be significant. For instance in Mexico's PSAHprogram (Pagopor Servicios Ambientales Hidrológicos), the annual payment to house-holds is about US$ 27 per hectare in regions where most households'daily earning are less than US$ 7.5, although areas with specific landcover can receive payments up to 30% higher. In Ecuador the incomeaccrued to the beneficiary households of the PROFAFOR PES programrepresents over US$ 21 per month per household, while in the PESprogram in Virilla, Costa Rica, the transferred amounts might representabout 16% a household's income (Miranda et al., 2003; Muñoz-Pinaet al., 2008). In addition, at the community level, the contribution of PEScan also be significant. For example, in Zimbabwe thePES-likeCAMPFIREprogram forwildlife conservation is estimated to constitute up to 24% ofall locally earned income and exceed all other forms of local income andgovernment grants (Bond, 2001). Table 2 compares the per householdannual transferred amounts of CCT and PES implemented in variouscountries in Africa, Asia and Latin America, where both types ofprograms can be found.

Third, the selection of the payment vehicle for CCT and PESresponds to the same factors, including project objectives, marketassessment, cost effectiveness, and beneficiary preferences. Bothsocial and environmental programs use cash or in-kind transfers, e.g.,food as payment vehicles. For example, the Mexican CCT programOportunidades or the Paraguayan Tekopora are exemplars in the use of

2 An important difference between PES and CCT is that PES can be also designed andimplemented as private schemes in which the buyers of the service directly pay theproviders usually at the local scale. Many local PES experiences exist and these arebeing extended. Examples include for example those from Pimampiro—Ecuador orSanta Rosa—Bolivia (Asquith et al., 2008; Echavarria et al., 2002).

cash as the main payment mechanism.3 By contrast, the VulnerableGroup Development project operating in rural Bangladesh and whichdistributes fortified food to disadvantaged women is an example ofusing in-kind transfers where payment receptors prefer food ratherthan cash and where the existing infrastructure of storage depotsfacilitate the logistics (Ahmed et al., 2004).

As regards PES, cash is the most common form of payment. Forexample in the PROFAFOR project in Ecuador, each beneficiarylandholder receives payment in cash based on the evaluation of thequantity of carbon sequestrated in their land (Wunder and Alban,2008). Other PES projects also involve in-kind payments, such asthose implemented in Santa Rosa, Bolivia, where one artificial beehiveis given each year to the payment receptors for forest conservation ina buffer zone around a national park (Asquith et al., 2008).

Despite the different nature of PES and CCT programs, cashtransfers are often preferred by the implementing agencies becausethey address information asymmetries by giving households discre-tion over how to best allocate the assistance received, be it on food,healthcare, housing or other needed services based on the condition-alities attached to the transfer. In addition, according to the WorldBank (2009), cash payments might better avoid the creation ofsecondarymarkets, reducing price distortions and theymight bemorecost-effective due to lower transaction costs and flexibility topolicymakers in adjusting the level of the transfer over time andacross populations.

Fourth, the frequency of the payment affects the final impact ofthe overall transfer. Both PES and CCT are designed with a regularand predictable frequency of payments to help beneficiaryhouseholds to reduce entry barriers to the programs and mitigateany risk from participation. For example, the payment receptors ofthe Colombian CCT ProgramFamilias en Acción receive a bi-monthlypayment for ensuring that a child achieves at least 80% of schoolattendance. In this typical case, in order to reduce entry barriers forchildren education either one third of the bimonthly allowedbenefit is retained in a savings account that families cannot accessuntil a week before enrolment for the next school year (Fiszbeinand Schady, 2009).

The schedule of actual payments in PES is commonly adapted totake landowners' investment constraints into account. For instance,an initial payment is normally provided for the services that currentland uses are already providing. These first installments help toremove potential entry barriers and capitalize landholders to financethe implementation of the desired new land management practices(e.g. Pagiola et al., 2005; Wunder and Alban, 2008).

In addition, PES and CCT programs play an important role as safetynets during times of economic hardship. There is some evidence aboutthe effect of CCT in mitigating risk, smoothing consumption andavoiding that the poor become poorer and increase the poverty gap(Handa and Davis, 2006) as well as the increasing role of PESpayments as part of the risk management strategy of landholdersbecause fixed, regular and predictable PES help to reduce incomefluctuation when returns of crop production vary with weather andmarket conditions (Pagiola, 2008).

2.4. Conditionality Criteria

The rationale for including conditionalities in both CCT and PES,also known as co-responsibilities (in the CCT literature jargon) is tocompel individuals to adjust their behavior towards the desiredprogram's goals. By imposing conditions or co-responsibility policymakers provide incentives for households to take actions that they

3 Recent evaluations of these programs suggest that the income effect of thesepayments can only explain up to 50% of their impact, and other program componentssuch as talking and awareness sessions have delivered unexpectedly high impacts(Perez-Ribas et al., 2008).

Table 2Parallel between PES and CCT implemented in selected countries of Latin America, Asia and Africa.

Program name Householdbenefitper yearUS$*

Targeting Conditionalities Monitoring Sources

Latin AmericaEcuadorCCT Bono de

DesarrolloHumano

360 -Proxy means -Bimonthly healthvisits

-No verification of compliance World Bank(2009)

−90% schoolattendance

PES PROFAFOR 252 -Geographic -Active plantationmanagement,including firecontrol andsurveillance, andkeeping outlivestock

-At the plot level, all contractedareas are visited at least onceannually

Wunder andAlban(2008)

-Certification companiesscrutinizethe annual carbon uptake.-Members would legally haveto reimburse the paymentsreceivedif they do not fulfill the terms.

ColombiaCCT Familias en

Accion552–768 -Geographic -Growth control

and developmentcheckups

-Bimonthly verification World Bank(2009)

-Proxy means −80% schoolattendance

-School principal monitoringattendance

PES SilvopastoralProject

607 -Geographic -Maintain orswitch to landuses that provideenvironmentalservices.

-Annual payments are madeafter land use changes havebeen monitoredin the field.

Rios andPagiola(2009)

-Self selectionof individualswith minimalfarm and herdsize criteria

-Output basedsystem. Paymentsare by increasingan EnvironmentalService Index

-Switch to land uses thatreduceservice provision would incurpaymentreduction

BrazilCCT Bolsa Familia 506–703 -Geographic -Vaccine

schedules, regularcheckups children

-Education bimonthly verificationusing information consolidatedby localmunicipalities

World Bank(2009)

-Checkupspregnant andlactating women

-Meanstesting

-85% schoolattendance

-Health, twice a year, using infothat health providers enteredinto a national database-Participation

parent–teachermeetings

PES Bolsa Floresta 517 -Geographic -Commitment tozero deforestation

-Annual monitoring bysatellite images and analyzedby partnering institutions

Viana (2008)

-Self selection -Participation indwellers'associations

-Yellow and red cardsa

Families who havedeforested a crop area upto 50% larger than thecrop area in 2007will receive a ‘yellow card’and must explain to theassociation the reasons for havingincreased deforestation. Thosewith a yellow card who continuedeforestation in thefollowing year will receive a ‘redcard’ and the payment will besuspended. If thenew crop area is extendedmore than 50% (compared tothe 2007 crop area) a red card willbe given immediately and paymentscease. The families given either twoconsecutive yellow cards or three inalternate years will be excludedfrom the program

(continued on next page)

2167L.C. Rodríguez et al. / Ecological Economics 70 (2011) 2163–2174

Table 2 (continued)

Program name Householdbenefitper yearUS$*

Targeting Conditionalities Monitoring Sources

Latin AmericaMexicoCCT Oportunidades 371–1068 -Geographic -Medical checkups -Bimonthly monitoring of

compliance using data collectedby local education and healthservice providers

World Bank(2009)-Attendance

health andnutrition lectures

-Proxy meanstest

-80% monthlyschool attendance,93% annually-Completiongrade 12 beforeage 22

PES PSA-CABSANiños Heroes

545 -Geographic Not being reneficiaryof any other PESprogram

-A monitoring process is consideredin the design but rarely done.

Corbera et al.(2009)

-Self selection. -Make projects complywith rulesestablished for small-scaleafforestationand reforestation projectsunder the Kyoto Protocol's CDM

-Participants might return paymentsin case of no compliance.

-A forest management plan,and shows that PES activitieswere additional (i.e., thedevelopment of land-useactivities for ES provisioncould not have been possiblewithout ES payments).

-Participants were entitled to adeferral in the application of sanctionsif they showed that failure to complywas due to an uncontrollable reason.-No sanctions have to date been imposed.

NicaraguaCCT Programme

Atencion a Crisis325–378 -Geographic -Health conditions but

no implemented-Bimonthly verification usinginformation collected from theservice providers

World Bank(2009)

-Proxy means test -85% school attendance-Deliver teacher transferto teacher

PES SilvopastoralProject

592 -Geographic -Maintain or switch toland uses that provideenvironmental services.

-Annual payments are made afterland use changes have been monitored in the field.

Rios andPagiola(2009)

-Self selection ofindividuals withminimal farm and herdsize criteria

-Output based system.Payments are by increasing anEnvironmental Service Index

-Switch to land uses that reduceservice provision would incur payment reduction

AsiaCambodiaCCT Cambodia

Education SectorSupport Project

60 -Geographic -Enrollment in schooland regular attendance.

-School monitors ongoingattendance and passing grade once a year

World Bank(2009)

-Scoring by acommittee

-Maintain a passing grade

PES BiodiversityConservationPayments

120–160 -Geographic -Stop hunting key species -Local monitoring by villagers Clements etal. (2010)-Certified by external agency

-Self selection -Abide a land use plan -Output based, no service no payment-Self enforcement within the community

IndonesiaCCT Programme

KeluargaHarapan

65–152 -Proxy means test -Medical checkups childrenages 0–6 and pregnant andlactating women

-Sporadic monitoring of compliancebut designed to be verified every 3 months

World Bank(2009)

-85% school attendancePES Tlekung and

CidanauWatershedpayments

42–132 Geographic based oncontribution tosedimentation

-Tree planting at a density of500/ha in identified critical landunder their individual ownership

-A team representing the involvedparties verify planting and maintenance

MunawirandVermeulen(2007)-Survival of seedlings and tree

maintenance-An external agency facilitate thecontracts, payments and compliance

-Put in place and maintainhigh quality terracing

-Output based: The payments aredistributed after verifying the target was fulfilled.

PhilippinesCCT Pantawid

PamilyangPilipino Program

202–342 -Proxy means test -Medical checkups forchildren and pregnant women

-Quarterly monitoring of complianceusing monthly data collected fromthe service providers

World Bank(2009)

-85% school attendancePES Bakun

WatershedProtection

31 Geographic -Adopt land managementplan and tree planting butno specific target

-Private agreement between theinvolved parts for monitoring purposes

Beria et al.(2009)

2168 L.C. Rodríguez et al. / Ecological Economics 70 (2011) 2163–2174

Table 2 (continued)

Program name Householdbenefitper yearUS$*

Targeting Conditionalities Monitoring Sources

AfricaKenyaCCT Cash transfer for

orphan andvulnerablechildren

252 -Geographic -Vaccine schedules, regularcheckup children growthmonitoring and vitamin Asupplement

-Compliance verified at least bimonthly for childrenconditions using information collected from serviceproviders and once a year for adult awareness sessions.

Pearson andAlviar(2009)

-Community target −80% school attendance-Adult attendance awarenesssessions

PES WildlifeConservationLease Program

306 Geographic, selfselection

-Retain ownership of their land -Violation of conditions involves termination of payments. Reto-o-Reto(2006).

-Leave land open, uncultivatedand unsubdivided

-Payments might be restored if fencing is removed World Bank(2008).

-Graze livestock sustainably -Independent monitoring and evaluation of programimplementation, lease compliance and impacts, withstakeholder participation

-Share both pasture and wateramong livestock and wildlife-Allow free movement oflivestock and wildlife.

MozambiqueCCT Bolsa Escola 21 -Geographic -90% School attendance -School attendance is registered by teachers and monthly

monitored by representatives of the Ministry of EducationMassingarelaand Nhate(2006)

-Household evaluationby Ministry ofEducation

PES NhambitaCommunityCarbon

34 -Geographic afterevaluation ofconditions for carbonsequestration

-Involvement in forest and firemanagement, forestrehabilitation or agroforestry

-Monitoring of compliance is carried out by communitytechnicians with support from the carbon broker technicalteam

Jindal et al.(2008)

-Restrict timber extraction to theamounts defined by a resourceinventory

-International agencies audit the carbon-Community/broker disputes are resolved by consultationand individual/broker disputes are resolved dependent onthe infraction based on guidelines defined in the contracts.

a Authors' calculations based on information from the sources.

4 PSA-CABSA stands for “Programa para Desarrollar el Mercado de ServiciosAmbientales por Captura de Carbono y los Derivados de la Biodiversidad y paraFomentar el Establecimiento y Mejoramiento de Sistemas Agroforestales”.

2169L.C. Rodríguez et al. / Ecological Economics 70 (2011) 2163–2174

would not ordinarily take on their own (Das et al., 2005). However,conditionalities can be costly to monitor and verify. In other words,they can be constrained by the capacity of the agencies to put theminto practice (Schubert and Slater, 2006).

In the Mexican CCT program Oportunidades, cash transfers aresubject to school attendance and health controls. While theseconditionalities can increase the demand for health and educationservices they can also potentially reduce the capacity of paymentreceptors to choose the desired providers, ultimately affecting thesuccess of the intervention since the quality of these services caninfluence the decision of payment receptors to drop out the program(Alvarez et al., 2008). In the case of PES, conditionalities canpotentially increase the supply of a given environmental service butat the same time also reduce the capacity of landholders to innovateor choose management practices.

Interestingly, Table 2 indicates that in most cases the estimatedtransferred amounts per household in terms of the level of benefitsfrom CCT is comparable to the transfers from PES programs. Thissuggests the high potential of PES to contribute to poverty alleviationinitiatives if properly targeted to achieve that goal. For example, inKenya, the Cash Transfer for Orphan and Vulnerable Children providesfamilies with about US$ 250 per year, with the condition of attendingregular medical checkups and sending children to school, while a PESdesigned to conserve wildlife and conditioned to environmentalprotection actions provide the average beneficiary family with apayment of US$ 306 per year, which is usually invested in medicinesand school fees (World Bank, 2008).

Table 2 also highlights the large differences regarding the levelof benefits provided by CCT when implemented in middle incomecountries like Mexico and Brazil, versus poorer countries such asNicaragua, Cambodia or Mozambique. The transferred benefitsfrom PES also show large differences by the type of environmental

service considered in the scheme. Programs paying for globalenvironmental benefits such as those involving carbon sequestra-tion, e.g. the Ecuadorian PROFAFOR or the Mexican PSA-CABSAestablished in 2004,4 or paying for biodiversity conservation e.g.,Cambodia biodiversity projects, provide larger amounts to theirpayment receptors than programs that provide local benefits suchas watershed protection. This is due to the linkages of the formerprograms with international carbon prices, or wealthy foreigntourists while the value of the latter program is constrained by thelower willingness to pay of local communities from developingcountries.

3. Towards Designing Unified Conditional Programs

Decision makers are becoming more aware of the linkages orcoupling between environmental and socio-economic systems, alsoknown as social–ecological systems. It is now recognized thatenvironmental degradation is a major barrier for poverty reductionand, at the same time, reaching environmental conservation goalsrequires progress in the eradication of poverty. Thus, in order to besuccessful, poverty reduction and environmental initiatives could inprinciple be linked together (see Sachs and Reid, 2006). However, theusual policy approach is to deal with poverty and environmentaldegradation in separate ways through either CCT or PES. The attemptof merely including social criteria within PES schemes in order tomake them look pro-poor have driven towards inefficient conserva-tion outcomes. While adding social variables to add to the sociallegitimacy of PES programs at the expense of efficiency is seen as an

2170 L.C. Rodríguez et al. / Ecological Economics 70 (2011) 2163–2174

acceptable approach in a second-best world, in other circumstances,this masks rent seeking behavior as discussed by Muñoz-Piña et al.(2011). In any case, practitioners and decision makers have to definethe equity/fairness criteria used in PES and the relative importancegiven to equity and efficiency in their programs (Pascual et al., 2010).We consider that the inclusion of social criteria within PES is animportant issue also as it can also open the door to the inclusion ofsome of the lessons learnt in monetary transfers from social programssuch as CCT, or even to design cash transfers with broaderenvironmental and social objectives.

An important limitation is the lack of recognition that despite theirdifferent objectives, both PES and CCT share a set of basic commonprinciples in their design. Development practitioners familiar with theuse of conditional cash transfers (CCT) to improve human capitalmight be interested in the general design of PES (Pattanayak et al.,2010) and vice versa. There is then a need to bridge the PES and CCTliterature to communicate the principles to a broader audience.

The conceptual and design similarities of PES and CCT initiatives,their potential cost effectiveness and growing evidence that environ-mental payments might have socio-economic impacts and vice versa,provide an interesting platform from which to propose the design ofeconomic incentives aimed at achieving both environmental andsocio-economic goals. The CCT literature recognizes that linkingprograms operating in the same geographic region might be cost-effective, reducing operating and transaction costs (Grosh et al., 2008;World Bank, 2009).The question is thus whether the reduction incosts of a unified scheme might justify its increase in complexity.

Of course, such a unified payment scheme would not be expectedto follow a simple “one size fits all” design, but rather it would need tobe adapted to reflect the differences in setting up both theconservation and poverty alleviation goals, their relative importanceand the particular environmental, socio-economic and politicalcontexts in which such unified scheme would be implemented orindividualized PES or CCT might be the first choice. In the reminder ofthe paper we refer to such potential unified scheme as “Payments forEcosystem Services and Poverty Alleviation (PESPA)”.

Based on Sections 2 and 3, the design of PESPA schemes wouldrequire to i) identify the payment receptors by a targeting exercise;ii) define the nature and level of the payment and iii) identifyconditionality criteria so that payments would be transferred ifpayment receptors fulfill a number of conditions aimed at promotingthe desired joint objective. The next section briefly develops somegeneral principles for PESPA that could be used as a set of ideas toguide the preparation of more detailed operational manuals bypractitioners required to implement unified payment schemes underparticular settings.

3.1. Targeting in PESPA

A key aspect in the design of a unified payment scheme would beto promote its program efficiency by transferring the availableresources to the set of payment receptors that generates the largestsocietal benefits. Demonstrating efficiency gains from targeting is nota simple undertaking considering that the poor are not a homoge-neous group. Poverty is a multidimensional concept that includessubjective components and there might remain large uncertaintiesabout the level of supply of environmental services after adopting theproposed land conservation actions (Matin and Hulme, 2003; Pascualet al., 2010). Thus inclusion or exclusion targeting errors might beexpected for both environmental and socio-economic variables, e.g.falling to exclude (include) the poor from biodiversity poor (rich)areas, respectively.

The combined social and environmental rationale of the unifiedscheme weakens the vision of uniform and universal provision ofbenefits to the wider population of a given area. This is becausealthough in the casewhere every household is eligible for a given level

of social protection, not everyone is likely to be a provider ofenvironmental services. Therefore a set of social and environmentaleligibility criteria should be developed in consultation with benefi-ciary communities in order to define the required rules to distinguishwho might be able to participate in the program and who might beeligible to receive payments, especially in regions where the incidenceof poverty is high and everyone could consider himself eligible forsupport (see Semu-Banda, 2009).

The possibility of a common targeting approach between PES andCCT can be exemplified by Fig. 1. The vertical axis indicates the level ofthe provision of environmental services. The horizontal axis indicatesthe level of well-being which can be measured by different indicators.Let's, for the sake of exposition, use income as the key proxy forwellbeing (though other dimensions of well-being could also be usedif available). PES are designed considering a metric comprised byenvironmental criteria that determines if land is suitable to beincluded in the intervention based on its potential level of provision ofenvironmental services or level of threat of degradation. Additionally,CCT are designed considering a metric comprised by a set of socioeconomic criteria to target the potential payment receptors of thesocial intervention, usually a segment of the poor population, but notthe poorest of the poor. The ultra-poor are usually ineligible for CCTbecause their level of deprivation and vulnerability makes them torequire other forms of social assistance initiatives such as nonconditioned payments and assets transfers, or social mobilizationinitiatives (Matin et al., 2008; Matin and Halder, 2004).

A horizontal threshold defined by the environmental metricdemarcates if land is eligible to receive the environmental paymentbased on the expected provision of services; a vertical line demarcatesby a cut-off point of the proxy means whether a household falls underthe poverty threshold and hence is eligible to be a CCT beneficiary. Inaddition, an extra vertical line based on an ultra-poverty line can alsodetermine the segment of the population that although being poor isusually ineligible for CCT. Thus, the horizontal line of the environ-mental metric intersects the vertical line of the cut off point of theproxy means and the ultra poverty line determining six regions.

Any land quality below the environmental metric line is, bydefinition, not eligible for PES since that land does not providesufficient environmental service values from a societal point of view.Thus, PES should only be implemented in regions 2–4. However, it hasbeen argued that payments to the poorest (region 2) and wealthiestpeople (region 4) will not generate substantive additionalities. Theseultra poor groups are so limited in every aspect that their capacity tonegatively impact the environment is negligible, hence lowering theeffectiveness of the payment (Wunder, 2007). For the wealthiestgroup, a payment will have low additionality since they might use theenvironment for recreation being unlikely to be involved inenvironmental degradation activities (e.g. Kosoy et al., 2007).

The area to the right of the line of the poverty threshold is bydefinition non eligible for CCT since these households are not seen asbeing poor. The area to the left of the ultra poverty line is not eligibleeither for CCT because the deprivation level of the households locatedin regions 1–2 requires other types of social assistance programs.Thus, CCT should be implemented in regions 3 and 6. While CCT inthese areas might be considered as pure poverty alleviation programswithout a priori environmental objective. There is evidence that inenvironmentally marginal areas of region 6, social payments such asthe ‘Cash for Relief Program’ in Ethiopia, have had positive environ-mental benefits since the payment receptors usually invest thetransferred money in rehabilitating land, restoring its productivityand helping to regenerate livelihoods (Cuesta, 2007; Standing, 2007).Nevertheless these environmental benefits are not expected frompayments implemented in severely degraded areas where therehabilitation of the land requires higher investments that areunaffordable with the received amount under a CCT scheme. Inareas like this, where environmental poverty traps are also associated

5 In practical terms many PES have been coincidently implemented in poor areasand therefore the poor are over represented among the beneficiaries (Wunder, 2008).Such PES schemes did not include specific poverty targeting mechanisms, but thefortuity of the coincidence of poverty-struck and environmentally rich areas mightexplain the unexpected welfare gains of these interventions. In these programs, asecond stage household targeting could increase the social benefits of the interventionbeyond the unexpected poverty reduction effects of single PES schemes.

Households are not beneficiaries of any of these programs

Only CCT applicableEligible for social protection programs

Payments with high impact on the environment and on poverty reduction

Only PES are applicable

Only PES are applicable

Pro

visi

on o

f env

ironm

enta

l ser

vice

s

Low provision

High provision

Income

Ultra poverty line Poverty line

Eligible CCT Not eligible CCT

Eligible PES

Not eligible PES

Not eligible CCT

PoorUltra Poor Better-off

1

2 3 4

56

Fig. 1. Unified targeting for environmental and social protection payments.

2171L.C. Rodríguez et al. / Ecological Economics 70 (2011) 2163–2174

with technological and infrastructure bottlenecks, larger investmentsand long term political support are required to improve wellbeing(see Matin et al., 2008).

Human wellbeing has environmental and socio-economic compo-nents and income itself might be a poor indicator of welfare if it is notaccompanied by adequate level of provision of environmental services(see: AIHV, 2007). Hence, identifying overlapping areas where CCT andPESmayoperate jointlymight help to select regions for designing unifiedpayment schemes. In this regard, investments in region 3 are likely togenerate the largest social impact for a given payment transfer since highenvironmental and poverty reduction effects are both expected ifappropriate social assistance and environmental policies are set togetherwith adequate information systems and a set of institutions. The latterwould regulate the transfer of benefits to the providers of environmentalservices and increase the demand for education and health services,facilitating project monitoring and coordination activities with delegatesof the environment health and education sectors.

The choice of a targeting mechanism to select payment receptorsshould take into account the transaction costs associated with designingand implementing the payment scheme. On the one hand, the cost oftargeting may be significant considering that both types of social andenvironmental information are required. On the other hand, the publicand private costs of administering the schememight be reduced becauseapplicants are applying for a single joint programandnot to two separatePES and CCT programs. The indirect costs of targeting arising when theselection criteria induce households to change their behavior in anattempt to become payment receptors of the program, e.g. purposivelychanging land use or altering labor patterns to reduce their income to beeligible, might be positive or negative depending of the relativecontribution of environmental or social funds to the overall transfer.

We suggest that a cost effective approach for unified paymentscould involve the coordinated use of both geographic and household/plot targeting. The initial step should consider a geographic target thatcombines geo-referenced layers containing socio-economic data andenvironmental information. The resultingmaps could prove helpful toselect regions with large abundance of the desired environmentalfeatures, e.g. high biodiversity, and high incidence of poverty.Within ageographic target region selection of beneficiary households might be

done using both proxy means to rank the participants for eligibility,and environmental metrics to rank the expected biodiversity benefitsof different plots. From these criteria, a unified PESPA should beimplemented if there are clear correlations between the provision ofenvironmental services and poverty incidence in a region, if thiscondition does not exist, independent PES and CCT programs shouldbe preferred instead.5

3.2. The Level, Frequency and Vehicle of the Payments

Section 2 suggests that the amounts transferred to the paymentreceptors of CCT and PES are comparable but the transfer rules differbetween the two schemes. Despite the increasing interest in the use ofCCT, most payment schemes are not at the scale required to have asignificant impact on poverty, but great advances have been achievedspecially in middle income programs such as those from Brazil andMexico (Grosh et al., 2008; World Bank, 2009). In PES, the transfersare usually closer to the minimum willingness to accept compensa-tion by the environmental service providers rather than themaximum willing to pay amount of the service receivers. To reachboth environmental and social objectives, a unified payment schemeshould define an appropriate transfer amount consistent with theneed to reduce the poverty level of the payment receptors and tocover the required investments in conservation.

The first consideration is that the transferred amount in a PESPAprogram should cover the direct cost to landholders' of participatingin the program, e.g. costs of sending children to school and costs ofimplementing certain land management actions, as well as theiropportunity costs, e.g. income foregone by children due to attendingschool and not working, and the revenues foregone by landholdersadopting conservation activities. However, due care should be taken

2172 L.C. Rodríguez et al. / Ecological Economics 70 (2011) 2163–2174

as these costs might not be directly aggregated since for example thedirect cost of labor required for land management activities might beat the same time the income lost through school attendance.

Based on the experience of successful PES and CCT, we suggest thatPESPA payments should be distributed based on provision rewardcriterion (see Pascual et al., 2010), using a variable benefit formulawith a base fixed payment but considering larger variable benefits forpoorer families and for the households that provide more environ-mental services, rather than inefficiently using the same level ofpayments for all the payment receptors.

The payment mode/vehicle is a critical component for the designof a unified scheme. Unless rejected by the payment receptorsthemselves, cash transfers might be initially preferred because of theirlower transaction costs and capacity to address information asym-metries. For example, considering an intervention aiming to reducepoverty and enhance environmental services in an agricultural area,PESPA could be transferred as an investment in relevant infrastructureor technology requested by the payment receptors, e.g., improveaccess to markets, enhance irrigation systems, or introduce preferredseeds or livestock breeds, that helps farmers to address entry barriersto alternative income generation activities. However, these indirectpayment alternatives might provide incentives for free riders andhave the same problems that made former conservation approacheslike Integrated Conservation Development Projects (ICDP) unsuccess-ful (see Wells et al., 1998). Therefore, PESPA might maximize itspotential by providing cash to payment receptors who can later investsuch direct payment based on their needs, preferences, and existingconditionalities. These direct payments are considered the most costeffective mechanism for conservation (Ferraro and Simpson, 2002)and are also preferred in poverty alleviation initiatives (Grosh et al.,2008).

For both CCT and PES, the regular and predictable frequency ofpayments helps beneficiary households to reduce entry barriers andmitigate risk. While in general, PES tend to include an upfrontpayment to cover for example the initial cost of environmentalconservation programs, CCT tend to include delayed payments orsavings mechanisms to encourage use of funds to deliver socialobjectives. In the context of PESPA, thus, an upfront payment shouldbe included in the design to cover part of the costs of theenvironmental component and the release of the remaining funds(remnant PES funds and the social assistance CCT component) shouldbe designed to deliver the objectives of the intervention. Theseobjectives would explicitly include timing resource transfers to fulfillthe defined conditionalities when required (e.g. medical checkupswhen starting to send children to school, reforest a slope before therain season, adopt tillage practices after the harvest months) and toassist households to smooth their income and consumption. This lastpoint requires that program designers should have good understand-ing of the poverty dynamics of the payment receptors and thedynamics of the environmental systems. That is, factors such asdrought frequency, seasonality, crop prices, health status, andseasonal unemployment need also be considered.

An existing problem in the CCT design is the lack of exit strategiesand the fact that payment receptors can easily fall again into poverty ifhuman capital investments do not generate sufficient short termreturns, and the CCT payment receptors are no longer eligible forpayments or the program ends (Yashine, 2008). A unifying schememight help ameliorate this problem because although the povertypayment component might phase out, the PES component might beorganized in a much longer term perspective as long as theenvironmental services are provided and service recipients areready to pay for them, e.g., fostering a market for ecosystem services,hence potentially reducing the negative impact of phasing out CCTcomponents on the payment receptors.

In summary, we suggest that the level of direct (cash) rewards ofPESPA schemes ought to cover the direct and opportunity costs to

landholders participating in the program by appropriately mixingboth up-front and regular transfers to maximize uptake of rewards bylandholders.

3.3. Setting Conditionalities in PESPA

As in individual PES and CCT programs, unified PESPA schemescould be linked to a set of environmental and social conditionsdesigned to adjust individual behavior towards delivery of theconservation and poverty alleviation objectives. From a projectperspective, conditionalities are expensive to implement and monitorand for unified payment schemes the expected costs could be higherthan for individual PES or CCT schemes. Participants may findthemselves exposed to a more diverse range and larger number ofconditions implying higher cost and potential difficulty in demon-strating compliance. In addition, the inclusion of both environmentaland socio-economic conditions can represent a challenge forimplementation in low capacity and poor infrastructure scenarios.For example, skills and experience in agencies monitoring compliancemay be insufficient or housed in separate agencies.

If costs are expected to be high it may be appropriate to includeinstitutional and capacity analyses of the region in the design ofconditions for a unified scheme. In particular, it is important to ensurethat conditions can practically be met. For example, requiring schoolattendance or medical check-ups will need sufficient schools andmedical facilities in the region. Similarly, conditionalities requiringafforestation should consider the availability of seedlings in thenurseries or the capacity of the landholders to produce the seedlings.

Alternatively, combined programs may find their costs are lowerthan the combined cost of aggregating costs of independent PES andCCT programs operating in the same region since the administrativeand transaction costs of the joint intervention might be shared andparticipating households may find it easier to demonstrate compli-ance within a single program rather than across several. Appropriateoperational arrangements may also provide a way to reduce costs orease monitoring through for example agreements between differentgovernment agencies, such as the ministries dealing with Health,Education and Agriculture. In some cases there could be opportunitiesto involve the private sector where they have higher/betterinfrastructure/capacity than state organizations, and they are able toprovide a service to the implementing agencies. Such public–privatealliances can help with capacity problems but should be subject toappropriate accountability and minimize opportunities for politicalmanipulation of the program.

4. Conclusions

Donors, research and development organizations and fieldpractitioners are constantly looking to refine the selection, designand implementation of programs to improve the cost effectiveness ofschemes to jointly promote environmental conservation and povertyalleviation (Muradian et al., 2010). In this context, numerous CCT andPES initiatives are being implemented at different scales with eithersocial or environmental objectives which have provided increasingevidence of positive side effects of social interventions on theenvironment as well as positive impacts from environmental in-terventions on social systems (Niesten and Rice, 2004; Standing,2007).

This paper reviewed a diverse set of existing PES and CCT programsand identified potentially positive livelihood improvements as a sideeffect of environmental initiatives and positive environmentalimpacts of social interventions. Despite the similarity of some keydesign aspects of both programs, the specialized literatures on bothtypes of programs have hardly met and hence policy makers have hadvery little exposure if at all to ideas as to how they can both becombined in their designs towards a unified payment.

2173L.C. Rodríguez et al. / Ecological Economics 70 (2011) 2163–2174

The goal of this paper is to start shedding some light on suchpotential by identifying key areas where such unified so-calledconditional program of payments for environmental and povertyalleviation services (PESPA) approach could be designed, as well assome issues that may undermine them.

The main points developed in the paper are that both PES and CCTprograms share many similarities in their conceptual designs. Bothcan be considered: i) market based interventions in which; ii) apayment amount is transferred to; iii) a number of payment receptorspreviously identified in a target exercise; if iv) they fulfill a number ofconditions aimed to internalize an externality.

To reach both environmental and social objectives, a unified paymentscheme should define an appropriate transfer amount consistentwith theneed to reduce thepoverty level of thepayment receptors and to cover therequired investments in conservation. Based on the experience ofsuccessful PES and CCT, we suggest that PESPA payments should bedistributedusing a variable benefit formulawith a basefixedpayment butconsidering larger variable benefits for poorer families and for thehouseholds that provide more environmental services.

In addition, the targeting mechanism should consider thecoordinated use of both geographic and household/plot targeting.Initially mapping the areas where poverty and high level of provisionof environmental services overlaps and then apply proxy means testsand environmental metrics to rank the households and selectpayment receptors that fulfill the poverty and environmental criteriaof eligibility. However, the inclusion of both environmental and socioeconomic conditions might be expensive and their implementationand verification might represent a challenge in low capacity and poorinfrastructure scenarios requiring the collaboration between differentgovernment agencies and the creation to public–private alliances.

These conclusions should be taken as a first attempt to developgeneric ideas towards developing a unified CCT-PES scheme. Ofcourse, the suggested design features of the unified payment schemewill need to be adapted to the specific circumstances found on theground. PESPA would not provide a silver bullet nor it is suggested itshould replace in every context already well functioning PES or/andCCT schemes. Otherwise we would just add to the growing frustrationand confusion by practitioners in the field when applying genericrecipes of CCT and PES programs. The goal instead is to provide a firstapproximation to understanding when and how such PESPA mayprovide positive outcomes through an integrated policy initiative. Werecognize that win–win situations regarding poverty alleviation andenvironmental conservation are difficult to find and that merelyincreasing the wealth of people might not solve environmentalproblems, but we also recognize the capacity of conditionedmonetarytransfers to generate changes in the behavior of the paymentreceptors when an appropriate level of incentives and relevantconditionalities are defined.

We believe that setting a dialog by environment and developmentpolicy makers on the different types of conditioned transfers isimportant to evaluate the potential tradeoffs and synergies ofintegrated PES and CCT schemes aimed to deal with poverty andenvironmental degradation. This paper will hopefully help place thePES and CCT literature under new perspective. One that creates afruitful dialog between development and ecological economists aswell as practitioners on the ground.

References

Ahmed, A.U., Rashid, S., Sharma, M., Zohir, S., 2004. Food aid distribution in Bangladesh:leakage and operational performance. IFPRI. Discussion Paper 173. Washington,D.C.

Alvarez, C., Devoto, F., Winters, P., 2008. Why do beneficiaries leave the safety net inMexico? A study of the effects of conditionality on dropouts. World Development36, 641–658.

Asquith, N.M., Vargas, M.T., Wunder, S., 2008. Selling two environmental services: in-kind payments for bird habitat and watershed protection in Los Negros, Bolivia.Ecological Economics 65, 675–684.

Australian Institute of Health andWelfare, 2007. Australia's Welfare 2007. Cat. No. AUS93. AIHW, Canberra.

Barrera-Osorio, Bertrand, M., Linden, L.L., Perez-Calle, F., 2008. Conditional cashtransfers in education: design features, peer and sibling effects evidence from arandomized experiment in Colombia. NBER Working Papers 13890. NationalBureau of Economic Research, Inc.

Beria, L., Laxman, J., van Noordwijk, M., 2009. Can rewards for environmental servicesbenefit the poor? Lessons from Asia. International Journal of the Commons 3,82–107.

Bond, I., 2001. CAMPFIRE and the incentives for institutional change. In: Hulme, D.,Murphree, M.W. (Eds.), African Wildlife and Livelihoods. The Promise andPerformance of Community Conservation. James Currey, Oxford, pp. 227–243.

Bulte, E.H., Zilberman, D., 2008. Special issue: payment for ecosystem services.Environment and Development Economics 13, 245–439.

Chapman, K., 2006. Using Social Transfers to Scale Up Equitable Access to Education andHealth Services. DFID Background Paper. Department of International Develop-ment, London.

Clements, T., Ashish, J., Nielsen, K., Dara, A., Seth, T., Milner-Gulland, E.J., 2010. Directpayments for biodiversity conservation: comparison of three schemes fromCambodia. Ecological Economics 69, 1283–1291.

Corbera, E., González Soberanis, C., Brown, K., 2009. Institutional dimensions ofpayments for ecosystem services: an analysis of Mexico's carbon forestryprogramme. Ecological Economics 68, 743–761.

Cuesta, J., 2007. On more ambitious conditional cash transfers, social protection andpermanent reduction of poverty. Journal of International Development 19,1016–1019.

Das, J., Do, Q.T., Ozler, B., 2004. Conditional cash transfers and the equity—efficiencydebate. World Bank Policy Research Working Paper No. 3280. The World Bank,Washington.

Das, J., Do, Q.T., Ozler, B., 2005. Reassessing conditional cash transfer programs. WorldBank Research Observer 20, 57–80.

Dasgupta, P., 2003. Population, poverty, and the natural environment. In: Mäler, K.-G.,Vincent, J. (Eds.), Handbook of Environmental Economics, Vol. I. North Holland,Amsterdam, pp. 191–247.

De Janvry, A., Sadoulet, E., 2004. Conditional cash transfer programs: are they reallymagic bullets? Agricultural and Resource Economics Update, Gianini Foundation ofAgricultural Economics 6, 9–11.

Department of Sustainability and Environment, 2008. BushTender: rethinkinginvestment for native vegetation outcomes. The application of auctions forsecuring private land management agreements. State of Victoria, Department ofSustainability and Environment, East Melbourne.

Duraiappah, A.K., 1998. Poverty and environmental degradation: a review and analysisof the nexus. World Development 26, 2169–2179.

Echavarria, M., Vogel, J., Albán, M., Meneses, F., 2002. Impact Assessment of WatershedEnvironmental Services: Emerging Lessons from Pimampiro and Cuenca inEcuador. EcoDecisión, IIED, Quito.

Engel, S., Pagiola, S., Wunder, S., 2008. Designing payments for environmental servicesin theory and practice: an overview of the issues. Ecological Economics 62,663–674.

Farrington, J., Slater, R., 2006. Cash transfers: panacea for poverty reduction or moneydown the drain? Development Policy Review 2006, 499–511.

Ferraro, P.J., Simpson, R.D., 2002. The cost effectiveness of conservation payments. LandEconomics 78, 339–353.

Fiszbein, A., Schady, N., 2009. Conditional Cash Transfers: Reducing Present and FuturePoverty. The World Bank, Washington, D.C.

Glewwe, P., Olinto, P., 2004. Evaluating the impact of conditional cash transfers onschooling: an experimental analysis of Honduras's PRAF Program. Final Report forUSAID. International Food Policy Research Institute, Washington, D.C.

Gökalp, Y., 2006. Conditional cash transfers in Turkey: motivation, design, achieve-ments, challenges, and the way forward. Third International Conference onConditional Cash Transfers Istanbul, Turkey. June 2006.

Grieg-Gran, M., Porras, I.T., Wunder, S., 2005. How can market mechanisms for forestenvironmental services help the poor? Preliminary lessons from Latin America.World Development 33, 1511–1527.

Grosh, M., del Ninno, C., Tesliuc, E., Ouerghi, A., 2008. For Protection and Promotion: TheDesign and Implementation of Effective Safety Nets. World Bank, Washington DC.

GTZ/BMZ, 2005. Social Cash Transfers: Reaching the Poorest. GTZ/BMZ, Eschborn.Handa, S., Davis, B., 2006. The Experience of Conditional Cash Transfers in Latin America

and the Caribbean. Development Policy Review 24, 513–536.IADB, 2002. Inter-American Development Bank. Press Release January 16th 2002.

available at http://www.iadb.org/exr/PRENSA/2002/cp1002e.htm.Imbach, P., 2005. Priority Areas for Payments for Environmental Services (PES) in Costa

Rica. Centro Agronomico Tropical de Investigacion y Enseñanza (CATIE), Turrialba,Costa Rica.

Jindal, R., Swallow, B., Kerr, J., 2008. Forestry-based carbon sequestration projects inAfrica: potential benefits and challenges. Natural Resources Forum 32, 116–130.

Karsenty, A., 2004. Des rentes contre le développement ? Les nouveaux instrumentsd'acquisition mondiale de la biodiversité et l'utilisation des terres dans les paystropicaux. Mondes en Développement 32, 59–72.

Kosoy, N., Martinez-Tuna, M., Muradian, R., Martinez-Alier, J., 2007. Payments forenvironmental services in watersheds: insights from a comparative study of threecases in Central America. Ecological Economics 61, 446–455.

Kosoy, N., Corbera, E., 2010. Payments for ecosystem services as commodity fetishism.Ecological Economics 69 (6), 1228–1236.

Martínez-Alier, J., 2002. The environmentalism of the poor. A Study of EcologicalConflicts and Valuation. Edward Elgar, Cheltenham, UK.

2174 L.C. Rodríguez et al. / Ecological Economics 70 (2011) 2163–2174

Massingarela, C., Nhate, V., 2006. Background Paper for the Chronic Poverty Report2008–09 The Politics of What Works: A Case Study of Food Subsidies and the Bolsa-Escola in Mozambique. Chronic Poverty Research Centre. February 2006.

Matin, I., Halder, S., 2004. Combining methodologies for better targeting of the extremepoor: lessons fromBRAC'sCFPR/TUPProgramme.CFPR-TUPWorkingPaperSeriesNo. 2.

Matin, I., Hulme, D., 2003. Programs for the poorest: learning from the IGVD program inBangladesh. World Development 31, 647–665.

Matin, I., Sulaiman, M., Rabbani, M., 2008. Crafting a graduation pathway for the ultrapoor: lessons and evidence form a BRAC program. Working Paper N 169. ChronicPoverty Research Centre.

Medeiros, M., Britto, T., Veras Soares, F., 2008. Targeted cash transfer programmes inBrazil: BPC and the Bolsa Familia. Working Paper No. 46. International PovertyCentre, UNDP.

Miranda,M., Porras, T.,Moreno,M., 2003. The social impacts of payments for environmentalservices in Costa Rica. AQuantitative Field Survey and Analysis of the VirillaWatershed.International Institute for Environment and Development, London.

Munawir, Vermeulen, 2007. Fair deals for watershed services in Indonesia. Interna-tional Institute for Environment and Development (IIED), London.

Muñoz-Pina, C., Guevara, A., Torres, J., Brana, J., 2008. Paying for the hydrologicalservices of Mexico's forests: analysis, negotiations and results. EcologicalEconomics 65, 725–736.

Muñoz-Piña, C., Rivera, M., Cisneros, A., Garcia, H., 2011. Retos de la focalización delPrograma de Pago por los Servicios Ambientales en México. Revista Española deEstudios Agrosociales y Pesqueros 228, 87–113.

Muradian, R., Corbera, E., Pascual, U., Kosoy, N., May, P.H., 2010. Reconciling theory andpractice: an alternative conceptual framework for understanding payments forenvironmental services. Ecological Economics 69, 1202–1208.

Niesten, E., Rice, R., 2004. Sustainable forest management and conservation incentiveagreements. International Forestry Review 6, 56–60.

Pagiola, S., 2008. Payments for environmental services in Costa Rica. EcologicalEconomics 65, 712–724.

Pagiola, S., Landell-Mills, N., Bishop, J., 2002. Making market-based mechanisms workfor forests and people. In: Pagiola, S., Bishop, J., Landell-Mills, N. (Eds.), SellingForest Environmental Services: Market-based Mechanisms for Conservation andDevelopment. Earthscan, London.

Pagiola, S., Arcenas, A., Platais, G., 2005. Can payments for environmental services helpreduce poverty? An exploration of the issues and the evidence to date. WorldDevelopment 33, 237–253.

Pascual, U., Muradian, R., Rodríguez, L.C., Duraiappah, A., 2010. Exploring the linksbetween equity and efficiency in payments for environmental services: aconceptual approach. Ecological Economics 69, 1237–1244.

Pattanayak, S.K., Wunder, S., Ferraro, P.J., 2010. Show me the money: do paymentssupply environmental services in developing countries? Review of EnvironmentalEconomics and Policy 4, 254–274.

Pearson, R., Alviar, C., 2009. Cash Transfers for Vulnerable Children in Kenya: FromPolitical Choice to Scale-up. United Nations Children's Fund (UNICEF), Policy,Advocacy and Knowledge Management, Division of Policy and Practice, New York.

Perez-Ribas, R., Irata, G.I., Vera Soares, F., 2008. Debating Targeting Methods for CashTransfers: A Multidimensional Index vs Income Proxy for Paraguay's TekoporaProgram. International Poverty Centre. IPC evaluation note 2.

PIDS, 2007. Conditional cash transfers: social assistance and human developmentcombined. Economic Issues of the Day, V 7, No 3. Philippine Institute forDevelopment Studies.

Porras, I., Grieg-Gran, M., Neves, N., 2008. All that Glitters: A Review of Payments forWatershed Services in Developing Countries. The International Institute forEnvironment and Development (IIED), London.

Rawlings, L., 2005. A new approach to social assistance: Latin America's experiencewith conditional cash transfer programs. International Social Security Review 58,133–161.

Reto-o-Reto, 2006. Conservation in Kitengela: keeping land open for people, livestockand wildlife. Policy Brief Number 1. ILRI, Nairobi.

Rios, A.R., Pagiola, S., 2009. Poor Household Participation in Payments for Environ-mental Services in Nicaragua and Colombia. World Bank. MPRA Paper No. 13727.

Rosa, H., Kandel, S., Dimas, L., 2003. Compensation for Environmental Services andRural Communities in the Americas: Lessons from the Americas and Key Issues forStrengthening Community Strategies. Programa Salvadoreño de Investigaciónsobre Desarrollo y Medio Ambiente, San Salvador.

Ruijs, A., Dellink, R.B., Bromley, D.W., 2008. Economics of poverty, environment andnatural-resource use. In: Dellink, Rob B., Ruijs, Arjan (Eds.), “Economics of Poverty,Environment and Natural-Resource Use”. Chapter 1. Springer, pp. 3–15.

Sachs, J.D., Reid, W.V., 2006. Investments towards sustainable development. Science312, 1002.

Schady, N.R., 2002a. Picking the poor: indicators for geographic targeting in Peru.Review of Income and Wealth 48, 417–433.

Schady, N.R., 2002b. Picking the poor: indicators for geographic targeting in Peru.Review of Income and Wealth 48, 417–433.

Schubert, B., Slater, R., 2006. Social cash transfers in low-income African countries:conditional or unconditional? Development Policy Review 24, 571–578.

Semu-Banda, P., 2009. Malawi: Separating the Ultra Poor from the Poor—Why? http://www.ipsnews.net/news.asp?idnews=46848 Accessed May 19th 2009.

Skoufias, E., 2005. PROGRESA and its impacts on the welfare of rural households inMexico. IFPRI, Research Report No. 139. Washington D.C.

Skoufias, E., McClafferty, B., 2001. Is PROGRESAWorking? Summary of the Results of anEvaluation by The International Food Policy Research Institute (IFPRI), WashingtonD. C.

Standing, G., 2007. How cash transfers boost work and economic security. UNDESAWorking Paper N 58.

Stoneham, G., Chaudhri, V., Ha, A., Strappazzon, L., 2003. Auctions for conservationcontracts: an empirical examination of Victoria's BushTender Trial. The AustralianJournal of Agricultural and Resource Economics 47, 477–500.

Suyanto, S., Khususiyah, N., Leimona, B., 2007. Poverty and environmental services: casestudy in Way Besai watershed, Lampung Province, Indonesia. Ecology and Society12, 13 [online] http://www.ecologyandsociety.org/vol12/iss2/art13/.

Vatn, A., 2010. An institutional analysis of payments for environmental services.Ecological Economics 69, 1245–1252.

Veras Soares, F., Peres Ribas, R., Issamu Hirata, G., 2008. Achievements and shortfalls ofconditional cash transfers: impact evaluation of Paraguay's Tekoporã Programme.IPC Evaluation Note No. 3. International Poverty Centre, UNDP.

Viana, V.M., 2008. Bolsa Floresta (Forest Conservation Allowance): an innovativemechanism to promote health in traditional communities in the Amazon. EstudosAvancados 22, 64. doi:10.1590/S0103-40142008000300009.

Wells, M., Guggenheim, S., Khan, A., Wardojo, W., Jepson, P., 1998. Investing inbiodiversity. A Review of Indonesia's Integrated Conservation and DevelopmentProjects. World Bank, Washington, DC.

Winters, P., Davis, B., 2009. Designing a programme to support smallholder agriculturein Mexico: lessons from PROCAMPO and Oportunidades. Development PolicyReview 27, 617–642.

World Bank, 2008. Kenya Wildlife Conservation Leasing Demonstration Project. GEFProject Brief Report 37471.

World Bank, 2009. Conditional cash transfers. Reducing Present and Future Poverty. AWorld Bank Policy Research Report. Washington.

Wunder, S., 2005. Payments for environmental services: some nuts and bolts. CIFOROccasional Paper 42. Center for International Forestry Research, Bogor, Indonesia.

Wunder, S., 2006. Are direct payments for environmental services spelling doom forsustainable forest management in the tropics? Ecology and Society 11 (2), 23[online] URL http://www.ecologyandsociety.org/vol11/iss2/art23/.

Wunder, S., 2007. The efficiency of payments for environmental services in tropicalconservation. Conservation Biology 21, 48–58.

Wunder, S., 2008. Payments for environmental services and the poor: concepts andpreliminary evidence. Environment and Development Economics 13, 279–297.

Wunder, S., Alban, M., 2008. Decentralized payments for environmental services: thecase of Pimampiro and PROFAFOR in Ecuador. Ecological Economics 65, 685–698.

Yashine, I., 2008. Why, when and how should beneficiaries leave a CCT programme.International Poverty Centre. Poverty in Focus 15, 8–9.

Zilberman, D., Lipper, L., Mccarthy, N., 2008. When could payments for environmentalservices benefit the poor? Environment and Development Economics 13, 255–278.