AnalysisIncentives for cooperation: The effects of institutional controls on common pool resource extraction in Cambodia
Highlights
► We use a common pool resource game to investigate resource appropriation behaviour. ► Promotion of self-organisation reduces individual extraction from a CPR. ► Incentive-based conservation policies should encourage local decision-making.
Introduction
Theory suggests that cooperation between unrelated individuals is costly and as a result, without externally generated incentives, rational actors should not cooperate even when it is in their common interest to do so (Hardin, 1968, Olson, 1965). Despite these predictions there is substantial evidence to suggest that cooperation is a common human behaviour, particularly if reciprocation can be expected. Much of this evidence stems from social dilemma experiments; from the success with which the simple tit-for-tat strategy negotiates an iterated prisoners' dilemma (Axelrod, 1980), to the initial positive contributions typically observed by participants in public goods experiments (Ostrom, 2000).
A number of different behavioural control mechanisms have been investigated. These include preferences to act fairly (Falk et al., 2000), to reciprocate (Berg et al., 1995, Fehr et al., 2002) and to abide by commitments or punish those who break them (Fehr and Gächter, 2000, Masclet et al., 2003, Ostrom et al., 1992). Such preferences arise when externalities are created which affect the welfare of others (Coleman, 1990) and are highly dependent on the social or institutional conditions in which people operate (Biel and Thørgersen, 2007).
One situation in which the behaviour of a single individual exerts externalities on others within the same peer group is the use of common pool resources (CPRs). Considerable research has focused on this subject, including social dilemma experiments (Cardenas et al., 2000), theoretical modelling (Rodriguez-Sikert et al., 2008) and field observations (Baland and Platteau, 1996, Ostrom, 1990, Wade, 1988). This work fits within the wider research field of analysing the sustainability of socio-ecological systems (SESs), which consist of complex interactions between resource users, governance systems and natural resources. Much of the evidence emerging from this growing field has highlighted the importance of institutional arrangements in achieving sustainable outcomes. In a study examining the factors affecting the sustainability of 130 community co-managed fisheries, for example, Gutiérrez et al. (2011) adapted a framework of analysis proposed by Ostrom (2009). They found that success was most strongly related to institutional factors at the community level, with less importance given to the characteristics of the resource or exogenous factors.
Given the significance of SESs to the livelihoods of many, particularly in developing countries, it is now common for policy interventions to be aimed at providing incentives to encourage cooperation amongst resource users. This represents a significant shift in thinking from the traditional ‘command and control’ models of resource management. Current policy focuses on the transfer of management rights to resource users and changing incentives, either on an individual or collective basis, to encourage a change of behaviour towards sustainable use. It is rare, however, for the design of any such interventions to be based on analytical studies of their potential effects on behaviour.
Few studies have used common pool resource games to investigate behaviour in the context of developing countries although this omission has begun to be addressed in recent years (e.g. Cardenas et al., 2000, del Pilar Moreno-Sànchez and Maldonado, 2010, Velez et al., 2010, Vollan, 2008). Consequently, the majority of experimental results come from games which have been played by subjects from so-called western, educated, industrialised, rich and developed (WEIRD) societies, to whom resource access dilemmas are remote (Henrich et al., 2010). It is now widely accepted that the local context of an SES can have a significant impact on behaviour measured in experimental games (Anderies et al., 2011). In a series of experimental games in 15 ‘small-scale societies’, Henrich et al. (2005) found evidence of wider behavioural variation than commonly reported from experiments involving students, concluding that a large part of the variation seen could be explained by differences in economic and social organisation within the societies considered.
A common conclusion of much of the research on the sustainability of SESs, drawing on the design principles for sustainable management of CPRs proposed by Ostrom (1990) and others, has been the importance of self-organisation – the ability of resource users to devise, monitor and enforce their own set of rules – in promoting sustainable management. This process requires significant cooperation between individuals at all stages and it is often only when a resource becomes scarce that there is sufficient incentive for users to cooperate. Further support was recently provided by a CPR game investigating behaviour in the context of marine protected areas in Colombia (del Pilar Moreno-Sànchez and Maldonado, 2010). This found that mean individual extraction for groups playing a co-management treatment, in which participants were able to discuss extraction rates between themselves following advice from a national park ranger, was significantly lower than during the baseline treatment in which no communication was allowed. Similarly, a recent study in 84 forest SESs across East Africa and South Asia found a strong correlation between jointly positive outcomes for conservation and local livelihoods and participation in resource governance, particularly in small forest fragments which representing the greatest challenges for conservation (Persha et al., 2011). On the basis of this evidence, it is expected that, contrary to predictions based on the model of rational self-interest, institutions which promote self-organisation amongst resource-users can reduce levels of individual extraction.
A key aspect of the relative decrease in individual extraction for conditions which promote self-organisation is the opportunity for social approval or disapproval, as this enables social penalties to be imposed on those individuals who do not cooperate. A number of social dilemma studies have found that providing participants with the opportunity to express approval or disapproval can increase individual cooperation (Gächter and Fehr, 1999, Rege and Telle, 2004). Such studies suggest that the opportunity for social sanctioning may lead individuals to regulate their own behaviour through feelings of guilt or increased self-belief (Lindbeck, 1997).
Sanctioning can also be applied by external agents through penalties for infringement of externally applied rules. There is a risk though that this can lead to a crowding out of pro-social behaviour. Cardenas et al. (2000) showed that the introduction of a weakly enforced penalty designed to limit resource collection had the perverse effect of increasing average extraction above baseline rates. Initial cooperation was higher than the baseline, but was lost over time as the application of enforcement penalties crowded out other-regarding behaviour. The evidence for this effect is, however, equivocal. In a further study, Cardenas (2004) found no significant difference in extraction behaviour between three treatments: two in which participants faced external sanctions and one in which regulations were endogenously derived. This suggests that individual extraction rates are partly dependent on existing social norms regarding cooperative behaviour, which vary depending on the particular context of the system investigated.
A difference in extraction behaviour under externally applied and endogenous rules has been found for positive incentives (Bowles, 2008, Deci et al., 1999). In Switzerland, for instance, a survey of support for the site of a proposed new nuclear facility found that the offer of incentives as compensation crowded-out feelings of civic duty, resulting in lower levels of support (Frey and Oberhozer-Gee, 1997). There is, however, some evidence to suggest that endogenously awarded incentives may be more effective at encouraging cooperation, although there remain questions with respect to longevity. In a study in which subjects could reward others after decisions had been made, Sefton et al. (2000) found that, although initial cooperation was higher than the baseline, in which there was no opportunity to reward, this cooperation dropped off to the same level as the baseline over time.
Our experiment was conducted in four villages located in two protected areas (PAs) in the Northern Plains landscape, Cambodia: Kulen Promtep Wildlife Sanctuary and Preah Vihear Protected Forest. Management of these two PAs forms the basis of biodiversity conservation within the landscape, a region recognised for its global conservation significance (Myers et al., 2000, Olson and Dinerstein, 1998, WCS, 2009). The main threats to achieving conservation goals in the area come in the form of hunting, habitat destruction and human disturbance by agro-industrial concessions, immigrants and local residents, and are exacerbated by human population growth, migration and improved road access.
Local villagers are highly dependent on natural resources which include small-scale freshwater fisheries, forest land, wildlife populations and a large variety of forest products, most notably liquid resin harvested from dipterocarp trees (McKenney et al., 2004). In many instances, despite harvest being restricted or banned altogether, these resources can be considered de facto open-access since institutional capacity for PA management is weak. All of the study villages are remotely situated and consequently, in reality, access to resources is largely limited to households from the respective villages.
Resource-based decision-making in the study villages is centred on individual households, with few collective decisions taken. Khmer people are typically risk-averse and avoid confrontation. Hence, conflicts over resources are avoided and claims to land or high value products respected, while any disputes that do arise are settled by village elders. As such most villagers look to village elders to make decisions which affect collective well-being. This study focuses on the harvest of fish, which makes up a significant component of household livelihood strategies for the majority of families in the study area. Fish are a widely accessible resource, with both men and women commonly taking part in harvesting, which is subject to few harvesting restrictions. Each village has a number of locations suitable for fishing and so, with the majority of catches consumed in the home, individual household effort and catch rarely become common knowledge. In some cases, however, those households that invest the most effort into fishing may acquire a reputation within the village and become the subject of peer pressure.
In 2002, the Ministry of Environment and Ministry of Agriculture, Forestry and Fisheries, with the support of the Wildlife Conservation Society (WCS), an international non-governmental organisation (NGO), instituted a series of pilot Payments for Environmental Services (PES) initiatives as a complement to PA management. Clements et al. (2010) describe and compare three such programmes which are being implemented in the same villages selected for this study. This comparison found that those programmes that were managed by local organisations achieved a greater level of understanding and support amongst local people and were more institutionally effective, leading to the conclusion that programmes which seek to empower local institutions may be more sustainable (Clements et al., 2010).
Whilst previous studies have tended to focus on the effect of single experimental conditions, in this study we examine levels of cooperation between resource users under varying settings, each designed to mimic different institutions commonly implemented as part of policy interventions. We are not aware of any other study that has considered the effects of such a broad range of potential institutional solutions to the commons dilemma within a single field context. Structuring the study in this way uniquely enables us to examine the relative influence of different institutions on the behaviour of a subject pool familiar with resource dilemmas and hence identifies those institutional arrangements most suitable for implementation as conservation interventions.
Section snippets
The Common Pool Resource Game
The game was framed as a resource extraction problem familiar to the participants in order to aid understanding and ensure relevance, following evidence suggesting that framing field experiments results in behaviour which more closely matches that of real-life (Lusk et al., 2006). It was explained that there was a communal fishing pond from which all participants could extract fish. Fishing is less controversial than hunting for other wildlife species or clearing land and so the game was framed
Theoretical Predictions
Hardin (1968) describes the choices faced by a rational being, who seeks to maximise their own utility, in the context of extracting from a common pool resource. As the benefits of extracting more are felt only by the individual whereas the costs are shared amongst all users, the rational decision is to extract as much as possible. As this strategy is shared by all resource users, the result is a tragedy of the commons. This simple model of behaviour posits that an individual will act
Statistical Analysis
Statistical models were constructed with the number of fish taken by an individual in each round as the response variable. All statistical analysis was carried out using the software package R version 2.9.0 (R Development Core Team, 2009). Due to the non-normality and hierarchical structure of the data we used generalised linear mixed effects models (GLMMs; Bolker et al., 2009). A logit link function was used to account for the binomial error structure of the response variable with Laplace
Results
Individual extraction during the control treatment is the baseline against which behaviour in other treatments is compared. In the initial round of the control treatment the average number of fish taken was 5.44 (s.d. = 3.11) and this increased with each subsequent round (Fig. 1). This follows the standard pattern observed in previous studies for treatments with no behavioural controls (Ostrom, 2000).
As is standard for such experiments, the pattern of individual extraction was erratic for most of
Discussion
In general, observed behaviour was markedly different from the predictions made under the rational, self-interested model but matched expectations from the findings of previous studies. For the control treatment, initial extraction was within the range of 40–60% of the maximum possible, which is the most commonly reported range from n-person dilemma experiments. In subsequent rounds the average extraction increased continuously, which again is consistent with common findings from iterated
Conclusion
This study is the first to test experimentally how multiple differing institutional structures for resource management affect individual extraction from common pool resources, in a developing country context of poor property rights and resource degradation. This setting is typical of the difficult circumstances within which conservation interventions take place. Our results are in line with previous studies, but bring wider insights. In particular we demonstrate the importance of promoting
Acknowledgements
We would like to thank Mark Gately, Ashish John, An Dara, Hugo Rainey, Tan Setha, Ear Sokha, Vann Ny and Nhem Sokheng of the WCS Cambodia Program and our research team Chim Sochinda, Phon Vannak and Nguon Sokha. Special thanks go to all the people who participated in the games. The project benefited immensely from the technical advice and experimental experiences of Abigail Barr, Devesh Rustagi and J.C. Cardenas and from the comments of Andy Gardner and Claire El Mouden on an earlier draft. Any
References (58)
- et al.
The challenge of understanding decisions in experimental studies of common pool resource governance
Ecological Economics
(2011) - et al.
Trust, reciprocity and social history
Games and Economic Behaviour
(1995) - et al.
Activation of social norms in social dilemmas: a review of the evidence and reflections on the implications for environmental behaviour
Journal of Economic Psychology
(2007) - et al.
Generalized linear mixed models: a practical guide for ecology and evolution
Trends in Ecology & Evolution
(2009) Norms from outside and from inside: an experimental analysis on the governance of local ecosystems
Forest Policy and Economics
(2004)- et al.
Local environmental control and institutional crowding-out
World Development
(2000) - et al.
Context matters to explain field experiments: results from Colombian and Thai fishing villages
Ecological Economics
(2011) - et al.
Direct payments for biodiversity conservation: comparison of three schemes from Cambodia
Ecological Economics
(2010) - et al.
Evaluating the role of co-management in improving governance of marine protected areas: an experimental approach in the Colombian Caribbean
Ecological Economics
(2010) - et al.
Psychological foundations of incentives
European Economic Review
(2002)
Collective action as a social exchange
Journal of Economic Behaviour and Organisation
External validity of a framed field experiment
Economics Letters
The impact of social approval and framing on cooperation in public good situations
Journal of Public Economics
Socio-ecological explanations for crowding-out effects from economic field experiments in southern Africa
Ecological Economics
A new look at the statistical model identification
IEEE Transactions on Automatic Control
Effective Choice in the Prisoner’s Dilemma
Journal of Conflict Resolution
Halting Degradation of Natural Resources: Is There a Role for Rural Communities?
Conserving tropical biodiversity amid weak institutions
Bioscience
lme4: Linear mixed-effects models using S4 classes
Policies designed for self-interested citizens may undermine “the moral sentiments”: evidence from economic experiments
Science
Model Selection and Multi-model Inference
At loggerheads? Agricultural expansion, poverty reduction, and environment in the tropical forests
Foundations of Social Theory
A meta-analytic review of experiments examining the effects of extrinsic rewards on intrinsic motivation
Psychological Bulletin
Appropriating the Commons: A Theoretical Explanation
Cooperation and punishment in public goods experiments
The American Economic Review
Strong reciprocity, human cooperation and the enforcement of social norms
Human Nature
The cost of price incentives: an empirical analysis of motivation crowding-out
The American Economic Review
Proximate causes and underlying driving forces of tropical deforestation
Bioscience
Cited by (71)
Impact of dynamic compensation with resource feedback on the common pool resource game
2024, Chaos, Solitons and FractalsImmediate backfire? Nudging sustainable food choices and psychological reactance
2023, Food Quality and PreferenceTortoise panopticon: Linkages between taboos and conservation management in Madagascar
2022, GeoforumCitation Excerpt :This may introduce a risk of novel, unsustainable dependencies. Indeed, multiple examples exist in which the introduction of reward systems results in erosion of conservation goals (Travers et al., 2011; Narloch et al., 2012). The underlying issue is sometimes identified as motivational crowding, where new incentives neglect or ‘crowd-out’ pre-existing motivations (Rode et al., 2014; Kaczan et al., 2019; Moros et al., 2019).
Experimental evidence on sharing rules and additionality in transfer payments
2021, Journal of Economic Behavior and OrganizationRegulatory stringency and behavior in a common pool resource game: Lab and field experiments
2021, Journal of Asian Economics