|
Article Excerpt
I. INTRODUCTION
This paper investigates how individuals perceive risks and uncertainties when making decisions about private and public goods. Using the framework of public goods games and laboratory experiments, we study whether--in payoff-equivalent situations--decision makers contribute more or less toward private and public goods when outcomes involve different forms of risks and uncertainties.
The economics literature usually applies the term "uncertainty" to describe situations when the probabilities of possible future events are unknown. The term "risk" on the other hand is used to describe situations when the probabilities of future events are known. (1) The presence of uncertainty and risk can be very important in the provisioning of public goods and in policy making. For example, classic public good problems like climate change and depletion of the ozone layer can be thought to differ in the degree of knowledge about the probabilities associated with the events. While climate change is undoubtedly a case of uncertainty since less precise probability estimates exist in either science or economics, the ozone layer depletion can be characterized more as risk since detailed scientific and economic predictions exist about the depletion of the ozone layer and its impact on human health (e.g., melanoma, cataracts). Risk and uncertainty exist in the private domain as well. For example, in the area of individual health, some estimates exist about the correlation between smoking and lung cancer (could be characterized as risk); however, research is still in its infancy on the connection between working with animals and being susceptible to diseases like bird flu (a case of uncertainty).
In addition to the risks and uncertainties mentioned above, the process of policy making involves further uncertainties due to the unknown information about the decisions of others. In the standard noncooperative experimental setting of public goods and common pool resource games, there is always uncertainty about the ultimate size of the public good or resource request, as subjects do not know how other group members will behave. Messick, Allison, and Samuelson (1988) refer to this uncertainty as "strategic uncertainty." (2)
An implicit assumption underlying these public goods and common pool resource dilemma games is that the optimal size of the public good or the carrying capacity of the commons is known and that there is no uncertainty associated with the benefits of the public or private good. (3) However, in many problems, decision makers do not know with certainty the optimal level of public goods of the carrying capacity of the resource. For example, continuing with the example of climate change, the optimal level of carbon dioxide emissions abatement and the costs and benefits of proposed mitigation strategies are not known with certainty. Messick, Allison, and Samuelson (1988, 678-79) introduced the terminology "environmental uncertainty" to distinguish this external factor from strategic uncertainty: "Environmental uncertainty refers to environmental variables that determine which group action is best, while [strategic] uncertainty centres on how other group members will respond ... The problem that is raised by the environmental uncertainty is the problem of optimality or efficiency, while the problem raised by [strategic] uncertainty is (...) coordination."
Several researchers have found that the distinction between strategic and environmental uncertainty is important in decision making. (4) Existing literature focuses mainly on uncertainty in common pool resource dilemmas where risk and uncertainty are defined in terms of threshold uncertainty. Messick, Allison, and Samuelson (1988), for example, incorporated a probabilistic destruction of the resource when the safe yield was surpassed. In their experiment, groups of four subjects participated in a single-trial task. If the group total request were below a certain threshold, each group member received his request. If the group total were above the threshold, a coin was flipped to determine whether the members would receive their requests. Messick, Allison, and Samuelson (1988) show that even when strategic uncertainty was absent (i.e., there was only one participant), the random size of the resource led to a suboptimal outcome. Rapoport et al. (1992) used mean-preserving spreads to model increasing environmental uncertainty regarding the size of the common pool resource. The authors found in a five-person common pool resource experiment that subjects dealt with the strategic uncertainty by requesting roughly one-fifth of the mean amount available. As the environmental uncertainty increased, however, subjects requested more than an equal share from the resource. Associated with the observed overexploitation of the common resources, research (Rapoport et al. 1992; Suleiman, Budescu, and Rapoport 1966) suggests that subjects' estimates of the resource size increase as the resource uncertainty increases. One explanation provided by Rapoport et al. (1992) is that people tend to perceive the variability and central tendency of probability distributions to be positively correlated. A second explanation is that when people are asked to provide their best estimate of an unknown resource, their estimate will be biased by what they hope for. Such a tendency has been labeled in the literature as "optimism," "outcome desirability bias," or simply "wishful thinking" (Hogarth 1987). (5) While there is extensive research in both the psychological and the economics literature on the effects of resource uncertainty in common pool resource situations under simultaneous and sequential protocols (Budescu, Rapoport, and Suleiman 1995a, 1995b; Gustafsson, Biel, and Garling 1999, 2000; Rapoport and Suleiman 1992; Rapoport et al. 1992; Suleiman and Rapoport 1988), there is considerably less research on the effects of risk and uncertainty in public goods situations. (6)
Wit and Wilke (1998), for example, investigated the effects of environmental uncertainty (regarding the provision threshold) and strategic uncertainty on contribution to step-level public goods. In the "low--environmental uncertainty" treatment, subjects were informed that the provision threshold would be randomly sampled from a uniform distribution defined over the range [800, 1,000], while in the "high--environmental uncertainty" treatment, the range was increased to [400, 1,400]. The strategic uncertainty was manipulated by giving subjects false information regarding the variance of contributions of others. The authors found lower contributions under high environmental uncertainty with the most dramatic drop in the contribution levels when high environmental uncertainty was coupled with high strategic uncertainty. Au (2004) compared voluntary contributions in a public goods experiment in the case where the provision point was fixed at m = 3 to the case where the provision point could take the values m = 2, 3, and 4 with equal probability. He found that the provision rate for the public good was significantly higher when the provision point was known precisely. Gustafsson, Biel, and Garling (2000) compared the voluntary contributions to public goods with the same expected provision threshold but different variances. In general, the average contribution was smaller in the high-variance group. While Wit and Wilke (1998) and Gustafsson, Biel, and Garling (2000) applied a simultaneous design, Au (2004) replicated these results using a sequential decision-making protocol (i.e., subjects made decisions sequentially). These findings are consistent with the results from common pool resource studies showing that higher environmental uncertainty reduces the rate of intensity of cooperative behavior.
Suleiman, Budescu, and Rapoport (2001), however, report that the effects of environmental uncertainty in the common pool resource and public goods games exhibit qualitative differences. While all studies in the common pool resource dilemma show that individual requests increase with environmental uncertainty, Suleiman, Budescu, and Rapoport (2001) find that the effect of threshold uncertainty is moderated by the threshold mean: contributions to the public good increased as a function of uncertainty for the lower threshold mean and decreased for the higher threshold mean. McBride (2006) and Gustafsson, Biel, and Garling (2000) also show that outcomes in public good games may be more sensitive to uncertainty and parameter changes compared to common pool resource games. The differential effect of environmental uncertainty in the two types of game was also documented by van Dijk et al. (1999, 111) who concluded that "environmental uncertainty may affect choice behaviour differently in public good dilemmas, than in resource dilemmas." (7)
Very little research has been conducted to examine the impact of endogenous probabilities in these social dilemma games. In the common pool resource experiments of Walker and Gardner (1992), the probability of destruction of the resource was an increasing function of total group exploitation of the commons, and such destruction ended the experiment. The authors found that in a decision environment with well-defined probability and significant opportunity costs of destruction, efficiency was very low (21%) and the resource was quickly destroyed (longest experiment consisted of six periods). Using members' individual effort in a team sport as a motivation, Dickinson (1998) conducted a public good experiment where he introduced environmental uncertainty as an endogenous probability of provision into the voluntary contributions mechanism--the provision of the public good (winning in the sports event) was conditional on the aggregate contribution level. The author found that the increase in probability as a function of contribution levels, per se, did not significantly affect contribution levels (only through the increase in marginal incentives to contribute). In another experiment, Dickinson (2001) measured the efficiency-improving properties of giving incentives in a public good game where monetary prizes were given to high contributors defined both in terms of absolute contributions and then in terms of relative ability (relative to initial endowment). Results show that such private incentives increased the efficiency levels by 10%-25%.
In this paper, we explore, using laboratory experiments, how environmental and strategic uncertainties when coupled together influence decisions in the context of the provision of public and private goods. This paper makes several contributions. While previous research has examined different aspects of risk and uncertainty in isolation and often in a common pool resource setting, this paper distinguishes between risk and uncertainty and between variability in public versus private goods in a standard public good setting. Complimenting the literature on common pool resource dilemmas where uncertainty is defined in terms of threshold uncertainty, this paper defines risk and uncertainty in light of different probability distributions. The treatments considered allow us to isolate strategic uncertainty from several forms of environmental risk and uncertainty using known (environmental risk treatments), unknown (environmental uncertainty treatments), and endogenous (incentives treatments) probabilities. In the baseline treatment, no environmental uncertainty exists and only strategic uncertainty influences the outcome. Throughout the rest of the treatments, strategic uncertainty (e.g., group size, communication, marginal incentives to contribute) is kept constant and environmental risk and uncertainty are introduced in both private and public goods settings. A within-subject design and a multiple-period game allow us to examine these areas of research in a systematic manner to gain further insights into different aspects of decision making.
The current design was chosen to allow for comparison with existing research but also to ensure that the effects of different kinds of environmental uncertainty, on the contribution levels, are isolated. While Wit and Wilke (1998) manipulated the strategic and environmental uncertainty simultaneously, we keep the strategic uncertainty constant across treatments in order to study the effects of environmental uncertainty. Also, while previous research defined the environmental uncertainty by increasing the range from where the threshold of the...
|
