Bridging the gap between laboratory experiments and naturally occurring markets: An inferred valuation method
Introduction
In the matter of a few decades, experimental economics has gone from infancy to a well-established field within economics. Since its inception, a question often raised is whether results obtained from experimental studies parallel real-world economies. Levitt and List [26] have argued, for example, that experimental results may fail to hold up outside the laboratory because, among other things, participants in a laboratory experiment know they are watched and their actions are being scrutinized. Unlike the objects studied in physical sciences, humans know they are a part of an experiment, a fact that has the potential to affect behavior in a variety of ways such as that attributed to the classic Hawthorne Effect. For example, List [29] found in a series of experiments that the pro-social behavior present in laboratory experiments was sharply attenuated in the field. He found that the theory of self-interest seemed to best describe field behavior whereas social preference theories related to fairness, trust, and reciprocity seemed to best describe laboratory behavior.
In addition to the observation that social preferences can drive a wedge between laboratory and field behavior, other research has pointed to people's knowledge and experience as a potential factor contributing to the lab–field behavior gap. List [30], [31], for example, found that the endowment effect frequently observed in laboratory experiments was much less apparent in the field among those with frequent trading experience. Neoclassical theory best described experienced traders in the field, whereas prospect theory best described inexperienced traders.
Despite the potential for people to behave differently in laboratory experiments and in the field, some studies have found that a lab–field behavior gap is absent in some circumstances [7], which naturally leads to questions regarding the nature and cause of the lab–field behavior gap.1 Others have argued that focusing on concerns about external validity mistakes the purpose of experimental methods, arguing that laboratory experiments test general theories and principles that should apply to all economies, those found in the field as well as those created in the laboratory, e.g., see [37]. Despite the compelling nature of such an argument, it has not gone unchallenged.
Studies such as [20], [29], [31] illustrate that suitability of economic theories, rather than being all-encompassing general principles, depends on the context in which the decision occurs, the experience of participants with the decision context, and whether people know they are taking part in an experiment. Indeed, Harrison and List [17] draw a conceptual distinction between various types of laboratory and non-laboratory experiments, and argue that despite our best attempts, contextual factors influencing behavior can never be completely controlled in an experimental setting. There is still much to debate, however, about the robustness of the observed differences in laboratory experiments and naturally occurring markets and about the seriousness of these findings for empirical and theoretical works alike.
In this paper, we explore the relationship between behavior observed in a laboratory market and that observed in a naturally occurring field market. We investigate the role of two factors that have a potential to drive a wedge between laboratory and field behavior: social concerns and inexperience with the trade good. A primary contribution of this study is the introduction of what we call an inferred valuation method, in which laboratory subjects predict others’ behavior in the field. We argue that inferred valuation has the potential to provide better predictions of field behavior if social concerns are the primary contributor to the lab–field behavior gap.2
Section snippets
Conceptual model
We assume that people know their value for a good with certainty but derive utility from normative or moral considerations.3
Inferred valuation
When there are strong normative motivations associated with a good, the lab–field behavior gap might be reduced by using a tool we refer to as inferred valuation. This section discusses the merits of this inferred valuation method. In particular, consider how an individual would respond to a question regarding how much another person is willing to pay for an increase in E in the field, where there is some monetary reward for accuracy. It is supposed that the analyst had available an unbiased
Methods
To test the predictions of the model outlined in Fig. 1, we investigated preferences for three new products that had previously been unavailable for purchase in the local grocery store where the field study was conducted. Each of the products possessed a normative dimension related to organic, “environmentally friendly,” and/or local production. Further, the products were chosen so that they differed in the degree of familiarity to consumers. The three products selected for study were as
Results
Before analyzing the self- and inferred valuations, it is important to further investigate how the goods studied varied along the key theoretical variables of interest: normative motivations and commitment costs. Recall that in relation to the lab–field gap, Fig. 1 implies specific hypotheses in two cases: (i) when commitment costs are low and normative motivations are high, willingness-to-pay in the lab is expected to exceed that in the field; and (ii) when commitment costs are high and
Conclusions
It is often questioned whether results from laboratory experiments accurately correspond to the world the experiments are constructed to model. This study represents an attempt to help answer this question. Key factors that have the potential to drive a wedge between laboratory and field behavior relate to concerns for how people perceive they are viewed by others, a desire by participants to please the researchers, people's experience with and knowledge of the traded good, and the ability of
References (45)
- et al.
An economic model of moral motivation
J. Public Econ.
(2003) - et al.
Specifying and testing econometric models for rank-ordered data
J. Econometrics
(1987) - et al.
Do players correctly estimate what others do? Evidence of conservation in beliefs
J. Econ. Behavioral Organ.
(2002) - et al.
Honestly, why are you driving a BMW?
J. Econ. Behavioral Organ.
(2006) - et al.
Value elicitation in laboratory and retail environments
Econ. Letters
(2003) - et al.
Availability: a heuristic for judging frequency and probability
Cognitive Psych.
(1973) - et al.
The illusion of courage in social predictions: underestimating the impact of fear of embarrassment on other people
Org. Behav. Human Dec. Processes
(2005) - et al.
A new explanation for the WTP/WTA disparity
Econ. Letters
(2001) Giving with impure altruism: applications to charity and Ricardian equivalence
J. Polit. Econ.
(1989)A theory of social interactions
J. Polit. Econ.
(1974)
Assessing the potential demand for electric cars
J. Econometrics
Incentives and prosocial behavior
Amer. Econ. Rev.
The external validity of experimental economics techniques: analysis of demand behavior
Econ. Inquiry
Is the experimental auction a dynamic market?
Environ. Resource Econ.
The Approval Motive: Studies in Evaluative Dependence
Feeling holier than thou: are self-serving assessments produced by errors in self- or social prediction?
J. Personality Soc. Psych.
Belief, Attitude, Intention and Behavior
Social desirability bias and the validity of indirect questioning
J. Cons. Res.
Anatomy of an experimental political stock market
Amer. Econ. Rev.
Vox populi
Nature
Cited by (84)
Valuing insurance against small probability risks: A meta-analysis
2024, Journal of Behavioral and Experimental EconomicsConsumer behavior in choosing microplastic contaminated seafood across different countries: The role of cultural and attitudinal factors
2023, Economic Analysis and PolicySocial desirability bias in the environmental economic valuation: An inferred valuation approach
2021, Ecological Economics