Adriaan R. Soetevent

Empirics of the Identification of Social Interactions: An Evaluation of the Approaches and Their Results

Over the last decade, the study of social interactions in economic decision making has become an important area of research. The main objective of this paper is to survey the extent to which recent empirical contributions have succeeded in overcoming the identification problems as first formulated by Manski (1993). This discussion is followed by a comparison of empirical studies in three key areas of research: neighborhood effects, substance use among teenagers, and peer effects among university roommates. Finally, I discuss questions like: Can economists restrict attention to a specific subcategories of social interactions? How do we define social groups, and what is the importance of social interactions for public policy?

Detailed Data and Changes in Market Structure: The Move to Unmanned Gasoline Service Stations

We illustrate the impact of detailed data in empirical economic research by considering how the increased data availability has changed the scope and focus of studies on retail gasoline pricing. We show how high-volume, high-frequency price data help to identify and explain long-term trends using original data for the Dutch retail gasoline market. We find that 22% of the observed increase in the highway/off-highway price gap can be explained by the trend towards more unmanned stations; another 13% can be explained by major-to-non-major re-brandings. In one of the first applications of event study analysis to non-financial price data, we show that the adjustment to the new, lower price level is almost immediate in case of manned-to-unmanned conversions but takes one to two months in case of major-to-non-major re-brandings. The impact of both events is asymmetric with no measurable price impact of changes in the opposite direction.

Trust and Recidivism; The Partial Success of Corporate Leniency Program in the Laboratory

An experiment is conducted were subjects interact repeatedly to examine the effect of a particular leniency program on cartel formation, cartel stability and cartel recidivism. The program leads to lower prices for three reasons. First, non-cooperators are more persistent in their behavior which effectively blocks cartel formation in their respective groups. Second, members of groups that do form a cartel defect more often thus reducing the average cartel lifetime. Third, the difference between the agreed-upon price and the undercutting price is larger. The leniency program does not however affect the probability that a dismantled cartel is re-established.

Bidding to Give in the Field: Door-to-Door Fundraisers Had it Right from the Start

This discussion paper resulted in an article in the Journal of Public Economics (2013). Volume 105, pages 72-85. In a door-to-door fundraising field experiment, we study the impact of fundraising mechanisms on charitable giving. We approached about 4500 households, each participating in either an all-pay auction, a lottery, a non-anonymous voluntary contribution mechanism (VCM), or an anonymous VCM. In contrast to the VCMs, households competed for a prize in the all-pay auction and the lottery. Although the all-pay auction is the superior fundraising mechanism both in theory and in the laboratory, it raised the lowest revenue per household in the field. Our experiment reveals two potential explanations for this anomaly. First, participation in the all-pay auction is substantially lower than in the other mechanisms while the average donation for those who contribute is only slightly higher. We explore various explanations for this lower participation and favor one that argues that competition in the all-pay mechanism crowds out intrinsic motivations to contribute. Second, the non-anonymity may have a negative effect: conditional on donating, households contribute less in the non-anonymous VCM than in the anonymous VCM. Among the non-anonymous mechanisms, the lottery raises the largest revenue per household. Notably, the method that scored best, the anonymous VCM, is the one most used by door-to-door fund raisers in the Netherlands.

Exploitation and induced tacit collusion; A classroom experiment of corporate leniency programs

Recent laboratory experiments support the popular view that the introduction of corporate leniency programs has significantly decreased cartel activity. We develop a classroom experiment that captures key features of leniency programs. Our treatments include an exploitable and a non-exploitable leniency program to highlight possible adverse effects of leniency programs that are too generous. We also examine to what extent a non-exploitable leniency program triggers tacit collusion. The experimental results show that if the efforts of the antitrust authority and the leniency program are directed exclusively to the most straightforward collusive scheme, subjects manage to switch to a more intricate form of coordination. This shift from overt collusion to tacit collusion can serve as the basis for a classroom discussion about the effectiveness of actual leniency programs.

From Overt to Tacit Collusion: Experimental Evidence on the Adverse Effects of Corporate Leniency Programs

Recent laboratory experiments support the popular view that the introduction of corporate leniency programs has significantly decreased cartel activity. The design of these repeated game experiments however is such that engaging in illegal price discussions is the only way for subjects to avoid the one-shot competitive equilibrium. Subjects in the experiment of this paper have multiple feasible Nash equilibrium strategies to avoid the competitive equilibrium. These strategies differ in the difficulty of the coordination problem they have to solve. The experimental results show that if the efforts of the antitrust authority and the leniency program are directed exclusively to the most straightforward collusive scheme, subjects manage to switch to a more intricate form of coordination. This shift from overt collusion to tacit collusion questions the acclaimed success of corporate leniency programs.

The Impact of Process Innovation on Prices: Evidence from Automated Fuel Retailing in the Netherlands

In the last decade, many European countries have seen a sharp increase in the number of automated fueling stations. We study the effect of this process innovation on prices at stations that are automated and their competitors using a difference-in-differences matching strategy. Our estimates show that prices at automated stations drop by 1.0 to 2.1% immediately after conversion and stabilize at this lower level. We find no indication of competitive spillover effects to neighboring sites at the conventional significance levels. Other than previous studies, our estimates do not reveal a difference in impact between early and later adopters of automation.

How Consumers use Gift Certificates

There are two important reasons for consumers to spend gift certificates differently than gifts in cash or non-gift income: a) they are forced to change their shopping pattern because of the conditions imposed by the issuer of the certificates, or b) they purposely separate gift certificates from other sources of income. The first reason implies a welfare loss, the second reason does not. We survey consumers who have just redeemed one or more gift certificates. For the gift certificate considered in our empirical application, we find that consumers are not constrained by the set of accepting merchants but that they do make some small changes in the timing of expenditures because of the certificates no-refund policy. About 14 percent of recipients separate their gift certificates from other income sources in order to buy a product they really love to have. Males tend to spend the certificates on ordinary items, whereas females are more likely to treat themselves by buying more personalized items.

Price-setting behavior in the presence of social interactions

Summary We consider a market with a profit-maximizing monopolistic firm. Utility-maximizing consumers either buy one unit of the good or none at all. The demand for the good is influenced by local social interactions. That is, the utility which a consumer derives from the consumption of the good depends positively on the fraction of other consumers in his own social group that consume the good. We first consider a benchmark case where the population of consumers is not segmented and constitutes one social group. We derive the optimal price and profit of the firm for this case. Next, we analyse the optimal price and profit for the case where the population of consumers is partitioned into two different social groups. Comparing the results for the cases with one and two social groups, it turns out that the partition into groups does not unambiguously gives the firm the opportunity to raise its price and increase its profit. The effects depend on a non-trivial interplay between the strength of the social interaction effect and the specific composition of the social groups.