Hi! I am a researcher in Applied Microeconomics focusing on Behavioral Industrial Organization and Cultural/Institutional Economics. I research how social relationships interact with economic institutions.

I will join the University of Leicester’s Business School starting September 2023.

You can reach me at paul.schaefer@uni-bonn.de.

You can find my CV here.

Job market paper

The Effect of Social Networks on Market Efficiency

I examine the impact of friendships on imperfectly competitive markets with substitutes and complements. I propose and test a model of friendships in these markets. In the model, people exhibit direct linear altruism towards their friends (like firms with common owners). The model predicts that friendships among sellers of substitutes increase prices and decrease efficiency, whereas friendships between sellers of complements decrease prices and increase efficiency. I invite pairs of friends to the laboratory and assign them different roles within a market experiment. Each individual chooses a price for different social networks within the same market. In some social networks, their friend sells a complement; in some, they sell a substitute; and in some, they do not participate. The estimated causal effects of friendships confirm the model’s predictions. A structural model with a homogeneous parameter for altruism among friends rationalizes the experimental data. I also investigate asymmetric social networks, the effect of price transparency, and the accuracy of people’s beliefs about their friend’s prices.

Working papers

Managing Bidder Learning in Retail Auctions (with Simon Schulten (his JMP))

Abstract When firms exploit behavioral biases it is natural to think that, eventually, consumers will learn to avoid their mistakes, which limits exploitation. Profit maximizing firms, however, have an incentive to undermine such learning. We study these learning dynamics in a multi-unit descending price auction setting with a simultaneous fixed price offer. We analyze 8 million bids from over 280.000 unique bidders in retail auctions. Consumers frequently bid more than the fixed price offered by the same seller. Depending on rival bidders actions, those overbids sometimes lead to paying more than the fixed price (overpaying). We argue overpaying increases saliency of the consumers’ mistake by making it payoff relevant and thereby may effect consumer learning. Indeed, bidders who overpaid subsequently overbid less often and are more likely to refrain from submitting a bid compared to bidders who overbid but did not overpay. Methodologically, we discuss identification of our treatment effects using causal graphs and show how these treatment effects identify a three-type structural model of bidder behavior with learning dynamics.


Norm Prevalence and Interdependence: Evidence from a Large-Scale Historical Survey of German speaking Villages (with Radost Holler)

Abstract We use large-scale survey data of German speaking villages from the 1930's to investigate drivers of cooperation, gender, and religious norms. Through geographic cluster analysis, we show that inter-regional variation explains only little heterogeneity in norms. Villages in the same physical and institutional environment still maintain different norms. We argue that local differences in the structure of social relationships can explain intra-regional heterogeneity in norms. We focus on a community's ability to transmit and enforce norms to derive theoretical links between correlates of community social relationships and the number of norms it maintains (norm prevalence). Empirically we find that: (1) norm prevalence is positively related to three correlates of community social relationships: religiously homogeneous villages, villages that border on other villages with a different majority religion, and villages with more within-village social gatherings; (2) villages with stronger community-level social relationships are also less likely to segment their reference group for the cooperation norm to smaller social units; (3) cooperation norms make other norms more likely.