Mark Dean

Job Market Candidate

New York University
Department of Economics
19 W. 4th Street, 6th Floor
New York, NY 10012
+1.646.266.5135
mark.dean AT nyu.edu

Peer Reviewed Publications

Other Publications

Current Research

Status Quo Bias in Large and Small Choice Sets - Job Market Paper

This paper introduces models of status quo bias based on the concept of decision avoidance, by which a decision maker may select the status quo in order to avoid a difficult decision. These models capture the experimental finding that the status quo is more frequently chosen in larger choice sets. This phenomenon violates the predictions of current preference-based models of status quo bias that assume a decision maker with a fixed status quo will make consistent choices. Using laboratory experiments, I show that subjects in large choice sets do exhibit behavior in line with decision avoidance, while in small choice sets, preference-based models offer a better explanation of behavior. These findings raise questions for advocated policies of "benign paternalism."

Link

Measuring Beliefs and Rewards: A Neuroeconomic Approach (with Andrew Caplin, Paul Glimcher and Robb Rutledge) - Job Market Paper

The neurotransmitter dopamine is central to the emerging discipline of neuroeconomics; it is hypothesized to encode the difference between expected and realized rewards and thereby to mediate belief formation and choice. We develop the first formal test of this theory of dopaminergic function, based on a recent axiomatization by Caplin and Dean. These tests are satisfied by neural activity in the nucleus accumbens, an area rich in dopamine receptors. Intriguingly, we find evidence for separate positive and negative reward prediction error signals, a novel empirical result suggesting that behavioral asymmetries in response to losses and gains may be encoded by activity in the nucleus accumbens. Our findings provide researchers with new methods for studying beliefs, learning, and choice.

Link

How Consistent Are your Choice Data? (with Daniel Martin)

This paper describes a fast and exact method for computing the Houtman-Maks (HM) Index - the largest subset of a choice data set consistent with acyclicality. This measure provides a metric for determining how close a set of choice data are to 'rationality', in the sense that they can be modeled as having been derived from the maximization of a complete preference relation. One reason that this measure has not been widely used is that it is extremely computationally intensive. We show that the problem of finding the maximal acyclic subset is isomorphic to a well-studied problem within computer science: the Minimum Set Covering Problem (MSCP). While the MSCP is NP-hard in the strong sense, there are a wide variety of algorithms built to solve this problem quickly and exactly for reasonably-sized data sets. This paper describes some of these algorithms and presents simulation results to demonstrate that our algorithm can be used to calculate the HM Index in under a second for cases that have previously been found insoluble.

Link

Information Search and the Choice Process (with Andrew Caplin)

Standard choice theory cannot explain behavioral phenomena such as framing effects and status quo bias. An important open question is when such behaviors result from changing preferences as opposed to incomplete information. We develop clean theoretical procedures for separating these causes by extending standard choice data to include data on the evolution of choice with time. Using such "choice process" data, we axiomatically characterize standard forms of search behavior such as that based on achievement of reservation utility, which is the optimal stopping rule for a decision maker with fixed disutility of search. In doing so, we expand the concept of revealed preference to the incomplete information environment. Unlike standard search theory, our axiomatic approach treats the cost of search, the order of search, and reservation utilities as outputs of the model rather than as inputs. Preliminary testing of our models is underway, and promises to increase understanding of the many choices that are anomalous from the view point of fully informed rational decision making.

Link

Over-Inference and Excess Sensitivity of Consumption

The Permanent Income Hypothesis (PIH) states that, for a rational consumer, consumption should only change in response to 'news', or unpredictable changes in the environment. However, several studies using panel data on consumption and income find that household level changes in consumption are negatively correlated with lagged income. This paper shows that such data can be explained if consumers consistently overestimate the permanence of shocks to their income. Evidence for such a bias in decision making is provided by Offerman and Sonnemans, who show that experimental subjects overestimate the correlation between sequential outcomes in a coin flipping experiment. Unlike previous explanations for excess sensitivity, bias in expectations formation can explain the finding of Japelli and Pistaferri that consumption of Italian households is sensitive to lagged actual income, but not income expectations. The paper concludes with an experiment which recreates consumption and savings decisions, and tests the hypothesis that the biases found by Offerman and Sonnemans also occur in this setting.

Link Coming Soon