This paper identifies and sets out to explain a set of striking similarities in boundedly
rational behavior across four domains: choice under risk, choice under ambiguity,
belief updating, and survey expectations about economic variables.
In each of these domains, behavior in experiments and surveys tends
to be insensitive to variation in objective probabilities, generating a
characteristic response pattern of compression towards 50:50 as in the classical
probability weighting function. Building on existing models of cognitive noise,
we formally propose that the unifying underlying mechanism is cognitive uncertainty:
people’s subjective uncertainty about what the rational action is,
which induces them to shrink probabilities towards a mental default.
We introduce an experimental measurement of cognitive uncertainty and show
that the responses of individuals with higher cognitive uncertainty exhibit stronger
compression of objective probabilities in choice under uncertainty,
belief updating, and survey expectations. Our framework makes additional
predictions that we test using exogenous manipulations of both cognitive
uncertainty and the location of the mental default. The results
provide causal evidence for the role of cognitive uncertainty in belief formation
and choice, which we quantify through structural estimations.
This paper studies why belief formation errors arise in a novel updating task that captures a key
feature of information structures in practice: people need to learn about a
state of the world from information that also depends on other unobserved states,
such as when inferring effort (X) from observed performance (S = X + Y) that is influenced by luck (Y).
The first part uses a series of laboratory and online experiments to
causally demonstrate a pervasive neglect of alternative causes in information structures,
leading to excessively sensitive and overprecise beliefs.
The second part explores the mechanisms behind this neglect.
Evidence from more than twenty treatments consistently shows that
inattention to alternative causes is not driven by excessive complexity,
computational errors or deliberate effort reduction.
Instead, people spontaneously form incomplete mental
models of the problem that determine how information is processed.
They are confident in their misspecified models and unaware of the resulting error.
These mental representations are not stable but context-dependent: cues that direct attention to the neglected part of the problem alleviate the bias.
Selected work in progress
Heterogeneity of Gain-Loss Attitudes and Expectations-Based Reference Points with Lorenz Goette, Alex Kellogg and Charles Sprenger
This project examines the role of heterogeneity in gain-loss attitudes for identifying models of expectations-based reference dependence (Kőszegi and Rabin, 2006, 2007) (KR). Different gain-loss attitudes lead to different signs for KR comparative statics. Failure to account for the known heterogeneity in gain-loss attitudes is a central confounding factor challenging prior tests of the KR model conducted under the assumption of universal loss aversion. We document heterogeneous treatment effects over gain-loss types in both an initial experiment and an exact replication. Recognizing heterogeneity over types allows us to both recover the KR model’s central predictions, and account for inconsistency across prior empirical tests.
Breaking Trust: On the Persistent Effect of Banking Crisis Experience with Tom Zimmermann
A popular narrative associates negative shocks to the banking sector with “crises of trust”. Exploiting cohort-variation within and across countries, we investigate this link systematically. We find a robust and sizeable effect of banking crises experience on interpersonal trust: Our benchmark estimate implies that experiencing a banking crisis 20 years ago decreases current trust by one-third as much as a recent personal traumatic experience. We provide evidence that the effect is not driven by income losses during crises but it is consistent with a breach of trust channel, relating our findings to a well-established phenomenon in the social sciences.
Intertemporal Altruism with Armin Falk and Philipp Eisenhauer