Philipp Strack's Contributions to Behavioral Economics and Information Economics
The 2024 John Bates Clark Medal is awarded to Philipp Strack in recognition of his contributions to be-havioral economics and information economics.This paper first introduces Strack's background,and then summarizes his contributions to studies of decision-making and behavioral economics,the measurement of information cost functions and application of these concepts,and the development of a new analytical approach to mechanism design.Firstly,in the fields of decision theory and behavioral economics,Strack and his collaborator expanded the theoretical model to study how the learning process,information collection,information update process,and decision-making speed af-fect the behavior of decision-makers in a dynamic environment,providing new explanations for the limitations of the tradi-tional decision theory in terms of model and empirical measurement.On the issue of intertemporal decision-making time in-consistency,they pointed out that the present bias of individual choices can be identified by observing the completion time of tasks,which can be used to explain the phenomenon that a present-biased person may often excessively delay the comple-tion of tedious tasks.Meanwhile,they constructed a general model to characterize the decision-making problem of individu-als with misinterpreting themselves.For example,a smoker with self-present bias predicts that he will quit smoking soon,but eventually smokes more.Or,an employer with implicit racial bias may think that he is treating all applicants equally,and then end up with a racially unbalanced team.They also studied how overconfidence distorts beliefs and leads to bad ac-tions with consequences not only at the individual level but also at a broader level,towards society,causing prejudice and racism.Secondly,based on the idea of Blackwell's experiment,they developed an axiomatic theory of information acquisition that captured the idea of constant marginal costs in information production and derived three axioms:(1)the monotonicity axiom(experiments with more precise information cost more);(2)the additivity axiom(the total cost of each independent experiment is the sum of the costs of the individual experiments);and(3)the dilution axiom(the cost of generating an ex-periment with probability half equals half the cost of generating it with probability one).Those axioms have a wide applica-tion in the field of rational inattention,the Wald model of sequential sampling,decision theory,dynamic information acqui-sition model,probability theory,and information theory.In the field of monetary economics and finance,these axioms pro-vide a complement to the limitations of the information cost function based on Shannon entropy.Furthermore,it also very helpful for the study of data markets if linking the information costs to economic stories.In addition,based on the definition of garbling in information structure,they proposed that information can be seen as a signal of state,so many concepts such as privacy protection can be formalized.They characterized all privacy-preserving signals under any state space and privacy set.These different privacy-preserving signals can be applied to research on statistical discrimination and fairness in algo-rithm design and optimal privacy protection disclosure in auctions,and identify price discrimination and the market segmen-tation conditions that prevent monopolists from implementing price discrimination.Thirdly,they used extreme points and majorization methods to show that many well-known optimal design and deci-sion problems share a basic common structure:all these problems can be reduced to the choice of an optimal element—that maximizes a given functional—from the set of monotonic functions that are either majorized by,or majorize,a given mono-tonic function.Their pathbreaking work on extreme points and majorization unlocks new doors for the theory of mechanism design.The theorems derived under this new analytical approach are equivalent to the incentive compatibility constraints in equivalence and optimality of mechanisms for revenue-maximizing rank-item auctions,matching contests,persuasion with preferences over the posterior mean,optimal delegation,and decision-making under uncertainty.At the same time,the theorems in the new framework can also be used to study equivalent results in the mechanism design,Myerson's original de-sign problem,and principal-agent problem with non-expected utility.Finally,they also contributed to the research areas including the design of auctions,organ exchanges,dynamic pric-ing of goods,optimal structuring of lockdowns in a pandemic,design of cryptocurrencies,and optimal taxation of goods with negative externalities(such as carbon emissions).
Philipp StrackBehavioral EconomicsInformation EconomicsMechanism Design
NETL
NSTL
万方数据
