Periodic Reporting for period 4 - BEHAVFRICTIONS (Behavioral Implications of Information-Processing Frictions)
Periodo di rendicontazione: 2022-12-01 al 2023-11-30
The BehavFrictions project studied micro-founded models of stylized facts from behavioral economics. This approach seeks to unify various behavioral biases, identify the domains in which these biases operate, and inform normative reasoning about them.
An example of this approach is the paper Attention Please!, co-authored with Olivier Gossner and Colin Stewart as part of this grant project, and published in Econometrica in 2021. In this paper, we proposed a novel mechanism that explains how directing the attention of a decision maker to a particular option increases the chance that she will choose this target option, even if the target is of low quality and her attention is therefore likely to lead to negative impressions. Our model is motivated by attention-grabbing techniques commonly used in marketing and advertising. The model could contribute to discussions on the regulation of advertising and marketing.
In response to the Covid crisis, I devoted part of my research in 2020 to Covid testing policies. Similar to information-processing frictions in individual decision-making, information-acquisition resources during the pandemic—Covid tests—were scarce at the societal level. Medical tests are mathematically conceptualized similarly to information-acquisition strategies in information economics—both are modeled as Blackwell experiments. I explored these conceptual similarities and authored three academic papers on epidemiological testing and related interventions.
During the final grant period, I explored models of stochastic growth that originate in the literature on growth-optimal investment portfolios (Kelly, 1956). The financial hedging studied in this literature can be understood as a passive substitute for active learning: the hedging agent distributes her wealth across many assets without actively learning about the assets’ future returns. However, ex post, the over-performing assets become over-represented in her portfolio. In the paper entitled Decision Theory and Stochastic Growth, joint with Arthur Robson and Larry Samuelson, and published in American Economic Review: Insights in 2023, we show that the outcome of optimal hedging is identical to the choice of a rationally inattentive agent who actively learns about the returns at an entropic cost of information. I am continuing to explore the connection between stochastic growth and models of information processing in a series of papers.
An example of this approach is the paper Attention Please!, co-authored with Olivier Gossner and Colin Stewart as part of this grant project, and published in Econometrica in 2021. In this paper, we proposed a novel mechanism that explains how directing the attention of a decision maker to a particular option increases the chance that she will choose this target option, even if the target is of low quality and her attention is therefore likely to lead to negative impressions. Our model is motivated by attention-grabbing techniques commonly used in marketing and advertising. The model could contribute to discussions on the regulation of advertising and marketing.
In response to the Covid crisis, I devoted part of my research in 2020 to Covid testing policies. Similar to information-processing frictions in individual decision-making, information-acquisition resources during the pandemic—Covid tests—were scarce at the societal level. Medical tests are mathematically conceptualized similarly to information-acquisition strategies in information economics—both are modeled as Blackwell experiments. I explored these conceptual similarities and authored three academic papers on epidemiological testing and related interventions.
During the final grant period, I explored models of stochastic growth that originate in the literature on growth-optimal investment portfolios (Kelly, 1956). The financial hedging studied in this literature can be understood as a passive substitute for active learning: the hedging agent distributes her wealth across many assets without actively learning about the assets’ future returns. However, ex post, the over-performing assets become over-represented in her portfolio. In the paper entitled Decision Theory and Stochastic Growth, joint with Arthur Robson and Larry Samuelson, and published in American Economic Review: Insights in 2023, we show that the outcome of optimal hedging is identical to the choice of a rationally inattentive agent who actively learns about the returns at an entropic cost of information. I am continuing to explore the connection between stochastic growth and models of information processing in a series of papers.
BehavFrictions supported the following publications:
1. On the Cost of Misperception: General Results and Behavioral Applications, with Olivier Gossner, Journal of Economic Theory
We explain perception biases as second-best adaptations in situations where perception mistakes are unavoidable but their distribution can be controlled. The framework provides unified explanations for the illusion of control, overprecision, and the overweighting of small probabilities.
2. Selective Sampling with Information-Storage Constraints, with Philippe Jehiel, Economic Journal
We model sequential learning of a decision-maker with bounded memory. The optimal procedure exhibits a bias toward news that confirms prior beliefs, while information contradicting the prior is more likely to be forgotten, consistent with confirmation bias.
3. Habits as Adaptations: An Experimental Study, with Ludmila Matyskova, Brian Rogers, and Keh-Kuan Sun, Games and Economic Behaviour
We experimentally test the dynamic rational-inattention model of Steiner, Stewart, and Matejka (Econometrica 2017). When incentives are serially correlated, repeating one’s own action is a good substitute for costly information acquisition, leading to habit formation. Our lab experiment supports this intuition.
4. Attention Please!, with Olivier Gossner and Colin Stewart, in Econometrica
We propose a mechanism explaining how directing attention to an item increases the likelihood that this target item will be chosen, even if the item is of low value and its inspection is likely to reveal adverse information. The effect arises when the choice process is based on approval learning—that is, when the decision-maker chooses the item once they believe it is good enough. In such cases, attention-grabbing increases the likelihood that the target of the manipulation will be approved before other items in the choice set.
5. Optimal Test Allocation, with Jeff Ely, Andrea Galeotti, and Ole Jann, Journal of Economic Theory
We characterize optimal epidemiological testing when an authority holds a portfolio of tests heterogeneous in their sensitivities and specificities, and the individuals to be tested differ in their pre-test infection probabilities and the costs of possible decision errors.
6. Rotation as Contagion Mitigation, with Jeff Ely and Andrea Galeotti, Management Science
We study organizations, such as schools, that seek to organize member attendance through rotation schemes. We derive the epidemiologically optimal rotation frequencies.
7. Merit of Test: Perspective of Information Economics, with Andrea Galeotti and Paolo Surico, Health Policy and Technology
This short note aimed at medical professionals explains the value of testing from the perspective of economic theory.
8. Boundedly Rational Demand, with Pavel Kocourek and Colin Stewart, Theoretical Economics
The standard notion of consumer surplus is based on the assumption that consumers engage in correct marginal reasoning given the prices. Instead, we assume that consumers may make erroneous choices and provide bounds on consumer surplus given the market data.
9. Decision Theory and Stochastic Growth, with Arthur Robson and Larry Samuelson, American Economic Review: Insights
We study wealth accumulation in a simple stochastic growth process. Depending on the nature of the stochasticity, the choices of the winners of the growth process—who eventually hold nearly all wealth—either coincide with the solution to a rational inattention problem or a related problem of wishful thinking.
1. On the Cost of Misperception: General Results and Behavioral Applications, with Olivier Gossner, Journal of Economic Theory
We explain perception biases as second-best adaptations in situations where perception mistakes are unavoidable but their distribution can be controlled. The framework provides unified explanations for the illusion of control, overprecision, and the overweighting of small probabilities.
2. Selective Sampling with Information-Storage Constraints, with Philippe Jehiel, Economic Journal
We model sequential learning of a decision-maker with bounded memory. The optimal procedure exhibits a bias toward news that confirms prior beliefs, while information contradicting the prior is more likely to be forgotten, consistent with confirmation bias.
3. Habits as Adaptations: An Experimental Study, with Ludmila Matyskova, Brian Rogers, and Keh-Kuan Sun, Games and Economic Behaviour
We experimentally test the dynamic rational-inattention model of Steiner, Stewart, and Matejka (Econometrica 2017). When incentives are serially correlated, repeating one’s own action is a good substitute for costly information acquisition, leading to habit formation. Our lab experiment supports this intuition.
4. Attention Please!, with Olivier Gossner and Colin Stewart, in Econometrica
We propose a mechanism explaining how directing attention to an item increases the likelihood that this target item will be chosen, even if the item is of low value and its inspection is likely to reveal adverse information. The effect arises when the choice process is based on approval learning—that is, when the decision-maker chooses the item once they believe it is good enough. In such cases, attention-grabbing increases the likelihood that the target of the manipulation will be approved before other items in the choice set.
5. Optimal Test Allocation, with Jeff Ely, Andrea Galeotti, and Ole Jann, Journal of Economic Theory
We characterize optimal epidemiological testing when an authority holds a portfolio of tests heterogeneous in their sensitivities and specificities, and the individuals to be tested differ in their pre-test infection probabilities and the costs of possible decision errors.
6. Rotation as Contagion Mitigation, with Jeff Ely and Andrea Galeotti, Management Science
We study organizations, such as schools, that seek to organize member attendance through rotation schemes. We derive the epidemiologically optimal rotation frequencies.
7. Merit of Test: Perspective of Information Economics, with Andrea Galeotti and Paolo Surico, Health Policy and Technology
This short note aimed at medical professionals explains the value of testing from the perspective of economic theory.
8. Boundedly Rational Demand, with Pavel Kocourek and Colin Stewart, Theoretical Economics
The standard notion of consumer surplus is based on the assumption that consumers engage in correct marginal reasoning given the prices. Instead, we assume that consumers may make erroneous choices and provide bounds on consumer surplus given the market data.
9. Decision Theory and Stochastic Growth, with Arthur Robson and Larry Samuelson, American Economic Review: Insights
We study wealth accumulation in a simple stochastic growth process. Depending on the nature of the stochasticity, the choices of the winners of the growth process—who eventually hold nearly all wealth—either coincide with the solution to a rational inattention problem or a related problem of wishful thinking.
Additionally, BehavFrictions supported the following working papers:
10. Risk Perception: Measurement and Aggregation, with Nick Netzer, Arthur Robson, and Pavel Kocourek, revise and resubmit in the Journal of European Economic Association
The paper explains risk attitudes described in prospect theory—such as the S-shaped utility function and the overweighting of small probabilities—using a model that features encoding and decoding processes involved in the perception of risky prospects.
11. Growth and Redistribution: The Hedging Perspective, with Larry Samuelson, a working paper
We characterize the wealth redistribution scheme that maximizes long-run economic growth using the modeling framework of Kelly’s betting. We highlight connections between growth-optimal portfolios and models of cognition used in machine learning and cognitive science.
10. Risk Perception: Measurement and Aggregation, with Nick Netzer, Arthur Robson, and Pavel Kocourek, revise and resubmit in the Journal of European Economic Association
The paper explains risk attitudes described in prospect theory—such as the S-shaped utility function and the overweighting of small probabilities—using a model that features encoding and decoding processes involved in the perception of risky prospects.
11. Growth and Redistribution: The Hedging Perspective, with Larry Samuelson, a working paper
We characterize the wealth redistribution scheme that maximizes long-run economic growth using the modeling framework of Kelly’s betting. We highlight connections between growth-optimal portfolios and models of cognition used in machine learning and cognitive science.