丁香园AV

Arthur Robson

Biological Basis of Economic Preferences

Robson鈥檚 research examines the biological foundations of economic preferences using two themes: (1) considering the neuroscience basis for bounded rationality in economic choices; and (2) examining attitudes to risk.

Together with Larry Samuelson of Yale University, one part of the project will extend the simple model of Robson and Orr PNAS 2021 to allow for a much more general demographic structure. This provides a general account of the equity premium puzzle (EPP). The EPP is a major challenge, not merely to macroeconomics, but to economic theory鈥攊t concerns the fundamental representation of individual preferences over risk and time. The most important contribution of the present approach is to allow as large a wedge as needed between attitudes to idiosyncratic and aggregate risks, while keeping reasonable attitudes to tradeoffs over time. A key component of the research will be a book on the biological approach to preferences and behaviour, also joint with Larry Samuelson, that will intuitively outline and integrate previous work on the biological basis of preferences and behaviour, along with material from other disciplines. 

David Freeman

Bracketing in Dynamic Risk-Taking

Most economic decisions involve a sequence of risk-taking choices, and behavioural economists have shown that people systematically make choices that cannot be accommodated by standard economic theory. Why do some people gamble in casinos even though it is a money-losing proposition on average? Why do some people invest only in bonds, avoiding the stock market, and almost certainly end up with less money in long run? Why do people tend, on average, be more prone to hold onto stocks that have declined in price? To explain each of these anomalies, behavioural economics uses different, mutually inconsistent sets of modelling assumptions about how people incorporate ("bracket") the past risks they faced and future risk-taking opportunities they will face into their current decision-making. This results in a misleading understanding of the individual biases that shape how people gamble and invest over their lifetimes, how these biases influence markets, and how regulation and fiscal and monetary policies can and should respond.

The proposed project will provide a new experimental design to obtain the evidence needed to build behaviourally-informed models of bracketing in dynamic risk-taking suited for positive and normative economic analysis of such decisions.  Clear guidance for economists looking to build their economic analysis and policy recommendations on empirically accurate assumptions about dynamic risk-taking, and will also provide an empirical basis for constructing normative models of preferences over dynamic risks suitable for welfare analysis.

Kevin Schnepel

Causal Determinants of Recidivism

This research project aims to discover whether and how employment opportunities and access to public assistance are causal determinants of reoffending (or "recidivism") for individuals with prior criminal convictions. The project will significantly advance the existing literature in this area and take on the following challenges:

(1)  Identifying the causal relationship between employment opportunities or public assistance and recidivism is very difficult given confounding factors such as unobserved individual motivation. This project provides novel causal estimates by using quasi-experimental econometric methodologies.

(2)  Causal estimates can be of little use to policymakers if they are based on a very specific context or small population. This project will provide causal estimates that are directly relevant to broad populations in British Columbia and more easily generalized to similar populations in other jurisdictions. The estimates will allow policymakers to engage in well-informed cost-benefit analysis when considering programs to help individuals escape cycles of criminal offending and incarceration.

(3)   We seldom have credible estimates on the causal role of employment opportunities and public assistance for outcomes specific to individuals with a prior criminal conviction. The studies here will provide estimates directly relevant to that demographic and can be used to predict the impact of expansions to social safety net programs or job opportunities on individuals with a prior criminal conviction.