شناسایی مجموعه اطلاعات عامل: یک برنامه کاربردی برای یک مدل چرخه عمر تحصیل، مصرف و عرضه نیروی کار

We adapt the insight of Cunha et al. (2005) to develop a methodology that distinguishes information unknown to the econometrician but forecastable by the agent from information unknown to both, at each point in an agent's lifecycle. Predictable variability and uncertainty have different implications in terms of welfare, especially when markets are incomplete. We apply our procedure in the context of an incomplete markets lifecycle model of consumption, labor supply, and schooling decisions, when borrowing limits arise from repayment constraints. Using microdata on earnings, hours worked, schooling choices, and consumption of white males in the US, we infer the agent's information set. We then estimate the model using the identified agent's information set. We find that 52% and 56% of the variance of college and high school log wages respectively are predictable by the agent at the time schooling choices are made. When we complete the market, college attendance increases from 48% to 59%, about half of this increase is due to uncertainty, and the other half because of the borrowing limits. To illustrate the importance of assumptions about what is forecastable by the agent, we simulate a minimum wage insurance policy under different assumptions about the information available to the agents in the model. When we allow for asymmetric information between the insurance institution and the individual, adverse selection turns profits negative. Consumer welfare, however, increases by about 28% when we give individuals access to their estimated information set regardless of asymmetries.