Research
-
Job Market Paper:
-
Academic 2-year colleges as a Stepping Stone
An important number of high-school graduates start their post-secondary educational careers at academic 2-year colleges even though returns to graduation are negligible. However, the returns to transferring to 4-year colleges are large: academic 2-year colleges act as a stepping stone in which agents learn about themselves in a cheaper and less demanding environment than 4-year colleges. This paper presents a model of educational choice that incorporates academic 2-year colleges together with 4-year colleges and work. Agents are initially uncertain about their innate ability to accumulate human capital. Pessimistic agents join the workforce, optimistic agents enroll in 4-year colleges and those in the middle enroll in academic 2-year colleges. Exams govern the accumulation of credits and provide information that update beliefs, inducing dropouts and transfers. The model is consistent with facts that are documented for two different data sets: (1) among those initially enrolled in academic 2-year colleges, more able agents are less likely to graduate, more likely to transfer, and less likely to dropout; (2) among those initially enrolled in 4-year colleges, more able agents are more likely to graduate and less likely to dropout or transfer; (3) there is a higher concentration of high ability students among transferees. A decomposition of returns shows that the dropout and transfer options account for 90% of the full return to enrolling in an academic 2-year college while the dropout option explains 70% of the full return to enrolling in 4-year colleges. Academic 2-year colleges are found to be close substitutes for 4-year colleges and thus the welfare effect of the availability of academic 2-year colleges is limited and is primarily driven by a slight increase in participation. The model is also able to reconcile low enrollment and graduation rates with high returns at 4-year colleges. The low graduation rate results from the interaction of learning and option value while the low enrollment and wedge in returns are explained by academic 2-year colleges.
-
Working Papers:
-
A property of the Kiyotaki and Wright (1989) model of commodity money as a medium of exchange is the multiplicity of dynamic equilibria as discussed in Kehoe et al. (1993). We modify the model to allow the meeting rate to depend on the length of the time period and focus on symmetric dynamic equilibria in a symmetric environment. We characterize the set of points in the state-payoff space that are consistent with equilibrium. With a time period of any fixed length, there is a large set of equilibria that includes cycles, sunspots, and other non-Markovian strategies, while in the continuous time limit there is a unique, rather simple, dynamic equilibrium. Despite the multiplicity, for short period lengths all equilibrium paths are well approximated by the unique equilibrium of the continuous time limit.
-
Returns to postsecondary education have been found to be unreasonably high relative to the low enrollment and graduation rates observed in data. The consensus at the time is that there exists underinvestment in postsecondary education. We model college enrollment and dropout decisions in a real options model similar to Miao and Wang (2007) where attitude towards schooling stems from the initial belief of students about their skill level and the option value of dropping out stems from the Bayesian learning process of one’s skill level. The option value of dropping out is a novel feature of this paper and allows for an easy and straightforward characterization. Our paper has four main results. First, our model shows that the mainstream models in the literature that assume risk neutrality overestimate the value of college, college enrollment and graduation rates because they omit the uncertainty regarding the outcome of college education. Second, we show that the option value of learning is much more important when agents are risk averse rather than risk neutral. Further, the effect of the option value is much more important at the margin so it becomes important for understanding drop-out rates. Finally, we show that ex-ante returns predicted by our model are smaller suggesting that uncertainty about future income stream can potentially solve the high returns to education puzzle.
-
Estimating a consumption-based asset pricing model with Markov regime switching for a wide set of countries, this paper finds that there are two groups: one where countries are subject to rare disasters that take the form of capital destruction, and another where countries suffer strong recessions. The United States belong to the second group. To evaluate the "Peso Problem" hypothesis, a conditional estimation is performed to find that no specification of the disaster state is able to generate enough risk to match the high return on equity and low return on bills observed in data. Test of model specification points in the same direction. The results of this paper provide that the Equity Premium puzzle is still not explained.
-
Work in Progress:
-
Dropping out from College: Determinants and Elasticities, 2010 (with Ali Ozdagli)
-
Learning and Occupational Transitions, 2009 (with Aspen Gorry)
A large literature on occupational choice has studied workers career patterns in a risk neutral environment. We construct an occupational choice model where agents have preferences that are averse to risk in their future consumption streams. In this environment, we are able to capture important features of both the bandit literature where workers choose riskier professions early in their career as in Johnson (1978) and the stepping stone mobility framework where agents use simple tasks to learn about their abilities and eventually move into more complex careers as in Jovanovic-Nyarko (1997). Our environment shows that risk can explain both the initial selection into careers based on beliefs about a worker’s ability and the latter patters of job switching between occupations as workers learn about their abilities.