Abstract: This paper estimates people's taste for living with own ethnic group neighbors using variation from a natural experiment in the Singapore housing market: the ethnic quotas. My approach combines policy-induced variation commonly used in reduced form settings with a structural framework à la Berry, Levinsohn, Pakes(1995). I find that all groups have strong preferences to live with other members of their ethnic group. More interestingly, the Chinese majority exhibit preferences that are inverted U-shaped so that once a neighborhood has enough Chinese neighbors, the Chinese would rather add a new neighbor from other ethnic groups.

Abstract: Seasonality is a basic fact of agrarian life in many developing countries. In this paper, we complement the literature on consumption smoothing by focusing on the less-studied problem of smoothing between seasons within an agricultural cycle. We present a novel model that relates seasonality in consumption to exponential and quasi-hyperbolic time preferences. The model identifies two crucial parameters for consumption patterns - discount factors and crop depreciation rates. Motivated by the model, we collected a unique dataset that combined time preference measures with detailed consumption and production information of more than 2000 households in Eastern Indonesia. We relate the model to the data and find suggestive evidence that both time preferences and crop depreciation rates play a role in the seasonality of consumption patterns. Based on these findings, we propose simple solutions that will be tested in future research.

Work in Progress

Abstract: Many desegregation policies take the form of quotas. This paper studies the impact of the ethnic housing quotas in Singapore that were designed to encourage residential desegregation amongst the three major ethnic groups, the Chinese, Malays and the Indians. I estimate the impact of the Chinese, Malay and Indian quotas on the price, quantity and quality of units sold. I find that all quotas significantly decrease the proportion of units sold. Malay-constrained units are 5% cheaper perhaps because the units sold are also of lower quality. The impact on the price and quality of Chinese- and Indian-constrained units are opposite. Chinese-constrained units are 7% more expensive even though the units sold are of significantly worse quality. Indian-constrained units are 2% cheaper even though the units sold are of a higher quality.

Abstract: Willingness-to-pay is an integral input for any welfare analysis. The two primary approaches to estimate willingness-to-pay (WTP) are hedonics (Rosen, 1974) and discrete choice models (McFadden, 1974). For many years, researchers have alluded to the apparent duality between both theories. The innovation in this paper is to show that the WTP derived from the hedonic model is a function of choice probabilities in the discrete choice model. The trick is to rely on the Implicit Function Theorem to define the gradient of the hedonic price function as a function of the share function in the discrete choice model. I find that the hedonic method estimates a weighted average of marginal utilities where higher weights are assigned to the consumers whose choice probabilities indicate a high degree of uncertainty regarding the choice. These are marginal consumers. As this choice becomes more certain, the weights start to decrease. These are inframarginal consumers whose choice probabilities are 0 or 1. Therefore, the hedonic method gives more weight to the preferences of the marginal consumer relative to the discrete choice approach. We can use these probability weights to analyze how WTP in the discrete choice model (an unweighted average) differs from WTP in the hedonic model (a weighted average). In the paper, I discuss how researchers can use this result in a variety of policy analysis.

Preliminary Projects

A Randomized Evaluation of Seasonal Food Consumption in West Timor, Indonesia, joint with Karna Basu