Working Papers

Claim Timing and Ex Post Adverse Selection. February 2013. [paper][online appendix]

Many health care treatments are not urgent and may be delayed if patients so choose. Because insurance coverage is typically determined by the treatment date, individuals may have incentives to strategically delay treatments to minimize out-of-pocket costs. The strategic delay of treatment — a particular form of moral hazard — can be an important source of subsequent adverse selection, in which ex ante identical individuals select insurance coverage based on their differing accumulation of previously delayed treatments. This paper investigates these forces empirically in the context of the missing market for dental insurance, with the goal of understanding how this source of asymmetric information may generate market unraveling and revealing lessons that apply to insurance markets more broadly. Using rich claim-level data, I present several simple tests of the hypothesis that people strategically delay claims and adversely select insurance coverage. These tests provide the key identication and motivation for a structural model which I develop and estimate to explicitly link the endogenous delay of treatment to the adverse selection it can cause. Using estimates from this context, I decompose adverse selection into that generated by traditional sources (heterogeneity in ex ante risk types) and that generated by delayed claims (referred to here as "ex post adverse selection"). The results show that ex post adverse selection is quantitatively important and can explain why the market for dental insurance has largely unraveled. More generally, the results suggest that features such as open-enrollment periods and contracting on pre-existing conditions may be helpful tools in overcoming market unraveling in insurance contexts where the timing of uncertainty is not contractible.

The Hated Property Tax: Salience, Tax Rates, and Tax Revolts. April 2013. [paper] (with Caroline Hoxby)

Because of the manner in which it is normally paid, the property tax is almost certainly the most salient major tax in the US. The property tax is also the least popular tax and the only major tax whose revenues have declined as a share of income. We hypothesize that high salience explains the unpopularity of the property tax, the level of the property tax, and prevalence of property tax revolts. To identify variation in the salience of the property tax over local jurisdictions and over time, we exploit conditionally random variation in tax escrow. Tax escrow is a method of paying the property tax that makes it much less salient, as we demonstrate using survey evidence. We find that areas in which the property tax is less salient are areas in which property taxes are higher and property tax revolts are less likely to occur.

The Effect of Insurance Coverage on Preventive Care . March 2011. [paper] (with Mark Cullen)

Although health insurance products that preferentially price preventive care are growing in popularity, little is known about how preventive care usage responds to changes in insurance coverage. Using health insurance claims data from a large company, this paper examines the implementation of an insurance benefit design which differentially increased patients' marginal price of curative care (non-preventive care) while decreasing the price of prevention. We examine the effect of the differential price change on the use of preventive procedures. We find that rural enrollees generally did less prevention while non-rural enrollees did more prevention in response to the benefit change.

Works in Progress

Externalities and Taxation of Supplemental Insurance: A Study of Medicare and Medigap (with Neale Mahoney)