doi:10.1016/j.jhealeco.2007.11.004 
Copyright © 2007 Elsevier B.V. All rights reserved.
Crowd-out 10 years later: Have recent public insurance expansions crowded out private health insurance?
Jonathan Grubera, b and Kosali Simonb, c, 
, 
aMIT Department of Economics, Cambridge, MA 02142, USA
bNational Bureau of Economic Research, Cambridge, MA, USA
cCornell University Department of Policy Analysis and Management, Ithaca, NY 14853, USA
Received 11 January 2007;
revised 28 October 2007;
accepted 1 November 2007.
Available online 29 November 2007.
References and further reading may be available for this article. To view references and further reading you must
purchase this article.
Abstract
Ten years ago, Cutler and Gruber [Cutler, D., Gruber, J., 1996. Does public health insurance crowdout private insurance? Quarterly Journal of Economics 111, 391–430] suggested that crowd-out might be quite large, but much subsequent research has questioned this conclusion. Our results using improved data and methods clearly show that crowd-out is still significant in the 1996–2002 period. This finding emerges most strongly when we consider family level measures of public insurance eligibility. We also find that recent anti-crowd-out provisions in public expansions may have had the opposite effect, lowering take-up by the uninsured faster than they lower crowd-out of private insurance.
Keywords: SCHIP; Medicaid; Crowd-out; Cost-Sharing; Take-up
JEL classification codes: I1
Table 1.
Literature on crowd-out
Table 2.
Descriptive statistics of selected variables
Note: Unweighted data from the SIPP 1996 and 2001 panels. From the 2001 panel, we exclude data after December 2002. Children are aged 0–18 years. Only 4th reference month observations are kept (one response per wave). States that are unidentified in the SIPP include North Dakota, South Dakota, Maine, Wyoming, and Vermont.
Table 3.
Tabulations by income group over time
Note: Calculations are based on authors’ tabulations of 1996 and 2002 SIPP data. Sample sizes are as follows: 5711 for the <100% FPL category; 5776 for the 100–200% FPL category; 5672 for the 200–300% FPL category; and 5389 for the 300–400% FPL category.
Table 4.
Crowd-out calculations from tabulations in Table 3
Note: Crowd1 assumes that the overlap is a move from private to public coverage; Crowd2 ignores the overlap.
Table 5.
Effect of eligibility for any public insurance on insurance status
Note: Standard errors are in parentheses. Each estimate is from a separate regression. *Indicates statistical significance at the 10% level; **indicates significance at the 5% level; and ***indicates significance at the 1% level. Number of observations is 405,389. All interactions refer to state × age, state × year and age × year.
Table 6.
Effect of family eligibility for Medicaid and SCHIP on insurance status (Months interaction, cost sharing interactions, continuous eligibility interactions and presumptive eligibility interactions)
Note: Standard errors are in parentheses. Each set of estimates is from a separate regression. *Indicates statistical significance at the 10% level; **indicates significance at the 5% level; and ***indicates significance at the 1% level. Number of observations is 405,389. All interactions refer to state × age, state × year and age × year.
Abstract
Purchase PDF (183 K)
E-mail Article
Add to my Quick Links
Cited By in Scopus (1)
Purchase PDF (293 K)
