Health insurance and individual labor market decisions.

Madrian, Brigitte C.

It is well accepted that health insurance distorts the demand for medical services. My research explores a further margin along which health insurance may affect behavior: by changing the labor market decisions of individuals. This distortion arises because in the current system of provision of health insurance in the United States, employers are the primary source of coverage for all but the elderly. As rising medical costs make health insurance an increasingly valuable component of employee compensation, we should expect coverage to be an important consideration in the labor market decisions of individuals.

There has been little previous research on the labor market effects of health insurance. However, there is growing interest in understanding this relationship, first because health insurance expenditures constitute a significant fraction of total employee compensation. Employers now spend more on health insurance than on any other employee benefit, including pensions. Health insurance expenditures are also the fastest growing component of benefit payments, increasing at an average rate of 15.6 percent annually from 1948-90.(1) Second, and perhaps more important, any health care reform that alters the current relationship between health insurance and employment has the potential to affect the labor market in significant ways.

The rationale for employer provision of health insurance is straightforward. By pooling their employees into large groups, employers can lower administrative expenses and reduce the risks of high health care costs faced by any individual employee. In addition, employer expenditures on health insurance are tax deductible, while individual expenditures generally are not. Given these cost advantages, it is not surprising that the delivery of health care in the United States has evolved into a system based primarily on employer provision of insurance.

Job-Lock

One significant disadvantage of employer-provided health insurance, however, is that it is not typically portable: when an individual quits his or her job, the insurance coverage associated with that job usually ceases as well. For many individuals, a change in insurers is inconsequential, but for some, relinquishing their employer-provided health insurance may be very costly. Exclusions on preexisting conditions are typical of almost all individual policies, and of many employer-provided policies as well.(2) In addition, half of full-time workers face length-of-service requirements before being eligible for any insurance.(3) Also, there is a growing trend toward medical underwriting, especially in small firms, in order to exclude serious ailments from coverage entirely. As a consequence, those with health problems may find themselves liable for many of their medical expenses that previously were covered by insurance. If these perceived costs of changing insurers are great enough, then the labor market decisions that individuals otherwise might make may be dictated instead by their needs for health insurance.

Exclusions for preexisting conditions and medical underwriting often are cited as causes of "job-lock": the tendency for individuals to stay in jobs they would really rather leave for fear of losing their health insurance coverage. While the popular press on several occasions has cited job-lock as a major problem with the current health care system, until recently there was no empirical evidence on the magnitude of this problem. This is in part because of the difficulty of identifying exactly what job-lock is and when it occurs.

Although it is impossible to observe directly whether individuals are locked into their jobs, in the population as a whole the extent of job-lock can be inferred by comparing the turnover rates of those who are more likely to be affected by it with the turnover rates of those who should not be affected by it. Job-lock should affect only those with health insurance, and the effect should be greater for those who have high expected medical expenses. If job-lock is important, the difference in mobility rates between those with high and low expected medical expenses should be greater for those with employer-provided health insurance than for those without it.

In recent research, I consider three different "experimental" groups to estimate the extent of job-lock: married men who have an alternative source of coverage in addition to employer-provided health insurance; heads of large families who are more likely to have high expected medical expenses simply because of the size of their family; and married men whose wives are pregnant.(4) I find that job-lock related to health insurance reduces the voluntary turnover rate of those with employer-provided health insurance by 25 percent, an effect that is both economically and statistically significant.

These results have been corroborated in a follow-up study done with Jonathan Gruber that examines the effect of continuation coverage mandates on job turnover.(5) Such mandates grant individuals the right to continue purchasing health insurance through their former employers for some period of time after leaving their jobs, and thus should reduce the extent of job-lock. We find that the availability of continuation coverage indeed increases the job turnover rate. Recent research by other individuals also has found evidence of job-lock,(6) although these results have been disputed.(7)

To the extent that health insurance does reduce mobility, there may be important consequences for economic welfare. First, it will directly affect the well-being of those who are locked into their current jobs. Second, and perhaps more importantly, job-lock may be a significant concern if there is a specific component of productivity that makes workers more productive in some jobs than in others.(8) The efficiency of the economy as a whole will suffer if individuals who would like to move to more productive jobs are constrained to keep their current positions simply to maintain their health insurance. The actual magnitude of the welfare loss associated with job-lock is something that has yet to be estimated empirically.

Retirement Decisions

Closely related to job-lock is the issue of how health insurance affects the retirement behavior of individuals. The underlying issues are the same: health insurance in the private market is much more expensive than the health insurance provided by employers, and individuals with preexisting conditions may find themselves unable to secure equivalent coverage if they retire from their job and give up the accompanying health insurance. However, the incentives facing older workers contemplating retirement are somewhat different from those faced by younger workers changing jobs.

First, all individuals become eligible for Medicare upon reaching age 65. Although Medicare is much less generous than most employer-provided policies, coverage is conditional only upon age and does not exclude preexisting conditions. Therefore, the costs of relinquishing employer-provided health insurance are diminished after reaching age 65. Second, many employers provide post-retirement health insurance to their retirees. Thus, the possibility of losing health insurance coverage should not be a deterrent to retirement for individuals who work in finns that offer this type of coverage.

Several recent papers suggest that health insurance is an important factor in the retirement decision. My own work and studies by Rogowski and Karoly both find evidence that the availability of employer-provided health insurance coverage after retirement is associated with early retirement.(9) Michael D. Hurd and Kathleen McGarry find that such health insurance is correlated with expectations of earlier retirement among those who are not yet retired.(10) Further work by Gruber and me finds that the availability of continuation coverage encourages early retirement as well as job turnover.(11)

The increased availability of both employer-provided retiree health insurance and continuation coverage may be important explanations for the trend toward early retirement that has been observed over the past several decades. The previously cited research suggests that these two sources of health insurance may account for between 10 and 50 percent of the decline in male labor force participation between 1960 and the late 1980s.(12)

The role of Medicare in the retirement decision is less well understood. My own research, as well as that of Robin S. Lumsdaine, James H. Stock, and David A. Wise, finds little evidence to suggest that the availability of Medicare helps explain the excess retirement that occurs at age 65, once the financial incentives associated with pensions and Social Security are taken into account.(13) This may be because Medicare is a vastly inferior source of health insurance, and therefore does not affect retirement, even though the availability of more generous health insurance might. Not only is the coverage provided by Medicare much less generous than what typically is provided by employer plans for retirees, but it also is available only to the individual, while employer-provided health insurance usually covers dependents as well.

1 J. S. Piacentini and J. D. Foley, EBRI Databook on Employee Benefits, Washington: Employee Benefits Research Institute, 1992.

2 P. Cotton, "Preexisting Conditions 'Hold Americans Hostage' to Employers and Insurance," Journal of the American Medical Association (1991), pp. 2451-2453.

3 U.S. Department of Labor, Bureau of Labor Statistics, Employee Benefits in Medium and Large Firms, Washington: Government Printing Office, 1989.

4 B. C. Madrian, "Employment-Based Health Insurance and Job Mobility: Is There Evidence of Job-Lock?" Quarterly Journal of Economics (February 1994), pp. 27-54.

5 J. Gruber and B. C. Madrian, "Health Insurance and Job Mobility: The Effects of Public Policy on Job-Lock," Industrial and Labor Relations Review (October 1994) pp. 86-102.

6 P. F. Cooper and A. C. Monheit, "Does Employment-Related Health Insurance Inhibit Job Mobility?" Inquiry (1993), pp. 400-416, and "Health Insurance and Job Mobility: Theory and Evidence," Industrial and Labor Relations Review (October 1994), pp. 65-68; and T. C. Buchmueller and R. G. Vailetta, "Employer-Provided Health Insurance and Worker Mobility: 'Job-Lock' or Not?" unpublished paper, University of California, Irvine, February 1994.

7 J. R. Penrod, "Health Care Costs, Health Insurance, and Job Mobility," unpublished paper, Princeton University, November 1993, finds mixed evidence in favor of job-lock; D. Holtz-Eakin, "Health Insurance Provision and Labor Market Efficiency in the United States and Germany," in Social Protection Versus Economic Flexibility: Is There a Trade-Off? R. M. Blank, ed. Chicago: University of Chicago Press, 1994, finds little evidence of job-lock.

8 B. Jovanovic, "Job Matching and the Theory of Turnover," Journal of Political Economy (1979), pp. 972-990.

9 B. C. Madrian, "The Effect of Health Insurance on Retirement," Brookings Papers on Economic Activity 1 (1994), pp. 181-252; L. A. Karoly and J. A. Rogowski, "The Effect of Access to Post-Retirement Health Insurance on the Decision to Retire Early," Industrial and Labor Relations Review (October 1994), pp. 103-123.

10 M. D. Hurd and K. McGarry, "The Relationship Between Job Characteristics and Retirement," NBER Working Paper No. 4558, December 1993.

11 J. Gruber and B.C. Madrian, "Health Insurance Availability and the Retirement Decision," NBER Working Paper No. 4469, September 1993.

12 B. C. Madrian, "The Effect of Health Insurance on Retirement," op. cit.; J. Gruber and B.C. Madrian, "Health Insurance Availability and the Retirement Decision," op. cit.

13 B. C. Madrian, "The Effect of Health Insurance on Retirement," op, cit.; R. S. Lumsdaine, J. H. Stock, and D. A. Wise, "Pension Plan Provision and Retirement: Men and Women, Medicare, and Models, "in Studies in the Economics of Aging, D. A. Wise, ed. Chicago: University of Chicago Press, 1994.