Studies have consistently shown that obese employees are paid less than normal-weight employees doing similar jobs, leading many people to attribute the gap to prejudice against workers based on their appearance.

But new research from Stanford University health economists adds another wrinkle to understanding these pay differentials: obese workers are paid less only when they have employer-sponsored health insurance.

These findings, just published in a working paper on the Web site of the National Bureau of Economic Research, suggest that employers-recognizing that obese workers are likely to have higher medical costs-compensate with lower pay for them. Given that employment-based health insurance requires that employees in the same plan make the same contributions to premiums, the employers adjust wages to account for the greater expense for obese workers' health care, according to the paper.

"A self-correcting mechanism is at work in the labor market," explained study co-author Kate Bundorf, MPH, PhD, assistant professor of health research and policy at Stanford and a fellow at the university's Center for Health Policy/Center for Primary Care and Outcomes Research. The study doesn't address whether the wage disparity is fair, she noted; it simply demonstrates that there are strong economic incentives for employers to adjust for the varying costs of providing medical benefits to different types of workers. "Our findings reinforce that these market forces are powerful," she said.

The findings also shed light on the question of who bears the cost of obesity-related health care. While it is often assumed that obese workers' medical expenses are passed on to their employers and normal-weight co-workers, the Stanford study indicates that obese workers are paying for it themselves through lower wages.

Understanding who bears the cost of obesity-related medical expenses has become more pressing, with a significant increase in the number of obese Americans. The proportion of American adults classified as obese rose from 12 percent in 1991 to 20.9 percent in 2001. Obese individuals are at much higher risk of chronic - and often costly - conditions such as heart disease, diabetes and hypertension. Annual medical expenditures are $732 higher on average for obese adults than for normal-weight adults, according to a recent study published in Health Affairs.

Bundorf and co-author Jay Bhattacharya, MD, PhD, assistant professor of medicine at the Center for Health Policy/Center for Primary Care and Outcomes Research, designed their study to find out who bears the brunt of obese workers' higher medical costs. In doing so, they also examined a broader, unsettled question in health economics: Who actually bears the cost of employer-sponsored health insurance-employers or employees? While many health economists assert that the costs of employer-sponsored health insurance are passed on to workers in the form of lower wages, the research findings on this question have been inconclusive.

To study both questions, the researchers compared the hourly wages of obese and non-obese workers with health insurance, adjusting for several factors including education, experience and job type. They found that obese insured workers earned significantly less per hour than non-obese insured workers - $3.41 less in 1998. When analyzing the wages in greater detail, they found that the gap is modest when these workers are young, but widens over time, meaning that this set of obese workers' pay rises more slowly than that of non-obese workers with employment-based health insurance.

The researchers then compared the hourly wages of obese and non-obese workers without on-the-job health insurance. This time, they found no significant difference in pay. This finding - that an obesity-related wage difference existed only for those with employer-sponsored health insurance - signaled that the obese workers' lower pay could be explained by their higher expected medical costs instead of outright prejudice.

To further test this hypothesis, the researchers examined whether a wage gap existed between obese and non-obese workers receiving other kinds of benefits, such as retirement plans or life insurance. They found no wage difference in those instances, thus reinforcing the idea that the pay adjustment results from greater health-care costs.

Bundorf noted that the study did not reach any conclusions about how the obesity-related disparity in pay comes about. "We don't think this is a conscious process where the employer says, 'OK, Jane is obese, and we're paying for her health coverage, so let's pay her this much less in wages,'" Bundorf said. But she added that the finding that the pay for obese insured workers rises more slowly than that of their normal-weight counterparts suggests that obese workers may be getting smaller and less frequent raises.

Aside from providing insight into the costs of obesity among workers, the study provides perhaps the strongest evidence to date that the costs of employer-sponsored health insurance are, in fact, passed on to workers through lower wages. By implication, insured workers should be just as alarmed by rising health-care costs as their employers are.

"When employers give you health insurance, they're not giving you something for nothing," Bhattacharya said. "It's coming out of your paycheck."

The paper is available to download at http://papers.nber.org/papers/w11303