The Problem: In March 2001, the longest economic expansion in our country's history ended. The recession that followed lasted until November 2001. Unemployment rose from 4.3 percent to 5.6 percent during that time and real per capita gross domestic product fell from $34,764 to $34,499.
What effect did the 2001 recession have on health coverage for individuals? Did women fare differently than men? Did children lose or gain health coverage during the recession? Did expansions in government health programs offset coverage losses in the private sector?
Little was known about how changes in the macroeconomic environment affect health coverage or even which measures best identify those effects. Gross state products (GSPs) and unemployment rates interact with private sector cutbacks and government-sponsored health programs in complex ways, making it hard to pinpoint cause and effect relationships.
Grantee Background: Kosali I. Simon arrived at Cornell University in Ithaca, N.Y., in 2001, soon after receiving her doctorate in economics from the University of Maryland. She wanted to understand what happens to health care coverage for people when the macroeconomic environment around them changes dramatically, as it did in March 2001.
A research grant from the Robert Wood Johnson Foundation's Economic Research Initiative on the Uninsured (formerly the Research Initiative on Health Insurance) allowed Simon to develop that understanding and build her career. "This program quickly became well known. I often cited the papers the program commissioned, papers that asked the 'big questions' about coverage and economics. So, when I saw the solicitation for studies, I applied.
"I grew up in this program. I applied for the grant soon after I started at Cornell and I finished it as I came up for tenure."
The Project: In the study, "Health Insurance Coverage and the Macroeconomy," Simon and her Cornell colleague John Cawley analyzed federal and state data to determine whether the 2001 recession affected the probability that individuals would have health insurance.
They used state unemployment rates and per capita gross state product to measure the macroeconomic environment. High unemployment and/or low gross state product are associated with a lower probability that employers offer insurance.
Because most insured non-elderly Americans obtain coverage through work, Simon separately explored the likelihood that people would receive insurance from any source and from employment during the recession.
Results: Simon found that "macroeconomic changes affect the insurance status of men, women and children in different ways," and that "looking only at the bottom line of coverage from any source without examining individual sources masks important trends."
In an article published in the Journal of Health Economics in 2005, Simon and Cawley reported:
- For men: A one percentage point increase in the state unemployment rate is associated with a 0.7 percentage point decrease in coverage from any source. "This meant that more than 900,000 adult men lost coverage during the 2001 recession."
- For women: A one percentage point increase in the state unemployment rate is associated with a 0.7 percent decrease in coverage from employment, but this decrease is offset by likely increases in coverage from other sources, generally Medicaid.
- For children: A one percentage point increase in the unemployment rate is associated with a 1.04 percentage point likely increase in coverage from Medicaid or the State Children's Health Insurance Program (SCHIP).
Simon found no correlation between per capita gross state product and the probability of coverage for men, women or children.
Simon and Cawley then estimated the number of Americans who lost health insurance due to macroeconomic changes during the recession and how many regained coverage due to macroeconomic changes during the recovery period. They found:
- About 984,000 Americans lost health insurance due to macroeconomic changes during the recession, but virtually no children lost coverage. About 908,000 of those who lost coverage were men and the remaining 75,000 were women.
- About 1.246 million Americans gained health insurance due to macroeconomic changes between November 2001 and March 2004, the recovery period. About 806,000 were men, 440,000 were woman and virtually none were children. This estimate does not include changes due to factors other than macroeconomic conditions, such as changes in health insurance markets, labor markets or society.
Involvement in the program yielded other results for Simon. She said "John [Cawley] and I wrote a paper about welfare reform and health coverage of women and children. That paper and our original paper were published in two edited volumes of Frontiers in Health Policy Research. We are now starting to study how the current recession affects health insurance coverage."
Simon reflects on her experience in the program. "The first six or seven years are critical in shaping research and academic careers. When you are starting out, having your research selected for funding and then having a chance to present your study and hear feedback is wonderful."
RWJF Perspective: RWJF established the Economic Research Initiative on the Uninsured in 2000 to generate information on the relationship between the labor market and health insurance coverage. RWJF also wanted to elicit new ideas about these relationships by attracting applied economists who had not previously focused on health concerns and by drawing upon new analytical frameworks.
"Given the number of uninsured people, the fragility of insurance for those who have it and the spiraling costs of care to individuals, businesses and the government, we needed the best information and the best ideas about ways to address the connections between work and health insurance. The Economic Research Initiative on the Uninsured was designed to fill gaps in our knowledge and understanding of these complicated relationships," said David C. Colby, PhD, RWJF's vice president for research and evaluation and program officer for the initiative.