The biggest player in the health-care debate right now isn’t Nancy Pelosi, Harry Reid, or even President Obama. It’s the Congressional Budget Office, which is responsible for estimating the costs of proposed legislation. After the director of the CBO testified on July 16 that none of the health-reform bills in the House or Senate would reduce the rate of increase in federal spending on health care, congressional efforts fell into disarray. Many policymakers began searching for a way to get costs below the CBO’s frightening estimate of $1.1 trillion over ten years. Others attacked the CBO, calling its estimates irresponsible.
The CBO is actually being kind to the would-be reformers. Its analysis likely understates—by at least $1 trillion—the true costs of expanding health coverage as current Democratic legislation contemplates. Over the last few months, my colleagues and I at the consulting firm Health Systems Innovations have provided cost estimates of health-care reform to both Republican and Democratic members of Congress, and we’ve posted these estimates on our website as well. We believe that the Democratic bills currently under consideration in the House and Senate would cost $2.1 trillion and $2.4 trillion, respectively—much higher than CBO’s figures.
The discrepancies between our estimates and CBO’s stem from our different assumptions about a key issue. The Democratic plans envision a government-run insurance program, modeled after Medicare, that will compete with private insurers. How many people would opt for coverage under this public insurance? We believe that both large and small employers would have powerful incentives to shift their employees out of private coverage and into the public plan. Like the Urban Institute, we estimate that roughly 40 million people would make the shift. CBO seems to assume, however, that large employers would use the public plan only sparingly and that only 11 million people would move from private to public insurance—which would, of course, result in lower costs.
Why the difference in these estimates? We believe that we have better data on this issue than the CBO, which uses simulation models of health-insurance plans based on much older health-plan data—typically from 2001 or even 2000. Our estimates are grounded in 2006 commercial-insurance data to which the CBO doesn’t have access (the data are not publicly available and the CBO didn’t make provisions to purchase them). These data reflect the advent of much cheaper, high-deductible health plans and limited-provider network plans. If the government modeled its public option on these inexpensive plans, the result would be cheap enough to lure far more people away from private health insurance than the CBO estimates.
Our model has a good track record. The last time government introduced a major health-insurance innovation was 2004, which saw the introduction of Health Savings Accounts. We used the same model to predict that 3 million people would adopt these HSAs by the beginning of 2006. Our estimate, which we published in the peer-reviewed journal Health Affairs, was spot-on, predicting the market response more accurately than most other models, which produced adoption-rate estimates at least one-third lower.
It’s possible, of course, that the CBO is right and that our estimates are too high. If the House or Senate bill passes, we should know who’s right by 2014–15, shortly after the bills take effect and costs start to explode.