On Health Plans, the Numbers Fly

New York Times

By KEVIN SACK

October 21, 2008

 

Economics, it is said, is the dismal science. Anyone paying close attention to the campaign debate over the economics of health care might wonder about the science part.

 

As Senators Barack Obama and John McCain battle over how best to control spending and cover the uninsured, they are both filling their speeches, advertisements and debating points with authoritative-sounding statistics about the money they would save and the millions of Americans they would cover.

But the figures they cite are invariably the roughest of estimates, often derived by health economists with ideological leanings or financial conflicts. Over time, these forecasts have become so disparate and contradictory as to be almost meaningless.

How many of the country’s 45 million uninsured would gain coverage under Mr. McCain’s plan to reconfigure the tax treatment of health benefits?

Consultants paid by Mr. McCain concluded that his plan would cover 27.5 million of the uninsured. But four health economists who looked into the McCain plan at the urging of David Cutler, a health care adviser to Mr. Obama, reached a far different conclusion. They estimated in a peer-reviewed article in the journal Health Affairs that the number of uninsured would grow by 5 million after five years.

How much would it cost for Mr. Obama to offer subsidized health insurance to those with low incomes?

Last week, the Lewin Group, a consulting firm, projected the cost to taxpayers at $1.17 trillion over 10 years. That was about 27 percent lower than the $1.6 trillion estimated by the Tax Policy Center, a joint project of the Urban Institute and Brookings Institution. And it bore little similarity to a $6 trillion estimate — using a broader measurement — put forth by HSI Network, a Minnesota consulting group that was paid $50,000 by the McCain campaign to assess both plans.

The campaigns acknowledge that the numbers are “all over the map,” in the words of Jay Khosla, a McCain adviser. But that does not keep them from selectively highlighting the most favorable ones (as when Mr. Obama says his plan will cut insurance premiums by $2,500 per family, or when Mr. McCain says his tax changes will leave 95 percent of Americans with more money).

Even the economists behind the forecasts say it makes them uncomfortable to hear candidates assert their numbers as indisputable fact, as if stating Derek Jeter’s batting average. What they are modeling, they emphasize, is ultimately unknowable. And the transformational nature of both candidates’ health care plans means that they can only guess at the future behavior of consumers, employers and insurers.

“Every candidate should say that these numbers were produced by my experts and they’re my best estimates but they’re not exact,” said Roger D. Feldman, a health economist at the University of Minnesota who directed the HSI studies. “But the campaign trail is not the time for ‘on the one hand, on the other hand.’ It’s a system where you paint things in black and white.”

Dr. Feldman and other economists said politics and relationships did not sway their science. But they said estimates could vary widely because of the assumptions they must factor into their formulas. Often they are flying blind because the campaigns, aware that details make the fiercest enemies, do not provide critical variables. Mr. Obama, for instance, has steadfastly declined to say how he would penalize employers who do not offer health coverage, an important component of his plan.

The economists are often left to use small-scale studies to predict how the candidates’ policies might affect the cost of coverage or the willingness of employers to provide it.

“The uncertainty surrounding what will happen under these policies is huge,” said John F. Sheils, senior vice president of the Lewin Group.

Sherry A. Glied, a Columbia economist and a co-author of the Health Affairs article about the McCain plan, said, “We are estimating what would happen in a world we’ve never seen.”

The more radical the restructuring, the economists said, the more they must assume. And the more they must assume, the greater the chance that ideology may drive methodology.

“It’s garbage in, garbage out,” said Uwe E. Reinhardt, a health economist at Princeton. “Every econometric study is an effort in persuasion. I have to persuade the other guy that my assumptions are responsible. Depending on what I feed into the model, I get totally different answers.”

Katherine Baicker, a health economist at Harvard, said economists’ views about the mechanics of markets were often shaped by their politics, or vice versa. “Certainly people who work for the campaigns have a strong motivation to see things one way or another,” she said, “but even those not involved in campaigns still come to the table with their own prior beliefs.”

Both candidates are surrounded by advisers with extensive backgrounds in health economics, many of whom could be in line for administration jobs.

For Mr. McCain, there are Douglas Holtz-Eakin, a former Congressional Budget Office director; Stephen T. Parente of the University of Minnesota; Thomas P. Miller of the American Enterprise Institute; Gail Wilensky, a health adviser to the first President Bush; Grace-Marie Turner, president of the Galen Institute; and Mr. Khosla, a former Congressional aide.

Mr. Obama receives advice from Mr. Cutler, David Blumenthal and Jeffrey Liebman, all of Harvard; Stuart Altman of Brandeis; Austan Goolsby of the University of Chicago; Jeanne M. Lambrew of the University of Texas; three campaign aides, Heather Higginbottom, Jason Furman and Neera Tanden; and a Senate office staff member, Dora Hughes. Campaign insiders suspect that if Mr. Obama is elected, a significant health-related position may go to Tom Daschle, the former Senate majority leader and an early Obama endorser who recently published a book on the subject with help from Ms. Lambrew.

The conflicts that devalue economic estimates can be both political and financial.

Mr. Obama, for example, has been claiming in speeches and advertisements that Mr. McCain would cut $882 billion in Medicare benefits to pay for his health plan. The number came from the Center for American Progress Action Fund, a Democratic-leaning group with close ties to the Obama campaign (Ms. Lambrew is a fellow).

“Consider the source,” Mr. Holtz-Eakin urged reporters last Friday.

A week earlier, Mr. Holtz-Eakin issued a news release trumpeting the HSI Network analysis of the McCain plan as “an independent assessment.” He did not mention that the campaign had paid for it (an Aug. 27 payment for $50,000 shows up in Mr. McCain’s disclosure filings) or that Mr. Parente is one of the firm’s owners.

Dr. Feldman, whose work is highly regarded, described himself as an Obama supporter and contributor, but he said he preferred Mr. McCain’s health plan. Though he acknowledged that the McCain campaign’s sponsorship was “certainly a potential conflict,” he said he hoped the study might advance a worthy proposal. “I wouldn’t sign off on these things if I didn’t support them,” he said.

A number of economists said voters would be wise to simply tune out all of the competing numbers and focus instead on the philosophical underpinnings of the candidates’ plans. Indeed, Dr. Reinhardt offered voters the same instruction he delivers to his students, that economics as practiced in the political arena is often “just ideology marketed in the guise of science.”

“I give a lecture on whether you can trust economists, and I tell them no,” Dr. Reinhardt said. “I tell them that if at the end of the year I tell you the time of day and you trust me, I have failed.”