So That's Why It's So Expensive

Economists have long believed that technology is the main reason that health-care costs are rising so rapidly. The endless stream of innovation, from new drugs to delicate tools for microsurgery, the theory goes, largely explains why medical spending has exploded from 5% of the U.S. economy in 1960 to 16.5% today. According to some studies, as much as 65% of that growth could be laid at the feet of tech.

Now a young economics professor at the Massachusetts Institute of Technology is challenging the conventional wisdom. After studying data going back to the 1960s, Amy N. Finkelstein has concluded that the real culprit for the rapidly rising cost of health care is the massive expansion of medical insurance over the past 40 years. Sure, new technologies play a role, but doctors, hospitals, and consumers adopt them so freely largely because insurance foots the bill. "Where does that technological change come from?" asks Finkelstein, 32, who lives in Cambridge, Mass., with her economist husband, Ben Olken. "I am trying to get inside that black box."

If Finkelstein is right, her work could change the way policymakers and the companies that pay for most medical care think about costs. For example, if individuals have to pay more for their care through high-deductible health plans, they may cut spending. Her theory could also spur the drive for evidence-based medicine, the effort of some reformers to encourage the use of only those treatments that have been proven to work (BW -- May 29).

Already, Finkelstein's analysis is shaking up views across the political spectrum. "This is pathbreaking work," says Joseph R. Antos, a health economist at the conservative American Enterprise Institute. Adds the more liberal MIT economist Jonathan Gruber: "This really changes the whole landscape in the way we think about health economics."

Why is insurance so important? One obvious reason, Finkelstein believes, is that consumers opt for more care if someone else pays for it. But the more significant effect may be that insurance guarantees a steady source of revenue for hospitals and other health providers. Such ready cash encourages them to build new cardiac-care centers and stock up on the latest high-tech equipment, knowing it will be paid for. "If you produce expensive new things for medical care, people will buy them," says Paul Ginsburg, president of the Center for the Study of Health System Change in Washington. He has found results similar to Finkelstein's by looking at medical spending patterns in 12 U.S. cities.

Finkelstein's breakthrough confirms a theory first advanced almost 30 years ago by Harvard economist Martin Feldstein. At the time he didn't have detailed health-cost data to prove his case. Then in 1987 a massive Rand Corp. study concluded that technology accounted for more than half of the rise in health-care costs. Insurance, Rand figured, increased costs by just 10%.

So Feldstein's theory gathered dust until Finkelstein discovered the proof by sifting through long-forgotten paper records in MIT's library. There, she found that hospital spending soared after the federal Medicare program began in 1966. "I thought, why am I getting such a large number," she remembers.

Finkelstein had the papers scanned and shipped to a company in Cambodia, where it took 18 months to turn the records into usable data. The story they told was dramatic. In regions such as the South, where most seniors had no insurance, health spending soared after Medicare. But in New England, where many already had coverage, Medicare had much less impact on costs.

Not everyone buys her conclusions. Some think she has overstated the importance of insurance. Others question whether her results apply to private coverage as well as Medicare. But she has prompted many experts to rethink their long-held views. And now that Finkelstein thinks she's figured out how much insurance has increased costs, she wants to find out whether all that extra spending has paid off with better care.

By Howard Gleckman

We Almost Lost the Nasdaq

(enter your email)
(enter up to 5 email addresses, separated by commas)

Max 250 characters

blog comments powered by Disqus