Whenever President-elect Obama is asked how he'll pay for his ambitious health care reform plans, he invariably talks about the $80 billion in annual savings he'll get from bringing computerized recordkeeping to doctors' offices and hospitals.

If only that were true. While there are benefits that might be had from using computers more widely in medicine, doing so won't save us any money and, in fact, will likely make things more expensive. There's even a chance that the quality of care might get worse along the way.

That's probably counterintuitive to anyone contemplating the wall of file drawers in a typical doctor's office. Medicine clearly has yet to join the rest of the world in going digital; no wonder, the thought goes, that U.S. health care is so expensive.

But while paper records certainly have their inconveniences--filling out your thousandth questionnaire, say--they play a very minor role in galloping health care inflation.

Instead, the heart of the problem is the U.S. fee-for-service system, in which doctors get paid to do things to people. The more technical and invasive the procedure, the more money they make. Doctors have responded in the expected Pavlovian manner, collectively shifting away from basic primary care toward expensive specializations that run up costs without necessarily improving medical outcomes.

As any chief information officer can tell you, adding computers to this sort of inefficient process only makes the inefficiency happen more quickly.

Much of what doctors or policymakers know about technology comes from vendors, who are busy guilt-tripping the medical sector about being slow to get with it. But more quietly, health care economists have been studying the actual impact of these systems. Their findings should disturb those who look to information technology for an easy fix.

The economic numbers are scary. October car sales were off by a third. Retail revenues (before subtracting inflation) fell 4.1% from a year earlier. Housing starts are at their lowest level in at least a half-century. About the only things that seem in high demand are "For Sale" signs and that perennial recession staple: spam.

At first blush it looks as if the "D" word is upon us. Not "depression" but "deflation"--the vicious phenomenon in which falling spending begets wage and price cuts, which beget further spending cuts in a debilitating downward spiral. That, anyway, is what the bond market is suddenly signaling (see chart).

A year ago investors priced Treasury Inflation-Protected Securities based on the expectation that consumer prices would rise 2% a year over the next half- decade. These days five-year TIPS are yielding 0.5 percentage points more than five-year Treasurys, implying prices will fall 0.5% annually. The last time U.S. prices fell consistently was in the midst of the Great Depression.

If prices merely flatline for a few quarters they could ignite "waves of bankruptcies," says Joseph Stiglitz, the Nobel Prize-winning economist. That's because many firms went into debt counting on a whiff of inflation to bail them out. Commercial real estate investors, for example, made heavily leveraged investments assuming they could hike rents. Deflation would turn such plans into financial disasters.

Merrill Lynch (nyse: MER - news - people ) chief economist David Rosenberg expects prices to fall at an annualized 1% or 2% a year from now, and lays 50-50 odds the drop will continue through the first half of 2010.

Related Quotes

MER

$13.22

+1.01

Get Quotes:

"It's going to be a very steady and significant decline," Rosenberg says. "You have deflation in commodities, labor markets, assets and credit."

His advice: Buy zero coupon 30-year Treasurys, which have been rising sharply as the economy slows. Rosenberg's is a cogent case in view of recent data. But the danger is that this bull market in Treasurys is about to end. Inflation might replace deflation. If that happens, those zeros will get clobbered.