Curbing medical inflation is no longer a plausible reason to pass health reform.

For more than a year, Democrats have been saying that health care reform is all about "bending the cost curve." This was a plausible claim while a strong public option remained possible. Now, however, evidence is mounting that without one—the straightjacketed public option in the Senate bill will be weakened still further if not discarded altogether (as seems more likely) to win a filibuster-proof majority—health reform won't curb medical inflation. It's time to stop saying it will.

According to the Congressional Budget Office, there isn't one single cost curve, but four. Really, though, there are only two: private health care spending and government health care spending. Health reform, the CBO has concluded, won't really bend either one.

Private spending. On Nov. 30, the CBO released an analysis of the "blended" Senate health reform bill's likely impact on private-sector premiums. For most of us—that is, for the 59 percent of Americans who receive health insurance through their employers—the bill's impact on premiums would be virtually nil. That isn't surprising, since almost nothing in the bill would change the way employer-based health insurers do business.

The chief exception is a 40 percent excise tax on "Cadillac" health plans valued at $23,000 or more for families. The Cadillac tax was excluded from the CBO's calculations because its full money-saving effects wouldn't be felt in 2016, the year for which the CBO made its calculations. For the 19 percent of employer-insured workers affected by the tax, the CBO estimates that premiums would be lowered 9 percent to 12 percent. The tax would lower rather than raise premium costs because it's assumed employers would reduce the value of their plans to avoid the tax. In this instance, then, lower premiums would also mean less coverage.

Health reform's major provisions concern the "nongroup" market in which individuals purchase insurance on their own. This is where insurers typically commit their worst abuses (like rescission), which the legislation would outlaw, and it's where those who are currently uninsured would in nearly every case satisfy the bill's "individual mandate" to acquire health insurance. According to the CBO, health reform would, by 2016, drive the average per-person premium up 10 percent to 13 percent.

There are good policy-based reasons not to get too bent out of shape about this. For one thing, the price would be driven up mainly because new regulations would improve the quality of coverage offered. Nearly 60 percent of all people purchasing nongroup insurance in the newly established exchanges would qualify for an income-based subsidy. For this group, the subsidies would reduce average premiums 56 percent to 59 percent below current levels. A further consideration is that individuals receiving subsidies would have the option of purchasing a less-expensive (and less-generous) policy, but because of the way the subsidies are structured, the CBO calculates most of them would instead opt for a slightly more expensive (and more generous) policy. According to the CBO, most unsubsidized customers would bypass the rock-bottom policy, too.

(In a Nov. 27 analysis, MIT health economist Jonathan Gruber concluded the Senate bill would drive nongroup premiums down, not up. I asked him via e-mail why his findings were at odds with the CBO's findings. He cited the CBO's "buying up" assumption as the reason. The rock-bottom policies that the CBO assumes most customers would pass up are, Gruber explained, 14 percent to 20 percent cheaper than comparable policies available today. (The White House says the same thing.)

There are, in sum, many reasonable explanations for why health reform would drive average nongroup premiums up, in the CBO's estimation, 10 percent to 13 percent. But to the extent this constitutes bending the cost curve, it's upward, not downward. Government subsidies to lower-income policyholders don't alter the cost; rather, they shift it to taxpayers.

The CBO has not, to my knowledge, estimated the effect health reform would have on nongroup insurance premiums if it contained a robust public option able to tie its hospital and doctor payments to Medicare rates. But a July 14 CBO analysis of the House bill as introduced, with a public plan whose rates matched Medicare's for hospitals and exceeded them by 5 percent for doctors, concluded that public-option (nongroup) premiums would be 10 percent cheaper than private nongroup premiums. I'm no economist, but surely this degree of competitive pressure would result in private nongroup premiums considerably lower than would exist in the absence of health reform. Alas, "Medicare plus five" had to be dealt away to win sufficient votes for House passage. It wouldn't stand a chance in the Senate.

Employer-based health insurance, as I noted before, is almost completely untouched by health reform. But back when there was some promise that health reform might contain a strong public option, it was possible to imagine that over time eligibility to participate in the nongroup exchanges might be broadened to let in those already eligible to receive health insurance through their employer, and that many in this group might seek relief from rapidly rising premiums by buying into the less-expensive public option.