The health care reform discussion is beginning-at last!-to get real. On June 9, the Senate Health, Education, Labor, and Pensions Committee released a draft bill, and the Congressional Budget Office published an estimate that the bill would cost $1 trillion over 10 years and leave 35 million uninsured.

To be more precise, the HELP Committee released fragments of a bill; it did not include extensions of Medicaid or mandates on employers that everyone anticipates it will add, nor did it specify how the added costs would be paid for. And CBO’s estimates referred only to fragments of the incomplete draft bill that was actually released.

But even that low degree of specificity was enough to send tremors through the health policy community. Then came a large after-shock–Chairman Max Baucus of the Senate Finance Committee announced a delay in issuing the committee’s bill because initial cost estimates were even higher–$1.6 trillion over 10 years.

After months of trading in generalities-high-flown goals and policy changes so vague that no one could possibly understand who would be required to do what differently from what they do now-the legislative language hit the table and the cost estimates hit the fan.

The estimates for the HELP bill were shocking. The $1 trillion estimate was bad enough. Worse was the spending rate at the end of the decade. The initial costs of any bill are close to zilch, because it takes a couple of years to get a plan going and a couple more before it is running full blast. For those reasons, the cost of the $1 trillion HELP bill translates into $200 billion in 2019. The $1.6 trillion estimate for the Finance bill implies outlays in the 10th year approaching $300 billion.

The Center on Budget and Policy Priorities points out that the initial estimate of the HELP bill was misleading because it ignored several important provisions in the bill and assumed that the individual mandate would be backed up by penalties so weak that enrollment in an insurance plan would be virtually voluntary. There is no reason to think that estimates of the Finance Committee were incomplete (they have not been released).

The combination of specific legislative language and CBO scoring is bringing into high relief two important aspects of the health reform debate.

The first is a reminder of how complex and nasty the tradeoffs are that have to be made if coverage is to be universal or close to it. The best estimate is that the system-wide cost today of a fully implemented universal coverage would be $125 billion to $150 billion a year. But that cost grows as the cost of health care increases. At the end of 10 years, the cost would be $200 billion a year. But that is system cost–added national health care spending, not budget cost–added spending by the federal government. Unless a draft bill finds a way of keeping every dollar of non-budget health care spending ‘in the game’ or forcing some increases in what businesses spend, the budget costs could be larger than the system costs.

The draft HELP bill was silent on how to pay for these added outlays. Sen. Baucus pledges that his proposal, when published, will be paid for. The hostile reactions of members of Congress, Democrats as well as Republicans, to the tax increases and spending cuts the president proposed as a reserve to pay for health reform suggest that paying for the bill is the biggest challenge. President Obama has said that health reform must be paid for. Where is the money to come from? Keep in mind that all those savings promised by drug companies, hospitals, and doctors through private actions will not count. CBO has made explicit that it will count only savings resulting from federal legislation.

The second aspect of current developments is a reminder that the administration’s calculated decision NOT to present a bill to the Hill has risks as well as benefits. So far, the administration has contented itself with enunciating broad principles in public and leaving the drafting of the final bill to lawmakers. This strategy has a large advantage. It avoids the result that befell the Clinton administration which drafted a detailed bill, complete to the last comma; those who had to vote on it had no stake in it whatsoever (except, of course, electorally, as they learned in 1994).

At some point, however, the administration must decide whether it is prepared to be more directive about what it wants and doesn’t want. It is not likely, for example, that the split among Democrats on whether to curb the exclusion of employer-financed health insurance from income and payroll taxes can be resolved without heavy White House involvement. This issue and others will not be resolved unless the president weighs in-and weighs in hard, twisting arms and using every bit of political leverage that the White House can muster.

Henry J. Aaron is the Bruce and Virginia MacLaury Senior Fellow at The Brookings Institution.