Ezra, among others, points to the CBO blog follow up which says that $86 billion of the new spending consists of continuing existing levels of spending in the Bureau of Indian Affairs and assorted other agencies.  In other words, just because they happened to stick this stuff in the health care reform bill, rather than somewhere else, doesn't mean we should attribute the cost to the health care bill.  A number of readers have mentioned this, so I think this is worth writing more about.

It's a fair enough argument, in one way, but it seems to me that we're getting entirely too cute with what "really" constitutes a cost of the health care bill.  The "doc fix", we're told, "has to happen anyway", so it shouldn't be counted--even though the permanent changes to the SGR, which are going to cost hundreds of billions, are very clearly being offered as a quid-pro quo in exchange for the American Medical Association's rather unenthusiastic support.  I've defended progressives on this, on the grounds that if it's in a separate bill, well, the CBO has to score it as a separate bill.

But the corollary to this is that it's in the bill, it's a cost of the bill--even if you think the government would have gone and done this anyway at some other time.  I mean, I'm happy to use the argument that we should think about previous appropriations levels, rather than what is or isn't in the bill itself . . . but then I think we have to include the doc fix as a cost of health care reform, because making it permanent is very clearly necessary to its passage, and furthermore (IMHO), would not likely be happening without PPACA--if they didn't have to buy off the doctors, they'd be doing it on a temporary basis, rather than scrambling to find the money for a permanent fix.  So I don't think this logic actually improves my opinion of the bill's costs.

And even if we did throw both out, we've still got another $30 billion of unexpected spending, not two months after the bill was passed.  We've got companies doing exactly what we were promised they wouldn't:  exploring the option of throwing their workers into the individual markets, at great cost to the rest of us.  We've got state insurance commissioners essentially commanding insurance companies to sell at a loss.

It's not that one expects the projections to be perfect.  But imperfect projections are supposed to have random error--you get surprises to the upside, and surprises to the downside.  All these errors run one way.  Though it's too early to tell, it makes me worry that the estimates might be biased--not in the way that we commonly use the word, which implies some sort of volitional, usually explicitly political thumb on the scale, but simply in the statistical sense that the method used to do the estimates may systematically produce projections that are lower than the true value. 

That wouldn't necessarily make it a bad method, either--there are reasons that we often use police counts for crimes, even though we know that many crimes go unreported; it's hard to estimate the incidence of unreported crimes, and so we use the easy-to-measure number in many contexts, even though we know it's too low.  So I want to make it clear that I don't think there's anything obviously wrong with the way the CBO is doing things.  It's just that the steady trickle of bad news makes me worried that there's worse to come.

Update:  Reader MarketKarma makes a good point

how about the government taking over student loans (and the budget savings associated therewith) being included in the bill?

Ezra didn't seem to have and issue with counting those savings

We want to hear what you think about this article. Submit a letter to the editor or write to letters@theatlantic.com.