Dec 28, 2012 at 3:01 pm
Here’s an important piece by Ezra Klein on recent budget debates stressing a couple of points that I hold near and dear.
First, there’s the “compared-to-what,” or budget baseline problem. Whenever we’re talking about revenue increases or spending cuts, we’re comparing those changes to some line in a budget that represents our existing plans for taxes and spending. So, if our “baseline” is to spend $1 trillion on Medicare over a bunch of years and we cut that by $200 billion…well, you get it.
But we are currently beset by so many different baselines—different possible spending paths—that it’s almost impossible to make a meaningful comparison. I was about to say “a comparison that anyone other than a seasoned budget wonk could understand.” But I’ve been in lots of meetings with seasoned wonks where we ourselves are confused by which baseline someone is using when they’re describing a plan.
Doesn’t CBO—the official score keeper of such matters—help here? Not as much as you think, because they have to use “current law” as their main point of comparison. In normal times, that works OK, but for years, they’ve had to assume, despite the fact that no politician supports this outcome, that the Bush tax cuts fully sunset as per the law (basically, they assume we go over a stay over the cliff). They have an alternative baseline, but that just becomes another in a sea of confusing options.
There’s got to be a better way. I know a lot of people wouldn’t pay attention to these details no matter what, but a lot more would if the debates were comprehensible. In that regard, the process is making it impossible for people to be informed consumers of critical information about how their tax dollars are or will be spent.
Ezra’s piece suggests one solution: just judge budget plans by the (seemingly) simple criterion of whether they reduce the debt/GDP ratio. But the implied simplicity is elusive. Whether a plan achieves that goal involves the same baseline issues invoked above—they’re just buried in a bunch of accounting that gets you to the final ratio.
I recall, for example, lots of budgets that ignored the AMT patch that everyone knew was coming (i.e., they left the cost of the patch out of the baseline). The out years in these budgets had much better looking debt/GDP ratios than they deserved, because they failed to reflect the cost of keeping the AMT from biting a lot more people.
Perhaps it would be best if CBO were instructed to come up with a plausible baseline that the analytic community would agree to use, but that’s probably not realistic—absent a law imposing jail time for using a non-sanctioned baseline, I’m sure folks would start slinging alternatives around pretty quickly.
I’m going to ask a bunch of my colleagues what they think is the best solution and I’ll report back to you. But my sense is we’re stuck with a surfeit of baselines. As a former bassist, this should please me…but it doesn’t.
The second point in the piece is even more important, as stated by CBPP’s Bob Greenstein:
The ultimate goal is to stabilize debt-to-GDP for the long term… However, I think that is not something we should be trying to do for future decades right now. For those decades, the key is health care, and we neither know what health-care costs will look like then — they’re slowing down now, but we don’t know if that’s permanent or temporary — and we don’t know enough about what changes we need to make to the health-care delivery system to slow costs. So stabilize the debt for the current decade and buy yourself time to figure out health care.
[See Peter Orszag’s graph and commentary here on this point re slower growth of health costs.]
What we need is an alternative baseline—D’oh!!—that shows just how much oxygen gets injected into the budget outlook if the recent slowdown is lasting (I’m working on it). And if so, the point isn’t just to coast along enjoying the better outlook but to figure out which reforms are generating the efficiencies in health care provision and spread them throughout the system.
Thank you for joining the conversation. Comments are limited to 1,500 characters and are subject to approval and moderation. We reserve the right to remove comments that: