Yes, Obamacare is Going to Explode the Deficit
I missed Jon Chait’s initial criticism of a GAO report requested by Senator Jeff Sessions a couple of days ago. That first post makes me wonder if Chait even read the report before using his jump to conclusions mat, but this second one, which is even more dismissive, caught my eye. As did its quote from liberal pediatrics professor Aaron Carroll, dismissing the whole thing: “It’s a worst-case-scenario. They looked at what would happen to the deficit if (1) we left in all the spending, (2) all of the cost control measures utterly failed, and (3) we removed all of the revenue streams/taxes.” Hmm. Is that really what they did?
The GAO report itself is here. It includes the normal longterm outlook, including an extended baseline, and an alternative scenario. The real debate here is over which scenario is more accurate. Does Chait really think the Medicare cuts are sustainable? (Does anyone?) The alternative is almost always more realistic, since it assumes the Doc Fix will happen and seeks out a more likely path for future spending. I don't think Sessions is doing anything cute here beyond extending the timeframe out in a much longer way. Chait's suggestion otherwise strikes me as a fundamental misunderstanding of how GAO works and what kind of requests they take (possibly just a blind spot because neither of them have ever worked on Capitol Hill, but that’s beside the point). Sessions apparently requested that GAO adopt their normal controls for technical and economic assumptions and try to figure out what Obamacare does under those preexisting scenarios.
Despite what Carroll writes, this isn't even the worst-case scenario. Just this past week, the Arkansas deal struck by HHS moved millions more dollars out the door for subsidies through the health insurance exchanges. If that deal spreads, as I hope it does, the worst case scenario for budgeting could be astronomically worse from Carroll's perspective. GAO won't do estimates any more in dollar terms, but they concluded Obamacare would add 0.7 percent of GDP to the deficit (page 19 of their report). The $6.2 trillion over 75 years Sessions claimed is consistent with GAO's estimate in terms of a deficit increase. If Arkansas’ deal spreads, I think that’s likely lowballing it.
GAO's alternate scenario represents what they deem to be more realistic and more likely because they believe it resembles more accurately the behavior of politicians. They weren't projecting some ridiculous apocalyptic scenario – indeed, it's consistent with what the Medicare trustees couch it as a more reasonable scenario, where after ten years, the cost controls are phased out gradually, with revenues returning to the same simplifying assumptions they use in all their longterm scenarios. That's what the CMS actuary and CBO assumes as well, not just GAO. On the spending side, the only thing the GAO assumes is the same thing CBO does in their alternative scenario – that the cap on the subsidies doesn't apply. Essentially, GAO and CBO are assuming Congress will respond to pressure that kicks in when the caps on premium subsidies through the exchange mean the subsidies cover less and less of rising premium costs. Politically, this seems to be a reasonable conclusion to me.
The key to this argument is a disagreement about revenues. Carroll and Chait are essentially arguing you can take a lot more revenue out of the American economy in the age of Obamacare than you have historically. After year ten, GAO always assumes the same thing: a forty year historical average of revenues. They do this for a reason that has nothing to do with partisanship: over the long term, the tax system has yielded about 19% of GDP (see Hauser's law). This doesn't mean there are no fluctuations – revenue is higher when income growth is higher and lower when income growth is lower, as well as higher and lower in the immediate aftermath of a policy change. But things even out over time. (Note: Jon Chait has previously rejected that fact as a "scam".)
And this brings us back to the central question: why would Obamacare increase the revenue yield over the long run when no other single piece of legislation has accomplished that? Or is it that Chait is assuming the inevitability of a new, more European tax system outside of typical scoring norms? How would the GAO model that, exactly? The problem is that it is Chait and Carroll, not Sessions, who want GAO to ignore historical norms, ignore the conclusions of the actuary and the CBO, and invent an alternate assumption which is closer to what they predict will happen. The burden is on Carroll and Chait to argue for revenues under Obamacare to be more than what they have been for the historical forty year average. All you have to do to make Obamacare not explode the deficit is to completely accept their rejection of historical scoring norms. In that scenario, who's really twisting things?
One final note: what I originally noticed about Chait's followup piece was that includes a back and forth with a GAO staffer about the report, which I take to be intended to lend a nonpartisan air to the conversation. Chait uses it to frame things as a dull-witted Republican Senator asking these poor nonpartisan green eye shades to warp the numbers his way. "I called Susan Irving, the lead author of the GAO report", he writes. Hang on – you don’t mean former Baucus legislative director and Mondale-Ferraro policy director Susan Irving? It's so interesting who ends up in these critical number crunching positions, isn't it?