Lobster Stuffed With Tacos

Matt Yglesias posts the following graph to illustrate that “the Obama-era surge in government spending is largely a figment of the conservative imagination.”

I sympathize with Matt’s argument, but charting only government spending on consumption and investment omits several other important budget items, not the least of which is government transfers (think Social Security, Medicare, Medicaid, etc.) which I’m guessing is where most conservative concern lies, rightly or wrongly. There’s also some other idiosyncrasies with this NIPA data, such as timing and treatment of depreciation, that make it an inappropriate substitute for budget numbers.

So why not just look at real total federal outlays?

So now the story becomes more interesting. It’s clearly true that the increase in real federal spending in 2009 was massive: almost 20% over 2008 real levels. Since this was Obama’s first budget, some of this increase was due to a continuation of Bush-era budgets, but some was undoubtedly also due to conscious policies by the White House such as the Recovery Act.

But when Charles Koch says that government spending has been uncontrolled “over the past several years”, the data contradict this.What you see instead is a small decrease in real spending in 2007, high growth in 2008 and extraordinary growth in 2009, followed by a large contraction in 2010. That’s not to say that government spending could spiral out of control in the future, if unchecked by future Congresses (more in a future post), but that’s not the point Mr. Koch was making.

In his slides selling the emerging debt ceiling deal to the GOP caucus, Speaker Boehner writes that the proposed “super committee” charged with coming up with an additional $1.5 trillion of deficit reduction over 10 years will be forced to use CBO’s current law baseline, making it “impossible… to increase taxes.” What does this mean, and is it true?

CBO’s current law baseline assumes that the 2001/2003 (commonly called the “Bush”) tax cuts – extended through the end of 2012 under a compromise deal last December between President Obama and the House GOP – will finally expire for everyone on December 31, 2012, at which point tax rates will revert to their pre-2001 levels (plus tax hikes passed as part of health care reform). There’s a lot of revenue riding on the expiration of the Bush tax cuts: CBO estimates that letting them expire for everyone will raise $2.5 trillion in revenue over a decade compared with extending them permanently for everyone, not counting additional deficit reduction from lower interest payments.

So when the “super committee” makes its deficit recommendations this coming November, the only revenue proposals that would score as actual deficit-reducers under a current law baseline would be those raising more revenue than would be true assuming the Bush tax cuts expire.

The GOP is betting that as a result, raising taxes will prove “impossible”. That’s probably because the only revenue raising Congress might have a stomach for would be tacked-on to fundamental tax reform of, say, the Bowles-Simpson type. But changing the fundamental nature of the income tax system effectively nullifies the expiration of the Bush tax cuts: after all, how can the 10% bracket, for example, rise to 15% percent if there is no more 10% bracket? In the earlier debt ceiling talks, Obama and Boehner were considering added revenues of $800 billion to $1.2 trillion; it was approaching the upper-end of that range that ultimately compelled Boehner to walk away. But keeping revenues in line with current law assumptions would require twice as much extra revenue.

Clearly, then, raising revenue under a current law baseline would face high political hurdles.

But it wouldn’t be “impossible.” First, the committee could of course just choose to raise marginal rates above their pre-2001 levels, though I’m guessing most variations of this idea would be DoA in a GOP House. Too bad too, because there’s a decent case to be made on the merits for raising the gas tax.

Second the committee could choose to eliminate or change tax expenditures without touching marginal rates and brackets, as long as they were tax expenditures that would otherwise have been permanent fixtures of the tax code. The mortgage interest deduction is one example; the exemption of employer-sponsored health insurance is another.

Third, the committee could come up with a brand new tax added-on to the current tax system. A carbon tax or a VAT would both fit this description.

So there’s actually plenty of accounting room for higher revenues under a current law baseline. The seminal question is whether there’s any political room in the era of the Tea Party.

UPDATE 1 (10:44pm): My friend Michael Linden writes that the White House is denying that the legislation constrains the committee to any particular baseline. Guess we’ll know more tomorrow.

Writers more qualified and skilled than I me will undoubtedly discuss the political viability of the Gang of Six’s deficit reduction proposal [PDF] today. Here, for example, is Ezra Klein.

For now, I just wanted to make two observations about how the summary document linked above characterizes their future budget assumptions and the savings from their proposal:

Slash our nation’s deficits by $3.7 trillion/$3.6 trillion over 10 years under CBO’s March 2011 baseline…

…

Reduce our publicly-held debt to roughly 70% of our economy by 2021.

1. The Gang of Six isn’t actually using the CBO’s March 2011 baseline.

In wonk-speak, saying “CBO’s March 2011 baseline” means something very specific. It means you’re assuming that current law won’t change. So under current law, the 2001/2003 tax cuts expire for everyone in January 2013, at which point marginal rates go up for a lot of people. Also, under current law the alternative minimum tax (AMT) isn’t indexed for inflation, and so more and more filers become subject to it over time, raising their tax liability. Finally, current law assumes we make deep cuts to Medicare physician reimbursements.

If this actually happened – and few observers believe that Congress and the president will actually let all these outcomes occur without intervening and changing the law – CBO projects that publicly-held federal debt will reach $18 trillion or 75.6% of GDP in 2021.

Here’s what’s odd, though: the plan claims to save $3.6 – $3.7 trillion off of the CBO baseline through 2021, and get our debt down to 71% of GDP that year (you can find the 71% number in one of the document’s later graphs).

But you don’t need to cut any where near $3.6 trillion over 10 years to get debt that low under CBO’s baseline. To get debt to 71% of GDP in 2021, in fact you’d only need to cut $1.1 trillion off of the CBO baseline (this already includes interest costs, so actual spending cuts or tax increases would be even less than this). You can confirm this yourself once you know that CBO projects nominal GDP to be $23.81 trillion in 2021. So debt reduction off the CBO baseline would be (75.6% – 71%)*$23.81 trillion = (4.6%)*$23.81 trillion = $1.1 trillion.

What gives? Clearly, the Gang of Six isn’t simply going off of CBO’s current law baseline. Instead, they’ve made assumptions about future policy and modifications to current law.

The graph on page 8 shows that whatever baseline they’re using assumes that publicly-held federal debt will reach 101% of GDP in 2021. After playing around with the numbers, my educated guess is that they’re making these modifications to the CBO baseline:

1. The 2001/2003 tax cuts are extended permanently for everyone.
2. The AMT is permanently patched for inflation.
3. Physician payment cuts under Medicare don’t happen.
4. Discretionary spending grows faster (by GDP instead of by inflation).
5. Troops in Iraq and Afghanistan are reduced to 45,000 by 2014.
6. A group of tax deductions, credits and exemptions that are supposed to expire – generally called “the Extenders” – are permanently extended.
7. Folds in the savings from the budget deal passed in April after CBO’s last baseline came out.

These aren’t unreasonable assumptions, but they’re a far cry from “CBO’s March 2011 baseline”.

2. You need a lot more than $3.7 trillion in cuts to get the debt reduction they’re claiming

Again, this is just simple math. The Gang of Six’s baseline assumes that debt reaches 101% of GDP in 2021. They claim their proposal reduces it to 71% of GDP, a reduction in debt equal to 30% of GDP. CBO projects that nominal GDP in 2021 will reach $23.81 trillion.

So, actual debt reduction over 10 years (and remember, debt is cumulative, so you only need to focus on the final year) is 30% x $23.81 trillion, or about $7.1 trillion.

Now, that $7.1 trillion includes both direct spending cuts/revenue hikes and the effect of lower interest on the debt, so perhaps the Gang of Six $3.7 trillion claim doesn’t include interest savings?

But that would mean interest savings are almost half of the 10-year debt reduction, and that’s just not possible under reasonable assumptions. If you front-loaded all the direct deficit reduction to maximize the savings from interest over 10 years (through compounding), you get about an 85:15 split between direct and interest savings through 2021, which would mean the upper-bound debt reduction from the Gang of Six’s plan would be about $4.4 trillion, well short of the $7.1 trillion they need under their own baseline. Since presumably the Gang of Six’s plan would be more gradual, $3.7 trillion in direct savings is going to yield less in interest savings and thus less in total debt reduction through 2021 than this upper-bound.

The bottom line is that something is amiss in their math. I want to emphasize that nothing here implies that anyone is being dishonest. This whole analysis rests on a question of baselines, which can get hairy and confusing very fast (especially for me). Also, the Gang of Six may have a political rationale for actually under-reporting debt reduction if it takes the spotlight off of revenue increases. This won’t get sorted out until we have a CBO score.