Republicans Eat Their "Budget Experts"!

Ross Douthat screams and leaps. His targets: Glenn Hubbard and John Cogan:

The Road To Serfdom?: Note that the increase [in domestic discretionary spending as reported by Cogan and Hubbard] looks roughly twice as shocking as it actually is because the chart-makers, John Cogan and Glenn Hubbard, decided to start with a baseline of $200 billion rather than zero.

They're honest enough to allow that a chunk of this increase is inflationary, and another chunk homeland-security related; what they don't show, though, is the growth of the U.S. economy during the same period, and how the Bush-era increase in discretionary domestic spending looks in historical context as a percentage of GDP. To his credit, [Peter] Robinson queried Cogan on this point:

Q: The chart shows the increase in spending in dollar terms. Haven't you been able to find a chart that shows the increase in spending as a proportion of GDP?

A: No, I haven't--not in the time I've had available for Googling this weekend, which, since I've been scrambling to get the family ready to go back East for a couple of weeks (we're off at 4.30 this very morning) amounted to a little under half an hour. Sorry about that. And I'll check in the from the beach when I can.

Um ... what? According to Cogan's bio, he's a professor in the Public Policy Program at Stanford University, and his "current research is focused on U.S. budget and fiscal policy, social security, and health care" - yet he can't find a chart showing one of the most relevant statistics to a debate about whether George W. Bush is a wild and crazy overspender? I know where to find those statistics right off the top of my head, and I'm a rank amateur: Just head to CBO.gov, click on Historical Budget Data [http://www.cbo.gov/budget/data/historical.shtml], and flip to page 8, where you'll discover that in 2001, when Bush took office, discretionary domestic spending accounted for 3.1 percent of GDP, and in 2007 it accounted for... 3.3 percent of GDP. In the years between, it rose as high as 3.6 percent of GDP, which is on the high side by post-Reagan standards (we averaged 3.25 percent a year in the 1990s), but way lower than in the profligate, post-Great Society Seventies, when we were spending as much as 4.8 percent of GDP a year on domestic programs.

The bottom line: The Bush years haven't been a small-government success story by any means, and fiscal conservatives have every right to be disappointed. But the road to serfdom this ain't. (Certainly Friedrich Hayek himself, who vigorously defended free markets without taking anything like the Norquistian position on the pressing need to drown the welfare state in a bathtub, wouldn't recognize it as such.)

Ross is, of course, right.

He is right on the substance: the complaint against George W. Bush's discretionary spending has been that the defense side has been stupidly overbloated (increases for the navy and the air force to keep them at parity with the expanded army) and the domestic side misallocated (farm subsidies for agribusiness), not that spending on non-homeland security domestic programs has grown unusually rapidly.

And he is right in his imputation of--let's call it "shoddy worksmanship" on the part of Cogan and Hubbard, especially the claim that they were unable to find the time to plot discretionary domestic spending ex-homeland security as a share of GDP.

As Ross says, it takes you two minutes to find the domestic discretionary share of GDP in 2008. It takes you thirty seconds to find that nominal GDP has grown from $8,304B in 1997 to $13,870B in 2007--an increase of 67%. It then takes another thirty seconds to take the $245B 1997 discretionary domestic spending number, multiply it by 1.67 to get $409B, and then add in $48B in increased homeland security spending according to AEI's Veronique de Rugy http://www.aei.org/docLib/20061214_FactsandFigures.pdf to get a 2007 number of $457B as the constant-share-of-GDP domestic discretionary baseline:

Comments

[bakho]

It should go without saying that GDP takes into account increase in population.
Spending as % of GDP is more closely related to spending per person.
More people takes more money to provide the same level of service to each person.
The argument that spending is less effective (incompetent managers, diversion of government program money to higher overhead private groups) can still be entertained.

[Tyrone Slothrop]

The explanations on the graph describing "baselines" are confusing, because they're amounts above two different baselines. At first blush, the Hubbard-Cogan baseline appears to be lower, which is counter-intuitive.

[derrida derider]

Cogan would have been better off saying bluntly "I chose the dollar base rather than the GDP base to highlight my case" rather than putting forward that ridiculous excuse.

bakho, google "Baumol's curse" or "Baumol effect". Governments are in the service provision business. Services take up an ever larger share of GDP as an economy grows, because productivity growth in services - especially government-type services - is less than productivity growth in agriculture and manufacturing. IOW it takes ever less human effort to produce food and widgets, so an ever bigger proportion of human effort is transferred to providing services to other humans. That's why *no* western government (not even Thatcher's or Reagan's) has succeeded in cutting spending as a share of GDP for more than a single year.

[bakho]

derrida deri- The Baumol effect is created by competition for labor. Labor competition has not been especially strong throughout most of the Bush years. It doesn't explain very much.

Spending as a percent of GDP went down through much of the 1990s contrary to your contention. This was due to reduction in military spending as % GDP (peace dividend). Reagan budgets were a complete mess.

[robertdfeinman]

Stories like this highlight why economists are held in such low regard.

A physicist was just censored for claiming the results of an experiment that couldn't be substantiated. The same thing happened with the cloning claims of a Korean researcher last year.

I've never seen any economist suffer professional consequences for "shoddy" (that is partisan) work. In fact there are plenty of institutions devoted to giving such people reliable employment so that they can continue passing off propaganda as scholarly work. If they work long enough at one of these misinformation factories they end up being rewarded by getting regular columns in the NY Times, or the Washington Post.

The problem is you can fool enough of the people enough of the time so that you have a nice career, but you can't fool Mother Nature as the world is suddenly finding out.

[David Graves]

I guess it comes back to the age-old requirement of my elementary school arithmetic teachers--show your work. BTW, what's the pay scale over there at Stanford? Or the McCain campaign?