Friday, January 25, 2013

Where the (Federal) Money Went

In today's column, Paul Krugman said that the current high federal budget deficit is due to the financial crisis -- specifically lower federal receipts and higher spending on unemployment insurance.
The truth is that the budget deficits of the past four years were mainly a temporary consequence of the financial crisis, which sent the economy into a tailspin — and which, therefore, led both to low tax receipts and to a rise in unemployment benefits and other government expenses. It should have been obvious that the deficit would come down as the economy recovered. But this point was hard to get across until deficit reduction started appearing in the data.
This seemed unlikely to me, but you know, he's right.

Here are federal receipts:

FRED Graph

Peak-to-trough the decline was $460 B. And here's unemployment payments:

Graph of Personal current transfer receipts: Government social benefits to persons: Unemployment insurance

Trough-to-peak the difference was $130 B. 

Adding them makes a budget change of $590 B, which if you subtract from today's deficit of $1.2 T gives about $600 B, not a lot higher than what Bush had it up (well, down) to, about $400 B):

FRED Graph

On a per capita, inflation-adjusted basis, total government spending (federal + state + local) has decreased 5.9% in the last 9 quarters.

1 comment:

Hohokam71 said...

Mr Appell,

I’m not following your reasoning here. By any chance, are you confusing deficit and debt? The Fed’s “FGRECPT-FGEXPND” graph that you cited is a quarterly “snapshot” of the yearly deficit, and not a measure of the total debt that was accumulated through prior deficits.

By looking at the Excel data that the Fed provides with the charts you cite, I think that I understand how you came to the “budget change” figure of ~$590B. The peak-to-trough difference in the federal receipt chart that you linked was a decrease of ~$460B (October 2007 to July 2009), and the trough-to-peak difference in unemployment payment chart you linked was an increase of ~$130B (June 2007 to January 2010). I have no problem with this: the government had to deal with an overall ~$590B reduction in its treasury accounts, due to reduced receipts and increased unemployment payments that can reasonably be attributed to the recession. So far, so good.
You then show that the difference between this year’s deficit of ~$1.2T (well, ~$1.1T per the Fed’s data used to make the chart, using the 3rd Quarter 2012 “snapshot”, but close enough) and the 3rd Quarter 2003 ”snapshot” of ~$450B (again, per the Fed’s data … you cited about $400B, but close enough) is close to the $590B “budget change” … but I just don’t see the significance. Now, if instead it were the case that the debt (not the deficit – that’s different; the deficit is essentially the “yearly change in debt”) was ~$1.2T today and the debt in 2003 was ~$450B, then you would have a very interesting point, since this would imply that the difference between these two figures could be accounted for largely by the ~$590B effects of the recession. But deficits are not debts; total debt incurred by deficit spending between 2003 and now was significantly more than ~$590B … per the U.S. Treasury Department’s database at http://www.treasurydirect.gov/NP/BPDLogin?application=np, I get an increase in debt of ~$9.2T (trillion, not billion) since 3rd Quarter 2003 and 3rd Quarter 2012.
Speaking respectfully, what point of yours am I missing here?

In any event, none of this really contradicts the specific Krugman quote that you cite … expecting to see the deficit itself go down due to a decrease in unemployment expenditures and an increase in receipts is one of the components used in models that predict future deficit and debt levels. From the data behind the Fed graphs that you cite, we have increased receipts by ~$465B since the “trough receipts” period (currently ~$5B over the pre-recession “peak receipts,”) and we are ~$90B under “peak unemployment expenditures.” I suppose that if the economy improves to pre-recession unemployment expenditures, we will “save” another ~$40B per year … but this is not a significant subtraction to either the deficit or the debt. Instead, receipts need to drastically increase (we need a growing economy and/or more tax revenue) to offset future deficits … or spending needs to be reduced, or both. So, the big political debate of how much taxes to raise or how much spending to cut persists.

Also, your statement “on a per capita, inflation-adjusted basis, total government spending (federal + state + local) has decreased 5.9% in the last 9 quarters” isn’t shown by the Fed graph that you link to. By pulling the data that the Fed used to form the graph, spending during the past nine quarters is up by ~$58B ($5,603B per April 1st 2010 shapshot, growing to $5,661B on July 1st 2012). Not a significant increase in total government spending, but not a decrease either. Where did you get the figure of a decrease of 5.9%? I don’t see it in how I interpreted the data you linked to, and I couldn’t find it after poking around via Google for a while … a 5.9% decrease in total governmental spending over nine quarters is a very significant datapoint.

Best Regards, JH