There is a lot of arguing going on about the dangers of increasing the deficit and debt by using federal government spending to end the present high level of unemployment. Whether this is innocent ignorance or a disingenuous, politically-motivated smokescreen, it is an irrelevant argument. An examination of the relevant economic data indicates that if this depression is allowed to continue at the present high level of unemployment, it will cause a serious, ongoing, automatic increase of hundreds of billions in the deficit and debt.

In fact, as shown below, each 1% of excess unemployment is automatically adding nearly $60 billion to the deficit. The bottom line is that, for our present 10%-unemployment situation -- compared to the 4% experienced (without significant inflation) back in 2000, before the Bush administration --

Ironically, this is almost half the amount of the much-fought-over and eventually-reduced one-time stimulus package of last year. And this excess spending might go on indefinitely in a "jobless recovery" if too little federal spending is applied to generate a quick reduction of unemployment. Both of these approaches involve deficit spending. But we have a choice in the quality and results of that spending:

The close correlation between excess unemployment and increases in the federal deficit can be seen clearly in Figure A's time-ordered scatterplot (and in Figure B's parallel movements of the two series):

The relationship illustrated by this scatterplot is between two series derived from Congressional Budget Office and Bureau of Labor Statistics data:

1. The Hi-Unemployment Deficit [1] -- the increase in the deficit attributable to excess unemployment. This increase is caused mainly by the "automatic stabilizers" -- the decreases in tax receipts, and increases in unemployment benefits, early retirement, etc., that result from rising unemployment.
2. Excess unemployment -- The difference between the "official" (commonly published) civilian unemployment rate [2] and the CBO's "non-accelerating-inflationary rate of unemployment" (NAIRU) [3]. (Caveats: It should be noted, as is widely understood, that the "official" unemployment rate seriously understates the impact of actual unemployment on the economy and society; also, the 1990s called into question the validity of the Phillips Curve assertion, implicit in NAIRU, that low unemployment inherently causes inflation.)

The chart shows three distinct periods in the four decades from 1970-2009 -- 1973-1982, 1988-2001, and 2002-2009 [4]. The regression lines for these periods indicate a high degree of correlation, with R values .96 or .97. The slopes indicate the deficit/unemployment relationship during that period; for the most recent, the value is .41.

This indicates that, between 1988 and 2009, each 1% change in unemployment above the NAIRU level (currently 4.8%) produced a corresonding change in the deficit of 0.41% percent of GDP. At that rate, with GDP currently at around $14.7 trillion, each 1% of excess unemployment is adding $60 billion to the deficit, or, as pointed out above,

Ironically, this is almost half the amount of the much-fought-over and eventually-reduced one-time stimulus package of last year. And this excess spending might go on indefinitely in a "jobless recovery" if too little federal spending is applied to generate a quick reduction of unemployment. Both of these approaches involve deficit spending. But we have a choice in the quality and results of that spending:

Notes and Data Sources

1. The Hi-Unemployment Deficit is calculated as the difference between the actual deficit and the CBO's "standardized" deficit, which is the deficit as it would be at a "standard" level of unemployment; in the context of a growing economy, we normalize this deficit by handling it as a percentage of GDP.
   Source: The Budget and Economic Outlook: Fiscal Years 2010 to 2020, January 2010
       Table F-1: Revenues, Outlays, Deficits, Surpluses, and Debt Held by the Public, 1970 to 2009 (bil.$)
       Table F-12: Cyclically Adjusted Deficit or Surplus and Related Series, 1970 to 2009 (bil.$)
2. Civilian Unemployment Rate
   Source: Bureau of Labor Statistics, via the Federal Reserve of St. Louis FRED data source
3. Non-acceleratng-inflation Rate of Unemployment -- the basis for the CBO's "standardized" deficit, mentioned above
   Source: CBO's The Budget and Economic Outlook: an Update, August 2009

4. It would be interesting to determine what changes in tax or relief policy caused these distinct and internally consistent periods (also shaded on the above time plot).