House Democrats propose to spend $550 billion of their two-year, $825 billion "stimulus bill" (the rest of it being tax cuts). Most of the spending is unlikely to be timely or temporary. Strangely, most of it is targeted toward sectors of the economy where unemployment is the lowest.
The December unemployment rate was only 2.3% for government workers and 3.8% in education and health. Unemployment rates in manufacturing and construction, by contrast, were 8.3% and 15.2% respectively. Yet 39% of the $550 billion in the bill would go to state and local governments. Another 17.3% would go to health and education -- sectors where relatively secure government jobs are also prevalent.
If the intent of the plan is to alleviate unemployment, why spend over half of the money on sectors where unemployment is lowest? Another 22.5% of the $550 billion would go to social programs, such as expanding food stamps and extending benefits for the unemployed and subsidizing their health insurance.
After subtracting what House Democrats hope to spend on government payrolls, health, education and welfare, only a fifth of the original $550 billion is left for notoriously slow infrastructure projects, such as rebuilding highways and the electricity grid.
The Obama administration claims the stimulus bill will "create or save three or four million jobs over the next two years . . . with over 90% [of those jobs] in the private sector." To prove it, they issued a report from Christina Romer, chairman of the Council of Economic Advisers, and Jared Bernstein, chief economic adviser to Vice President Joe Biden. Its key estimates, however, were simply lifted from an outdated paper by Mark Zandi of Moody's economy.com.
Mr. Zandi's current estimates have government employment growing by 330,400 over two years as a result of the House bill (compared with 244,000 in Bernstein-Romer paper). Yet even that updated figure still amounts to only 8.3% of total jobs added, even though state and local governments are to receive 39% of the funds ($214.5 billion). Spending $214.5 billion to create or save 330,400 government jobs implies that taxpayers are being asked to spend $646,214 per job.
Does that make sense?
Simulations with his macroeconomic model, according to Mr. Zandi, reveal that "every dollar spent on unemployment benefits generates an estimated $1.63 in near-term GDP." By contrast, such "multipliers" simulate that tax cuts for business or investors would add only 30-38 cents on the dollar.
But econometric models are parables, not facts. The big multipliers for transfer payments and tiny multipliers for capital taxes in Mr. Zandi's model reveal more about the way the model was constructed than about the way the economy works. If model builders make Keynesian assumptions, their model will generate Keynesian results. Yet as Harvard economist Robert Barro recently pointed out on this page, contemporary academic economic research does not support the multipliers used to justify the House stimulus bill.
In the March 2006 IMF Research Bulletin, economist Giovanni Ganelli summarized recent International Monetary Fund research on fiscal policy. Several studies find that reductions in government spending "can have expansionary effects, since they can contribute to a consumption and investment boom owing to altered expectations regarding future taxation."
A 2002 study of U.S. data by Roberto Perotti of Università Bocconi did find that the effect of debt-financed spending increases was somewhat positive, but the multiplier effect was much less than one. A 2004 IMF study of recessions in advanced economies likewise found that "multipliers are unlikely to exceed unity." A 2006 study of U.S. data by IMF economist Magda Kandil found the effect of "fiscal expansion appears insignificant on aggregate demand and economic activity."
In December 2008, the National Bureau of Economic Research published "What are the Effects of Fiscal Policy Shocks?" by Andrew Mountford of the University of London and Harald Uhlig of the University of Chicago. "The best fiscal policy to stimulate the economy," they report, "is a deficit-financed tax cut," and "the long term costs of fiscal expansion through government spending are probably greater than the short term gains."
That's because "government spending shocks crowd out both residential and non-residential investment," while "the [positive] response of consumption is small and only significantly different from zero on impact" (i.e., temporarily). But suppose all of these recent studies were mistaken, and the House Democrats' spending spree worked as advertised. We're still left with three million jobs added or saved at a cost of $825 billion -- $275,000 per job.
In short, a growing body of evidence suggests that a dollar of extra spending is likely to lift nominal income by less than a dollar, arguably much less. Several studies suggest the multiplier may be less than zero after a couple of years, because private investment (including housing) eventually falls by more than government spending rises. Another $550 billion of deficit spending on top of a deficit already above $1 trillion is likely to prove more dangerous than helpful to an economy already overloaded with risky debt.