Barro: Stimulus Hurts Economy

Robert Barro is a Harvard economics professor and Hoover Institution senior fellow. Barro concludes last year’s stimulus program will generate $600 billion of public spending at the cost of $900 billion in private expenditure; in sum, “a bad deal.”

Here’s how Barro arrived at his cost/benefit analysis of a package originally estimated at $787 billion but now priced at $862 billion:

1. We need an empirical model based on the history of past fiscal actions in the U.S. or other countries.

2. Burro uses long-term U.S. macroeconomic data to estimate the effects on GDP of increased government purchases (the spending multiplier) and increased taxes (the tax multiplier).

3. The spending results are based on wartime defense outlays, which show large positive values in the early stages of wars (extra spending of 26% of GDP in 1942) and large negative values in war aftermaths (27% of GDP in 1946).

4. Burro uses the defense-spending multiplier because such spending can be precisely estimated from the available data and it provides a reasonable gauge of nondefense government purchases.

5. He arrives at a stimulus spending multiplier of 0.4 within the same year and about 0.6 over two years, meaning that if the government spends $300 billion in 2009 and 2010, GDP will rise by $120 billion in 2009 and $180 billion in 2010.

6. For the tax multiplier, Burro uses a newly constructed measure of average marginal income-tax rates, estimating that an increase in marginal tax rates reduces GDP around minus 1.1. Hence, an increase in taxes by $300 billion lowers GDP the next year by about $330 billion.

7. Burro notes that Christina Romer, Obama’s Council of Economic Advisers chair, had with her husband done the research on tax multipliers, developing tax multipliers of even larger magnitude. Romer, however, hasn’t worked on spending multipliers, so shouldn’t assume spending values above one that that don’t conform to Burro’s work.

8. Burro estimates the 2009-10 fiscal-stimulus package, $300 billion of added government purchases in 2009 and 2010, goes back down to its 2008 level of zero in 2011, even though the added spending might be permanent. He also assumes taxes do not change in 2009-10, meaning the stimulus is deficit-financed. He then estimates GDP rises by $120 billion in 2009 and $180 billion in 2010—compared to a baseline of no stimulus package, while other parts of GDP fall by $180 billion in 2009 and $120 billion in 2010.

9. Such a deal looks good because we "buy" the added government outlays by paying 60 cents on the dollar in 2009 (losing 180 in private spending to get 300 in government spending) and 40 cents on the dollar in 2010.

10. But the added $600 billion of government spending leads to a correspondingly larger public debt we must pay for by raising taxes. Assuming a tax multiplier of -1.1, then applying with a one-year lag, Burro shows the higher taxes reduce GDP by $330 billion in each of 2012 and 2013.

11. That means the path of incremental government outlays over 2009-13 in billions of dollars is +300, +300, 0, 0, 0, which adds up to +600. The path for GDP is +120, +180, +60, -330, -330, adding up to -300.

12. The projected effect on other parts of GDP (consumer expenditure, private investment, net exports) is -180, -120, +60, -330, -330, which adds up to -900. So over five years, the fiscal stimulus package is a way to get an extra $600 billion of public spending at the cost of $900 billion in private expenditures.