Following up on my previous post, one way of ferreting out the likely effectiveness of an infrastructure spending program is to consider the results of previous programs. Fortunately or unfortunately as the case may be, we have a fairly recent example in the American Recovery and Reinvestment Act of 2009 (ARRA). As you might guess there’s some dispute over just how effective it was.
John Taylor, for example, wrote this for the National Bureau of Economic Research (NBER):
The implication is not that ARRA has been too small, but rather that it failed to increase government consumption expenditures and infrastructure spending as many had predicted from such a large package. A consideration of the counterfactual event that there had not been an ARRA supports the hypothesis that state and local government borrowing would have been higher and purchases would have been about the same in the absence of ARRA.
or, in other words, it didn’t achieve much at all. Daniel J. Wilson on the other hand, writing for the Federal Reserve Bank of San Francisco (PDF) comes to a somewhat different conclusion:
Based on my preferred measure of spending, announced funds, the results imply that its first year ARRA spending yielded about eight jobs per million dollars spent, or about $125,000 per job. Extrapolating from that marginal local effect to the national level, the estimates imply ARRA spending created or saved about 2.1 million jobs, or 1.6 percent of pre-ARRA total nonfarm employment, in that first year.
Some of that priciness might be explained by the significant number of the “jobs saved” having been on jobs in the financial sector, whether private or public. Was saving 800,000 jobs in the financial sector a social good or not? I think not but YMMV. Finally, writing for the Federal Reserve Bank of Philadelphia (PDF) Gerald A. Carlino arrives at a conclusion closer to my own than either of the foregoing:
Did the Recovery Act work? The evidence suggests the economy did indeed grow more than it would have without the stimulus but likely not as much as it might have with a different type of stimulus. In particular, the evidence suggests that direct measures — tax relief for households and firms, and programs such as Medicaid that target families with low incomes, little wealth, and a limited ability to borrow — have contributed more to GDP growth than direct federal expenditures or project aid to state and local governments. To the extent that the federal government implements its stimulus spending through transfers to state and local governments, perhaps that aid should target lower-income households and states that bear the brunt of the economic downturn.
Just as there’s a difference between the United States of 1935 and that of today, there’s a difference between that of 2010 and that of today and IMO our economy, more dependent if anything on imports than it was then, is likely to benefit less from an infrastructure spending program while our trading partners are likely to benefit more. Unless the construction of the program is very, very deliberate, something in which the Congress has shown little interest.