BERKELEY ― Former U.S. Treasury Secretary Lawrence Summers had a good line at the International Monetary Fund meetings this year: governments, he said, are trying to treat a broken ankle when the patient is facing organ failure. Summers was criticizing Europe’s focus on the second-order issue of Greece while far graver imbalances ― between the EU’s north and south, and between reckless banks’ creditors and governments that failed to regulate properly ― worsen with each passing day.
But, on the other side of the Atlantic, Americans have no reason to feel smug. Summers could have used the same metaphor to criticize the United States, where the continued focus on the long-run funding dilemmas of social insurance is sucking all of the oxygen out of efforts to deal with America’s macroeconomic and unemployment crisis.
The U.S. government can currently borrow for 30 years at a real (inflation-adjusted) interest rate of 1 percent per year. Suppose that the U.S. government were to borrow an extra $500 billion over the next two years and spend it on infrastructure ― even unproductively, on projects for which the social rate of return is a measly 25 percent per year. Suppose that ― as seems to be the case ― the simple Keynesian government-expenditure multiplier on this spending is only two.
In that case, the $500 billion of extra federal infrastructure spending over the next two years would produce $1 trillion of extra output of goods and services, generate approximately seven million person-years of extra employment, and push down the unemployment rate by two percentage points in each of those years. And, with tighter labor-force attachment on the part of those who have jobs, the unemployment rate thereafter would likely be about 0.1 percentage points lower in the indefinite future.
The impressive gains don’t stop there. Better infrastructure would mean an extra $20 billion a year of income and social welfare. A lower unemployment rate into the future would mean another $20 billion a year in higher production. And half of the extra $1 trillion of goods and services would show up as consumption goods and services for American households.
In sum, on the benefits side of the equation: more jobs now, $500 billion of additional consumption of goods and services over the next two years, and then a $40 billion a year flow of higher incomes and production each year thereafter. So, what are the likely costs of an extra $500 billion in infrastructure spending over the next two years?
For starters, the $500 billion of extra government spending would likely be offset by $300 billion of increased tax collections from higher economic activity. So the net result would be a $200 billion increase in the national debt. American taxpayers would then have to pay $2 billion a year in real interest on that extra national debt over the next 30 years, and then pay off or roll over the entire $200 billion.
The $40 billion a year of higher economic activity would, however, generate roughly $10 billion a year in additional tax revenue. Using some of it to pay the real interest on the debt and saving the rest would mean that when the bill comes due, the tax-financed reserves generated by the healthier economy would be more than enough to pay off the additional national debt.
In other words, taxpayers win, because the benefits from the healthier economy would more than compensate for the costs of servicing the higher national debt, enabling the government to provide more services without raising tax rates. Households win, too, because they get to buy more and nicer things with their incomes. Companies win, because goods and workers get to use the improved infrastructure. The unemployed win, because some of them get jobs. And even bond investors win, because they get their money back, with the interest for which they contracted.
So what is not to like? Nothing.
How, you might ask, can I say this? I am an economist ― a professor of the Dismal Science, in which there are no free lunches, in which benefits are always balanced by costs, and in which stories that sound too good to be true almost inevitably are.
But there are two things different about today. First, the U.S. labor market is failing so badly that expanded government spending carries no resource cost to society as a whole. Second, bond investors are being really stupid. In a world in which the S&P 500 has a 7 percent annual earnings yield, nobody should be happy holding a U.S. government 30-year inflation-adjusted bond that yields 1 percent per year. That six-percentage-point difference in anticipated real yield is a measure of bond investors’ extraordinary and irrational panic. They are willing to pay 6 percent per year for “safety.”
Right now, however, the U.S. government can manufacture “safety” out of thin air merely by printing bonds. The government, too, would then win by pocketing that 6 percent per year of value ― though 30 years from now, bondholders who feel like winners now would most likely look at their portfolios’ extraordinarily poor performance of over 2011-2041 and rue their strategy.
By J. Bradford DeLong
J. Bradford DeLong, a former assistant secretary of the U.S. Treasury, is professor of economics at the University of California at Berkeley and a research associate at the National Bureau for Economic Research. ― Ed.