Payroll-Tax Cut is the Working Part of Jobs Plan

Originally published in Bloomberg News

September 12, 2011
Edward Glaeser (Glimp Professor of Economics, Harvard University)

In response to our sputtering economy, President Barack Obama presented a new stimulus package last week, hopefully titled the American Jobs Act.
More than half of the plan’s $447 billion cost comes from cuts to the payroll taxes, which fund Social Security and Medicare. This could increase the incentives to hire and work, and get money to poorer Americans. But to avoid making our perilous fiscal situation worse, those cuts should be paid for with future increases in the retirement age, and the plan’s proposed spending on infrastructure and unemployment insurance should be substantially reduced.
The U.S. has been battling this downturn since February 2008, when Congress passed its first tax-cutting stimulus bill. When Obama replaced President George W. Bush in 2009, he pushed through the far larger American Recovery and Reinvestment Act, which has disbursed more than $700 billion and now has an estimated total cost of $840 billion. Tax cuts were the largest single item in that package, too, but it also contained substantial aid to states, distributed through Medicaid, education support and extra spending on unemployment insurance.
The Council of Economic Advisers estimates that the Recovery Act “raised the level of GDP substantially relative to what it otherwise would have been and has saved or created between 2.4 and 3.6 million jobs as of the first quarter of 2011.” That assessment may be optimistic, but I still believe significant tax-cutting stimulus made sense in 2009. Two and a half years later, we need to be more cautious.
Ballooning Debt
The national debt has expanded, and the world wonders whether the U.S. will be able to pay its bills. The European banking crisis may put new pressures on our financial system and our budget. The downturn has now revealed itself to be a long, painful crisis that couldn’t be overcome even with more than $700 billion of medicine.
In contemplating another stimulus package, we should restrict ourselves to interventions that carry the biggest benefit relative to cost. That’s why the president is right to emphasize payroll tax cuts, which get money into the hands of ordinary Americans, and have little potential for public waste. They also create stronger incentives for people to work and for companies to hire.
The downside is that lower payroll taxes hurt our long-term fiscal situation, but there is an easy remedy for that. We can create a quid pro quo in which lower payroll taxes are paid for with an offsetting increase in the age at which people can start drawing Social Security. If the age increase occurs many years from now, the reduction in the payroll tax can be budget neutral and wouldn’t hurt the current economy.
Offsetting Cuts
Raising the retirement age was always going to be part of any sensible entitlement reform, and this is the right time to start paying for current tax cuts with future benefit cuts. Inserting an offsetting retirement-age increase into the American Jobs Act would be a good way to show the world that the U.S. is getting serious about its finances.
Yesterday, the president suggested that he was going to pay for this plan with a potpourri of tax changes, including eliminating tax deductions for corporations and the wealthy. Some of these measures surely make sense, but paying for a tax reduction with an increase on the wealthy feels more like redistribution than real stimulus. It would be cleaner to pay for the tax cut by raising the retirement age, and then turning to ending some deductions in a larger overhaul of the system.
Transportation and Schools
Another good way to demonstrate seriousness about weighing costs and benefits would be to strip from the plan the $80 billion allocated to surface transportation and modernizing schools. The president seems convinced that the nation has a huge infrastructure gap and that federal spending on surface transportation is the right way to remedy for the shortfall.
Yet the best evidence the plan can muster for our infrastructure shortage is that our system "now receives a grade of ‘D’ from the American Society of Civil Engineers." The organization’s report is essentially a lobbying document that calls for $2.2 trillion of spending without the slightest whiff of cost-benefit analysis. Would the president similarly just accept assurances from the American Medical Association that we need to spend more on services doctors provide?
But my primary disagreement with the president isn’t over the state of our infrastructure: Many roads, bridges and airports do need repairs. I reject the idea that most infrastructure and its maintenance should be paid for with federal tax revenue.
State Costs
In most cases, public-works improvements overwhelmingly benefit the residents of individual states. When states have to pay for spending out of their own budgets, we are far less likely to see white-elephant projects such as bridges to nowhere and monorails that glide above empty streets.
In almost all cases, infrastructure should be paid for by its users, not with our tax dollars. Certainly, John F. Kennedy International Airport in New York could use an upgrade. But its many well-heeled travelers are perfectly capable of paying for that themselves through airport fees. Likewise, motorists are able to pay for bridge maintenance with tolls. If the president is genuinely frightened about global warming, then his first principle should be to do no harm; that means ending subsidies for drivers through additional surface-transportation spending.
Infrastructure Bank
The proposal for an infrastructure bank is less objectionable because it involves only $10 billion in spending. It also holds more promise. If the bank helps move the U.S. to a better public-works model, in which new highways are paid for with private money and funded by tolls, then that $10 billion would be money well spent.
The president also wants to spend $30 billion modernizing schools. The case for this proposal is supported by evidence suggesting that California home prices increase with school facility spending. A very fine paper by Stephanie Cellini, Fernando Ferreira and Jesse Rothstein, compares "school districts where referenda on bond issues targeted to fund capital expenditures passed and failed by narrow margins."
They argue, quite plausibly, that whether a school bond issue received 51 percent of the votes or 49 percent of the vote is essentially random, and as a result, it is possible to examine the long-run impact of an investment in school facilities by comparing the near winners with the near losers. They find that property values increase in the winning districts by about 7 percent after five years.
California Exception
Yet there are many reasons not to see this study as justification for $30 billion in federal spending. First, the authors’ estimates can barely rule out the possibility that the spending had no effect on property values. Second, they found no impact of this spending on academic outcomes. Instead, more construction usually means spending on the consumption side of schooling, items like snazzier stadiums, not the far more important investment side. Third, their results are from California, where school spending has often been severely restricted by Proposition 13 and the Serrano v. Priest ruling. Additional spending on school facilities may yield far less benefit in other states.
And most importantly, there is a big difference in projects selected and paid for by local residents and those bankrolled by the federal government. I believe that most school districts will pick good projects when they are paying for them. Once the federal government is involved, there is every reason to expect the high-school equivalent of bridges to nowhere.
Unemployment Insurance
We should also be careful about increasing unemployment insurance. Common decency argues in favor of helping jobless Americans. Unfortunately, such benefits also create incentives not to work. Bruce Meyer wrote a classic paper showing that unemployed workers are substantially more likely to get a job just when their benefits run out.
The president has put his jobs plan on the table, and now it is up to the Republicans to respond. They should prune the spending on public works and limit increases in unemployment insurance, but they should support the payroll tax cuts at the center of the proposal. Those reductions should be made budget neutral, by offsetting them with future increases in the retirement age.