Think of it as steering between the Scylla and Charybdis of ineffective government spending.
On one side are the rocky shoals of gridlock, where nothing gets done and Congress can’t re-calibrate a program in response to new problems or changing conditions. On the other side is political risk, where government policy veers back and forth every time a different party comes to power.
For Renee Bowen, an economist and game theorist at Stanford Graduate School of Business, one way to navigate the rival hazards is with mandatory spending programs that have some built-in flexibility.
“Mandatory” programs, such as Social Security and Medicare, are funded year after year under a formula that can only be changed by new legislation. A “discretionary” program, such as the budget for NASA, has to be appropriated every year.
The good thing about discretionary spending is that it can be adjusted to deal with new circumstances or changes in political preferences. The problem is that it is unpredictable, and it can swing widely. That makes planning difficult and undermines a program’s effectiveness.
By contrast, the advantage of a mandatory program is that it provides security and predictability — a plus from the standpoint of efficiency. The problem is that conditions always change over time, so a program on autopilot can gradually run off course.
In a new paper, Bowen and three co-authors use game theory to argue in favor of mandatory programs that contain some built-in flexibility.
Bowen has been using game theory and economic modeling to identify the best budget strategies for maximizing the public’s welfare for a given amount of spending. She has found that mandatory spending can be more efficient than discretionary programs. She has also explored “dynamic compromise,” finding that it can be easier to reach a deal in larger legislatures. And she has written about “dynamic coalitions,” where lawmakers settle for inferior policies because they put top priority on keeping their coalition together.
The common thread in Bowen’s research involves looking for ways to get the best policy outcome both at the time a law is passed and in the future.
In the new paper, written with Ying Chen of Johns Hopkins University, Hülya Eraslan of Rice University, and Jan Zápal of the Center for Economic Research and Graduate Education–Economics Institute in Prague, Barcelona Graduate School of Economics, and Institute for Economic Analysis of the Spanish Council for Scientific Research, Bowen develops a model based on game theory to show that mandatory spending with a dash of flexibility can reduce both gridlock and political risk — and get better public value for each dollar spent.
The key is to be able to adjust a mandatory spending program so that it continues to meet the same goals even as conditions change.
One simple example from actual experience is the federal government’s program for extended unemployment benefits. Ever since the early 1970s, federal law has allowed workers to collect an extra 13 to 20 weeks of jobless benefits if they live in states where unemployment climbs above a specified level.
This is a “state-contingent” formula for spending, meaning that the spending can go up or down depending on the state of the economy.
“What’s important in the theory is that we’re conditioning the spending on something that’s observable — in this case, the unemployment rate,” says Bowen.
The flexibility is particularly important in an environment of political gridlock. During the most recent recession, when joblessness soared to almost 10 percent and long-term joblessness climbed to the highest level in many decades, Republicans and Democrats fought bitterly and often inconclusively to provide additional help to those hit hardest.
The mandatory part of the program can increase the effectiveness of the spending because it reduces political risk and increases the security of people who either have lost their jobs or worry about losing them. That allows them to spend the money a bit more freely, which in turn helps stimulate a weak economy.
The flexible part of the program, meanwhile, makes it more adaptable to unexpected changes in the economy.
But Bowen says there are other advantages to flexibility; namely, that it can actually reduce political gridlock. That’s because elected lawmakers know that the impact of a program will stay constant even if the party in power changes. If they can reach an initial agreement about the right level of spending for a program, a built-in formula for adjustments can reassure both sides that they aren’t giving away the keys to the store.
Bowen and her colleagues make it clear that the adjustment formula can call for cuts as well as increases in spending. Indeed, the federal program for extended jobless benefits effectively calls for a reduction in spending when a state’s employment rate returns to healthy levels.
“The polarization and the fighting will always be there,” Bowen concedes. “But this can take some of the warring out of decisions. If you do it right today, you don’t have to wrangle about it going forward.”
In the context of American wrangling, Republicans could get some assurance that spending would come down when conditions improve, while Democrats would have assurance that additional benefits would be available if conditions get worse — and even if, as now, they are not the party in power.
Indeed, elected officials have an incentive to agree on a formula that makes sense for the future. Each party knows it may not be in charge after the next election.
Bowen and her colleagues say this is a way of getting “dynamic efficiency,” meaning that it will be on target not only at the time it is created but in subsequent years as well.
Renee Bowen is assistant professor of economics at Stanford Graduate School of Business and the Arch W. Shaw National Fellow at the Hoover Institution.
