A Note on the Deficit—and a Question for Political Scientists or Sociologists

This is being submitted to the Facing Up Blog Carnival on the $9 trillion debt.

One of the economic issues for which journalistic and public discussion generates the lowest ratio of light to heat—it wouldn’t be easy to pick the winners—is the issue of the federal government’s deficit spending. Even worse, as I’ll try to explain below, deficit spending is almost entirely a phony issue.

I’m reminded of this by a recent report that the Democrats in Congress would like to spend more money than President Bush wants. This follows several years’ worth of the Democrats energetically arguing that deficits and debt are bad. A representative statement from them is, “Instead of policies that help America's working families, he [President Bush] continued to tout the economic policies that have led to record deficits and passed on trillions in debt to our children.”

But both major parties have reversed themselves at least once on this issue. During the 60s and 70s, the Democrats happily accepted deficits. Sometimes they cited Nobel laureate Paul Samuelson, who wrote in his best-selling text that—I paraphrase from memory—in a healthy, growing economy, funding worthwhile investments through deficit spending posed no problem at all and could continue without problem indefinitely. Republicans, on the other hand, tended to be horrified—just horrified—by deficits. They were labeled the party of Scrooge and of “green eyeshades”. Once Ronald Reagan was elected, however, the Republican Party did not seem to worry about deficits much, while the Democratic Party began to.

Behind these apparent inconsistencies, there is a simple logic. Broadly speaking, Democrats want to increase spending but don’t want to increase taxes (taxes on the “rich”, on “windfall profits”, and a sin tax or two notwithstanding), and they don’t mind deficits that are created or increased in this manner. Republicans want to cut taxes but don’t want to cut spending (and they are willing to increase spending on “roads to nowhere” and defense) and don’t mind deficits created or increased in that manner. And each party objects to the type of deficit the other party is content with. (And each party pursues a version of “pleasure today, pain tomorrow”, arguably one of the serious defects of our—or any—political system.)

The real issue between our modern parties is, therefore, one of the main issues in our politics since the Founding: the size and scope of government, particularly the federal government. It’s an issue that has been debated especially ferociously since—take your pick—the Progressive Era, the Great Depression, or the Great Society.

Economists think that this is a debate well worth having. While we disagree sharply on the taxing and spending policies that are best, we agree that the proportion of the economy commanded by government has large effects on economic variables, and on our quality of life generally.

But on the issue that seems to command much more attention from journalists and the public, how much of a given amount of government spending is financed through borrowing—how big the yearly deficit is—at least some economists believe it is, at most, of minor importance. We believe this for three reasons.

1. There are serious problems with the way the deficit is usually measured. It’s economically reasonable to adjust for the rate of inflation, for the change in the government’s assets, and for the stage of the business cycle. Most importantly, any changes in the government’s unfunded liabilities—such as Social Security, government pensions, and contingent liabilities—should be included, but in popular discussions, almost never are. See this.
2. Even if it were properly measured, the deficit—and indebtedness more generally—should always be compared—but usually isn’t—to the economy’s ability to pay. Here’s Greg Mankiw (Principles of Macroeconomics, Third Edition, p. 514):

“Critics of budget deficits sometimes assert that the government debt cannot continue to rise forever, but in fact it can. Just as a bank officer evaluating a loan application would compare a person’s debts to his income, we should judge the burden of government debt relative to the size of the nation’s income. Population growth and technological progress cause the total income of the U.S. economy to grow over time. As a result, the nation’s ability to pay the interest on the government debt grows over time as well. As long as the government debt grows more slowly than the nation’s income, there is nothing to prevent the government debt from growing forever.”

Note that this is the same point that Paul Samuelson made in the 1960s.

3. An economic hypothesis, Ricardian Equivalence (named for the economist David Ricardo), posits that how government spending is financed has no real economic effects. An overview of the theory is here . A well-known and heavily-cited review of the theory and evidence for the hypothesis is by my colleague, John Seater, “ Ricardian Equivalence ,” Journal of Economic Literature, 31(1), March 1993. Seater concludes:

“Theoretically, we can be almost certain that Ricardian equivalence is not literally true; it simply requires too many stringent conditions to be believable. Nevertheless, equivalence appears to be a good approximation. Although some of the early empirical literature sent conflicting signals, recent work generally supports Ricardian equivalence. . . . Empirical success and analytical simplicity make Ricardian equivalence an attractive model of government debt’s effects on economic activity.” (p. 184)

A more recent survey by Robert Ricciuti, “ Assessing Ricardian Equivalence,” Journal of Economic Surveys, 17(1), February 2003, concludes virtually the same way:

“Even if it is difficult to believe that there is a one-to-one relationship between tax-cuts and the increase in the size of bequests, optimizing individuals appear to follow Ricardian equivalence, at least in an approximate way. Indeed, debt neutrality is nothing but a consequence of some widely used concepts and methods in modern macroeconomics. We think that more refined econometric work—in particular along the line of stochastic models—may cast more light on the field. Still the onus of the proof lies on those who support debt neutrality.” (p. 74)

Brian Goff and Robert Tollison note that (“Explaining Deficits: 1889—1998,” Economic Inquiry, 40(3), July 2002), Ricardian Equivalence makes popular discussion puzzling. “We do not attempt an explanation of why deficits, which seemingly exhibit no or small real effects (Ricardian equivalence), could lead politicians and voters to treat deficits as if the Ricardian hypothesis were rejected.”

Even a number sometimes touted as more meaningful, the debt-to-GDP ratio, yields the American public and policymakers little useful information. Information at Wikipedia for a large group of countries shows that the U.S.’s current value is not too far from France’s—some Democrats believe that the U.S. should try to do what France does (but not on nuclear power and apparently not on debt)—and those of Germany, Canada, and Portugal. It’s not even too far from the ratio of that supposed paragon of fiscal probity, Switzerland. And it’s rather substantially below Italy’s and Japan’s.

I believe politicians, journalists, and the public should pay attention to specific spending and taxing proposals and to the total amount of government spending, rather than commentating and thinking almost uselessly about the deficit and the debt.

Why they don’t is a question for political scientists or, maybe, sociologists.

UPDATE (11/14 at 3:00 p.m.): An embarrassingly large number of mistakes have been corrected, thanks to careful reader Michael Greenspan.

Comments

[kyle N]

The main difference between Repubs and Dems on the deficit is that the Repubs want to grow out of it. (let it remain a small percentage of GDP by virtue of growth). While the Dems want to inflate it away. (thus helping borrowers at the expense of investors).

I don't really like big deficits but I will take the former over the later any day.

[cactus]

"The real issue between our modern parties is, therefore, one of the main issues in our politics since the Founding: the size and scope of government, particularly the federal government."

Might it be possible to narrow it down further - I think its a question of the scope, not so much the size. Among both Dems and Reps, there were Presidents who saw fit increase the size of the government by various measures, whatever they happened to say in their speeches, and there were some who tried to cut the size of government by various measures. But... Republicans have generally had certain issues where they like to throw money, and Democrats have had a different set of issues where they like to throw money.

"Even if it were properly measured, the deficit—and indebtedness more generally—should always be compared—but usually isn’t—to the economy’s ability to pay."

My preference when looking at growth rates of GDP, whenever possible, is to look at a sort of Net real GDP per capita: real (GDP - change in the debt) per capita. After all, since GDP = C + I + G + NX, you can always pump up GDP by borrowing like mad and pumping up G. The problem is... eventually that has a cost.

kyle n,

"The main difference between Repubs and Dems on the deficit is that the Repubs want to grow out of it. (let it remain a small percentage of GDP by virtue of growth). While the Dems want to inflate it away. (thus helping borrowers at the expense of investors)."

Evidence for that statement? Because I don't see it that way.

[Rich Berger]

Cactus-

I suppose you might prefer your formula if you were a Keynsian, but I don't think the record supports it. For example, we seem to have grown pretty well when the deficits were falling in the mid-80's and the mid-to late 90's, and also since 2003.

[Ironman]

Here's a little different way to look at it - the growth of the national debt in recent years doesn't just track GDP growth, it would also appear to track U.S. population growth as well. We can see this if we factor population into the debt-to-income ratio.

What this shows us is that over time, the primary driver of increases in the national debt is real national crisis - in the chart from the link above showing the Debt per Capita to Income index (DTIP) from 1900 to 2005, we see the largest upward spikes coinciding with World War I, the Great Depression, and especially World War II. Outside these events, we see that the index has tended to hang around an index value of 2. Right now, I believe it's about halfway between 2.1 and 2.2 - pretty much on par with where it's been since 2000, and lower than where it was for much of the 1990s:

http://tinyurl.com/3dkzgj

Tipping my hand for an upcoming post, this pattern goes back much further. Believe it or not, the Civil War spike is nearly identical in magnitude to World War II, but took a lot longer to come down. There are likewise spikes coinciding with the Mexican War and a much larger one for the War of 1812. And even that doesn't compare to the biggest one, if you go back to the nation's earliest days following the Revolutionary War.

Here's where this matters. Outside national crisis, what paces government spending is both economic growth, which drives government revenue, and population growth, which largely drives government spending. While there's nothing that says that the government couldn't maintain the level of debt at a lower level with respect to these figures (say at a DTIP of 1.0 instead of 2.0), maintaining a higher level with wasteful spending would act to help insulate elected officials from the impact of having to raise debt at a rate faster than population and economic growth or taxes to have to deal with lesser national crises.

For lower-level crises, this higher level of debt would allow wasteful spending to be reallocated without having to increase taxes. Or taking on massive amounts of new debt. And, as a bonus in the absence of such crises, politicians can use the excess available funds to support their re-election efforts by rewarding their campaign contributors or buying votes through pork barrel projects in their districts.

The downside is that when a lower-level crisis hits the national stage, a lot of politicians resent having to deal with it since it prevents them from fully pursuing their preferred venues for wasteful spending. Especially if they might otherwise have a physical structure named after themselves or get credit for some 'improvement.' And this is why we hear about deficits so often.

[cactus]

Rich Berger,

I think growth that isn't paid for by government borrowing is better than growth that is. (I.e., a growth rate of X is better if it occurs in a time of falling deficits than in a time of increasing deficits.)

[kyle N]

Cactus, Republicans supported high growth/ low inflation policies since the early 1980's. While the democrats have always supported deficit spending and loose money policies of some sort.
Clinton was a bit of a deviation on this, but only because he had a republican congress most of the time.

In more recent years there was a large increase in deficit spending, but that has to some extent been reigned in. Partly due to growth. Now the Democrats running for president are selling the same old high tax snake oil. Especially Edwards and Obama. They feign concern about the deficit, but their spending schemes are so fantastic there is no reason to believe them. Hillary is a bit more sane, but if either of those two yahoos got into office with both houses of congress it will be Jimmy Carter, and seventies style stagflation all over again.

[Chris]

Cactus, if you're interested in the amount available for people to consume, you can just look at the growth in real consumption or maybe Y - G. Taking GDP minus deficits makes no sense in this context. C or Y - G would be different indicators of people's ability to consume since, well, it measures their actual consumption. I don't understand why one would rather look at Y - (G + Transfers - Taxes). If the government takes resources out of the private economy, it raises taxes or issues debt. Either way the same resources come out of the private economy; the opportunity cost of things that go boom or bridges to nowhere will be the same.

In order to look at the government's solvency, the change in the public debt-GDP ratio is the right thing to look at. That hasn't changed much over the past year to year and a half (about 40% if my recollections are right), indicating that current policy, if things don't change much, is almost sustainable. It's a purely backward-looking measure of the government's finances, but that's much easier than trying to make forecasts about fiscal policy 40 years out. I find it a little hard to believe that the present value of future deficits is infinite, which is what most of these attempts to estimate "unfunded liabilities" end up finding. How those "unfunded liabilities" actually do get funded, now THAT scares me.

[F. Fain]

You can slice and dice it anyway you choose, but it sure looks like we're a society living beyond its means. We don't make anything here anymore. How long can we continue to consume more than we produce? Isn't one of the first laws of economics that there is no free lunch?

[cactus]

Chris,

I'm thinking Ricardian equivalence. People know they have to pay it back at some point. As you said, it comes out of the private sector sometime.

kyle N,

Carter's policies may not have been very good, but he also ended his term with smaller deficits than he inherited. (Follow the link I provided, and note that no Rep president has done that since Ike.) And Newt wasn't there to help. So somehow or other, something the two most recent Dem Presidents pulled off but Rep Presidents don't is Republican policy? How does that work?

[Chris]

Exactly. The Bob Rubin / Clinton types look at debt issuance as resources taken from the private economy but forget that taxes do the exact same thing. The Bushies get it wrong but in the opposite direction. Ricardian equivalence says that these things should wash out in an accounting sense; there are still the incentive effects and whatnot which work a little bit on labor supply or capital demand in the longer run. But we can talk about that stuff another time. Barro's point was that both the old Keynesians and the Hoover-wannabes got the accounting wrong in an aggregate sense. Deficits have no direct effect on the wealth of the private sector. The first-order issue is G, not the financing of it. Apart from defense, most of G is a state issue anyway.

I'll agree that Carter gets a bum rap. He deserves some of the blame for gas lines and windfall profit taxes, letting Miller anywhere near the Fed, distortionarily high marginal tax rates, and a dour pessimistic personality. He deserves a lot of credit for sane fiscal policy in the aggregate, a fair amount of regulatory reform, and appointing Volcker in the long run. He's certainly better than Nixon and LBJ on these fronts.

This has turned into a cool discussion. I've given up commenting on some other sites because they turn into instant flamewars. It does kind of puzzle me how discussions about fiscal policy in particular provoke this kind of reaction. It's almost like people want to pick teams and fight for the sake of picking teams and fighting. The difference between a debt-GDP ratio of 30% vs. 40% is just not that interesting economically.