Deficit Projections Are Often Way Off the Mark

By Polina Vlasenko, PhD

The fairly modest federal deficit projected for the next few years has made fiscal discipline a less urgent policy concern than it was only a couple of years ago. But can we believe the projections? Does the fact that the deficit is projected to remain fairly small mean that the deficit is likely to remain fairly small in reality?

History indicates there is often a wide gulf between the deficit projections offered by the U.S. Office of Management and Budget, and what actually ends up happening. And that can have major implications when it comes to policy.

The deficit has shrunk considerably, and is expected to remain tame, according to recent federal budget deficit data, and projections released with the 2016 budget (see chart 1). In fiscal year 2014 the federal budget deficit was 2.8 percent of GDP. And it is projected to stay around 2.5 percent of GDP through 2020. This is close to the average deficit of 2.6 percent of GDP since 1960, and much lower than the deficit was only a few years ago, when it exceeded 8 percent of GDP.

Source: Fiscal Year 2016 Budget of the U.S. Government, Office of Management and Budget

Comparing past projections to what actually materialized, however, shows that they are rarely accurate.

Chart 2 compares the federal budget deficit projections for 2005-2010 (released in early 2005) to the actual outcomes. Huge differences are apparent, especially in 2008 and 2009. The Great Recession vastly deteriorated the federal deficit, but projections did not expect the recession (they never do).

Source: Fiscal Year 2006 Budget of the U.S. Government, Office of Management and Budget

Without a major disruption of the Great Recession, the projections for the next five-year period, 2010-2015, do not look quite so flawed (see Chart 3). But they are still far from accurate. The actual deficit differs from the projected one by as much as two percentage points of GDP, which is a substantial deviation.

In the world of fiscal policy, the difference between, for example, a deficit of 5 percent of GDP and a deficit of 7 percent of GDP (which is what the projected and actual deficit was in 2012 for the projection released in 2010) is not trivial. At a 7-percent-of-GDP deficit, fiscal discipline would command much higher urgency in policy circles.

Moreover, the projections appear to imply that the deficit would converge to a long-term trend of about 4 percent of GDP, a substantially larger deficit than what we saw in 2014 and what is now believed to be the long-term trend of the deficit. The long-term fiscal picture looks much less rosy in a world where the deficit stays at around 4 percent of GDP (assumed in the projections made in 2010) than in a world where the deficit stays at about 2.5 percent of GDP (assumed in the projections released earlier this year).

Source: Fiscal Year 2011 Budget of the U.S. Government, Office of Management and Budget

The poor accuracy of deficit projections arises from the fact that the economy is too complex to forecast its trajectory several years ahead with any degree of accuracy. Deficit projections, as so many other macroeconomic forecasts, tend to produce a trajectory where the economy (and the deficit) converges to what the forecasters believe to be some long-term equilibrium value. Our ability to forecast disruptions (recession, crises, and the like) is very limited – these things almost never show up in any projections, but they do show up in life.

Now let’s look at the deficit currently projected for 2015-2020 one more time (chart 4).

Source: Fiscal Year 2016 Budget of the U.S. Government, Office of Management and Budget

Given how inaccurate past projections have been, how confident should we be in today’s projections?

In the past 10 years, the actual deficit differed from the projections by an average of 2.3 percentage points of GDP. The current projection for 2015-2020 is for a federal deficit between 2.3 and 3.2 percent of GDP. With an average error of 2.3 percentage points, this means that the actual outcome is likely to be anywhere between a balanced budget and deficit of 5.5 percent of GDP. This is a very wide range of possibilities. Moreover, it is clear that no recession is assumed in these projections. How confident can we be that there will be no recession in the next five years?

Budgetary projections appear to be a poor guide to the future budget deficit. Implicitly assuming that we should not worry about the budget deficit because projections show it to be shrinking puts way too much faith in the abilities of those who create the projections. A better approach would be to carefully plan the fiscal affairs of government, including planning for the possibility of adverse effects not included in the projections, and to do so at all times, not only when deficit projections are dire.


Sign up here to be notified of new articles from Polina Vlasenko, PhD and AIER.