“Barack Obama said the deficit has fallen at the fastest rate in 60 years. … the president used an acceptable approach and his numbers are accurate. There are no statistical tricks in play.”
Eyes wide open, PolitiFact falls for the oldest statistical tricks in the book.
President Obama made the speech containing his reference to the shrinking deficit on July 24, 2013 in Galesburg, Ill., addressing the economy. The line comes in a string of statements following the president’s assurance that America, under his leadership, has things going the right direction:
Now, today, five years after the start of that Great Recession, America has fought its way back. We fought our way back. Together, we saved the auto industry; took on a broken health care system.
… And our deficits are falling at the fastest rate in 60 years.
PolitiFact ruled the last line “True”:
Barack Obama said the deficit has fallen at the fastest rate in 60 years. While economists vary on how to best measure that decline, the president used an acceptable approach and his numbers are accurate. There are no statistical tricks in play.
Our experts warned against reading too much into Obama’s claim. While Obama did not spend much time one way or the other on the significance of the decline in the deficit, he didn’t mention ongoing concerns over the national debt.
Factcheck.org also weighed in on Mr. Obama’s claim.
The White House Office of Management and Budget provides tables showing the raw numbers.
Analyzing the Rhetoric
The key to PolitiFact’s opinion that Mr. Obama employed no statistical tricks rests on its finding that the White House chose a reasonable means to measure the drop in the deficit.
Reasonable means to an end?
PolitiFact reports the method the administration used to back its claim:
The White House press office told us that the proof is in the numbers. They tracked the deficit as a percent of the entire economy, measured as the Gross Domestic Product.
In 2009, the first year of Obama’s presidency, after tax cuts and new spending, the deficit was 10.1 percent of GDP. In 2012, the deficit declined to 7 percent of GDP. So that’s a decline of 3.1 percentage points.
We have no problem with using the drop in percentage points of GDP as a measurement of a falling deficit. It’s to some extent a relative measurement and appropriate for apples-to-apples comparisons.
But this comparison will never achieve apples-to-apples status.
Let’s first dispense with the ridiculous fig leaf PolitiFact uses to excuse the administration’s claim:
Peuquet and others raised a second statistical issue — choosing the endpoints for the comparison. It’s a common device to pick the range that best proves your point.
But in the case of deficit reduction, no one quibbled with the idea that longer periods are more revealing, and Obama used all the years he has available. His first year in office was 2009 and the numbers for 2013 are not yet in.
Mr. Obama “used all the years he has available”? The claim makes no sense. It’s perfectly legitimate to compare how one president reduces deficit numbers compared to those of a predecessor. PolitiFact claims 2009 in Obama’s name even though fiscal year 2009 goes on the Bush administration’s ledger. If 2009 is available to Obama then so is 2008. But using the deficit spike from 2009 is the key to obtaining the 3.1 percentage point reduction figure.
“Longer periods are more revealing”
Following PolitiFact’s reasoning about statistical windows, using the numbers from 2008 as a baseline instead of 2009 should reveal more. It’s a longer period by a year. We see the deficit jump from 3.1 percent of GDP in 2008 to a figure of 7 percent in 2012, an increase of 3.9 percentage points.
PolitiFact is feeding its readers statistical mumbo-jumbo. Using endpoints that represent a longer duration may prove more revealing. But not when the comparison is apples and oranges or cherry picked. Using just the endpoints leaves out the middle of the story.
The importance of deficit reduction
It’s normal for governments to run a deficit. But there are two commonly accepted strong reasons for reducing the deficit.
- The deficit is too high
- The debt is too high
When is the deficit too high? Perhaps there is no objective point at which a deficit is too high. Much depends on the demands of the day, such as whether a nation is at war and needs to spend to ensure its own continued existence. The European Union has economic guidelines for its member nations that we can use as a point of comparison. Member nations are expected to keep deficits under 3 percent of GDP and are required to act to decrease the deficits exceeding that level or may face disciplinary action.
In the past 66 years since 1947, the U.S. budget deficit has been under 3 percent of GDP 46 times. Over two-thirds of the time during that period, by the EU measure, the United States has had no strong reason for cutting the deficit.
When is the debt too high? Again, we can use the EU’s rules as an estimate. The EU sets its warning limit at 60 percent of GDP. From 1953 through 2009, the U.S. federal debt as held by the public has run below 60 percent of GDP. From the postwar period through 2009, the debt has not been a driving motivation for decreasing the deficit, again using the EU’s measure. The debt exceeded the 60 percent threshold in 2010 and has increased every year since.
President Obama’s talking point serves as positive spin on the fact that, under his administration, we’ve had both the deficit and the debt in dangerous territory simultaneously for the first time since the early 1950s. The United States, using the EU standard, has not had a strong statistical motivation for lowering the deficit and the debt since the 1950s. Comparing deficit reduction achievements since 2009 to deficit reduction efforts after the World War II debt dropped below 60 percent of GDP amounts to an apples-to-oranges comparison.
More flies in the ointment
As we discussed at length in a fact check from May of this year, the 2009 budget year was quite unusual. We estimate that no less than $250 billion in FY2009 spending came in the form of loans to banks and automakers. The banks and the automakers paid back the bulk of that money, and government accounting rules count the repaid money as “negative spending.”
The deficit spending in 2009 makes the deficit look smaller in subsequent years as the government recovers its loans. As a result, the graph resulting from the OMB’s deficit figures carries over a distortion from the effects of negative spending. In FY2012 alone, ProPublica reports government-sponsored institutions Fannie Mae and Freddie Mac returned $18.4 billion to the government.
On top of the distortion from negative spending we can add the distortion from the economic stimulus package the president signed in 2009. The CBO says the bill boosted government spending by $494 billion from 2009-2011, with $114 billion of that spent in 2009.
The 2009 deficit spike occurred from revenue lost to the recession combined with bailout loans, stimulus spending, a high cost of living adjustment and increased assistance to the unemployed. Spending in each category eases naturally as the economy improves, and the payback of bailout loans helps mask later spending. Using 2009 as the baseline year when comparing Obama’s spending reductions thus qualifies as obvious cherry picking once we look at the unusual factors in play.
“Barack Obama said the deficit has fallen at the fastest rate in 60 years.”
It’s true that Obama said it. We’ve fact-checked the claim separately.
“The president used an acceptable approach and his numbers are accurate.”
The approach to the numbers we find acceptable (FactCheck.org disagrees), but the budget numbers don’t tell the whole story. The president is misleading his audience with an incomplete picture and a cherry-picked statistic.
“There are no statistical tricks in play.”
The Texas Sharpshooter fallacy and the apples-to-oranges comparison count as statistical tricks. Both are in play.
Clarification Aug. 12, 2013: In the “Reasonable means to an end?” section, we used the phrase “the 3.1 percent reduction figure.” We meant to refer to the 3.1 percentage point reduction figure. The line has been changed accordingly.
Greenberg, Jon. “Obama Says Deficit Is Falling at the Fastest Rate in 60 Years.” PolitiFact. Tampa Bay Times, 25 July 2013. Web. 11 Aug. 2013.
Obama, Barack. “Remarks by the President on the Economy — Knox College, Galesburg, IL.” The White House. The White House, 24 July 2013. Web. 11 Aug. 2013.
“Stability and Growth Pact.” European Commission. European Commission, n.d. Web. 11 Aug. 2013.
“Historical Tables.” The White House Office of Management and Budget. The White House, n.d. Web. 11 Aug. 2013.
“Table 1.3—Summary of Receipts, Outlays, and Surpluses or Deficits (-) in Current Dollars, Constant (FY 2005) Dollars, and as Percentages of GDP: 1940–2018.” The White House Office of Management and Budget. The White House, n.d. Web. 11 Aug. 2013.
“Table 7.1—Federal Debt at the End of Year: 1940–2018.” The White House Office of Management and Budget. The White House, n.d. Web. 11 Aug. 2013.
White, Bryan W. “FactCheck.org Says Federal Spending Has Increased “far More Slowly” under Obama than under Bush.” Zebra Fact Check. Zebra Fact Check, 7 May 2013. Web. 11 Aug. 2013.
“Bailout Tracker Tracking Every Dollar and Every Recipient: Fannie Mae.” ProPublica. Pro Publica Inc., n.d. Web. 11 Aug. 2013.
“Bailout Tracker Tracking Every Dollar and Every Recipient: Freddie Mac.” ProPublica. Pro Publica Inc., n.d. Web. 11 Aug. 2013.
Elmendorf, Douglas W., and N. Gregory Mankiw. “Government debt.” Handbook of macroeconomics 1 (1999): 1615-1669. Web. 11 Aug. 2013.
Checherita, Cristina, and Philipp Rother. “The impact of high and growing government debt on economic growth.” An empirical investigation for the Euro Area. Frankfurt: European Central Bank Working Paper Series 1237 (2010) Web. 10 Aug. 2013.
Rankin, Neil, and Barbara Roffia. “Maximum sustainable government debt in the overlapping generations model.” The Manchester School 71.3 (2003): 217-241. Web. 10 Aug. 2013
Roubini, Nouriel. “Debt sustainability: How to assess whether a country is insolvent.” Stern School of Business, New York University, mimeo (2001). Web. 10 Aug. 2013.