Excel Spreadsheet Error, Ha Ha! Lessons from the Reinhart-Rogoff Controversy
At this point everyone who follows economic policy debates knows about the famous Reinhart-Rogoff spreadsheet error uncovered by a University of Massachusetts graduate student. When the error is corrected, there is nothing resembling the growth falloff cliff associated with a 90 percent debt-to-GDP ratio that had been the main takeaway from the initial paper.
There has been an interesting response from the mainstream of the profession. On the one hand, they have been quick to rebuke those on their political left for making a big deal out of a silly mistake (here, here and here). They have also assured everyone that it really doesn’t matter anyhow since no one actually used the Reinhart-Rogoff work as a basis for policy. Both points raise interesting issues.
Of course the fact that two well-known Harvard economics professors made a silly spreadsheet error should not be a big deal. However the beauty of a spreadsheet error is that it is something that everyone can understand.
We all know what it means to enter the wrong number or add the wrong columns. That doesn’t require advanced training in economics.
The silly spreadsheet error was important in the debt debate controversy because it allowed for a real debate. Ordinarily Harvard economists don’t engage their less credentialed colleagues at places like the University of Massachusetts. (Hey, they never even responded to my e-mails requesting their data.)
Unfortunately, even the best reporters at the most prestigious news outlets rarely feel sufficiently knowledgeable to challenge pronouncements from prominent economists. This means that the profession must rely on internal policing to weed out bad arguments. The Reinhart-Rogoff 90 percent cliff was widely accepted policy wisdom for more than three years, which suggests the internal policing in the economics profession is pretty damn weak.
Of course the more fundamental point that came up in the wake of Excelgate is that Reinhart-Rogoff show nothing about causation. Efforts to examine the direction of causation show that it goes almost entirely from slow growth to high debt (here and here) not from debt to slow growth as is generally implied in the policy debate.
And, those who ever learned accounting would recall that debt is only half of a balance sheet. We would have to consider assets also if we really wanted to tell a story about how debt could impact growth.
These points were made to a large swath of the public not because of the internal policing of the economics profession, but because of an Excel spreadsheet error. For this reason, those who care about honest academic and policy debates should be celebrating the spreadsheet error. If this embarrasses important people in the economics profession, that’s because it is a profession that deserves to be embarrassed.
The other side of the argument from the mainstream of the profession, that Reinhart-Rogoff was not actually the basis for policy, also deserves ridicule. The essence of their case is that those pushing austerity would have done so whether or not they had the Reinhart-Rogoff studies. In other words, politicians wanted to push cuts in government spending because they wanted to push cuts in government spending, not because Reinhart-Rogoff’s paper convinced them that these cuts were necessary to support growth.
The fact that politicians respond to political pressures, not academic arguments is undoubtedly true. That is how politicians get elected.
This is why everyone’s bull***t detectors should be blasting off the charts every time they see a story about how a politician is acting based on their political philosophy or ideology. Politicians are acting based on the demands of their political supporters. They don’t get elected based on their beautiful political philosophies.
When the mainstream economists tell us that the pushers of austerity would have done so even without Reinhart-Rogoff, this is not news. However, the Reinhart-Rogoff story was hugely important in selling the austerity case to the larger public. It is much easier for politicians to say that we have to cut Social Security for widows and Head Start for children in order to avoid two decades of stagnation, than for them to say that these cuts are necessary in order to ensure that their campaign contributors don’t have to pay more in taxes.
And the Reinhart-Rogoff story was used with great success towards this end. In fact, the 90 percent debt-to-GDP threshold became a fixture in the Washington budget debate after it was included in the report by the co-chairs of President Obama’s deficit commission, former Senator Alan Simpson and Morgan Stanley Director Erskine Bowles.
In short, the story of the Reinhart-Rogoff error tells us a great deal about how the elites use economists and the prestige of the economics profession in order to impose their will on the public. The internal policing of the profession is essentially non-existent and even the best reporters do not feel competent to challenge the claims of top economists.
That is not a pretty picture of the state of economic policy debates, but the first step toward making it better is recognizing where things stand now.
Dean Baker is a macroeconomist and co-director of the Center for Economic and Policy Research in Washington, DC. He previously worked as a senior economist at the Economic Policy Institute and an assistant professor at Bucknell University.