Skip to main content

A simple Excel calculation error puts a famous economic study under scrutiny

A simple Excel calculation error puts a famous economic study under scrutiny

Share this story

Excel 2013 stock
Excel 2013 stock

Back in 2010, economists Carmen Reinhart and Ken Rogoff published a report that stated countries with a debt-to-GDP (gross domestic product) ratio greater than 90 percent would find it significantly more difficult to grow their economy. Their work was widely cited, and was used as a major part of Paul Ryan's "Path to Prosperity" budget, but a new study shows that its data is flawed thanks to a mistake many office-workers are familiar with — an incorrect Excel formula. The study from three University of Massachusetts researchers claims a number of flaws in Reinhart and Rogoff's earlier work, one of which is an Excel coding error that omits data from five countries in the analysis (Next New Deal shows an example of the mistake in its analysis of the UMass study).

Reinhart and Rogoff initially defended their study in the days immediately following the release of the UMass report, but today a response posted on the Financial Times shows them admitting the Excel error was an accidental oversight that does affect their study's results. "On the first point, we reiterate that Herndon, Ash and Pollin [the three UMass researchers] accurately point out the coding error that omits several countries from the averages in figure 2. Full stop. HAP are on point," write Reinhart and Rogoff. They went on to say that the mistake "leads to a notable change in the average growth rate for the over 90 percent debt group" — the group that politicians and economists have often cited as being in danger for difficult economic growth thanks to this study.

Check your calculations before you hand in your work

The UMass study found other potential flaws in the 2010 report including unconventional statistical weighting and selective data omission, but thus far Reinhart and Rogoff are largely standing by their work. While they did admit the Excel coding error, their statement said that they "object to in the strongest terms" the accusation of selective data omission. They also claim their weighting approach "has been followed in many other settings." All in all, Reinhart and Rogoff stand by their work despite the mistake they made: "Do Herndon et al. get dramatically different results on the relatively short post war sample they focus on? Not really. They, too, find lower growth associated with periods when debt is over 90 percent." Whether or not the study is still valid will be up to economists to debate, but the mistake certainly reinforces the importance of checking your work.