Carmen Reinhart and Kenneth Rogoff are two economists from Harvard University who published a major paper called Growth in a Time of Debt in 2010. This paper, tending to prove that a high level of debt was causing less economic growth, was widely cited to justify budget cuts. Problem: there was a formula mistake in the Excel spreadsheet the two economists used...
Reinhart and Rogoff analyzed historical data from 1945 to 2009, and the correlation between debt and economic growth. They wrote in 2010 that countries with debt above 90% of their GDP experienced on average a 0.1 percent negative growth of their GDP.
But three economists from UMass, Herndon, Ash and Pollin, claimed that the paper was flawed and that there were some significant mistakes in the Excel spreadsheet used to reach that conclusion. In fact, the UMass economists claimed that the GDP still grows significantly when government debt is high. They actually found, using a different method, that the economic growth would be around 2.2 percent. Reinhart and Rogoff admitted that their spreadsheet was accidentally omitting 5 rows in an Average formula, and claimed that the corrected result would be a positive growth but of just 0.2 percent.
Reinhart and Rogoff considered that the dispute was beside the point, which was that in any case, economic growth was significantly slower in periods of high debts than in periods of low debts. Yet, correlation is not causation and one should not necessarily conclude that high debt slows economic growth. It could indeed also mean that slow growth causes higher debt.
In any case, the problem is that their research was cited a number of times by militants of budget cuts and a strict reduction of government debt, even US House Budget Committee Chairman Paul Ryan.
So what does this story tells us? Aside from the economists debate on whether or not government should cut spending, this shows that the simplest Excel formula can weaken your entire argumentation. Once a mistake has been found, it does not matter anymore if your argumentation is correct or false, your credibility is shaken and people will stop trusting your conclusions.
For Excel users, this should be a good story to keep in mind before sharing results of an Excel model. Always bulletproof your calculations to protect your credibility and reputation!