A seminal paper from Harvard has been found to contain major errors. This paper: Growth in a Time of Debt, authored by Reinhart and Rogoff, has been used as evidence by politicians to support decisions in making austerity cuts during the economic crisis. Decisions which have been made based on false information.
To add insult to injury, the mistakes in the paper were not due to computer error and nor were they intentional. Rather it was down to plain and simple clumsiness – while using excel and highlighting cells for a formula the left mouse button was released a second too soon; yet somehow this went unnoticed. Or it did go by unnoticed until Thomas Herndon chose to use the paper for his homework assignment. The assignment was to choose a study and replicate the findings. At first Thomas thought he had made the mistake, as did his professors, but no matter how many times he tried he could not replicate the results. This proved to be an enigma to Thomas’s professors as well, who eventually advised that he contact the Harvard professors – Reinhart and Rogoff. After much correspondence the Harvard professors sent Thomas the spread sheets containing the raw data so Thomas could see for himself. When Thomas received the spreadsheets what lay before him was astonishing – 5/20 countries, Australia, Austria, Belgium, Canada and Denmark, under analysis had been missed off the key calculations! This was to analyse the average growth of Gross Domestic Product, otherwise known as the monetary value of a countries finished goods and services produced within a given time, in countries with high public debt. Also in other parts of the study, Australia, Canada and New Zealand were missing parts of their data altogether!
With the calculations and statistics corrected and with the data complete the study has been re-published by Herndon and his professors. The main difference between this paper and its predecessor is that the correlations and associations are not as strong as they first seemed to be. Daniel Hamermesh, professor of economics at Royal Holloway, University of London, comments “I don’t think jobs were destroyed because of this but it provides an intellectual rationalisation for things that affect how people think about the world. And how people think about the world, especially politicians, eventually affects how the world works.”
To finish off, the moral of all this, if you haven’t already guessed, is be careful and vigilant when using excel and working with data – as students this is an important message for us all.