Excel & data
One Formula Range, Big Consequences: The Reinhart-Rogoff Spreadsheet Error
A formula that skipped a few rows in an influential economics paper became a famous case study in spreadsheet risk.
The case
In April 2013, three researchers at the University of Massachusetts Amherst, Thomas Herndon, Michael Ash and Robert Pollin, published a critique of a widely cited economics paper by Carmen Reinhart and Kenneth Rogoff on public debt and economic growth. Working from the original authors' spreadsheet, they found that one average formula covered rows 30 to 44 where it should have covered rows 30 to 49, leaving five countries out of that calculation.
What the error changed, and what it did not
The critics reported that this formula problem alone shifted the published average growth figure for the highest-debt group by about 0.3 percentage points. They also pointed to other issues, including the selective exclusion of some data and an unusual weighting method. With their corrections, they reported average growth of about 2.2 percent for countries with public debt above 90 percent of GDP, compared with the originally published figure of minus 0.1 percent. Reinhart and Rogoff responded that they had described an association rather than a cause, and that lower growth at high debt levels remained visible in the corrected results.
The practical lesson
The point is not who was right about economics. It is how easily a range error hides in a large sheet. A formula can look correct, return a plausible number, and still leave out data. Three habits reduce the risk: use structured tables so that formulas expand with the data, have a second person review key formulas, and keep a separate check, such as a count of rows included, next to every important total.
In our own work
When MayTech Labs builds an Excel tool, key totals sit next to their own checks, and the workbook is built on named tables rather than hand-drawn ranges. It is a small discipline that prevents a large class of mistakes.
