Harvard economists Ken Rogoff and Carmen Reinhart came under heavy fire last week after a study in which they claimed economic growth plunges once a country's debt exceeds 90% of gross domestic product was found littered with math errors.
It turns out the study may still have been flawed even if the math were correct, because GDP is an imprecise moving target. And it's about to move even more.
The Bureau of Economic Analysis announced a series of changes to the way it calculates GDP that will "boost" the size of the U.S. economy by about 3%, or $400 billion.
Who knew growth was that easy? "How will YOU spend your Statistical Revision Income?" quipped economist William Easterly.
In short, research and development will now be counted as a capital investment, rather than a cost of making goods. Investment in "artistic originals" will be counted as a fixed investment. And the amount of money pension plans promise to pay retirees will be counted as wages, not just the amount of cash companies pay into defined benefit plans. (For more detailed explanations, see here).
Most of the overall change to GDP comes from the R&D accounting shift. The rest of the changes will each have a negligible impact.
"We are carrying these major changes all the way back in time -- which for us means to 1929 -- so we are essentially rewriting economic history," Brent Moulton of the BEA told The Financial Times. This is why they call it a soft science.
None of this is new. As the economy changes, the BEA updates how it calculates the size of the economy from time to time. It added investment in computer software to its calculations in 1999. One of the main reasons for the changes is to adopt account policies consistent across the globe to make comparing different economies apples to apples. "Most other developed economies, including those of Europe, will have incorporated most of the major changes ... into their economic accounts by 2014," it writes.
To me, there are three takeaways.
One, every calculation that uses GDP as a denominator will now change. Debt-to-GDP, profits-to-GDP, government spending-to-GDP ... all of those figures will be adjusted down. The 3%-bump to GDP is large enough that it will impact the validity of some arguments. Are profits as a share of GDP really at an all-time high? Maybe not after these changes.
Two, there will be a gaggle of critics who say these changes are being made for political purposes. That's nonsense -- the changed methodologies were first presented in a 2008 paper, before the current administration was elected. The statisticians calculating these figures have a grossly difficult and imprecise job. What would be shameful is realizing the calculations are wrong and not doing anything about it in the name of "consistency."
Third, this is yet another reminder of how fallible our economic data is. Data can help push you toward one direction, but -- especially in economics -- it shouldn't ever give the false sense of watertight truth.
