Measuring GDP Growth

Only rarely do I think that Kevin Drum would benefit from having studied economics more in college. He almost always seems to understand economics issues better than most professional economists.

Therefore I jump on an error, which, I suspect must be very widespread if Drum made it.

He wrote

the contribution of the airline industry to GDP probably hasn’t changed an awful lot since 1973. But its contribution to the increased wellbeing of the median person has increased tremendously. In 1973, the average schmoe simply never flew: it was too expensive and flights weren’t always very convenient. Today, nearly anyone can afford to fly at least occasionally. So this is one area where pure measures of GDP probably understate the benefit to the median person.

Actually no. It is one area where the pure measure of GDP isn’t quite that simple.
When we try to use the growth of GDP per capita to get some hint of benefit to the median person, we look at real GDP. Drum’s guess is that the huge increase in the volume of services provided by the airline industry and the huge increase in the price multiply out to agrowth in dollars of value added similar to the growth of overall nominal GDP. For the sake of argument, I will pretend that the share of airline value added in GDP is exactly the same now as it was in 1973.

This absolutely does not imply that the contribution of the airline industry to measured real “pure GDP” growth is average. That’s not how it’s done. The BEA calculates the growth nominal value added by sector. It calculates the change in a value added deflator by sector (often it’s pretty much made up). It calculates the growth of real value added by sector. Then it takes an average weighted by the shares of the sectors in GDP. Rapid growth of volume of a sector and rapid decline in prices implies rapid measured growth of real value added which contributes a lot to measured real GDP growth. Nominal value added in a sector can shrink compared to GDP and yet the sector can contribute more than the average sector to GDP growth.

Another way of looking at it is Real GDP is nominal GDP divided by a price index — the gDP deflator. The GDP deflator is a weighted average of lots of prices, including E1AIT1 (*.xls warning)

Air transportation E1AIT1 PPI Domestic scheduled passenger air transportation

The airline industry may have made an average contribution to the growth of nominal GDP, but the low increase (or decrease) of E1AIT1 caused the growth of the estimated GDP deflator growth to be low and estimated read GDP growth to be high.

Real pure GDP estimates ignore practically everything but sure don’t ignore prices.