I am not saying this, the BEA states that on their website but they went ahead and extrapolated a large fall in GDP for second quarter four months into the future. Below are the details.
The way second quarter GDP was reported has left the impression in the minds of many people not familiar with the math that the economy shrank 32.9% in the second quarter. I saw numerous tweets claiming that by individuals but also by mainstream financial media.
Although the BEA release was technically correct, many focused on the -32.9% and took it out of context.
This part of the release
Real gross domestic product (GDP) decreased at an annual rate of 32.9 percent in the second quarter of 2020…
quickly turned into this
Real gross domestic product (GDP) decreased at rate of 32.9 percent in the second quarter of 2020…
maybe because in the minds of many people “annual” is an irrelevant qualifier.
But it is very relevant. In my humble opinion, the BEA should not have reported GDP as an annual rate, not because I think so, but because they state so in their website for cases when series are volatile:
For some volatile quarterly series, such as corporate profits, BEA publishes percent changes at quarterly rates rather than at annual rates, because the comparability issue mentioned above is less important and because annual rates of change may be misleading.
This is in fact too technical; what does it mean “because annual rates of change may be misleading” and why didn’t they follow their own rules in the case of GDP?
Let us first look at the actual numbers in my Tweet below:
Annual rate: (((GDP quarter 1/GDP quarter 2)^4)-1)*100
GDP quarter 1 = 19010
GDP quarter 2 = 17205
Annual rate (plug in above) = -32.9%
but change = -9.5% (GDP 2/GDP 1 -1) *100@pinebrookcap to get him bored.
— Michael Harris 📈 (@mikeharrisNY) July 30, 2020
Therefore, the drop in GDP from the previous quarter was 9.5%. The annual rate is a practice of BEA as explained in their website:
For ease of comparison, BEA publishes percent changes in most quarterly estimates at annual rates. … A quarterly percent change at an annual rate shows what the percent change would be if the quarterly rate continued for four quarters. It is computed by compounding the quarterly rate for four quarters.
So basically, as noted above, the -32.9% assumes a decline of -9.5% for four quarters in a row, or equivalent. The following questions arise:
- Is the above a reasonable assumption (-9.5% for four quarters or equivalent)?
- Why projecting into the future GDP numbers under current environment?
- Why BEA did not follow their guidelines of reporting only the quarterly change due to the high volatility?
The public is not supposed to understand these technicalities. The impression that is left now to many is that the economy shrank 32.9%, instead of -9.5% in the second quarter. A Google search will convince you this is the case. Part of the media ceased the opportunity to engage in sensational reporting. Apart from any political preferences, selective reporting of facts about the economy hurts the image of the country not only internally but also externally.
Reference. BEA website