Sign up for email updates!

Don't miss out on what matters. Sign up for email updates!

Stay informed! Sign up for e-mail updates:

Monday, September 10, 2018

BEA Needs to Explain Its Growth Number

Updated to highlight the difference between annual and annualized growth. As my article explains, the former is the actual year-to-year growth rate; the latter, which is what the BEA press release is about, is not a year-to-year growth rate. 

As a political economist, having spent all my professional career working with macroeconomics and economic policy, I rely heavily on national accounts data from the world's top statistics agencies. One of them is the Bureau of Economic Analysis (BEA) of which I only have good things to say. 

Therefore, I was surprised to find that on August 29, the BEA reported:
Real gross domestic product (GDP) increased at an annual rate of 4.2 percent in the second quarter of 2018 (table 1), according to the "second" estimate released by the Bureau of Economic Analysis. In the first quarter, real GDP increased 2.2 percent.
The problem is that I fail to find support for this number in the BEA's own GDP database. 

The report refers to the BEA's Domestic Product and Income data, Table 1.1.1, where the BEA reports seasonally adjusted, real annual growth rates, but from preceding period. Since the table provides quarterly GDP data, this means the period prior to the latest reporting period - in other words the 4.2-percent growth rate is from Q1 2018 to Q2 2018. 

Actual annual growth rates can also be reported quarterly, but that means comparing GDP for the same quarter. This growth rate is available in the BEA's Table 1.1.6, according to which GDP in Q2 in 2018 was $18,514.6 billion. When compared to to $17,995.2 billion in Q2 in 2017, we get an annual, inflation-adjusted growth rate of 2.89 percent. 

I have reported annual growth rates by quarter in an earlier article. There, I explained that the annual growth rate is steadily increasing, and has been doing so since Obama's last months in office. However, again, the latest number is 2.89 percent, not 4.2 percent. 

Other tables validate my concerns regarding the BEA's press release:

  • Table 1.1.3, which reports GDP growth by quantity indexes, has growth at 2.88 percent, year to year for Q2; the quantity index for Q2 2017 was 111.1 and 114.3 for Q2 2018;
  • Table 1.1.5, which reports GDP growth in current prices, puts year to year growth for Q2 at 5.44 percent, so there has not been a mistake in confusing real and nominal growth;
  • Table 1.1.6, again, comes out to a 2.89-percent growth rate;
  • Table 1.1.4, which reports price indexes for GDP, puts growth in prices at 2.42 percent; with the usual account for proximate estimates, this number goes well with the number reported in 1.1.3, giving us the current-price number in 1.1.5; again, no mistake there either.

A couple of questions from blog readers made me realize that I should boil down the problem as follows: 

1. There is something called annual growth, which is from, say, Q2 2017 to Q2 2018. This growth rate is the actual annual growth rate, and the only one that genuinely measures how the economy grows year to year.
2. Then there is the annualized growth rate, which is an estimate of annual growth, by quarter, and measures how much that annualized rate grows from one quarter to the next. This is not a measurement annual growth rate, but is being used as such in the media.

Why is the annualized rate being used as the annual rate? Because the the BEA's press releases do not explain the difference. Either that, or someone has intentionally written the last two to make it look like the economy is growing at a genuine annual rate of 4+ percent. 

I find it hard to accept the latter explanation. For now, I will chalk this up to sloppy writing by someone in the BEA's press release office.

No comments:

Post a Comment