What G.D.P.’s Cousin Can Tell Us About the Economy
Gross domestic product is a great measure of the economy, but it would be a gross exaggeration to say it’s the only one. The Bureau of Economic Analysis also puts out gross domestic income, value added by industry and gross output by industry, to name three alternatives. The variety of indicators “puts powers in the hands of users to select the statistics that work best for their needs,” a bureau spokesman, Thomas Dail, wrote to me last week.
I want to zero in on gross output by industry because it’s a strange beast. First of all, it’s much bigger than gross domestic product. Last year, when the U.S. G.D.P. was around $25.5 trillion, gross output was around $46 trillion (neither figure adjusted for inflation).
Gross output is bigger than gross domestic product because of deliberate double counting. Gross domestic product is the market value of goods and services produced in the United States. Gross output is the sum of G.D.P. and all the intermediate inputs into G.D.P., such as energy, raw materials, semifinished goods and purchased services. It sums up the sales that companies make, without subtracting their costs of goods sold.
To put it another way, gross output includes business-to-business (or B2B) commerce that gets netted out in the gross domestic product. (Here is the government’s FAQ.)
For example: A company that makes kayaks decides to buy the fittings from another company instead of making them itself. That wouldn’t affect gross domestic product because there’s no increase in the net output of goods and services. But it would make gross output go up because there’s a new B2B purchase involved.
It’s at least conceivable that gross output is a leading indicator of the economy. When the economy is expanding, companies that are scrambling to meet stronger demand might turn to other companies for help, which would raise the volume of transactions (and hence gross output) even faster than it would raise G.D.P. Conversely, when businesspeople foresee trouble ahead, they might retrench by bringing more work in house, reducing the volume of B2B transactions and shrinking gross output. G.D.P. isn’t affected because the same amount of work is being done, just by someone else.
Gross output’s predictive power is hard to gauge, though, because it began as a quarterly series only in 2005. Gross output adjusted for inflation peaked in the same quarter as gross domestic product in 2007, when the deep recession of 2007-9 began. It peaked in 2019, before G.D.P. did, but that doesn’t mean much because the 2020 recession was caused by the pandemic, which nobody saw coming.
I looked into gross output at the urging of Mark Skousen, an economist at Chapman University in Orange, Calif., who has built an entire website devoted to the measure. Skousen likes to cite Finn Kydland, a Nobel laureate economist who has recommended that others in the field “seriously consider” Skousen’s unconventional approach. Skousen also pointed me to the work of David Ranson, the president and head of research at the economic research firm HCWE & Co., who wrote last year that intermediate inputs into G.D.P. tend to rise and fall slightly ahead of G.D.P. itself.
Skousen has created his own version of gross output that’s even bigger than the government’s because it has even more deliberate double counting. It counts total sales for wholesalers and retailers rather than netting out their cost of goods sold, as the government does. His gross output number for this year’s first quarter was more than $58 trillion at an annual rate.
“Business spending is floundering,” and “recession is still in the near future,” Skousen wrote in an analysis of the gross output data for the first quarter, which came out on June 29, along with revised government estimates of G.D.P.
I’m sympathetic to Skousen’s view that a recession could be near, but I’m not fully convinced about the value of gross output as an indicator.
Skousen points out that gross output got a plug in a 2006 book, “A New Architecture for the U.S. National Accounts,” by three prominent economists, Dale Jorgenson of Harvard, Steven Landefeld of the Bureau of Economic Analysis and William Nordhaus of Yale. They wrote: “Gross output is the natural measure for the production sector, while net output is appropriate as a measure of welfare. Both are required in a complete system of accounts.”
I asked Landefeld and Nordhaus for their latest thoughts on gross output. (Jorgenson died last year.) I’d say they were lukewarm. Nordhaus wrote in an email that gross output and other components “are helpful for a complete set of accounts. But I do not believe that gross output should replace a value-added concept such as G.D.P. as the central measure of our national accounts.”
Landefeld, who has since retired, told me in a phone interview that he thought gross output was more useful on the industry level than as a measure of the entire economy. Even then, he said, “conceptually, I think value added by industry is the preferred measure.”
Still, I give Skousen credit for raising awareness of a fairly obscure data point. He likes to cite Humphry Davy, a British chemist who died in 1829, who once wrote: “Nothing tends so much to the advancement of knowledge as the application of a new instrument.”