GDP shrank last quarter. Are economists right to reassure us?
We won't have a demand-side recession like 2008, but we shouldn't be sanguine either.
Real gross domestic product (GDP) decreased at an annualized rate of 1.4% in the first quarter of 2022, according to a preliminary Bureau of Economic Analysis (BEA) estimate released last week.
After the BEA released its numbers, the White House Council of Economic Advisors (CEA) wrote: “the U.S. economy contracted over the first three months of the year, primarily reflecting drags on growth from inventories, net exports, and government purchases. While GDP growth was negative, a measure of economic growth that removes these volatile elements grew solidly.”
Ben Casselman, an economics reporter at the New York Times, characterized things in a similar way, writing that “the negative number masked evidence of a recovery that economists said remained fundamentally strong.”
So how reassuring should we find these reassurances? The partisan-brained answer is either “very” or “not at all.” But the truth is somewhere in between. Properly understood, these reassurances are a bit like a doctor telling you “at least we know you don’t have cancer.” It’s good to hear, but you should be concerned that the conversation has taken this turn in the first place.
The White House and economic journalists who stressed this message ran a legitimate diagnostic tool and came to a legitimate result—we aren’t in danger of a demand-side contraction like that of 2008. But this style of analysis reflects a demand-constrained view of the world that was more appropriate in the 2010s than the 2020s. Today we have plenty of demand, but production has been battered by shocks to supply. And that requires a different style of analysis.
Net exports, or exports minus imports, is the most important part of the White House argument. Per Table 2 of the BEA release, net exports contributed negative 3.2 percentage points to the measured negative 1.4 percent annualized growth in GDP. If you chose to exclude the trade component of the GDP calculation, you’d get a positive number. (About positive 1.8.)
This might seem like cherrypicking to you—looking at only favorable components of an aggregate to make yourself feel better.
But that’s not actually what the White House or the New York Times were doing. Instead, they were transforming GDP into a different measure—a measure of how much Americans were purchasing, rather than how much Americans were producing. Here’s how that works.
GDP is calculated in a peculiar way. Although it is a measure of the value of what we have produced, we find it easiest to measure by starting from what we consume. It’s easier to trace final spending than every bit of a value-added chain. So economists at the BEA sum up private consumption purchases, private investment, and direct government spending. Then, in order to transform this measure of spending into a measure of production, they add all of the things that we produced but did not consume (exports) and subtract that which we consumed but did not produce (imports). This gets you the introductory macroeconomic definition of GDP: C + I + G + NX, or consumption plus investment plus government plus net exports.
GDP is essentially a reverse-engineered metric of production. So what the economists excluding trade from the calculation are doing is re-reversing it, to get back to what we spent, rather than what we earned or built.
Why are they doing this? Because over the last 15 years we’ve most often been more worried about too little spending. People had low incomes and couldn’t buy things, and as a result firms got fewer sales, and hired fewer people, and those unemployed people had low incomes, making it hard for them to buy things. And around and around it went.
This kind of vicious cycle is terrible for the economy, so it was legitimate for economists to worry about it. Subtracting net exports from GDP is a quick way to judge whether spending is strong or not. However, this is just a diagnostic for a specific problem. It’s not a clean slate of health for the whole economy.
Trade deficits suck money out of domestic accounts
Right now,of course, few people are worried about inadequate demand. They’re worried about the economy’s capacity to meet the extremely strong demand we’ve had since last year. We’re worried about whether producers can navigate the fallout of COVID-19 in the U.S. or the Russian invasion of Ukraine or the shutdowns of Chinese cities.
And low production will eventually come back to haunt us. We have a limited ability to spend more than we produce by running trade deficits. But to finance this extra spending, we need to give financial assets to foreigners in exchange.
So a trade deficit slowly sucks assets out of the domestic economy. You can actually see some of this effect visually, in checking account data from the JPMorgan Chase Institute. People got extra money from COVID-19 relief, and it is slowly disappearing from their accounts. Some is going to other accounts not measured here, like corporations or brokerage accounts, but some of it is leaving the country.
There’s a way to stop this problem: if the government passes a big deficit-expanding bill, it can prop up people’s balance sheets. (The chart, in fact, demonstrates this.) Government should run deficits and buoy private balance sheets when inflation and interest rates are low. But with inflation and interest rates rising, they will be disinclined to do so right now.
Instead, the high-spending low-production situation will be forced to resolve itself once the public starts running out of money.
Can it be resolved through higher production? Somewhat, of course. Employment is still a tad off of its record highs. 80 percent of working-age Americans have a job, and we should be able to boost that to 81, where we were before the pandemic. And perhaps some of our supply-side problems will resolve themselves. This could help us “earn” our spending levels again, and stop the outflow of cash.
But I think it’s a tall order to solve the problem entirely through higher production and earnings. Instead, some of the problem will resolve itself through stagnant real consumption. American households will run out of extra pandemic cash, and find that a new round of stimulus isn’t forthcoming. This will force them to slow their consumption or even cut back in real terms, as foreigners, flush with dollars from America’s 2022 import binge, increase their consumption. At that point, GDP will rise faster than domestic purchases and our trade deficit will fall again.
Take care with accounting identities and causal inferences
The role of the trade term in the GDP account is easy to misunderstand, or at least, easy to mis-describe. A lot of economists like to argue about the best way to describe it to get across the key point: trade doesn’t causally add or subtract from GDP, rather, the trade term is a quick way to turn a consumption measure into a production measure.
But it takes quite a bit of real estate to explain this. (I used about 120 words.) Inevitably, writers who are primarily interested in summarizing the recent figures rather than explaining the GDP calculation come up with shorter explanations.
Ben Casselman’s description included a passage with the following text: “the ballooning trade deficit, meanwhile, took more than three percentage points away from G.D.P. growth in the first quarter. Imports, which are subtracted from gross domestic product because they are produced abroad, have soared in recent months.”
The economist Noah Smith disliked this description. “Took” is arguably too active a verb to describe the trade deficit’s role, making it seem like a cause of low GDP when it’s arguably more like a consequence. And, more importantly, imports aren’t merely “subtracted” from GDP because they're also components of consumption, investment, and government spending. So they’re added and subtracted, leaving no net effect.
I sympathize with the need to state things succinctly; under many circumstances, I’d say Smith is being persnickety, and we don’t need to rehash the exact nature of how trade deficits work in GDP every time the figure is released.
But the especially high trade deficit this quarter is noteworthy, and deserving of extra care in its description. And judging by the already red-hot job market, we can’t resolve it by simply “making the goods here” instead. We need real production and growth, not just demand.
So I would be wary of gauging the economy on consumption-based measures that exclude trade. The dwindling bank accounts shown above are a real cost of our consuming more than we produce.