Imports do not subtract from GDP. Reasoning about the economy from an accounting identity does not lead to useful knowledge. Historically times of increasing imports are also times of increasing GDP. https://www.noahpinion.blog/p/imports-do-not-subtract-from-g...

A reasonable question here is "did companies spend down petty cash to import goods in Q1" (no change to GDP), or "did companies stop investing and producing domestically in order to afford to import more goods in Q1" (reduced GDP).

> Imports do not subtract from GDP.

Whether or not that's true in reality, it is true in the report that caused the headline.

From the article:

> A logistical consideration makes Wednesday’s report difficult to interpret: Imports subtract from the Commerce Department’s calculation of GDP, since they represent spending on foreign-made goods and services.

They do subtract as an accounting identity, but reasoning from that makes no sense. My net income can be calculated as how much I consume plus my net savings. I = C + NS. So withdrawing from savings "subtracts from net income" in the same sense - ie, it has no explanatory power. If I do a home remodel, spending a bunch of saved up money, the way to understand that situation is not "my income would have been higher if I had not withdrawn from savings".

But the point is that this can't be responsible for the reported contraction, correct?

Because imports are only subtracted from the higher spending on imports in the first place.

Conceptually, imports are not part of GDP. Imports are never subtracted from GDP. But to calculate GDP, part of that is to take total spending (a number larger than GDP) and then subtract spending on imports.

So the 0.3% contraction in GDP has to be a real thing. Unless there's some other deeper technical discrepancy, or some kind of time lag where spending on imports and the import subtraction are being counted in different quarters?

Importing into inventories absolutely does decrease GDP. If you look at the Q1 stats some of them are normal, but imports and inventories had huge weird effects and government contribution was negative for the first time in a while.

Noah Smith is a galaxy-scale idiot, by the way.

Importing into inventories counts as both investment and import, resulting in no net change to GDP. Consumers importing for themselves counts as both import and consumption, resulting in no net change to GDP

One way of counting gdp is to count everything sold and then subtract imports. This is because imports are inherently included in everything sold (if I buy a widget for 10usd from abroad and use it to make an 11usd product, that’s 1usd of gdp, not 11).

So when imports are subtracted that doesn’t mean gdp is “short” imports. It means it isn’t affected either way by imports.

Right, but those $11 products have not been sold yet b/c companies were trying to import as much as possible to avoid tariffs.

I've seen this in coffee - roasters buying a year's worth of packaging.

Yeah, but that doesn’t matter here: we already counted the 10 when they bought them. So when we subtract the 10 because they’re imports that doesn’t mean importing a lot will lead to a net negative.

If a company 10x their imports to get ahead, we’d count the 100 worth of buying, and then subtract 100 because that was all imports. You still get zero effect on gdp from the import itself. Which is correct.

The point of the subtraction is to make imports zero impact, zero impact when big, zero impact when small.