A while ago I dug into the mysterious world of GDP accounting and wrote an article (one of my personal favorites) showing the absurdities into which “inventory adjustments” could lead. On a related note, I sent the following email to some guys I work with about the recent revision to the 3rd quarter GDP figures:
The reason we care about inventories at all, is that they calculate GDP by counting spending. So to keep it simple, let’s say consumers and businesses spend $1 trillion buying vehicles. Right there, that goes into GDP and raises it by $1 trillion.
But now they have to adjust it. What if $100 billion of those vehicles were made in foreign countries? Well we don’t want to count that as part of US GDP. So we subtract $100 billion from the original figure. That’s what it means when the BEA says “imports subtract from GDP.”
What about inventories? Well, if the private inventories of vehicles in US car dealerships went down by $300 billion during the period, then we don’t want to include them in this period’s GDP figure, since they were obviously produced in a previous period. So if inventories drop during a period, then we subtract the change out of the original GDP figure too.
Now the really tricky thing. Strictly speaking, the BEA isn’t reporting the level of GDP, but rather the headlines all talk about the percentage rate of *growth* in GDP. And then they break that down and say which components contributed to the growth.
This leads to some weird things. For example, in the advance estimate press release
it says this about inventories:The change in real private inventories subtracted 1.08 percentage points from the third-quarter change in real GDP after subtracting 0.28 percentage point from the second-quarter change. Private businesses increased inventories $5.4 billion in the third quarter, following increases of $39.1 billion in the second quarter and $49.1 billion in the first.
So if you read that carefully, it is saying that even though the BEA (at the time) thought private inventories had *increased* $5.4 billion in the third quarter, that number was still making them knock off 1.08 percentage points from the growth rate they reported. In other words, they would have otherwise reported a growth rate of 3.58 percent, but then because inventories only grew $5.4 billion, they instead originally only reported a GDP growth of 3.58 – 1.08 = 2.5 percent.
The answer to this weird thing is that even though (they originally thought) inventories grew in the third quarter, they didn’t grow as much as they had in the second quarter (when they grew by $39.1 billion). So the slowdown in the rate of growth of private inventories, was dragging down the number they could report for total GDP growth.
Anyway, it’s a subtle little point, but I know most people don’t quite get that. They just paraphrase the BEA press release about inventories and I’m quite sure most have no idea how it really works.