You may have noticed that I post a decent number of comics from Saturday Morning Breakfast Cereal. While I do know Zach, it wasn’t until recently that I started meddling in the cartoon-generating process. In general, it’s more amusing to see what non-economists come up with on their own, but sometimes I just can’t help myself. So here goes…
The original question:
Econ question: If I wrote a program wherein I and another person constantly traded $1 for nothing, back and forth, does GDP go to infinity?
(For the record, Scott Adams, the creator of Dilbert, actually has a degree in economics, yet I am far more impressed by the issues posed by thoughtful non-economist cartoonists.) Of course this sounds absurd, but there are plenty of things in macroeconomics that sound absurd, so I figured this was worth thinking about for a second. Or an hour. Here’s what I came up with:
So here’s some general information on GDP:
I think the part that is most relevant to the question at hand is the notion of “produced.” So, used goods don’t count in GDP unless there is some value-added service associated with selling them, and even then it’s only the value-added part that counts in GDP. If I modified your scenario slightly so say that two people kept paying each other a dollar for a widget of some sort, that widget would only count in GDP the first time it was sold to an end user consumer. Your scenario is a bit more complicated I think, however (perhaps unintentionally). If the dollar going back and forth is exchanged just because, then it would be considered a repeated gift or transfer and not included in GDP because nothing was actually produced. If, on the other hand, the dollar is each time exchanged for some intangible service (a smile, wink, nod, etc.), then there is in fact a dollar of GDP generated with every transaction, since the service is new each time rather then the reselling of a used service. (I’m not even sure what a used service would be exactly, but it sounds kind of dirty.)
Your twitter commenter is right that digging a hole and then filling the hole back in would add to GDP with no net change in the state of the world, and one could make two arguments about this. One argument is that this is a shortcoming in the notion of GDP, similar to the broken window fallacy, which you can read about here:
Another argument is that, because people were willing to pay both to have the hole dug and filled in, that both activities must have had worth to people. Unfortunately, there is likely also a negative externality imposed on the guy who wanted the hole dug when the hole is filled back in and vice versa, which would counteract the value that is counted in GDP to some degree.
I read a thing a few days ago that I can’t seem to find but is very relevant to this situation. Suppose guy 1 pays $50 to punch guy 2 in the face. Suppose further that guy 2 pays guy 1 $50 to then punch him in the face. The article or whatever’s conclusion was that both guys had no more money than before but had black eyes, so they were worse off but $100 of GDP was created. It’s a cute example, but it isn’t quite correct because it ignores the warm fuzzies or whatever that the guys got from doing the punching, which they had to have gotten since they were willing to pay $50 to do it. When this is taken into account, each guy got a service that they valued at at least $50 plus a black eye. Because both men were willing to accept $50 to get punched, it must be the case that the cost of getting punched was less than $50 to both men. Therefore, both men were made better off on net by this set of transactions, but not by the full $100 of GDP that was created. The reason is that GDP doesn’t take into account wealth destruction, whether it be black eyes or loss of buildings due to earthquakes and the like.
In other words, GDP calculations can be a little absurd on the surface but aren’t totally absurd once you think a little. So apparently this is why Zach was asking his question:
He had actually shared the storyline with me earlier, and I made the following point:
I suppose that the motion of electrons counts as a different service each time, and the nerds are willingly giving the money each time, so as much as I want to say that this doesn’t count as GDP, I’m leaning towards allowing it. You could also just have the computer make a trivial beeping noise or something to count as the service rendered. The important part is that it has to be clear that nerd 2 is actually providing the service worth $1 to nerd 1 and vice versa, otherwise you’re just describing paypal with a zero commission, and it’s the paypal fees that count in GDP, not the money transferred itself. Perhaps describe the scenario as nerd 1 and nerd 2 each installing a program on the machine to accept the dollar in return for a beep?
After this exchange, I sent Zach a copy of Greg Mankiw’s favorite textbook, though I’m not sure how much it helped, since this is well outside of textbook and into thought exercise territory. In other words, it would make a great exam question if you want to give your students ulcers.