GDP is high in booms and low in recessions (relative to trend). That is (roughly) how we currently define "booms" and "recessions".
GDP (roughly) measures the volume of trade in newly-produced final goods. But that is a narrow measure of trade. A lot of goods get traded that are not newly-produced. People buy and sell old houses, for example, where by "old" I mean "not produced this year". And those trades are not counted as part of GDP.
There is nothing wrong with excluding sales of old houses from GDP. Because GDP is supposed to measure the volume of production and not the volume of trade. But there is something peculiar in defining booms and recessions and the business cycle in terms of the volume of production, rather than the volume of trade. We produce goods and trade the goods we have produced, presumably because it makes buyer and seller better off, which is why production matters. But trade in old goods, like old houses, presumably makes buyer and seller better off too, so trade in old goods matters too.
In a monetary exchange economy, all goods, both newly-produced and old goods, get bought and sold for money. If something goes wrong with money, we should expect to see trade in all goods, both new and old, get disrupted. Some mutually advantageous trades in newly-produced goods don't get made; some mutually advantageous trades in old goods don't get made either. The decline in production we observe in a recession is just one symptom, albeit an important symptom, of a general decline in trade of both newly-produced and old goods. And production only gets disrupted because trade gets disrupted. We stop producing goods if we can't sell them. Those goods produced for our own use are the exception that proves the rule. Anecdotally, home production increases during a recession.
What do we trade? We trade newly-produced goods. We also trade old goods. We also trade IOUs. When I sell an IOU, we call it "borrowing". When I buy an IOU, we call it "lending". Borrowing and lending are trade, and those trades happen for the same reason all trades happen -- because the buyer and seller are presumably better off as a result of the trade. Of course there are exceptions, but then there are exceptions to all trades. Everyone has regretted buying or selling something, whether or not it was an IOU.
If you look at business cycles this way, as a trade cycle, in which the volume of trade rises in booms and falls in recessions, it is totally unsurprising that the volume of borrowing and lending should also rise in booms and fall in recessions. Because borrowing and lending is just another type of trade. The same monetary problem that disrupts trade in newly-produced goods, and disrupts trade in old goods, also disrupts trade in IOUs. Because, in a monetary exchange economy, all those goods -- newly-produced goods, old goods, and IOUs -- are traded for money.
What would be surprising and in need of explanation would be if trade in IOUs did not follow the same cyclical pattern as trade in other goods.
Neil Irwin tells me that trade in IOUs is increasing in the US. (HT Mark Thoma). He's right that it's good news. I'm just re-framing what he is saying.
Yes, some IOUs are used as money. Which means that some trades in IOUs create money. And this, depending on the monetary policy followed by the central bank, may create a positive feedback loop. But the job of a good central bank is to replace that positive feedback loop with a negative feedback loop, just like a good thermostat. (I'm talking about "banks", in case you missed it.)
[After this afternoon, I will be taking a couple of weeks away from the blog.]