Keynes' Theory of Recessions

Written by Paul Krugman

Imagine for a moment an economy that is chugging along at full employment. All factories are working at full shifts, all workers have jobs. (In reality this never happens, because there is always enough friction and uncertainty in the world to ensure that some factories are closing and some workers can't find jobs; but that isn't important at this stage of our story.) Alongside the smoothly functioning "real" economy will be a smooth financial flow, as firms earn money from their sales, pay out their earnings in wages and dividends, and households spend these receipts on new purchases from the firms.

But now suppose that for some reason each household and firm in this economy decides that it would like to hold a little more cash. There are a number of reasons, which we need not go into, why people hold some of their funds in the form of currency or bank deposits backed by currency rather than in other assets that yield higher returns. The important point is that sometimes they decide that they would prefer to hold more cash than they were holding a little while ago. Keynes argued, in particular, that this happens when businessmen lose confidence and start to think of potential investments as risky, leading them to hesitate and accumulate cash instead; today we might add the problem of jittery households who worry about their jobs and cut back on purchases of big-ticket consumer items. Either way, each individual firm or household tries to increase its holdings of cash by cutting its spending, so that its receipts exceed its outlays.

But as Keynes pointed out, what works for an individual does not work for the economy as a whole, because the amount of cash in the economy is fixed. An individual can increase her cash holdings by spending less, but she does so only by taking away cash that other people had been holding. Obviously, not everyone can do this at the same time. So what happens when everyone tries to accumulate cash simultaneously?

The answer is that income falls along with spending. I try to accumulate cash by reducing my purchases from you, and you try to accumulate cash by reducing your purchases from me; the result is that both of our incomes fall along with our spending, and neither of us succeeds in increasing our cash holdings.

If we remain determined to hold more cash, we will react to this disappointment by cutting our spending still further, with the same disappointing result; and so on and so on. Looking at the economy as a whole, you will see factories closing, workers laid off, stores empty, as firms and households throughout the economy cut back on spending in a collectively vain effort to accumulate more cash. The process only reaches a limit when incomes are so shrunken that the demand for cash falls to equal the available supply.

That is Keynes's story. It's pretty simple, yet it has caused an astonishing amount of confusion; somehow even intelligent people seem to find it difficult and abstract. As it happens, however, we can illustrate the essence of Keynesianism with an example that is, literally, childishly simple.

Infantile Keynesianism

Nobody can hold a picture of the reality of the U.S. economy, with its hundreds of thousands of firms and hundred million workers, in his or her head. Inevitably we must rely on models-that is, some kind of simplified representation that we hope gets at the essence of the question we are trying to answer. Part of what distinguishes professors and policy entrepreneurs is the kind of model that they prefer. Policy entrepreneurs generally prefer their models in the form of metaphors: for example, they may describe the U.S. economy as being like a corporation, competing in the world marketplace. Professors generally prefer their models mathematical. Each of these preferences has disadvantages: mathematical models can be constricting, leading you to ignore what you haven't figured out how to represent as an equation; metaphorical models can all too easily create a false impression of understanding and sophistication, and those who rely exclusively on metaphor often fail to notice that their fine phrases are covering crude conceptual or factual mistakes.

In the physical and biological sciences, however, there is a third kind of model: the experimental model. My personal favorite example is the work of the great meteorological theorist Carl-Gustaf Rossby, who found that the essential features of global weather can be simulated by placing a pan filled with water on a slowly rotating turntable and gently heating its rims. Experimental models can allow you to see, on a small scale, the essence of events that occur in the vastly larger and more complex real world.

Unfortunately, experimental models are hard to come by in economics, for obvious practical and moral reasons. There is a limited literature on experimental auctions and other market mechanisms, some anti-poverty schemes have been tried out with pilot projects, and so on; but how could you experimentally produce recessions and recoveries?

Well, it turns out that a group of Washington D.C. professionals inadvertently created a sort of experimental macroeconomy during the 1970s. Their unhappy experiences are described in a whimsical article by Joan and Richard Sweeney in the Journal of Money, Credit and Banking (February 1977), entitled "Monetary Theory and the Great Capitol Hill Baby-Sitting Co-Op Crisis."

Here's the story. A group of young professional couples with children formed a baby-sitting co-op: that is, they set up an arrangement to look after each other's children. Any such arrangement requires some method to make sure that the burden gets fairly allocated. What this co-op did was introduce a self-regulating bookkeeping system, by issuing scrip: coupons with one hour of baby-sitting. Every hour of baby-sitting would involve a transfer of a coupon to the baby-sitters from the baby-sittees.

If you think about it a bit, you will realize that such a system requires that there be a fair amount of scrip in circulation. Couples cannot predict exactly when they will want to go out, or when they will be free to baby-sit for others, so on an average they will want to keep a reserve of coupons available in case they want or need to go out several times before earning some more scrip by sitting for others.

After the co-op had been in existence for some time, it got into trouble. For reasons that we needn't get into, the number of coupons in circulation per couple became rather low. This had peculiar consequences.


Other Political and Economic Writings | Top of Page | Back