Wednesday, September 23, 2009

The Pardox of Thrift

I'm trying to understand Keynesian business cycle theory. I think I get this much. There's a shock to the system that causes consumers to worry about the future. In order to smooth consumption, consumer spend less and save more. Spending less decreases the demand for goods and services. The resulting drop in demand for goods and services causes a drop in demand for people providing those goods and services, resulting lower GDP and unemployment.

Classical economics says that the creators of goods and services will respond to lower demand by lowering prices, and at the lower price they will be able to sell what they could have earlier. Similarly, workers will respond to lower demand by working for less.

Keynesians believe that prices and wages are sticky--that it takes time for people to understand that the labor that used to be worth more in nominal dollars, is actually worth less now in nominal terms. Thus, the necessary adjustment does not take place for some time, and that results in prolonged periods of unemployment and depressions.

Keynesians view the solution to this problem as increasing aggregate demand, or the total demand in the system for goods and services. If demand falls because of some shock to the system, it can be restored by government stepping in to supplement the demand, by spending money.

I'm having a hard time understanding the idea that saving reduces consumption, also called the paradox of thrift. Russ Robert and Steve Fazzari go a couple rounds on this point, here. I've now listened to this podcast twice, and I still don't quite get it. And I've been thinking about this occasionally since I first studied Keynes in college some years ago.

I do understand on one level that if a dollar comes to me, and I put 20 cents under my mattress, and spend 80 cents, and then someone else does the same thing (saving 20% and spending 80%) that will result in X number of dollars of economic activity. If you "save" less, and spend more, like say you save only 10%, and spend 90%, there will be more economic activity.

But "saving" in this instance is putting the money under your mattress. Saving, at least in common parlance, also means putting the money in the bank. The bank, however, is going to lend all but a small portion of that money out. So suppose you put a dollar in the bank instead of spending 90% of it, the bank will lend most of it out, and the person who get that loan will spend the money on consumption. It seems to me that, saving, then is actually mostly investing, which is also a specific type of consumption, which does not, in fact, reduce consumption, triggering the paradox of thrift.

So despite Fazzari's efforts, I don't understand the paradox of thrift.

UPDATE: ask and ye shall receive. Here is an article about Keynes by Richard Posner. Still not sure I understand the paradox, though.

4 comments:

Brett said...

This is something I've wondered about as well. I listened to most of the Robert and Fazzari conversation during my prep the other day (while entering grades as well of course). I thought the example of the family not eating at the restaurant leading to increased investment was interesting. Fazzari kept retorting (and seemed to imply that it was the crux of the argument of why spending is important) that the restaurant will decrease their investments by the same amount so investment isn't actually increased. But I wish Roberts would have brought up the fact that fact that this situation would not decrease investment either. The family is just investing in something other than the restaurant. So if you break even, why is spending the key?

Another aspect of thrift that I feel is germane (which they did not touch on) is that a family with savings would likely have more consistent spending. During a economic slowdown, even if their income is decreased, if they have a sizable savings they will be able to continue consuming at near the same rate as before. I see the fact that US households had severely negative savings rates as a major exacerbation of the current recession.

Anonymous said...

I'm stuck on the same thing you mention in your first paragraph.

I'm thinking Fazzari's implicit answer is that, when the family spends the $5 on the meal, the $5 is spent, so you have $5 of economic activity by the time the restaurant puts $5 in the bank. When the family saves the $5 by putting it in the bank, you don't have that $5 of consumption, so when the family puts its $5 in the bank, it's already behind.

Anonymous said...

To you second point, I think Fazzari would respond--and I think the general idea behind the paradox of thrift is--that, when everyone starts saving, income is reduced such that, not only does income fall, but total saving actually falls too.

Brett said...

Sorry I'm slow getting back to this, but I had more thoughts about the family/restaurant example.

Fazzari claims that if the family spends the $5, their savings goes down, but the restaurant's balance increases by the same $5 (I'm inferring this from what he said about the family saving the money) and in the mean time the act of exchanging goods/services has stimulated the economy.

To me it seems like there's always going to be a loss of overall savings (like losing energy just about whenever it's transferred) because the $5 decrease for the family isn't a $5 profit for the restaurant.

On second thought, maybe he'd say that what the restaurant loses in operating costs just gets distributed to its employees, suppliers, etc. So the entire $5 could end up on several parties' balances.

I'm thinking that for this to work you would have to consider "total money" to be a constant (like the energy of the universe). But what about interest, inflation, money being printed/destroyed, etc.

If this total money were constant, I could see how spending money would drive everything. Yet saving (investing) still seems just like letting someone else spend your money, so what difference does it make. I guess Fazzari would say investing gives someone else the opportunity to spend the money but doesn't guarantee that they will.

I think I need to read a book about this.