DeLong Smackdown Watch: Walras' Law and Say's Law Edition
Nick Rowe asks:
What Does Cutting-Edge Macroeconomics Tell Us About Economic Policy for the Recovery?: You see, Say(1803) (lovely way to express this, by the way) was very nearly right. Suppose we start in equilibrium, then there's a sudden desire to stop buying newly-produced goods and buy land instead. Either the price of land rises to equilibrium or it doesn't. If the price of land rises to equilibrium, then people stop wanting to buy land and return to buying newly-produced goods. If the price of land stays fixed (it's sticky, or whatever) people cannot buy land because nobody is willing to sell. So they have to buy something else with their income instead, or else hoard money.
Ultimately there are only two things an individual can do with his income, if everybody else is trying to do the same thing: buy newly-produced goods, where there are plenty of willing sellers in a general glut; or hoard money, by not buying things, which nobody else can stop you doing.
And:
What Does Cutting-Edge Macroeconomics Tell Us About Economic Policy for the Recovery?: OK Brad, a challenge for you:
A general glut means an excess supply of newly-produced goods. You say that a general glut can be caused by an excess demand for financial assets: which could be money, bonds, or safe assets. Is it theoretically possible for a general glut to be caused by an excess demand for something that is neither a financial asset nor a newly-produced good? For example, could it be caused by an excess demand for: land, old houses, old books, antique furniture etc.? Or, what about intermediate cases, like an excess demand for gold, where new and old gold is identical, but new production is very small and inelastic compared to the existing stock?
My position is that a general glut can only be caused by an excess demand for the medium of exchange. An excess demand for any of those other assets can only cause a general glut if it spills over into an excess demand for the medium of exchange. The distinction between financial and non-financial assets is irrelevant. Why should it matter?
I'm trying to smoke out your inner quasi-monetarist!
The way Say expresses it in 1803 is roughly as follows: nobody makes anything unless they intend to use it or sell it, and nobody sells anything unless they intend to buy something with the proceeds of the sale. Thus, "by the metaphysical necessity of the case," as John Stuart Mill was to put it, there has to be the purchasing power to buy everything offered for sale--there can be particular gluts of commodities, but every market in which there is excess supply must be balanced by another in which there is excess demand.
This is an anticipation of what we now call Walras' Law: that the sum across all markets of all excess demands must equal zero. And it was John Stuart Mill who pointed out the hole in Say's argument: you can have an excess supply of all currently produced goods and services if you have an excess demand for financial assets, specifically for money. As Mill put it:
Although he who sells, really sells only to buy, he needs not buy at the same moment when he sells.... [I]t may very well occur that there may be... a very general inclination to sell with as little delay... accompanied with an equally general inclination to defer all purchases.... It is true that this state can be only temporary and must even be succeeded by a reaction of corresponding violence... [but] this is no more than may be said of every partial over-supply.... It must, undoubtedly, be admitted that there cannot be an excess of all other commodities and an excess of money at the same time. But those who have... affirmed that there was an excess of all commodities never pretended that money was one of these commodities.... What it amounted to was that persons in general... liked better to possess money than any other commodity. Money, consequently, was in request, and all other commodities were in comparative disrepute...
Where I think Mill's explanation is incomplete is in his reference to "money... in request, and all other commodities... in comparative disrepute." When money is in request and all other commodities in disrepute, one of those commodities is corporate bonds. Hence people should be dumping corporate bonds as they try to build up their cash holdings. Hence the prices of corporate bonds should be low. But in 2001 we had relatively high prices of corporate and government bonds--low interest rates all across the board--so that there wasn't a shortage of money in particular. What there was was a shortage of savings vehicles for carrying purchasing power from the present into the future, a shortage of bonds and other assets that serve as such savings vehicles--and money is one such vehicle. And today we have relatively high prices for government bonds and other safe assets. What there is a shortage of safe assets--and money is one such safe asset.
This matters because the monetarist cure for a downturn--for a general glut of currently-produced commodities--is to expand the money stock via open market operations, via purchases of short-term government bonds for cash. When the key excess demand in financial markets is an excess demand for money, that works: you expand the supply of money and reduce the excess demand for cash and so people are no longer scrambling to cut back on their spending on currently-produced goods and services to build up their cash balances.
However, things are different when the key excess demand is a demand for savings vehicles or for safe assets. Here, I think, the zero bound on interest rates is crucial. Bonds can go to par at which point they become perfect substitutes for cash and cannot go any higher. Other safe nominal assets can go to par at which point they become perfect substitutes for cash and cannot go any higher. And that zero-bound property is what gets the economy stuck in a general glut.
When the key excess demand is a demand for savings vehicles--for bonds--open-market operations don't work: you buy bonds for cash but you haven't done anything about the excess demand for bonds, bonds go to par and cannot go any higher and there is still an excess demand for bonds, and so people keep scrambling to cut back on their spending on currently-produced goods and services to build up their bond balances. And when the key excess demand is a demand for safe high quality assets, open-market operations still don't work: you buy safe government bonds for cash but you haven't done anything about the excess demand for safe assets, all safe assets go to par and cannot go any higher and there is still an excess demand for bonds, and so people keep scrambling to cut back on their spending on currently-produced goods and services to build up their safe asset balances.
But suppose the excess demand were for some other non-currently produced asset--Nick's examples are land, old houses, old books, and antique furniture--could that produce a general glut? Walras Law would say yes, I think. But it would also say that the monetarist cure--buy bonds for cash--would alleviate the problem: create an excess supply of money and people will try to spend down their cash balances and that will boost demand for currently-produced goods and services, leaving you with an excess demand for old furniture and an excess supply of money.
It is the fact that the key excess demand is for a broader financial asset class than mere cash money that is, I think, the key to why the monetarist cure is likely to fail--or at least to be supplemented by policies that make sure that not just the supply of liquid cash money but the supply of safe assets in general or savings vehicles in general increases in order to cure a downturn.
