Previous article: Reply to Polyconomics - Part 2
Meanwhile, Wanniski defines the source of the monetarist error as viewing money “not as a medium of exchange, but as a storehouse of value”; while Goldman explains that their error is to consider money as a medium of exchange when “an enormous amount of U.S. currency is held as a store of value.”
This appears to illustrate the philosophical precept that “truth is single, but error is multiple.”
In his eagerness to argue that the World Dollar Base cannot predict inflation, Goldman insists as a matter of theory that there can be no measure of money for which people’s demand is a stable function of anything.
If this is true, Goldman has proven not too little, but too much – for Wanniski. Namely, Goldman will have disproved Say’s Law. The question is not whether people hold money as a store of value, but whether this demand is relatively predictable. If the demand for money were as unstable as Goldman says, then there would always be a huge and erratic gap between total supply and total demand for goods – caused not by a flawed monetary regime, but by an inherently unstable market economy. In that case, Say’s Law could never be even “roughly correct.” There could never be any periods of price stability – ever. Fortunately, Goldman is wrong.
As it happens, I recently examined the “missing currency” question, including the Federal Reserve studies (“The Mystery of the Missing Money: Not So Mysterious?” LBMC report, December 1990). It turns out that most U.S. currency has always been unaccounted-for by the authorities, as far back as we have records (at least since 1890). I also show that there is nothing new or unpredictable about the recent behavior of U.S. currency in circulation. This is confirmed by the evidence we examined earlier in this report.
Goldman capsulizes his objection to LBMC’s theories by saying that “the quantity of money may or may not determine the general price level, depending on whether it is accompanied by greater quantities of goods, services or financial assets” (p. 1 emphasis in original).
This can be true, even in the short run, only if “the general price level” excludes the prices of financial assets – that is, if the “general” price level is not general. But as we have seen, Rueff’s and LBMC’s definition of excess money already includes the quantities of goods, services and financial assets. So Goldman’s argument cannot apply to LBMC.
But it’s hard to see how it applies even to Milton Friedman. Friedman’s theory is almost entirely about the demand for money, not the supply of money.
In its simplest form, monetarist theory describes the demand for money as a function of the stock of nonmonetary wealth (W: not, as in Rueff’s formulation, the flow of exchanges of non-monetary wealth, Q). This means that as people’s nonmonetary wealth increases, they will want to hold more money. Leaving aside the catch-all residual factor (H), then
L = f(W)P.
But wealth is equal to income or output (Y) capitalized at the rate of return (i), or Y/i. So the monetarist often describe the demand for money as:
L = f(Y,i)P.
In this version, the demand for money is directly proportional to the price of output (P), positively related to real output or income (Y) and inversely related to rate of return on assets (i). So, according to the monetarists, the velocity of money with respect to income (V = PY/M) is supposed to vary positively with the interest rate, and inversely with the price of financial assets.
So Goldman’s chief theoretical objection has been fully anticipated in both Rueff’s and Friedman’s theories, though in different ways. The important question is whether either theory is a useful description of reality. But this cannot be settled merely by pontificating: there is no way of telling without research to test their predictive power. But Goldman gives us no evidence one way or the other. Our research indicates that Rueff’s theory predicts inflation, but the monetarist theory does not.
But does Goldman’s statement make any sense? Goldman says that excess money “may or may not” bid up the price of commodities, depending on whether it is absorbed by financial assets.
This argument is interesting, not because it is true, but because it concedes that there can be such a thing as “financial inflation.” In more than a decade, I have never heard Wanniski explain a rise in the stock market as the result of excess money.
However, Goldman’s argument requires that the process simply stop there. We have an excess supply of money that bids up the price of Good A (financial assets) while leaving the price of Good B (commodities) unchanged. Since Goldman’s “general” price level includes only Good B, not Good A, he calls this a shift in the “velocity” of money. But it’s merely a change in relative price of Good A and Good B.
The belief that everything then stops bespeaks a curious view, not so much of how money works, as of how anything in the economy works. Goldman’s argument is like saying that if you drop a stone into the Grand Canyon, and don’t instantly hear it hit bottom, it must have stopped in mid-air.
Suppose there is a shift in demand to any Good A from any Good B. Other things being equal, the price of A rises, and the price of B falls. In Goldman’s argument, there’s an end to it.
Except, that’s not the end of it. Goldman implicitly assumes that the supplies of Good A and Good B are unchanged. This means their producers are indifferent to profit incentives; a strange doctrine for a “supply-sider.”
It costs something to produce Good A and Good B. Suppose, for simplicity, that the sole cost of producing one unit of Good A, or one unit of Good B, is an hour of unskilled labor. Then the price of a unit of Good A must have equaled the price of a unit of Good B to start with. But, due to the shift in demand, the price of A has risen while the price of B has fallen. Therefore, A will be selling above its cost of production. Therefore producers of Good B will reduce output and lay off workers, while producers of Good A will increase output and hire more workers. The rise of supply of A will cause the price of A to fall, and the fall in the supply of B will cause the price of B to rise. And this must continue until the price of Good A once again equals the price of Good B, because the cost of producing one unit of each is still one hour of labor.
Next installment: Reply to Polyconomics - Part 4