In general, for an argument like Professor Fama's to work, there has to be some finite resource that is being fully utilized, so as to impose a binding constraint on the economy. That is, when the government borrows, it must be using up something – some actual, definite thing, not just a vague "funds" (which could mean any number of things depending on how we interpret it). The government must be using up some limited resource that is no longer available to businesses seeking to invest. What is that resource?
As I understand Greg Mankiw's interpretation, the limited resource is labor. In the classical model to which Greg refers, the availability of labor is what usually constrains an economy, the reason you cannot do more of one thing without doing less of something else. Now Professor Fama says explicitly that his argument applies "even when there are lots of idle workers." On the face of it, that would seem to contradict Greg's interpretation.
But perhaps Professor Fama is referring to frictional unemployment, and perhaps he believes in a theory in which recessions are associated with increased frictional unemployment. For example, today's unemployment could just reflect the difficulty in reassigning all the people that have been laid off in construction, finance, and other industries related to the mortgage boom. I can think of a number of empirical arguments as to why that's not the case, but the position is logically sound and does not rely on any assumptions that are inherently unreasonable. If that's what Professor Fama has in mind, I wish he would be clearer about it.
Nick Rowe has a different interpretation. He thinks the finite resource is money. If that's the intended interpretation, then there is an overwhelming empirical case against Professor Fama, as he will perhaps realize if he clarifies what he is trying to say. Money is not a finite resource today: there is nothing to stop the Fed from printing more money to finance any additional federal deficit, thus leaving plenty of money for those who want to use it for investment. (The argument goes beyond this, and I'm going to retell, in different words, the story I take Nick to be telling. Our argument shall be all things to all men, that we might by all means save some.)
Even if the Fed refuses to finance the deficit, and the money supply is fixed, the empirical case is still overwhelming, once you appreciate the nature of money and the relevance of a zero interest rate. The critical point is that money, even if it is limited in quantity, is a reusable resource. Money isn't like paper towels, where you use them once and then have to throw them away, and if my wife uses up all the paper towels and I can't get to the store then the dishes will have to sit in the drainboard. Money is more like cloth towels. If my wife uses up all the cloth towels, I can just put them in the washing machine and the dryer and use them again. Similarly, my wife can use money to buy a hamburger, and to the burger cook, it is as if the money had already been cleaned and dried. (Ah, yes, laundered money!) Even though my wife has already used the money, the cook can immediately go and use it again to buy something else.
Arguments that the quantity of money matters rely on some mechanism that limits the number of times money can be reused in a given year. The usual assumption (a controversial one, to say the least, but one we can accept for the sake of argument) is that people will hold money in proportion their incomes, regardless of the interest rate. In that case, if you try to raise aggregate income (for example, by a stimulus program), there won’t be enough money to go around. The excess demand for money will cause interest rates to rise until someone reduces their demand. The classic example is a business that is contemplating building a factory. When the interest rate rises, the factory becomes more expensive to finance, building it is no longer profitable, and the business decides not to build it. As that sort of thing happens across the economy, the demand for construction is less than it would have been, construction workers are laid off, and aggregate income goes back down to where it was before the stimulus program. Since we have assumed that the supply of money is fixed, and the demand for money is proportional to income, aggregate income has to go down to exactly where it was before the stimulus program in order to equate supply with demand.
But the critical point now is that the process also works in reverse, but it runs up against a brick wall when the interest rate gets to zero. Suppose incomes drop for some exogenous reason (like, for example, that housing prices collapse and throw the banking system into disarray). When incomes drop, the demand for money goes down. Therefore interest rates go down, and a bunch of businesses suddenly want to build factories.
So far, so good, but suppose that demand for commercial construction (and all the other demand that results from lower interest rates) doesn't create enough income to replace that which was lost. In theory, interest rates should go down even further, but suppose the interest rate goes all the way down to zero, and there still isn't enough aggregate income. There could be a very large excess supply of money, but interest rates can't go down any further, and thus incomes won't go up any further, and there is nothing to increase money demand and relieve that excess supply. (And please note that the interest rate on 3-month T-bills Tuesday was approximately zero.)
Now suppose the government institutes a stimulus program to raise incomes. As incomes rise, the demand for money increases. And then what happens? Well, nothing. There is an excess supply of money, and part of that excess supply gets used up by the new demand, but some of it remains – provided the stimulus program is not too large – and the interest rate remains at zero, and there is no reason for anyone to reduce investment, and there is no offsetting decline in income: aggregate income has risen; the stimulus has worked.
But suppose the stimulus program is too large. In that case you can think of the stimulus as being in two parts. The first part is just enough to use up the excess supply of money, and that part will raise incomes by some amount. The second part will create an excess demand for money, and ultimately it won't raise incomes any further. Overall, therefore, incomes will rise to a certain level and no further. But that certain level is still higher than where they were before the stimulus program. Thus the stimulus program has again been successful in raising incomes.
QED, if Professor Fama is using the word "funds" to mean "money" in the literal sense, I wonder if he will explain what he does mean.
I'll conclude with another point concerning the savings-investment equation that Professor Fama uses. In the National Income and Product Accounts, that equation holds more or less by definition. To the extent that there is causation involved, that causation seems to go from investment to savings rather than the other way around. In other words, any increase in investment immediately and automatically creates the increase in savings to finance it. In the national accounts, savings is a residual calculated by subtracting consumption from income. Consumption comes from the product side of the accounts; income comes from the income side. When an increase in investment takes place, it is entered as an increase in income on the income side, and it is entered as an increase in investment on the product side. In other words, income increases, but consumption does not. By definition, therefore, savings increases. So whenever a business chooses to invest, savings must necessarily increase as a result.
I'll leave you to ponder that argument. To be honest, I don't really buy it. I think there is an inherent flaw in national income accounting that allows a bit of Keynesian sophistry, and perhaps I'll write about that in the future. I'd rather fall back on my earlier argument about how the people in the chain from those who receive the government stimulus end up saving the total amount of the stimulus. I don't see how that argument can be refuted – again, unless Professor Fama means something different from what he says. And if he means money, I think Nick and I have pretty much buried his argument.
DISCLOSURE: Through my investment and management role in a Treasury directional pooled investment vehicle and through my role as Chief Economist at Atlantic Asset Management, which generally manages fixed income portfolios for its clients, I have direct or indirect interests in various fixed income instruments, which may be impacted by the issues discussed herein. The views expressed herein are entirely my own opinions and may not represent the views of Atlantic Asset Management.