Paul Krugman writes:
The Doctrine of Immaculate Crowding Out: I’ve written before about the doctrine of immaculate transfer in international macroeconomics, which is a common fallacy but not, I’ve suggested, one that rises to zombie status. There is, however, a somewhat related doctrine — call it the doctrine of immaculate crowding out — which has now, I’d argued, achieved true zombiehood. That is, it keeps coming back no matter how many times you kill it. The most recent example came from John Boehner’s Wall Street talk, where, as Bloomberg puts it,
Boehner’s statement in his Wall Street speech that government spending “is crowding out private investment and threatening the availability of capital” runs counter to the behavior of credit markets. “Look at interest rates. Look at capital spending,” said Nariman Behravesh, chief economist of IHS Inc., a research firm based in Englewood, Colorado. “It’s very hard to come to a conclusion that there’s any kind of crowding out.”
Well, yes. If you believe that government spending has to crowd out private spending by actually changing incentives, namely by raising interest rates, you have to confront the fact that rates are historically very low, even for business borrowers. But it’s now an article of faith on the right that government spending must crowd out private spending, no evidence is necessary. And one must say, alas, that this view has been promulgated by supposedly serious economists. And the thing is, at this point no amount of facts and logic will dislodge that article of faith. It’s pretty hard to kill a zombie.
Let me underscore how truly remarkable a claim this is--that increases in government spending right now have no effect on demand, employment, and production because they crowd out private investment spending.
Ever since at least David Hume's essay "Of the Balance of Trade", the consensus of economists has been that nominal demand depends on propensities to spend: if some agent in the economy decides to spend more, then nominal GDP will rise. In Hume's essay it is an increase in monetary gold generated by a trade surplus that induces an increased pace of spending. In Milton Friedman's work it is an increase in the money supply that does the work of inducing money-holders to spend faster as they try to return their balances to normal. In this, it would seem to us, the government's money is as good as anybody else's.
There have been a number of economists over the years who have claimed that there is something about government purchases that is in some way special: that while greater private propensities to spend, especially those induced by expanding liquidity stocks, boost economy-wide nominal aggregate demand and--in the short-run in which prices and wages are somewhat sticky--real production, expanded debt-financed government purchases, for some reason or other, do not. But note who these skeptical economists do not include. They do not include even the libertarian fringe of old-style nineteenth-century classical economists. They do not include the likes of Jean-Baptiste Say, who wrote in his very first book:
[A] benevolent administration can appropriately make provision for the employment of supplanted or inactive labor in the construction of works of public utility at public expense, as in construction of canals, roads, churches, or the like..."
And they do not include the likes of Frederic Bastiat, who wrote in his essay "What Is Seen and Not Seen":
There is an article in the Constitution which states: "Society assists and encourages the development of labor.... through the establishment by the state, the departments, and the municipalities, of appropriate public works to employ idle hands..." As a temporary measure in a time of crisis, during a severe winter, this intervention on the part of the taxpayer could have good effects... as insurance. It adds nothing to the number of jobs nor to total wages, but it takes labor and wages from ordinary times and doles them out, at a loss it is true, in difficult times...
The liberal classical economists were, it turns out, effectively unanimous in their belief that government spending during a downturn could relieve unemployment and boost production.
As I see it, those who have in the past argued that the government's spending is special have taken one of four roads:
Some have argued that there is a hard potential-output constraint on the economy--that even in the short run real production is limited by resources and technologies, and increases in government purchases that raise nominal aggregate demand merely raise prices and have no effect on real GDP.
Some have argued that there is a hard cash-in-advance constraint on nominal demand: increase government purchases and you increase government borrowing; increase government borrowing and you increase interest rates; the rise in interest rates depresses both investment spending and, through its effects on household wealth, consumption spending and so neutralizes any effect of increased government purchases on spending.
Some appear to have misinterpreted Robert Barro's "Ricardian equivalence" doctrine that government bonds should not be net wealth to apply to debt-financed increases in government purchases as well as to debt-financed reductions in taxes.
And many have pointed out that if inflation expectations are adaptive and the short-run Phillips Curve is linear, then decreases in unemployment below baseline this year must be offset by equal and opposite increases above baseline in the future if the long-run inflation rate is to remain constant.
First, I see no evidence at all, anywhere, of a hard potential-output constraint on the economy right now that means that government purchase-induced increases in nominal GDP would raise prices but not production or employment. If there were, there would be inflation. There is no inflation.
Second, I see no evidence at all, anywhere, of a hard cash-in-advance constraint on nominal demand. Even when interest rates rise sharply when government borrowing increases, that attenuates but does not eliminate the net effects of increased government purchases. As John Hicks wrote in 1937, it is highly implausible to argue that there is a strict cash-in-advance constraint that keeps the velocity of money from rising when nominal interest rates rise:
The direct sacrifice made by a person who holds a stock of money is a sacrifice of interest; and it is hard to believe that the marginal principle does not operate at all in this field.... The demand for money [and hence the velocity of money] depends upon the rate of interest!...
And there are at least two senses in which claims that the interest elasticity of money demand is zero are simply beside the point. The income-theoretic foundations of the demand for money make it highly implausible to argue that the government's ratio of necessary cash holdings to spending are not immensely lower than the private sector's. And it would be a brave economist indeed who would argue right now that U.S. Treasury interest rates have been increased to any significant degree by the--mammoth--U.S. Treasury borrowing of the past three years, and that this non-existent increase has had any depressing effect on investment or consumption spending.
(There is, however, one very important situation in which there is complete interest-rate crowding out. There is complete interest rate crowding out when times are normal and the central bank wishes to make it so. That day may come. But that is not this day.)
Third, Robert Barro's doctrine of "Ricardian equivalence"--which Ricardo did not believe, writing that "the people who paid the taxes never so estimate them... do not manage their private affairs accordingly.... It would be difficult to convince a man... that a perpetual payment of £50 per annum was equally burdensome with a single tax of £1000"--was never supposed to apply to the government purchases side. A bond-financed increase in government purchases of ΔG this year reduces private consumption spending this year by rΔG, not by ΔG, and so raises nominal spending.
Fourth, it is certainly true that under adaptive expectations net benefits from expansionary fiscal policy hang on nonlinearities in the short run Phillips curve, otherwise net benefits depend on expected inflation being well-anchored. But if ever the short run Phillips curve were nonlinear or inflation were well-anchored, it would be now.
Thus any theoretical argument that we should presume that expansionary fiscal policy has no net effect on demand and production would have to be a new argument--to take a form that, to my knowledge, no economist of reputation no has advanced before.
And, indeed, Boehner and his tame economists right now are making such an argument. It is that uncertainty about future government policy has depressed investment spending. Moreover, this uncertainty has depressed business expectations and so depressed investment spending without depressing foreign exchange speculators' expectations and depressing the value of the dollar. It has depressed business investment without depressing bondholders' expectations and causing a rise in treasury yields. It has depressed business expectations and so depressed marginal Q without noticeably depressing average Q and this the stockmarket: stockholders' expectations appear relatively nouyant given economic slack and the downward risks to forecasts.
That the expectations of business entrepreneurs about future policy uncertainty would be so completely disjoint from the expectations of foreign exchange speculators, bondholders, and stockholders appears less than fully plausible to me. Moreover, when you ask entrepreneurs whether they are unusually depressed about uncertainty about future government policy, they say: "no." They say they are depressed about slack demand.
Note that I am not relying in any way on the efficient market hypothesis. I am not claiming that any set of market expectations are rational, or fully reflect information. I am simply observing but if future policy uncertainty is unduly depressing investment, it is doing so in a very unusual way that leaves no fingerprints on any aspects of asset prices whatsoever. I am pointing out the unlikelihood of the claim that entrepreneurs are disjoint from and do not talk to stockholders, bondholders, and foreign exchange speculators.
The economy needs budget balance in the long-run--tax increases and spending cuts, overwhelmingly in government health care programs, to be put in place in the 2020s if not starting in 2015. But for the next three years what the economy needs is for government spending to be pulled forward from the future into the present, and for taxes to be pushed back from the present into the future.