Why QE2 worked
‘To economists reading this essay in 2010, perhaps the most remarkable single fact to note about monetary policy at the end of the interwar period” — the author, Columbia University economist Charles Calomiris, is of course also talking about the end of the Great Depression — “is that its architects were, for the most part, quite pleased with themselves. Far from learning about the errors of their ways during the interwar period, Fed officials congratulated themselves on having adhered to appropriate principles, and to the extent that they were self-critical, it was because they thought that they had been too expansionary.”
Almost all economists today agree that monetary policy during the Depression, especially its early stages, was disastrously tight, indeed that this contractionary policy is the principal reason the Depression became Great. But perhaps we should not judge the central bankers of America in the 1930s too harshly. For one thing, as Calomiris notes, smart economists are still arguing about the precise nature of the Fed’s policy mistakes. He himself presents evidence against the received view that the Federal Reserve precipitated the “recession within a recession” of 1937 by raising banks’ reserve requirements and thus discouraging lending.
More important — and more disturbing — is that it is not at all clear that we have learned from the mistakes of the 1930s. Those central bankers believed that money was easy because interest rates were low and the monetary base (the supply of money under the Fed’s control) had expanded. They worried that further easing would reduce confidence in the dollar. British economist R. G. Hawtrey, writing in the late 1930s, described the climate of opinion in his country at the start of the decade: “Fantastic fears of inflation were expressed. That was to cry ‘Fire! Fire!’ in Noah’s Flood.” The economy was actually deflating, not inflating. Under the influence of the “real-bills doctrine,” some central bankers believed that the money supply should respond only to traders’ need for credit. Anything else would only fuel speculative excess.
Today’s inflation hawks employ the same reasoning that those firefighters did. And they are not wholly wrong. Easier money can lead to a destabilizing run on the currency. Inflation can be associated with low real interest rates and an expanded monetary base. But not always: Not in the 1930s, and almost certainly not today, either. The late Milton Friedman, perhaps the most famous inflation hawk of his generation, spotted the fallacy in his analysis of 1990s Japan: Low interest rates can also be a symptom of an excessively tight monetary policy that has choked off opportunities for growth. A looser policy, by increasing expectations of future economic growth, could actually raise real interest rates.
To see why changes in the monetary base are also an unreliable guide to whether money is loose or tight, I’m afraid it’s necessary to look at an equation. Friedman and others familiarized us with the equation of exchange: MV=PY. What that means is that the money supply (M) times the speed with which money changes hands (V, for velocity) must equal the price level (P) times the size of the real economy (Y). If velocity holds constant and the money supply goes up, either prices must go up or the economy must expand or both.
If, on the other hand, velocity drops — if people have an increased desire to hold money balances — then either prices must drop or the economy must shrink or both. And prices, especially wages, are sticky. They won’t automatically and evenly fall in response to a shock. So at least some of the reduction in PY, and maybe a lot of it, will have to come from real economic contraction. What this suggests is that if velocity drops unexpectedly, stabilizing the economy will require increasing the money supply to make up for it. Another way of putting it is that if the demand for money balances increases, the money supply has to grow to accommodate it. If the Fed does not increase the money supply, its inaction in the face of changing economic conditions amounts to passive tightening.
The money supply is itself the product of the monetary base (B) and the “money multiplier” (m), which measures how changes in the base are being converted into changes in commonly used monetary assets such as checking and money-market accounts. So — I promise this is the last equation — BmV=PY. David Beckworth, a conservative economics professor at Texas State University who maintains a blog, has shown that at the height of the financial crisis in late 2008, velocity dropped significantly and the money multiplier fell off a cliff. The monetary base grew a lot too: The inflation hawks are right about that. But it grew enough to offset only the fall in the money multiplier. It didn’t offset the fall in velocity.
Beckworth’s interpretation: The Fed was increasing the base to save the financial system, not the economy overall. At the same time the Fed was injecting money into the financial system, it instituted a policy of paying banks interest on their reserves — a policy that made them less likely to lend out those reserves (and thus kept m down). That policy helped the banks’ solvency but not the economy. The financial system would almost certainly have benefited more from a broader policy: A rising tide lifts all banks. Both m and V remain well below their pre-crisis levels.
Beckworth is among those economists who argue that Fed policy should aim to stabilize the growth of nominal spending — roughly, the total value of the economy in current dollars, which is equal to PY (and therefore to MV and BmV). That policy is superior to trying to grow the base at a steady rate, a much-discussed idea in the past, because it allows the base to change in response to changes in the money multiplier and velocity. It is superior to trying to hold inflation constant because it allows the price level to respond to changes in productivity. It would create a stable environment in which economic actors could make their decisions and contracts.
Josh Hendrickson — an economist at the University of Toledo, and like Beckworth a right-leaning blogger — has shown that the Fed did a pretty good job of stabilizing the growth of nominal income at roughly 5 percent per year during the “great moderation” that lasted from the mid-1980s until the current recession. (Although Beckworth notes that growth was slightly above trend during the housing boom, for which he faults the Fed.) Most debts — notably, most mortgage debts — are contracted in nominal terms, with no inflation adjustment. If people are used to 5 percent growth in nominal incomes each year and make their arrangements accordingly, then an unexpected drop will make their debt burdens heavier and also make them reluctant to make plans for a suddenly uncertain future.
That’s what happened during the recent crisis. Scott Sumner — yet another right-of-center econoblogger, this one based at Bentley University — often notes that in late 2008 and early 2009, we saw the sharpest fall in nominal income since 1938. In his view, much of what we think we know about the recession of 2007–09 is wrong. Not only has money not been loose since the crisis began, but tight money is the fundamental reason the recession was so severe and the recovery has been so halting. He argues that it was more fundamental than the housing bust, since residential-construction employment started falling long before the crisis hit.
An alternative theory of the crisis goes something like this: While a recession may have been inevitable, it was the Fed’s passive tightening that made it a disaster. The recession began in late 2007, although many observers knew it only after the fact. The Fed passively tightened mildly in mid-2008. In the fall of 2008, the financial crisis caused velocity (and the money multiplier) to drop dramatically — in part, perhaps, because political and financial leaders were scaring everyone. The Fed did not act aggressively enough to accommodate the increased demand for money balances, and what had been a mild recession became a severe one.
As panic subsided, velocity stopped falling, and the economy then began to recover. But in mid-2010, the eurozone crisis resulted in a flight to the dollar. Increased demand for dollars again had a contractionary effect, and the Fed took months to respond to it. Finally, in the late summer, it began letting it be known that it would dramatically increase the money supply — an initiative called quantitative easing, or QE2, the first QE having been the injection of money into the financial system in late 2008 — and then, in the fall, it followed through.
A number of justifications were offered for QE2. The most plausible was that if people expected it to move nominal income to a permanently higher path, asset values would increase to reflect this higher expected income stream. Both consumers and investors would then spend more money. The demand for money, in other words, would decline at the same time as the supply increased, restoring equilibrium.
As the market became convinced that the Fed planned to act, both expectations of inflation and expectations of real growth increased: the former indicated in the spreads between inflation-indexed bonds and non-indexed bonds, the latter in real interest rates. Nominal income thus moved closer to trend, if not as much as it would have with a bolder Fed initiative. (An explicit announcement that the Fed is willing to do what it takes to restore the trend might itself change expectations enough to make great exertions by it unnecessary.) Stocks picked up too. QE2, though flawed, worked. It began to work even before being formally implemented.
Conservative politicians and conservative pundits nonetheless blasted it. They seized on the rise in long-term interest rates as proof it was not stimulating the economy: the precise kind of confusion about the meaning of interest-rate changes that Friedman had warned against. They blamed QE2 for a boom in commodities prices, largely ignoring the role of rising Asian demand for those commodities. And they fretted about runaway inflation, even though, in the aftermath of QE2, the spreads mentioned above were forecasting inflation rates below 2 percent for years to come — a lower average inflation rate than in each of the last five decades.
The argument for stabilizing the growth of nominal income implies that if the Fed overshoots its target in one year it should undershoot it the next, and vice versa: The key is to keep to the targeted long-term trend. Any rational person would want as much of the growth in nominal income to be made up of real economic gains, and as little of it of inflation, as possible. But a modest uptick in inflation that helps to bring nominal income back to trend is better than staying below trend.
There is a reasonable argument — made by Beckworth, among others — that over time the Fed should gradually reduce the average growth of nominal income from 5 percent per year to something closer to 3 percent. The idea would be to move the economy to a lower average inflation rate, or even to a mild and gradual deflation. But such a move should neither be abrupt nor begin in the midst of a shaky recovery, let alone a crisis.
In warning about inflation, conservatives are crying “fire” in, if not Noah’s flood, at least a torrential rain. It may be that they are stuck not so much in the 1930s as in the 1970s — the time when conservatism forged much of its current outlook on economics, and a time when monetary restraint was badly needed. Conservatives also tend to think that loosening monetary policy is a kind of intervention in free markets, and therefore to be suspicious of it. But this is an error. Professor Hendrickson points out that in a system of free banking, with competitive note issue rather than a central bank, the desire for profit and the need for solvency would lead to the supply of banknotes roughly equaling the demand. In a fiat-money regime such as the one under which we, for better or worse, live, a central bank’s withholding of a sufficient supply of money is just as much of an intervention in the economy as its overproduction of it.
The economy, post-QE2, seems to be on the mend. But events overseas — a renewed eurozone crisis, or trouble in the Middle East — could again boost demand for dollars. If so, will the Fed accommodate that demand? Or will it be dissuaded by the vehement criticism its efforts to date have already drawn from conservatives? Congressional Republicans are at the moment blocking the confirmation of Peter Diamond, a Nobel Prize–winning economist, to the Federal Reserve’s board of governors because of their opposition to the inflation they believe is just around the corner.
We are not likely to see a second Great Depression. But it would be tragic if conservatives, moved by hostility to the Fed, led it to repeat, even on a smaller scale, the worst mistakes in its history.