How to Think about QE3

By Arnold Kling 

Here are three possible outcomes.

Alan Blinder, a prominent Democratic economist, once complained that when economists are in broad agreement with high confidence (on the issue of free trade, for example), they are never consulted. Instead, journalists and laymen ask economists to make pronouncements only on topics where there is little confidence and strong disagreement.
Recently, I have encountered this “Murphy's Law of economic advice” in the form of people asking me what I think of QE3, which is the name given to the Federal Reserve's latest initiative to stimulate the economy by purchasing long-term debt instruments, including mortgage-backed securities. (“QE” stands for “quantitative easing,” and the “3” refers to the fact that this is the third such initiative undertaken by the Fed since the financial crisis sent the economy into recession.)


When it comes to QE3, while there is no solid professional consensus on which to draw, I can boil the issue down to two questions. First, will QE3 raise inflation? That is, will inflation be higher than it would have been without QE3?
Second, will the inflation-unemployment tradeoff be favorable? That is, will we get a large gain in employment with relatively little pain in terms of higher inflation? Or will we get a lot of pain for little or no gain?

Foreign Currency Intervention

Quantitative easing is a relatively new central bank tactic. My view of how it works is based on what economists have found studying the intervention in central banks in markets for foreign currency, something that has been going on for a long time.

Quantitative easing looks a lot like sterilized intervention, which many economists think is ineffective.

Suppose that the Fed wanted to weaken the dollar relative to foreign currencies. It could take steps to create dollars and to use those dollars to purchase securities denominated in, say, euros. The intent would be to drive up the price of euros relative to that of dollars. China has been accused of using these sorts of tactics to drive down the value of its currency. Presumably, the goal of driving down the value of your own currency relative to foreign currencies is to make your exports more competitive.
Foreign currency intervention can take three forms: Unsterilized intervention, sterilized intervention, and sterilized intervention with a credibility kicker. It will help to understand how these differ, because quantitative easing looks to me a lot like sterilized intervention, which many economists think is ineffective.
If the Fed were to undertake an unsterilized intervention to weaken the dollar, it would purchase foreign securities using its power to create money (bank reserves) at the stroke of a pen (or keystroke of a computer). This would increase the supply of money in the United States. With more money chasing the same amount of goods, there would be higher inflation.
Sterilized intervention is an attempt to lower the value of the dollar without adding to the money supply or to inflation. With sterilized intervention, the Fed would offset its purchase of foreign securities by selling some of its portfolio of U.S. Treasury securities, so that there would be no net increase in the U.S. money supply.
Most economists are skeptical that sterilized intervention works. They argue that when the Fed makes no net change in the amount of money outstanding, the markets pretty quickly undo the effect on the relative currency values. Without changing the U.S. money supply, the Fed changes nothing about the fundamental relationship between the value of the dollar and the value of the euro. When the Fed exchanges some of its Treasury bonds for euro-denominated bonds, the markets adjust to neutralize the impact. For example, perhaps some companies that formerly issued dollar-denominated bonds now issue euro-denominated bonds, so that the overall market supplies remain unchanged. Alternatively, some investors who formerly held euro-denominated bonds switch to dollar-denominated bonds, without requiring any price change.
Suppose, however, that while it undertakes sterilized intervention, the Fed announces that it is determined to see the value of the dollar decline. This is what I mean by sterilized intervention with a credibility kicker.
If sterilized intervention includes a credibility kicker, then companies will not be so willing to issue euro-denominated bonds and investors will not be so willing to buy dollar-denominated bonds. When the Fed's commitment to lowering the value of the dollar has credibility, the actions of private investors serve to reinforce rather than counteract the Fed's goal.
I have been speaking as if a credibility kicker can be created simply by making an announcement. However, announcements are not necessarily believed. In fact, some sorts of announcements are rarely believed. When a country wants to reduce inflation and strengthen its currency, the mere announcement of that intent is almost never believed.

When the Fed's commitment to lowering the value of the dollar has credibility, the actions of private investors serve to reinforce rather than counteract the Fed's goal.

When will a central bank's announcements be believed? One theory is that a credibility kicker for sterilized intervention has to include a threat, or perhaps even a commitment, to undertake unsterilized intervention. That is, investors have to believe that when push comes to shove, the Fed will allow the money supply to increase in order to achieve its goal of lowering the value of the dollar. To the extent that investors need a commitment to be demonstrated, then to take advantage of a credibility kicker the Fed has to undertake unsterilized intervention.

Quantitative Easing as Sterilized Intervention

Quantitative easing differs from foreign currency intervention in that the Fed is intervening in the market for domestic securities. It exchanges short-term domestic securities for long-term domestic securities, including mortgage-backed securities. The idea is that lower costs of mortgages and other forms of long-term borrowing will serve to increase spending.
Since the financial crisis began, the Fed has undertaken massive purchases of domestic securities. Other things equal, this would have produced a doubling or tripling of the money supply. However, the Fed has taken offsetting measures, including paying interest on bank reserves. As a result, the money supply has not increased anywhere close to proportionately to the rise in Fed purchases of securities. Because banks have dramatically increased their holdings of excess reserves, the Fed's balance sheet activities have amounted to sterilized intervention.
The latest round of purchases was accompanied by announcements that appeared to signal an intention by the Fed to continue to intervene until employment starts to increase more rapidly. Thus, I view QE3 as sterilized intervention with an attempt at a credibility kicker.
Because quantitative easing is unprecedented, it will be difficult for markets to assess the credibility kicker. It is hard to tell how much the Fed wants to see excess reserves continue to balloon versus how much new money it wishes to inject into the economy.
If investors are not convinced that the path for money growth has changed in any fundamental way, then markets may soon settle back into their previous pattern. In that case, QE3 will be, like sterilized intervention in currency markets, a futile gesture.

The Phillips Curve, Revisited

Suppose that QE3 does achieve the objective of raising inflation relative to what would have taken place without QE3. Will this be a good thing?

If investors are not convinced that the path for money growth has changed in any fundamental way, then markets may soon settle back into their previous pattern.

In the 1960s, economists believed that there was a tradeoff between unemployment and inflation. If we had 5 percent unemployment and 2 percent inflation, then an increase of 1 percentage point in inflation, to 3 percent, would reduce the unemployment rate by about 1 percentage point, to 4 percent. A diagram representing this tradeoff is called the Phillips Curve.
In the 1970s, the Phillips Curve went haywire. Instead of seeing unemployment go down as inflation went up, we saw both unemployment and inflation rise together. As a result, the Phillips Curve was abandoned by many academic macroeconomists.
Since 2008, many macroeconomists have begun to speak again in Phillips-Curve terms. That is, they believe that additional inflation would help to reduce unemployment.
It is possible that these latter-day Keynesians, or latter-day Phillipsians, are correct. Moreover, it is possible that they are correct that relatively small increases in inflation would produce relatively large increases in employment.
However, I personally do not believe in Keynesian macro or Phillipsian macro. (See part three of my essay series on “a million mutinies.”) My guess is that if the Fed does succeed in raising the rate of inflation, the effect on employment will be minimal.

Three Scenarios

Overall, the possible outcomes are as follows:

1. If QE3 operates like sterilized intervention, we will see some short-term reductions in the cost of certain types of long-term borrowing, but these effects will soon dissipate. In the end, the economy will proceed along the same path as it would have without QE3. I put the probability of this at 20 percent.

2. If QE3 operates like unsterilized intervention or like sterilized intervention with a credibility kicker, and the Phillipsians are correct, then QE3 will put us on a slightly higher inflationary path with significantly lower unemployment. I put the probability of this at 30 percent.

3. If QE3 operates like unsterilized intervention or like sterilized intervention with a credibility kicker, but the Phillipsians are incorrect, then QE3 will put us on a significantly higher inflationary path with little effect on unemployment. I put the probability of this at 50 percent.

Unfortunately, assessing the actual result will be difficult, even after the fact. It will be tempting to compare the state of the macroeconomy 12 to 18 months from now with the state of the economy today. However, the relevant comparison is between the state of the economy 12 to 18 months from now and what it would have been without QE3. This is not an experiment that we are capable of running.
Arnold Kling is a member of the Financial Markets Working Group at the Mercatus Center of George Mason University. He writes for econlog, part of the Library of Economics and Liberty.