Link to the Column on Quartz

Here is the full text of my 23d Quartz column, “QE or Not QE: Even Economists need lessons in quantitative easing, Bernanke style,” now brought home to supplysideliberal.com. It was first published on May 14, 2013. Links to all my other columns can be found here.

If you want to mirror the content of this post on another site, that is possible for a limited time if you read the legal notice at this link and include both a link to the original Quartz column and the following copyright notice:

© May 14, 2013: Miles Kimball, as first published on Quartz. Used by permission according to a temporary nonexclusive license expiring June 30, 2014. All rights reserved.

Martin Feldstein is an eminent economist. In addition to being a prolific researcher, he served as head of US president Ronald Reagan’s Council of Economic Advisors, and made the National Bureau of Economic Research (NBER) what it is today—an institution that Paul Krugman called “the old-boy network of economics made flesh.” (I am one of the many economists who belongs to the NBER.) But Feldstein was wrong when he wrote in the Wall Street Journal last week, “The time has come for the Fed to recognize that it cannot stimulate growth,” in an op-ed headlined “The Federal Reserve’s Policy Dead End: Quantitative easing hasn’t led to faster growth. A better recovery depends on the White House and Congress.”

“Quantitative easing” or “QE” is when a central bank buys long-term or risky assets instead of purchasing short-term safe assets. One possible spark for Feldstein’s tirade against quantitative easing was the Fed’s announcement on May 1 that it “is prepared to increase or reduce the pace of its purchases” of long-term government bonds and mortgage-backed securities depending on the economic situation. This contrasts with the Fed’s announcement on March 20 that had only pledged as if the Fed would either keep the rate of purchases the same or scale them back, depending on circumstances. Philadelphia Fed Chief Charles Plosser described this as the Fed trying “to remind everybody” that it “has a dial that can move either way.”

So the Fed sounds more ready to turn to QE when needed than it did before.

Feldstein’s argument boils down to saying, “The Fed has done a lot of QE, but we are still hurting, economically. Therefore, QE has failed.” But here he misunderstands the way QE works. The special nature of QE means that the headline dollar figures for quantitative easing overstate how big a hammer any given program of QE is. Once one adjusts for the optical illusion that the headline dollar figures create for QE, there is no reason to think QE has a different effect than one should have expected. To explain why, let me lay out again the logic of one of the very first posts on my blog, “Trillions and Trillions: Getting Used to Balance Sheet Monetary Policy.” In that post I responded to Stephen Williamson, who misunderstood QE (or “balance sheet monetary policy,” as I call it there) in a way similar to Martin Feldstein.

To understand QE, it helps to focus on interest rates rather than quantities of assets purchased. Regular monetary policy operates by lowering safe short-term interest rates, and so pulling down the whole structure of interest rates: short-term, long-term, safe and risky. The trouble is that there is one safe interest rate that can’t be pulled down without a substantial reform to our monetary system: the zero interest rate on paper currency. (See “E-Money: How Paper Currency is Holding the US Recovery Back.”) There is no problem pulling other short-term safe interest rates (say on overnight loans between banks or on 3-month Treasury bills) down to that level of zero, but trying to lower other short-term safe rates below zero would just cause people to keep piles of paper currency to take advantage of the current government guarantee that you can get a zero interest rate on paper currency, which is higher than a negative interest rate.

As long as the zero interest rate on paper currency is left in place by the way we handle paper currency, the Fed’s inability to lower safe, short-term interest rates much below zero means that beyond a certain point it can’t use regular monetary policy to stimulate the economy any more. Once the Fed has hit the “zero lower bound,” it has to get more creative. What quantitative easing does is to compress—that is, squish down—the degree to which long-term and risky interest rates are higher than safe, short-term interest rates. The degree to which one interest rate is above another is called a “spread.” So what quantitative easing does is to squish down spreads. Since all interest rates matter for economic activity, if safe short-term interest rates stay at about zero, while long-term and risky interest rates get pushed down closer to zero, it will stimulate the economy. When firms and households borrow, the markets treat their debt as risky. And firms and households often want to borrow long term. So reducing risky and long-term interest rates makes it less expensive to borrow to buy equipment, hire coders to write software, build a factory, or build a house.

Some of the confusion around quantitative easing comes from the fact that in the kind of economic models that come most naturally to economists, in which everyone in sight is making perfect, deeply-insightful decisions given their situation, and financial traders can easily borrow as much as they want to, quantitative easing would have no effect. In those “frictionless” models, financial traders would just do the opposite of whatever the Fed does with quantitative easing, and cancel out all the effects. But it is important to understand that in these frictionless models where quantitative easing gets cancelled out, it has no important effects. Because in the frictionless models quantitative easing gets canceled out, it doesn’t stimulate the economy. But because in the frictionless models quantitative easing gets cancelled out it has no important effects. In the world where quantitative easing does nothing, it also has no side effects and no dangers. Any possible dangers of quantitative easing only occur in a world where quantitative easing actually works to stimulate the economy!

Now it should not surprise anyone that the world we live in does have frictions. People in financial markets do not always make perfect, deeply-insightful decisions: they often do nothing when they should have done something, and something when they should have done nothing. And financial traders cannot always borrow as much as they want, for as long as they want, to execute their bets against the Fed, as Berkeley professor and prominent economics blogger Brad DeLong explains entertainingly and effectively in “Moby Ben, or, the Washington Super-Whale: Hedge Fundies, the Federal Reserve, and Bernanke-Hatred.” But there is an important message in the way quantitative easing gets canceled out in frictionless economic models. Even in the real world, large doses of quantitative easing are needed to get the job done, since real-world financial traders do manage to counteract some of the effects of quantitative easing as they go about their normal business of trying to make good returns. And “large doses” means Fed purchases of long-term government bonds and mortgage-backed bonds that run into trillions and trillions of dollars. (As I discuss in “Why the US Needs Its Own Sovereign Wealth Fund,” quantitative easing would be more powerful if it involved buying corporate stocks and bonds instead of only long-term government bonds and mortgage-backed bonds.) It would have been a good idea for the Fed to do two or three times as much quantitative easing as it did early on in the recession, though there are currently enough signs of economic revival that it is unclear how much bigger the appropriate dosage is now.

Does QE work? Most academic and central bank analyses argue that it does. (See for example, work by Arvind Krishnamurthy and Annette Vising-Jorgenson of Northwestern University, and work by Signe Krogstrup, Samuel Reynard and Barbara Sutter of the Swiss National Bank. ) But I am also impressed by the decline in the yen since people began to believe that Japan would undertake an aggressive new round of QE. One yen is an aluminum coin that can float on the surface tension of water. Since September, it has floated down from being worth 1.25 cents (US) to less than a penny now. Exchange rates respond to interest rates, so the large fall in the yen is a strong hint that QE is working for Japan, as I predicted it would when I advocated massive QE for Japan back in June 2012.

Sometimes friction is a negative thing—something that engineers fight with grease and ball bearings. But if you are walking on ice across a frozen river, the little bit of friction still there between your boots and the ice allow you to get to the other side. It takes a lot of doing, but quantitative easing uses what friction there is in financial markets to help get us past our economic troubles. The folks at the Fed are not perfect, but they know how quantitative easing works better than Martin Feldstein does. If we had to depend on the White House and Congress for economic recovery, we would be in deep, deep trouble. It is a good thing we have the Fed.