Questions for Joe Gagnon: What is Quantitative Easing, Does it Work, and Should We Do QE3?
In the wake of recent bad employment numbers, people are looking for options to boost the economy. Some have proposed another round of quantitative easing to get us back to the normal rate of growth. I interviewed Joe Gagnon, Senior Fellow at the Peterson Institute for International Economics, about this program. He recently spoke at the Future of the Federal Reserve conference, co-hosted by the Roosevelt Institute and New America Foundation, about what we need to do now to get the economy re-started. He outlines some basics about monetary policy, discusses whether the first rounds of QE worked, and address criticisms of the program and whether a third round of QE will help.
I: How Monetary Policy Works
Mike Konczal: What is a recession, and what does the Federal Reserve normally do in reaction to one?
Joe Gagnon: A recession is when there isn’t enough spending in the economy to keep everyone fully employed. People are thrown out of work and inflation starts to trend downward. What the Federal Reserve does, what monetary policy does, is to lower interest rates to make it easier for people to borrow and less attractive for people to save, which encourages businesses to borrow and invest. That spending gets people working again.
MK: Why does a lower interest rate help us recover from a recession?
JG: If the interest rate, which is the cost of borrowing, is lower, then it is cheaper to borrow. That makes it easier for businesses to borrow. They are then more willing to borrow to build factories, hire workers and build stuff. It also means households are more willing to borrow to buy a house. It’s all about making it more attractive for people to borrow and spend. And conversely, it makes it less attractive for people to save, because they are earning less on their savings. So they are more likely to spend it.
MK: If that’s the case, why doesn’t the Federal Reserve lower the short-term rate?
JG: The short-term interest rate the Fed controls is essentially at zero now. It cannot make the interest rate negative because if you were a saver looking to put money in the bank, and the bank was offering you a negative interest rate, you would say, “No thanks, I’ll hold cash instead,” and stuff your mattress with dollar bills. Dollar bills earn a zero rate of return, and you’d be better off holding that.
MK: Since the Federal Reserve is buying other assets, does QE have the same kind of effect regarding interest rates?
JG: Yes it does. The way to think of monetary policy is that it tilts the playing field between savers and spenders. Normally it just works on short-term interest rates, which encourages corporations to invest more and savers to save less. Quantitative easing impacts longer-term interest rates, but this also has the same effect on savers and spenders. It makes it easier to borrow and less attractive to save.
MK: So did QE2 work? And if so, how can you tell?
JG: It did work. I think QE2 had two elements. One element was of moderate importance, one element was of minor importance. The moderate one is that QE2 convinced markets that the Federal Reserve would not allow deflation or a double dip recession to happen. This is good because it inspired confidence and kept inflation expectations from falling any further. That was the most important step, because it convinced financial markets that the United States wouldn’t turn into Japan, which they were worried about. The element of minor importance was that it lowered long-term bond rates a little bit. It takes a lot of purchases to move these interest rates even a little bit, and QE2 wasn’t big enough to move them dramatically. It’s not nothing, but it is small in the scheme of things.
MK: Why is deflation bad, and what would it mean to be the next Japan?
JG: The interest rate that matters for households and businesses is the real interest rate, which is the nominal interest rate minus the inflation rate. Why is that? Because when prices of everything are going up, when you pay back a loan at a certain interest rate some of that is just eaten up by inflation. The lender doesn’t get in real terms — in terms of goods and services — as much back as he lent because of this inflation. Higher inflation lowers the real rate of interest. Deflation raises the real rate of interest. If deflation gets big enough, when you hit the zero bound, you have a positive rate of interest and get into deflationary spiral, where the real rate of interest is choking the economy and choking the economy makes the deflation worse, which is a vicious cycle.
Japan has been in a mild version of that. It hasn’t accelerated, but it hasn’t gotten out of it either.
MK: So how do you know that QE has worked? What kind of studies are conducted, and how do they draw their conclusions?
JG: I have a paper that looked at two things. When the Fed made announcements on QE1, what happened to bond yields? Yields on the things the Fed was buying went way down, but yields on things the Fed wasn’t buying also went down. All yields went down. So that was one piece of evidence. As for the other piece of evidence, we looked back thirty years and ran a regression of how the government’s net supply of long-term bonds affects bond yields. We found when the government issues more long-term bonds, bond yields increase. When the government buys back long-term bonds, bond yields go down. QE, really, is like the Treasury buying back long-term bonds and issuing short-term bonds. There’s a long history of this, including in non-crisis times. So for both pieces of evidence, when the government buys long-term bonds and issues short-term bills, it can push down the yield curve.
MK: So QE2 helped with the job growth of the past year?
JG: It definitely contributed by relieving businesses’ fears of a double-dip recession and deflation. This helped with some growth in the economy. But really, not enough. The Fed has not been aggressive enough, it has been too timid.
II: Addressing Criticisms of Quantitative Easing
Mike Konczal: I want to talk about some criticisms of QE that have come up. There’s an argument that QE generally can provide a floor on how bad the economy will go, but can’t, by itself, get us back to full employment and trend growth. What do you make of that argument?
Joe Gagnon: There’s no reason to think that there’s any limit to the effectiveness of quantitative easing or monetary policy. There’s no theoretical or practical reasons to think that monetary policy will stop working. They’ve just been too timid.
MK: What about the argument that QE has caused massive inflation?
JG: Look at measures of underlying inflation. Wage inflation is the most important thing. Think about slow-moving prices that are inside the US economy.
MK: Wage inflation is not spiraling.
JG: Right. Commodities have spiked, but oil has come back. If monetary policy was to respond to commodities, you’d have really unstable policy. You don’t want to ignore any price, but you want to smooth through the noisy ones. Wages are an important one. Monetary policy works through the labor market. If inflation is too high, we throw people out of work to cool the economy and keep wages low. And if inflation is too low, you want to hire more people to get the economy going faster. If you look at wages there’s no worry about any future inflation.
MK: There are two other criticisms of QE2 that go in different directions. One is that the new capital has just sat on banks’ balance sheets and not impacted the recovery. The other is that, with rates being so low, QE just helps create asset bubbles. How would you address these?
JG: They are both different. The first is rather easy. We know that the banks aren’t lending the money out. There’s little the Fed can do about that. So QE2 wasn’t aimed at the banks. It would be good if the banks lent that money out, because we wouldn’t then have to do so much QE. But the banks aren’t. Given that they aren’t, we need QE3 to push down other prices to work through the bond market or the foreign exchange markets.
MK: Can you talk a bit more about these alternative channels?
JG: The basic channel is the bond market. The Fed is buying up long-term bonds and that pushes interest rates down on those bonds. That’s what makes it attractive for people to borrow. The market then does some arbitrage. The equity market looks at the bond market and says “oh, well, long-term bond rates are low, so we are going to discount future dividends and profits differently.” This makes the value of stocks more attractive and raises their values. And this encourages businesses to invest. Also, international investors look at rates of return in other countries, and they say “these other countries have higher rates of return than in the U.S.,” so that pushes the dollar down. These aren’t direct channels of monetary policy, but they are linked.
MK: Some might interpret that as saying QE deliberately creates a stock bubble, which makes them nervous.
JG: Well, there’s two important things to keep in mind. First of all, the harm of a bubble arises almost entirely if it’s leveraged. We have to make sure, through financial regulations, that people are not borrowing to buy stocks. And they aren’t as far as we know. The tech bubble wasn’t leveraged, and when it burst it had little effect. The housing bubble was leveraged, and it had a major effect. If we have an equity bubble, and equities don’t seem to be priced unusually high, but even if we did, we’d want to make sure it wouldn’t cause harm when it unwinds.
The second is that it’s not clear that it is even a bubble if monetary policy is working through interest rates, as monetary policy always does. You need equities to be priced highly when there’s a recovery, you want to encourage people to invest. Moreover, by creating a healthier economy, monetary policy can increase the fundamental value of equities, which by definition is not a bubble.
MK: Some like Raghuram Rajan and Thomas Hoenig have taken the concern about bubbles further and argued that rates being “unnaturally low,” specifically short-term rates at zero for too long, creates conditions for moral hazard and distorts asset prices. How do you respond to this?
JG: The cause of unusually low interest rates right now is two things: first, households and small businesses are repairing their balance sheets and do not want to borrow more; and second, developing economies — led by China — are funneling massive amounts of government money into the US and European economies. The correct response of monetary policy is to push interest rates as low as possible. Rajan and Hoenig are confusing cause and effect.
MK: Recently, Rajan has argued the morality of monetary policy, saying that QE2 hurts “the patient and uncomplaining saver.” What do you say to the argument that QE (and monetary policy generally) is being too unfair to savers?
JG: This is always an effect of monetary policy, which benefits borrowers when the economy is weak and savers when the economy is strong. Savers do not have any right to a specific rate of return and they are free to spend the money or invest in physical capital or equities if they want a higher rate of return. In any event, the distributional effects of monetary policy are much smaller than those of fiscal policy, which transfers from future generations to today’s generations.
MK: One last argument against QE. The economist Richard Koo looks at the US recession and, comparing it to Japan and the Great Depression, says we suffer from a debt overhangthat has devastated the balance sheets of households and firms.
JG: I totally agree with him.
MK: His critique then follows that QE, and monetary policy more generally, encourages people to take on more debt, but since everyone has too much debt, QE can’t help. Interest rates are at record lows, yet consumers are de-leveraging, implying that consumers want to shed debt regardless of how cheap it is. How would you address this?
JG: People don’t want to borrow as much as they normally would. But QE helps by allowing people to repair their balance sheets. Households can refinance their debts into lower rates. Corporations are issuing long-term bonds at record-low rates. They are paying off older, higher-yield debts. This repairs their balance-sheets and increases their value. For households, if you can get a lower rate that reduces your payments. That’s a huge improvement to you.
Once you repair your balance sheet, you are prepared to spend sooner. So even if QE doesn’t immediately raise borrowing as much as it would in normal times, it is making it easier for you to repair your balance sheet and get the economy to a place where you’ll spend faster. It’s absolutely all the more essential because of the balance sheet nature of the recession. You want to raise asset prices to make people feel better off. You want to lower interest rates so people can refinance and repair their balance sheets. QE is the best way we have of addressing Richard Koo’s concerns.
III: The Potential for QE3
MK: What should QE3 look like?
JG: A lot of the benefit of telling the markets that you aren’t going to allow deflation is already out there. You could re-enforce that, but the major effect is out there already. They would need to do a bigger number. There’s no point in doing it unless it’s at least $1 trillion dollars.
MK: The floor is set. The market is reassured against fears of deflation. Can QE3 return to trend and full employment?
JG: While QE2 had good effects, it was too timid. A QE3 needs to be bigger than QE2 — you want to signal a larger amount. A trillion dollars sounds like a big number, but it isn’t like a trillion dollar tax cut. All it is is a swap of two different assets. Buying one kind, selling another.
MK: Are the effects of QE3 amplified by a short-term stimulus in the form of infrastructure spending and employee tax cuts?
JG: Absolutely. The end of the payroll tax cut, the winding down of the stimulus spending, and the termination of extended unemployment benefits are conspiring to create a large fiscal drag on the economy in 2012. I would recommend extending or even enlarging the payroll tax cut for 2012 and renewing extended unemployment benefits for at least 12 months. QE3 can help to ensure these actions have the best possible effect.
As for infrastructure spending, I think it needs to be analyzed in two categories. First, anything that can really be built (and not just planned) in 2012 could be viewed as near-term stimulus in place of (or in addition to) a payroll tax cut. Second, longer-term infrastructure projects (many of which I support) need to be based on long-term needs and in the context of a long-term fiscal plan to stabilize our national debt, not the near-term state of the recovery. The dividing line between the near term and the long term is probably 2013, with spending in 2013 still contributing usefully to near-term recovery.
MK: Let’s say this all goes sideways. What kinds of risks is the Federal Reserve taking on by doing more QE?
JG: We are not taking on a lot of risks. People see large numbers and they get scared. The only risk here is a small risk on future profits of the Federal Reserve. Because of QE1 and 2, they are making record profits. They are making these profits by buying bonds that yield 3 or 4% and funding them with near 0% money. They hand it back to the Treasury every year. Down the road, if this works, and the economy starts growing, then the Fed will have to ease off the accelerator and start raising interest rates. Then, in the new world that we are in, the Fed is going to have to fund these QE assets at the new higher interest rate. If short-term rates go high enough, the Fed could make a loss, but it would only be after years of excess profits. That’s the biggest risk, and it isn’t that big.
Mike Konczal is a Fellow at the Roosevelt Institute.