[size=1.3em]Large-scale asset purchases

Let me now turn to the second form of unconventional monetary policy, large-scale asset purchases. The goal of large-scale asset purchases, or LSAPs, is the same as for conventional policy actions and forward guidance: to drive down longer-term interest rates, and thereby boost economic growth. How do LSAPs work? First, let me tell you when they wouldn’t work. In a hypothetical world of perfect financial markets, LSAPs would have essentially no effect on asset prices or the economy. In such a world, the price of an asset depends solely on its expected future returns, adjusted for risk. If the price of a specific asset deviated from this level, arbitrageurs would swoop in to take advantage of the discrepancy, knowing that the price would inevitably return to its proper level. Suppose the Fed were to step in and buy large amounts of an asset class, say, for example, Treasury securities. In that case, other investors would freely sell their holdings and rebalance their portfolios accordingly. But, asset prices would not change at all. And there would be no impact on the broader economy.

The reason LSAPs work is that financial markets are not perfect. Decades ago, James Tobin and Franco Modigliani pointed out that markets are to a certain degree segmented. Some investors, such as pension funds, have “preferred habitats” for their investments. For example, a pension fund might prefer longer-term securities to hedge its longer-term liabilities. Thus, the supply and demand of assets in these habitats can affect prices because that pension fund is not going to start buying short-term securities just because the prices of longer-term securities rise.

Now, if the Fed buys significant quantities of longer-term Treasury securities or mortgage-backed securities, then the supply of those securities available to the public falls. As supply falls, the prices of those securities rise and their yields decline. The effects extend to other longer-term securities. Mortgage rates and corporate bond yields fall as investors who sold securities to the Fed invest that money elsewhere. Hence, LSAPs drive down a broad range of longer-term borrowing rates. And lower rates get households and businesses to spend more than they otherwise would, boosting economic activity.

LSAPs can also affect interest rates by signaling that the central bank is determined to ease monetary conditions (see Bauer and Rudebusch 2012, Christensen and Rudebusch 2012, and Krishnamurthy and Vissing-Jorgensen 2011). Effectively, the central bank is putting its money where its mouth is. Thus, LSAPs reinforce forward guidance. For this reason, I view these two types of unconventional monetary policy as complementary.

The use of LSAPs goes back to a 1961 initiative with the catchy name of Operation Twist, an effort by the Fed and the Kennedy Administration to drive down longer-term interest rates. More recently, in late 2008 and 2009, the Fed purchased over $1.7 trillion of longer-term Treasury bonds and mortgage-backed securities, a program often referred to as QE1. In November 2010, the FOMC announced an additional $600 billion of longer-term bond purchases—QE2. And, two months ago, we got QE3 when the FOMC announced that the Fed would buy an additional $40 billion in mortgage-backed securities every month until the outlook for the job market improves substantially.

Other central banks have also carried out large-scale asset purchase programs. The Bank of Japan began a large-scale asset purchase program in 2001. In its most recent program, launched in 2010, it has bought roughly $1.1 trillion in Japanese government bonds and other assets. In March 2009, the Bank of England announced an LSAP program that was later raised to the equivalent of roughly $600 billion in purchases mostly of British government bonds. Both of these central banks have continued and expanded their asset purchase programs in the past year.

[size=1.3em]The effects of unconventional monetary policy on the economy

A great deal of research has analyzed the effects of forward policy guidance and large-scale asset purchases on financial conditions and the economy. As I mentioned before, forward policy guidance has proven to be effective at lowering expectations of future interest rates (see Swanson and Williams 2012 and Woodford 2012). Similarly, the evidence shows that LSAPs have been effective at improving financial conditions as well.

To be precise, the estimated impact of a $600 billion LSAP program, such as QE2, is to lower the 10-year Treasury yield by between 0.15 and 0.20 percentage point (see, for example, Williams 2011, Krishnamurthy and Vissing-Jorgensen 2011, Hamilton and Wu 2012, Swanson 2011, Gagnon et al. 2011, and Chen, Curdia, and Ferrero 2012). It is around the same magnitude as the effects of forward policy guidance, and about how much the yield on 10-year Treasury securities typically responds to a cut in the fed funds rate of three-quarters to one percentage point (see Chung et al. 2012 and Gürkaynak, Sack, and Swanson 2005). So, by that metric, LSAPs have big effects on longer-term Treasury yields.

By pushing down longer-term Treasury yields, forward guidance and LSAPs have rippled through to other interest rates and boosted other asset prices, lifting spending and the economy. For example, mortgage rates have fallen below 3½%, apparently the lowest level since at least the 1930s. Thanks in part to those rock-bottom rates, we’re at long last seeing signs of life in the housing market. Likewise, cheap auto financing rates have spurred car sales. And historically low corporate bond rates encourage businesses to start new projects and hire more workers.

In addition, low interest rates help to support asset prices, such as the value of people’s homes and their retirement funds. All else equal, households are more likely to consume if their wealth is growing rather than falling. Stronger asset prices support consumption because they make people feel wealthier and more confident. And that in turn helps boost the economy.

Finally, although it’s not our main intention, these unconventional policies have also had an effect on the dollar versus foreign currencies. When interest rates in the United States fall relative to rates in other countries, the dollar tends to decline as money flows to foreign markets with higher returns. One estimate is that a $600 billion program like QE2 causes the dollar to fall by roughly 3 or 4% (see Neely 2011). That helps stimulate the U.S. economy by making American goods more competitive at home and abroad.

I’ve argued that forward guidance and LSAPs invigorate the economy by lowering interest rates and improving financial conditions more generally. But just how big are these effects? That’s not easy to answer. Financial markets react instantly to FOMC announcements, so it’s relatively easy to gauge the financial impact of any policy move. By contrast, monetary policy actions affect economic growth, employment, and inflation gradually over time. Thus, the broad economic effects of monetary policy are not immediately obvious. Moreover, data on unemployment and gross domestic product are only collected monthly or quarterly. Many factors besides monetary policy affect these variables. In any particular data release, it’s devilishly hard to separate the contribution of monetary policy from other factors.

To control for these other factors, a researcher must use a macroeconomic model. In some of my own research with staff at the Federal Reserve Board, we used the Board’s large-scale macroeconomic model, which has hundreds of economic relationships built in, for this purpose (see Chung et al. 2012). We estimated that the Fed’s $600 billion QE2 program lowered the unemployment rate by about 0.3 percentage point compared with what it would have been without the program. We also estimated that the program raised GDP by a little over half a percentage point and inflation by 0.2 percentage point. When we considered the combined effects of QE1 and QE2, we found that these programs had a peak effect of reducing the unemployment rate by 1½ percentage points. In addition, we found that these programs probably prevented the U.S. economy from falling into deflation.

Other researchers using different macroeconomic models have found roughly similar effects, although there is a lot of uncertainty surrounding these estimates (see Chen, Curdia, and Ferrero 2012, Kiley 2012, Fuhrer and Olivei 2011, Baumeister and Benati 2010, and Curdia and Ferrero 2011). Part of the uncertainty stems from the fact that changes in longer-term interest rates due to LSAPs may be atypical. That is, they may affect the economy differently than do changes in longer-term interest rates in normal times. That would make the past relationship between longer-term interest rates and the economy less informative for estimating the effects of unconventional monetary policy.