The Fed’s New Policy Tools

We had to throw out our textbook descriptions of how monetary policy is implemented after the fall of 2008, as the Fed turned from its traditional tools to active use of large-scale asset purchases. A number of studies have now been conducted of the potential efficacy of these new policy tools. I surveyed some of the new studies last October. Today I’d like to discuss three new papers that have come out since then.

Let me begin with a little background. Prior to the fall of 2008, the focus of monetary policy was to choose a target for the fed funds rate, which is the interest rate banks charge each other for overnight loans of Federal Reserve deposits. In normal times, this rate was extremely sensitive to the quantity of those deposits created by the Fed, enabling the Fed to achieve its target for the fed funds rate with relatively modest additions or withdrawals of reserves. But by the end of 2008, the Fed had driven the fed funds rate essentially to zero and began paying interest on reserves. Since then, banks have been content to hold an arbitrarily large amount of excess reserves, and the overnight rate has been as low as it could go. In other words, the traditional tools of monetary policy have become completely irrelevant in the current setting. 

The Fed has therefore been trying to find other ways to stimulate the economy by buying longer term assets. Hess Chung and colleagues expressed the idea this way:

A primary objective of large-scale asset purchases is to put additional downward pressure on longer-term yields at a time when short-term interest rates have already fallen to their effective lower bound. Because of spillover effects on other financial markets, such a reduction in longer-term yields should lead to more accommodative financial conditions overall, thereby helping to stimulate real activity and to check undesirable disinflationary pressures through a variety of channels, including reduced borrowing costs, higher stock valuations, and a lower foreign exchange value of the dollar. In many ways, this transmission mechanism is similar to the standard one involved in conventional monetary policy, which primarily operates through the influence on long-term yields of changes to the current and expected future path of the federal funds rate.

Although the transmission mechanism may be similar to conventional monetary policy, the implementation is quite different. To make a significant change in the supply of Federal Reserve deposits, in normal times the Fed only needed to buy or sell a few billion dollars worth of T-bills. But to make a significant change in the market’s available supply of long-term Treasury securities, the purchase needs to be in the hundreds of billions of dollars, and even then, there are debates as to whether there would be any noticeable effects. Hence the interest in empirical studies of exactly what the effect of such operations appears to be.

Eric Swanson of the Federal Reserve Bank of San Francisco has a new analysis of Operation Twist, which was an effort by the Kennedy Administration in 1961 to change relative yields by having the Treasury preferentially issue less long-term debt and the Federal Reserve preferentially buy more long-term debt. Although the magnitudes seem small by current standards, Swanson argues that they amounted to 4.5% of U.S. Treasury-guaranteed debt at the time, which is actually bigger in percentage terms than the magnitudes associated with QE2. Swanson found that there were unusually large declines in long-term yields on the days of significant Operation Twist announcements, and that the size of the effect is similar to what would have been predicted to happen based on other empirical studies of the potential effects of such operations.

Federal Reserve economists Diana Hancock and Wayne Passmore have a new study that focuses on the Fed’s purchases of mortgage-backed securities since 2008. They regard the risk premium on MBS to have been unnaturally low during the housing boom, but shot way up as the market for new MBS broke down in the fall of 2008. The spread was brought quickly back down by the Fed’s announced intention to purchase large quantities of MBS.


Yield spread between mortgage-backed securities and U.S. Treasuries of comparable maturity, in percentage points, 4-week moving average. Source: Hancock and Passmore (2010).

Hancock and Passmore attribute these effects not so much to the Fed’s absorbing the stock of outstanding MBS, but instead to creating a functioning market for newly issued MBS.


Source: Hancock and Passmore (2010).

A third study by John Williams of the Federal Reserve Bank of San Francisco and Federal Reserve Board economists Hess Chung, Jean-Philippe Laforte, and David Reifschneider has recently been updated to include an assessment of the effects of the Fed’s large-scale asset purchases. As initially envisioned, the Fed was going to allow its holdings of MBS to shrink gradually as a result of prepayment and repayment of the underlying mortgages, implying a future time path for Fed holdings such as that represented by the black line below. Last August, the Fed announced that it would instead replace maturing MBS with longer-term Treasuries in order to prevent its holdings from contracting, consistent with a path such as the blue line in the figure below. Then in November the Fed announced QE2, in which it intends to buy an additional $600 billion in longer-term Treasuries, consistent with a future path such as that indicated in red below.


Projected MBS, Treasury, and agency holdings by the Federal Reserve for phase 1 (black), phase 2 (blue), and phase 3 (red) plans. Source: Chung et. al. (2011).

Chung and coauthors then used rough estimates derived by Gagnon and colleagues of the effects these actions could have on longer-term interest rates, and fed those effects into the FRB/US model (a large-scale model the Fed sometimes uses to describe the economy) to find predicted implications for a variety of variables of interest. The figure below summarizes the results of their calculations. For example, the upper-left panel shows that the initial MBS purchases resulted in a 10-year Treasury yield that was almost 50 basis points lower than it would have been in the absence of such purchases. The panel also shows that when those purchases were augmented by the August reinvestment and the November QE2, the combined effect would be a yield almost 60 basis points lower than it would have been otherwise. The authors also conclude that real GDP in 2012 would be 1% lower without phases 2 and 3 and almost 3% lower if the Fed had done nothing (upper right panel). The authors estimate that there would be 3 million fewer people employed on private payrolls next year if the Fed had not acted, and that inflation would have been 1% lower, that is, the policy helped avoid deflation.


Difference between time path under phase 1 (black), phase 2 (blue), or phase 3 (red) plans and that with no Fed large-scale asset purchases for 6 different macro variables. Source: Chung et. al. (2011).

It should be emphasized that these calculations are the result of putting together a series of multipliers, no one of which we know with any precision. The numbers are better interpreted as effects that one could plausibly claim rather than magnitudes we can be at all sure about. Still, they are consistent with the policy recommendation that it was better for the Fed to give this a try rather than stand by and do nothing.

It should also be emphasized that the authors study the effects only of the Fed’s actions in isolation. My assessment is that the actions by the Treasury more than 100% offset the effects on security supplies of the Fed’s QE2.

But even if one is unconvinced that the Fed was able through these actions to provide the significant benefits claimed in the above analysis, there is another channel discussed by Chung and coauthors which may have been one of the most important ways in which QE2 was beneficial:

Finally, the Federal Reserve’s asset purchase program could potentially have stimulated real activity by changing public perceptions about the likely longer stance of monetary policy, conventional and unconventional; for example, it may have led market participants to expect that the FOMC would respond more aggressively to high unemployment and undesirably low inflation than was previously thought. In a similar vein, initiation of the program may have diminished public perceptions of the likelihood of extreme tail events, such as deflation, potentially lowering risk premiums and increasing household and business confidence, thereby raising agents’ willingness to spend.

Whatever else you may say, QE2 did seem to have an effect on public perceptions. And that has always been one of the most important elements of the conduct of conventional monetary policy as well.

Originally published at Econbrowser and reproduced here with permission.