Katherine Femia,* Jeff W. Huther,** and Andrea Tambalotti
The Federal Open Market Committee (FOMC) recently announced its intention to extend the average maturity of its holdings of securities by purchasing $400 billion of Treasury securities with remaining maturities of six years to thirty years and selling an equal amount of Treasury securities with remaining maturities of three years or less. The nominal size of this maturity extension program, at $400 billion, is smaller than the $600 billion of purchases during the second round of large-scale asset purchases (LSAP 2) completed in June 2011. The two programs are more comparable in size, however, once we consider the characteristics of the securities expected to be purchased and sold under the maturity extension program. In this post, we explain what this means and why it matters.
In the FOMC’s words, the maturity extension program, often referred to as operation twist, “should put downward pressure on longer-term interest rates and help make broader financial conditions more accommodative.” Some of the channels through which the maturity extension program should affect financial conditions are very similar to channels at work with LSAP 2. In particular, both programs remove interest rate risk from the market, leading investors to adjust their portfolios in ways that put downward pressure on longer-term interest rates. This so-called portfolio balance effect was discussed, for instance, by Chairman Ben Bernanke in his Jackson Hole speech last year. (For a survey of several channels by which purchases or sales of assets by the Federal Reserve may affect investors’ behavior and some estimates of the quantitative effects of recent FOMC programs on interest rates, see this recent paper by Krishnamurthy and Vissing-Jorgensen.) The key difference between the maturity extension program and LSAP 2 is the way the purchases of longer-term securities are funded. In the case of LSAP 2, the asset purchases were financed by issuing bank reserves. In the case of the maturity extension program, the purchases of longer-term securities will instead be funded by selling shorter-term securities, leaving the overall size of the Federal Reserve’s balance sheet little changed.
Under the portfolio balance view of these interventions, what matters is the amount of duration risk removed from the market. Duration can be thought of as a measure of the change in price of a security when interest rates change. When interest rates change, securities with higher duration, which also tend to have longer maturity, are subject to larger capital gains or losses than securities with less duration. Investors seek the additional return associated with duration risk for a variety of reasons. With the Federal Reserve buying longer-term securities and thus reducing the amount of duration risk available to the market, investors obtain duration risk by purchasing other securities. These portfolio adjustments should keep interest rates lower than they would otherwise be, not just on the securities being purchased, but also on other assets.
To measure the amount of duration risk involved with the FOMC’s portfolio shift, analysts may use the concept of ten-year equivalents. The ten-year equivalents of a given portfolio are the amount of ten-year Treasury notes that an investor would have to buy to be exposed to the same amount of duration risk contained in the portfolio. For example, if the duration of a twenty-year Treasury security is twice as large as that of a ten-year Treasury note, and the Federal Reserve bought $1 billion of that twenty-year security, we would say that the Federal Reserve purchased $2 billion in ten-year equivalents.
Based on the ten-year equivalent measure, the maturity extension program should remove roughly the same amount of duration risk from the market as LSAP 2. More specifically, we expect the net shift in duration from the combination of the securities purchased and sold during the maturity extension program to be similar to the duration of a ten-year Treasury security. Accordingly, the $400 billion program is expected to remove about $400 billion in ten-year equivalents from the market. The amount of securities purchased during LSAP 2 was larger in terms of par value ($600 billion), but the average duration of those securities was smaller than that anticipated under the maturity extension program. As a result, LSAP 2 also removed about $400 billion in ten-year equivalents.[1] This conclusion is consistent with similar calculations carried out by several market participants (see, for instance, this WSJ.com article). For a survey of some of the available estimates of the effects of these programs on interest rates, we point the reader to this Econbrowser post from last year and to a more recent follow-up that focuses explicitly on operation twist.
LSAP 2 |
Maturity |
|
Purchases |
$600 billion |
$400 billion |
Sales |
0 |
$400 billion |
Net duration |
5 ½ years |
9 – 10 years |
Ten-year equivalents |
$400 billion |
$400 billion |
The above calculations reflect the immediate impact on the duration risk removed from the market and taken onto the Federal Reserve’s portfolio. Some of this effect will erode over time, however, because the maturity extension program will affect the amount of debt that the Treasury has to issue to private investors In particular, under the maturity extension program, the Federal Reserve will have sold $400 billion of Treasury securities with maturities of three years or less. Those holdings, when they matured, would have been rolled over into new Federal Reserve holdings at Treasury auctions. With these securities no longer held by the Federal Reserve, the Treasury will have to issue a larger amount of debt to private investors, thereby reintroducing some of the duration risk that was initially absorbed by the maturity extension program. Based on current issuance patterns, the additional financing need created by the maturity extension program ($400 billion of securities maturing over the coming three years) would require the private sector to absorb roughly $275 billion in ten-year equivalents between January 2012 (the maturity date of the earliest maturity extension program sales) and June 2015 (the maturity date of the latest sales).
*Katherine Femia is a financial/economic analyst in the Markets Group of the Federal Reserve Bank of New York.
**Jeff W. Huther is an assistant vice president in the Markets Group of the Federal Reserve Bank of New York.
Disclaimer
The views expressed in this blog are those of the author(s) and do not necessarily reflect the position of the Federal Reserve Bank of New York, or the Federal Reserve System. Any errors or omissions are the responsibility of the author(s).
Thanks for your comment. The effectiveness of the maturity extension program must indeed come through some connection between the yield curve and the average maturity of the debt held by the public. The relationship between net government supply and yields is driven by many factors. We highlighted the program’s projected effects on one factor that may influence yields. We fully realize that other factors will cause larger movements in yields over the course of the program. Note, however, that at least some case studies of changes in the maturity composition of the debt held by the public (for instance, Operation Twist in 1961) have identified yield effects. See, for instance, Eric Swanson’s study, available at http://www.ericswanson.us/papers/lta.pdf. It is true that the Treasury could decrease the average maturity of the debt it issues to the public and, through the portfolio balance channel, push interest rates lower than they would otherwise be. In contrast to the FOMC’s mandate to promote price stability and full employment, however, the Treasury seeks to finance the national debt at the lowest cost to the taxpayer over time.
I fail to see why this is considered monetary policy since the balance sheet of the Fed remains unchanged in size. Exactly the same impact on the average maturity of the debt could be had by the Treasury changing its issuance mix-although this would reverse existing policy to lengthen the average maturity of the debt, which has risen from 46 months in October 2008 to 59 months in June. If Twist is to be effective in lowering long rates (flattening the curve) there must be some relationship between the curve and the average maturity of the debt but none exists as far as I can see from running the analysis on monthly data. Any impact of Twist on yields was, imho, a result of increased market risk premiums to the Fed talking about significant downside risks in the FOMC statement.