Each $1 trillion of bond purchases from the Federal Reserve, as part of quantitative easing (QE), will create the equivalent of a 75 basis point (0.75%) interest rate cut, estimates Goldman Sachs’ Jan Hatzius in a new note.
Moreover, each 75 basis point equivalent interest cut could increase U.S. GDP by 1.2% over the course of two years, and Goldman expects that a new round of QE will entail $2 trillion of purchases, though this will vary relative to how the Fed sees the economy perform.
So the first point is that QE can still have a stimulative effect, Goldman says.
Yet most interestingly, Jan Hatzius also explains why the proper way to visualise QE isn’t as ‘a printing press’, but rather trading short-term for long-term U.S. government debt, thus reducing the average maturity of all U.S. government debt outstanding:
Goldman’s Jan Hatzius:
Second, and more importantly, the “money printing” aspect of QE2 is less meaningful than widely believed. In a situation where the zero bound on nominal short-term interest rates is binding, the best way to think about QE is not via the sharp increase in the monetary base, but via a shortening of the average maturity of the government debt. The reasoning is explained in a recent blog post by Paul Krugman, as well as a paper by James Hamilton and Cynthia Wu. (See Paul Krugman, “How to think about QE2,” October 23, 2010, and James Hamilton and Cynthia Wu, “The Effectiveness of Alternative Monetary Policy Tools in a Zero Lower Bound Environment,” October 2, 2010).
To understand this, think of the decision to create bank reserves in order to buy long-term bonds as a composite of two decisions: (1) a decision to create reserves to buy overnight Treasury bills, and (2) another decision to sell overnight Treasury bills to buy long-term bonds.
The effect of step (1) is likely to be approximately zero. This is because bank reserves and overnight bills are essentially perfect substitutes as long as short-term interest rates are close to zero. Both are highly liquid and safe assets available on demand that carry a negligible interest rate. Hence, exchanging one for the other is unlikely to have a significant impact on the economy, at least to a first approximation.
This leaves step (2). One way to assess the potential effectiveness of the sell short/buy long policy is to consolidate the government sector by treating the Federal Reserve and the US Treasury as one single entity. If so, it is clear that a decision to sell short-term bills and buy long-term bonds is equivalent to a decision by the federal government to reduce the average maturity of the government debt by selling more short-term bills and fewer long-term bonds and notes.
Goldman then attempts to estimate how another round of QE could affect the government bond market:
How much will the average maturity fall? To get a rough number, note that the outstanding marketable Treasury debt is about $8.5trn, and the monetary base is about $2 trillion. If we exclude the $800bn in Treasury securities held by the Fed, this means that the sum of Treasuries and monetary base held sectors other than the Fed is just under $10trn. If Fed officials create $1trn in additional monetary base to purchase $1trn in Treasury securities with an average maturity of 5 years, the average maturity of the Treasuries-plus-base aggregate declines by 6 months; if they purchase $2trn, the average maturity declines by 12 months.
How dramatic is such a shift? To get a sense, note that the average maturity of the marketable Treasury debt –by far the largest component of the Treasuries-plus-base aggregate above—has fluctuated between 47 months and 71 months over the past three decades. The biggest 1-year changes over this period were a 7-month drop in 2003 and a 6-month rise in 2009; the biggest 2-year changes were a 9-month rise in 1985 and a 13-month drop in 2003. These numbers indicate that QE2 would be equivalent to a significant decline in Treasury debt maturities, but they are also consistent with our view that a lot of LSAPs may be needed to make a large difference to the bond term premium and hence economic activity.
(Via Goldman Sachs, QE2 as a Shortening of Treasury Debt Maturities, Jan Hatzius, 26 October 2010)