In Advance Of Bernanke's Speech At The Boston Fed
Readers have already likely had the chance to read the official Fed mouthpiece's bulletin on what to expect out of Bernanke's speech tomorrow at the Boston Fed. Since as we have disclosed previously anything that comes out of the WSJ on the topic of the Fed, gets Calvin Mitchell's stamp of pre- and post, approval we are positive the propaganda spin is in place: after all, can't make the Fed seem "too transparent." So while we are on the topic, here is Goldman's Sven Jari Stehn to confirm just what is best for the bankers: here it goes "For example, Glenn Rudebusch’s analysis—which assumes that Fed purchases
have larger effects on the economy than we estimate—implies that the
Fed would need to buy around $2tr of additional assets to compensate for
the zero bound."Did Goldman just informally double its QE2 expectations, and implicitly bring the 30 Year rate to zero, now that the Fed will have to buy every single treasury out there within its SOMA limit (oh yeah, that 35% SOMA cap - we give it 6 months).
Fed Chairman Bernanke’s speech tomorrow on tools and objectives of monetary policy will provide an opportunity to shape the policy outlook. The speech is likely to include comments on additional asset purchases as well as other options such as price level or nominal GDP level targeting.
- An additional approach to achieve a further drop in fed funds rate expectations and boost financial conditions would be for the chairman to discuss “optimal monetary policy” models which suggest that the “warranted” fed funds rate is currently deeply negative.
- Our “optimal policy models” have two implications. First, interest rate hikes will not be appropriate for a long time to come. For example, under our own economic forecasts it might take until 2015 or longer before a rate hike became appropriate (although we emphasize that this is a scenario projection built partly on assumptions about fiscal policy rather than a formal forecast). Second, substantial asset purchases would be required to make up for the constraints imposed in the interim by the fact that the funds rate is at the zero bound—which is why we expect the Fed’s expected purchases of US Treasuries eventually to cumulate to $1tr and possibly a lot more.
Fed Chairman Bernanke will speak at a conference in Boston tomorrow morning, providing an opportunity to discuss his views on the monetary policy outlook. He clearly intends to do this, as the title of his speech is “Monetary Policy Objectives and Tools in a Low-Inflation Environment.” In his speech the chairman is likely to include comments on additional asset purchases as well as other options such as price level or nominal GDP level targeting. However, as discussed in Tuesday’s Daily Comment we think that it is unlikely that the Fed will adopt price or nominal GDP level targeting any time soon. Rather, we see their mentioning as a signal that the Fed has plenty of options left to boost the economy.
An additional approach to provide guidance to markets and achieve a further drop in fed funds rate expectations, would be for the chairman to discuss “optimal monetary policy” models (see “Sealing the Case,” US Views, October 11, 2010). We have repeatedly argued that such models imply that the “warranted” federal funds rate—i.e. the rate appropriate if no zero lower bound existed—is deeply negative and should remain well below zero for a long time to come. This conclusion is robust across frameworks and forecasts used to derive it.
The starting point of this analysis is the so-called “Taylor rule”, which relates the fed funds rate to inflation and economic slack in the economy. Our version of this rule, which relates the funds rate to the concurrent unemployment gap (the difference between actual and structural rates) and core PCE inflation, suggests that the Fed would have cut the funds rate to -4¾% were it not for the zero lower bound. Glenn Rudebusch of the San Francisco Fed similarly concluded that the warranted funds rate is around -5% (see “The Fed’s Exit Strategy for Monetary Policy,” FRBSF Economic Letter, Number 2010-18, June 14, 2010.) Our preferred rule, however, is a “forward-looking” specification which links the funds rate to expectations of future inflation and unemployment. (For details see “The "Warranted" Funds Rate: Is It Really Negative?”, US Daily Comment, March 10, 2010.) Given our outlook of falling inflation and further increases in unemployment, this rule points to an even more negative rate of -6½%. While the FOMC’s latest (June) forecasts imply a funds rate not as negative as ours, these projections are in flux. We expect a meaningful downgrade of these forecasts (which will be published at the November FOMC meeting) which should result in a warranted rate which is 1-2 percentage points less negative than that implied by our own economic forecasts.
This Taylor rule approach, however, likely overstates the need for additional monetary stimulus because the economy is also receiving boosts from unconventional monetary policy and, at least until recently, from expansionary fiscal policy. Earlier this year we therefore constructed an estimate of the overall macroeconomic policy stance, which takes into account not only the position of conventional monetary policy but also the Fed’s quantitative easing and fiscal policy. Our analysis found that the overall policy stance can be expressed as a weighted average of the real fed funds rate (with a 40% weight), the impact of Fed MBS purchases on the mortgage/Treasury spread (40% weight) and the cyclically adjusted budget balance (20% weight). In the spirit of the Taylor rule, we then related this measure of the overall policy stance to expectations of inflation and the unemployment gap as a description of how overall macro policy has behaved in the past. (See “No Rush for the Exit,” Global Economics Paper, No. 200, June 30, 2010.) Again our conclusion was that although the current overall policy stance is very easy by historical standards it is not quite as easy as past behavior would suggest. In other words, while the Fed purchase program and the fiscal stimulus helped provide a substantial amount of stimulus, the response did not quite compensate for the fact that the federal funds rate hit the zero bound. This analysis suggests that the current warranted funds rate would be -3¾%—a rate somewhat less negative than that implied by our Taylor-rule analysis. (Again this warranted rate would probably be 1-2 percentage points less negative if we used the FOMC’s yet-to-be-released new economic forecasts.)
These models have two important implications. First, interest rate hikes will not be appropriate for a long time to come. For example, under our own economic forecasts it might take until 2015 or longer before a rate hike became appropriate (although, as discussed in the Global Economics Paper cited above, this is a scenario projection built partly on assumptions about fiscal policy rather than a formal forecast) [TD: yes, yes, we get it].
Second, substantial asset purchases would be required to make up for the zero bound. This is, of course, the reason why we expect the Fed to eventually buy more (and possibly a lot more) than $1tr of longer-term assets. A while ago we estimated that the Fed would have needed to buy an additional $5tr or more in 2009 to compensate for the fact that the funds rate could not be cut to its warranted level (see “What Does It Take to Beat the “Zero Bound”?”, US Daily Comment, April 13, 2010). The fact that policymakers did not adopt such an aggressive approach suggests that they were concerned about the unwanted side effects of such a policy, including the potential for renewed Fed-induced asset bubbles or the risk of large losses on the Fed’s asset portfolio. An “optimal” policy which takes these costs into account would therefore not attempt to compensate fully for the zero bound (see our Global Economics Paper, cited above). For example, Glenn Rudebusch’s analysis—which assumes that Fed purchases have larger effects on the economy than we estimate—implies that the Fed would need to buy around $2tr of additional assets to compensate for the zero bound (see paper cited above).
By presenting his own version of this type of analysis, chairman Bernanke could probably achieve a further drop in fed funds rate expectations and boost financial conditions. In fact, discussing “optimal policy models” might be a preferable way to ease financial conditions than talking about price level or nominal GDP level targeting, because the implied policy does not require a change in the Fed’s targeting framework (as it is based on historic Fed behavior). And such a discussion would not be a “commitment” because it would be based on uncertain economic forecasts. But a detailed explanation by Bernanke of just how long the funds rate might stay near zero could be a valuable complement to a November QE2 announcement.
Alternatively, the Fed chairman can just print a check for $100 trillion and LBO the world using another $900 trillion in 0% perpetual UST consols, AAAA+ rated by both S&P and Moody's. Ultimately, it will pretty much achieve the same task.