Back in May when we presented our humble and succinct analysis on what the preliminary 1.8% GDP looked like, we said "Ex the now traditional inventory build [of 1.2%], Q1 GDP growth was sub 1%" basically being the only party who said that aside for the "old faithful plug" better known as the traditional BEA fudge to get GDP to whereever the administration wants it, growth was where it ultimately ended up being: 0.4%. And the kicker? The primary cause of the downward revision was, you guessed it, Inventories, which imploded from 1.31% to 0.32% (see chart). In other words, the next time we are skeptical about government data in any format, believe us, and not "them." Which also goes for our skepticism when it comes to the predictive ability of one Goldman Sachs, most notably our take on Goldman's December 1 2010 "watershed" report in which Hatzius said: "This outlook represents a fundamental shift in the thinking that has governed our forecast for at least the last five years... Five years ago, we became very pessimistic about the US economic outlook...So why do we now expect growth to pick up? In a nutshell, it is because underlying demand has strengthened significantly... After a deep downturn from 2007 to mid-2009 and near-stagnation from mid-2009 to mid-2010, underlying demand is now accelerating sharply. Currently, it is on track for a 5% (annualized) growth rate in the fourth quarter." Total and utter fail. Our summary then was also rather spot on: "Much more hopium inside. This is unfortunate. Jan Hatzius used to have credibility." Indeed, after waiting for so long, the firm once again capitulated per its most recent report released last night: "Our forecasts for 3%-3.5% growth in Q4 and 2012 are under review for probable downgrade." So with apologies for the self-backpatting, this brings us to the topic of this post. As we have said for over a year, the catalyst for QE3 will be none other than Goldman. Which is convenient because the title of Goldman's report is "The Fed's Easing Options." Pretty much as subtle as it gets: a month ahead of Jackson Hole, Goldman, aka the Federal Reserve's superior, has not only also admitted the other theme was have been pounding the table on, namely that 2011 is a carbon copy of 2011, but has also listed out the entire menu of options for former Goldmanite Bill Dudley to present to his inferior, Ben Bernanke. Let's dig in.
The Fed’s Easing Options
In a remarkable parallel to last year, Fed officials head into their August meeting amidst weak growth and questions about the possibility of further monetary easing. There are of course major differences between then and now, which Chairman Bernanke has been quick to point out in public statements. Most importantly, inflation is much higher—the core CPI increased at an annualized rate of 2.5% over the last six months—and inflation expectations are at levels consistent with the Fed’s mandate.
Nevertheless, growth is running well below trend— GDP grew by just 0.8% annualized in the first half of the year—and recession risks have increased. Some measures of underlying inflation have also started to cool. If downside risks materialize, further easing might well become appropriate.
Consistent with these risks, comments from Fed officials have shifted slightly in an easing direction over the last month. For example, minutes of the June FOMC meeting reported that some participants thought “it would be appropriate to provide additional monetary policy accommodation” if growth remained slow.
Chairman Bernanke also listed easing options in his semi-annual monetary policy testimony to Congress (formerly known as “Humphrey-Hawkins”) and Chicago Fed President Evans recently said he could support further easing if growth disappoints.
With many arrows already launched, what remains in the central bank’s quiver? Here we walk through the main easing options and our sense of the Fed’s willingness to use them. They fall into three main groups: 1) communication, 2) asset purchases, and 3) interest rate policy.
Communication: Even More Extended?
Because asset values imbed investors’ expectations about the future, monetary policy can influence current financial conditions by changing expectations. In part this is achieved by following predictable (perhaps loosely rule-based) policy over time. Today, when US activity weakens, funds rate expectations and longer-term interest rates reflexively fall, because investors have observed consistent behavior from the Fed and understand its reaction function. This in turn eases financial conditions and supports growth—monetary policy on autopilot.
Central bankers also shape market expectations through public communication (“Fed speak”). In earlier research, then-governor Bernanke argued that managing investor expectations through communication can be a valuable tool when policy rates have reached their lower bound:
“Even with the overnight rate at zero, the central bank may be able to impart additional stimulus to the economy by persuading the public that the policy rate will remain low for a longer period than was previously expected. One means of doing so would be to shade interest-rate expectations downward by making a commitment to the public to follow a policy of extended monetary ease.”
The Fed has implemented this type of “forward guidance” a number of times (Exhibit 1), most recently through its “extended period” pledge: “The Committee continues to anticipate that economic conditions … are likely to warrant exceptionally low levels for the federal funds rate for an extended period.” Our research has found that this language has roughly the same effect on financial conditions as a 30bp funds rate cut.
Changes in official communication would likely be the first step in any renewed easing. In particular, we believe an attractive option would be the use of some kind of forward guidance for the size of the Fed’s balance sheet (“extended period” currently only refers to the level of the funds rate). Bernanke fielded a question on this issue at the June FOMC meeting press conference and responded: “It’s something we have on the table, something we’ve thought about.”
The idea came up again in his Humphrey-Hawkins testimony and in a speech by Brian Sack—head of the New York Fed’s Markets Group—last week. The purpose of this type of language would be to push back investor expectations for when the Fed will allow its balance sheet to shrink. A recent survey showed that 72% of economic forecasters expect the balance sheet to start shrinking later this year or in the first half of 20124. Therefore, announcing that the balance sheet will remain the same size “for an extended period” would likely change market expectations.
Based on New York Fed estimates, pushing back expectations for the start of the decline in the balance sheet by one year would be the equivalent of removing expectations for one 25bp rate hike (assuming all assets were allowed to run off, which amounts to a rate of about $250bn per year).
Other changes in official communication are also possible. In his Jackson Hole speech last year, Bernanke mentioned two ideas: committing to keep policy rates unchanged for a specific time (like the Bank of Canada did in 2009-2010) or tying rate changes to specific developments in the economy (like the Bank of Japan did during its quantitative easing experiment in 2001-2006). President Evans discussed similar ideas in his recent remarks.
We believe these options are on the table but that they are unlikely to be enacted. First, funds rate expectations are already extremely low, and further language changes would therefore need to be quite significant to lower market rates much further (Exhibit 2). Second, some Fed research finds that the Bank of Canada’s commitment language had a similar effect as “extended period”, implying that the Fed would gain little by moving in that direction. Third, Bernanke noted at Jackson Hole that “it may be difficult to convey the Committee’s policy intentions with sufficient precision and conditionality.”
A final set of options relate to communication about the Fed’s mandate. In particular, Fed officials have frequently raised the prospect of more clearly quantifying their inflation objective. Most committee members seem in favor of this idea, but a number of issues appear to have held back implementation, including questions about the dual mandate and uncertain benefits relative to the current system. We do not think an inflation target is around the corner but also cannot rule it out. A few officials have also raised the idea of price level targeting. We think this is a realistic prospect in a deflationary scenario, but not before that point.
Asset Purchases: Composition is Key
Asset purchases have played a major role in the Fed’s response to the recession and financial crisis and would likely be a component of any future easing. One obvious option would be a large-scale Treasury purchase program like the one just concluded. Although there is still significant debate about the impact of quantitative easing (QE) on the economy, earlier purchase programs did appear to ease financial conditions and raise inflation expectations.
If the Fed were to restart QE, our research suggests that each $1 trillion of asset purchases would be roughly equivalent to a 75bp cut in the funds rate. Fed officials appear to see a larger effect, with each $1 trillion in Treasuries purchased substituting for a 67-200bp cut in the funds rate.
One potential drawback of additional QE is that the Fed’s balance sheet is already quite large. Bernanke has expressed concern that further expansions of the
balance sheet could lead to “an undesired increase in inflation expectations”. Perhaps because of a desire to avoid further growth in the absolute size of the balance sheet, recent public comments suggest the Fed might prefer to change the composition of its assets.
In the Fed’s view as well as ours, QE works through a portfolio rebalancing channel: by buying securities and issuing reserves, the Fed reduces the overall supply of risky assets available for the private sector and thereby raises the equilibrium price of those assets. In the specific case of Treasury purchases, the Fed removes duration (interest rate) risk from the aggregate private sector portfolio and takes it on its balance sheet.
If the portfolio rebalancing framework is correct, then what matters is not the face value of the Fed’s balance sheet but the amount of duration risk it holds. Exhibit 3 demonstrates the difference for the Fed’s holdings of Treasury notes and bonds (the Fed’s total portfolio also includes Treasury bills, agency debt and agency MBS). The chart shows the face value of the Fed’s holdings and our estimate of its holdings in 10-year equivalent terms—that is, the amount of 10-year
Treasuries that would equate to the same duration risk.
At present the Fed owns $1.6 trillion in Treasury notes and bonds and around $1 trillion in 10-year equivalent duration. The estimate for 10-year equivalents is significantly lower because the weighted-average duration of the Fed’s Treasury portfolio (roughly 5 years) is lower than the current duration of a 10-year Treasury note (about 8.5 years).
One idea—mentioned in Brian Sack’s recent remarks—would be to keep the overall size of the Fed’s portfolio unchanged, but to increase its duration risk by buying longer-term securities. The Fed is already buying Treasuries regularly in order to reinvest paydowns from its MBS holdings (a policy in place since August 2010). A simple way to lean policy in a slightly easier direction would be to weight those purchases toward longer maturities. Currently the Fed allocates 6% of its purchases to nominal Treasuries with 10-30 years remaining maturity. By increasing this share, it could raise the amount of duration risk it is taking on its balance sheet with the flow of purchases. Even if the total face value of the Fed’s portfolio remains unchanged, an increase in its aggregate duration risk should be considered an easing of monetary policy.
Currently the flow of purchases for MBS reinvestment is relatively low—just $14bn per month according to the latest schedule—and the incremental duration risk the Fed could add to its portfolio through these purchases is relatively small. Therefore, as a more aggressive step, the committee could consider changing its reinvestment policy for the Treasury portfolio. For example, we believe the Fed has the authority to rebalance its reinvestment of maturing securities at Treasury auctions toward longer-duration securities (e.g. 10-year notes instead of 3-year notes).
Beyond additional purchases of Treasuries or changing the composition of its holdings, the Fed’s choices are limited. Agency MBS is one option, but the Fed appears inclined to move toward a Treasuries-only portfolio over time. We doubt the committee would move back to MBS purchases without signs of significant distress in the mortgage market. Purchases of private sector assets could be more effective from a portfolio rebalancing perspective, but the scope for these types of purchases is limited by the Federal Reserve Act (Exhibit 4).
Interest Rate Policy: A Dry Well
A final easing option often listed by Fed officials is a cut in the interest on excess reserves (IOER) rate—the rate that banks earn on deposits held at the Fed above levels mandated by regulation. The IOER rate is to zero (or perhaps even a negative value).
We see little merit in this option, and continue to believe that the Fed is unlikely to use it. First, the benefits are likely to be very small. While the IOER rate is 0.25%, the effective federal funds rate is only 0.08% (Exhibit 5). Thus, the maximum impact from cutting the IOER rate to zero is just 8bp.
Second, cutting the IOER rate down to zero could be harmful to market institutions. Chairman Bernanke made this argument himself in Q&A at the July 2010 Humphrey-Hawkins testimony:
“The rationale for not going all the way to zero has been that we want the short-term money markets like the Federal funds market to continue to function in a reasonable way, because if rates go to zero, there will be no incentive for buying and selling Federal funds overnight money in the banking system. And if that market shuts down, people don’t operate in that market, it will be more difficult to manage short-term interest rates when the Federal Reserve begins to tighten policy at some point in the future.”
We expect that Fed officials would hold the same view about cutting the IOER rate today. Finally, as we argued in a recent article, calls for cutting the IOER rate are partly motivated by the premise that banks are holding excess reserves because of an unwillingness to lend. In fact, the level of reserves is controlled entirely by the size and composition of the Fed’s balance sheet, and says nothing about banks’ incentives or their willingness to lend. Cutting the IOER could theoretically stimulate activity by easing financial conditions and boosting loan demand, but it would not affect the quantity of reserves in the banking system.
Where We Stand
In recent remarks Fed officials have not appeared ready to ease policy. As Bernanke said on the second day of his Humphrey-Hawkins testimony: “we’re not prepared at this point to take further action”. He listed three reasons: 1) inflation is higher than last year, 2) inflation expectations are close to the Fed’s target, and 3) the Fed is forecasting a rebound in growth, whereas last year “the recovery looked like it was stalling.” For the Fed to ease it would likely need to see: 1) significant risks of recession or 2) continued soft growth, lower core inflation and perhaps falling inflation expectations. President Evans said that Q3 growth will be critical to his thinking:
“It’s obvious that the third quarter has to show improvement and it ought to show a high likelihood of sustained improvement … If we continue to have weakness in the third quarter, it’s going to be harder to plausibly sustain this idea, ‘The next six months is going to get better.’ We’ve been saying that for quite some time now.”
If the Fed were to shift its policy stance, we believe the process would start with a change in communication—most likely related to the timeline over which the balance sheet would shrink—and be followed by changes in the composition of its assets. In our view these options already have reasonably high odds. An expansion of the balance sheet (QE3) looks less likely unless the outlook deteriorates significantly further.
Changing communication or the composition of the balance sheet would largely be symbolic. Our quantitative estimates suggest the benefits to growth would likely be small.
And as for the strawman that the US economy will be weak, Goldman conveniently provides the following report in the second part of its analysis explaining the horrible Q2 numbers. Also, for the record, we are confident, Goldman's Q3 GDP target of 2.5% will absolutely not be met.
1. We forecast that real GDP will pick up modestly from its anemic pace in the first half of the year to 2.5% (annualized) in Q3. Our forecasts for 3%-3.5% growth in Q4 and 2012 are under review for probable downgrade. Although oil prices have stabilized somewhat after their surge earlier this year, they are likely to remain a meaningful drag on consumer spending and business investment. And fiscal tightening—already more substantial in H1 than we expected—is apt to increase in 2012. The inability of policymakers to agree on a measure to lift the federal debt ceiling has damaged consumer confidence, not to mention our own confidence in a broader-based normalization in the economy.
2. Continued high unemployment, with only a marginal drop in the jobless rate to 8.8% by year-end 2012. Our forecast is for second-half growth roughly at trend, so we expect the unemployment rate to end the year at its current level of 9.2%. Growth of 3¼% in 2012 would bring it down, but only slightly.
3. Core inflation peaks around the end of the year. We expect the core PCE price index to accelerate further to 1.9% yoy by the fourth quarter, from 1.2% now. Two factors contributing to the recent acceleration—a surge in vehicle prices and pass-through from higher commodity inflation—should begin to ease later in the year. Rising rent inflation—caused in part by a decline in homeownership and surge in demand for apartments—is likely to remain a source of inflation pressure in the major price indexes. Overall, however, we see headline and core inflation decelerating throughout 2012.
4. No Fed rate hikes before 2013. Data disappointments have clinched our longstanding call for 2011 and made it even more likely for 2012 than we previously thought. With the jobless rate far above the Federal Open Market Committee’s “mandate-consistent” 5%-6% range and drifting higher, and core inflation still below the comparable “2% or a bit less” standard, the near-term question is how seriously the Fed will consider the easing options discussed on the previous pages.
5. Yields on 10-year Treasury notes reach 3¾% by year-end 2011 and 4¼% by year-end 2012. This forecast presumes recent weakness in US activity proves temporary; in that case, we believe that many participants in the financial markets will turn their attention back to higher inflation.
And for those who really want to know the specifics, including the dirty secrets, of the Fed's next steps, we once again present the June 24-25, 2003 Vince Reinhart FOMC Minutes Appendix 1 which has all you need to know and more.