This editorial is written by Justin Scott Katiraei, Ben Sprung-Keyser, Andreas Schaab, Eugene Wang, Asfandyar Nadeem, and Bob Wu. While serving as Co-Editor-in-Chief of this publication, Katiraei also works with his co-authors to represent the Harvard Economics Department at the Federal Reserve. Katiraei, Sprung-Keyser, and Schaab have been together ranked by the Federal Reserve System as First Place in the United States among university students in monetary policy, receiving an invitation to present economic policy recommendations before the Federal Reserve Board of Governors in Washington D.C. as well as the Federal Reserve Bank of Boston.
We have now entered our sixth year of anemic economic growth in this country. Many Americans wonder when the “crisis” will finally end, particularly amid a backdrop of such severe political polarization, the likes of which this nation has not seen since the Civil War. Following the introduction of divided government in 2011, the opportunity for a new round of fiscal policy stimulus has all but evaporated, putting great pressure on the Federal Reserve to act where Congress did not. The Fed has responded with unprecedented, yet in this publication’s view, very appropriate monetary policy measures to encourage hiring and restore growth. Nonetheless, there is still more work to do.
After thoroughly considering the real economy and the factors that currently inhibit the recovery, this publication supports the Federal Open Market Committee’s recent decision to modify its forward guidance policy by adopting an explicit target of 6.5% unemployment and 2.5% inflation before the Federal Funds rate is to rise. We also argue for the continuation of MBS purchases, discussing the conditions that merit their termination, and finally, we advocate decreasing the interest rate paid on excess reserves. Ultimately, we are optimistic about the direction of the economy while still cognizant of high business uncertainty.
Although U.S. GDP growth was strong in the second half of 2011, the recovery seems to have weakened over the course of 2012. It is no surprise, therefore, to find that labor market conditions remain weak. While unemployment dropped in 2012, changes in part-time workers and labor force participation make these numbers difficult to gauge. Payroll employment may have increased, but these are only moderate gains once population growth is taken into account. Payroll employment clearly remains below a level consistent with a robust recovery.
The question then becomes, why have labor markets not yet rebounded? After all, productivity is on the rise and firms are in good financial standing. We believe that the primary reason is business uncertainty about aggregate demand. This publication has identified three important downsides risks that have caused delays in new hiring and consequently reduced aggregate demand. First, although the risk of a European collapse has diminished, it still poses a serious threat to U.S. exports, global supply chains, and U.S. financial stability. Second, there is a substantial risk that regardless of whether the Eurozone collapses, recessions abroad will weigh on exports. Growth has slowed in key emerging economies like China, and the S&P estimates that the Eurozone will contract by 0.8% this year. Finally, there is significant uncertainty about abrupt fiscal changes as a result of endless patch-up solutions in Washington. Real time policy uncertainty, as measured by Baker, Bloom and Davis, is near its highest level since the 2011 debt ceiling debate. This is particularly important because they estimate that the rise in uncertainty from 2006 to 2011 cost the economy 2.5 million jobs.
Taken together, these three downside risks, as well as concern over regulatory uncertainty, help to explain why the Kansas City Fed Manufacturing Survey and the ISM index are near their lowest levels since the recovery began. Firms are delaying hiring and investments until uncertainty subsides.
Although there is weakness in labor markets and high uncertainty, this does not mean that the economy is without bright spots however. The housing market is gaining momentum and has the potential to drive growth in 2013. New home inventories have fallen below their long-run average of six months, while both Case-Shiller and new home construction are on the rise. Household deleveraging, which had weighed on consumption since the start of the recession, seems to be abating. This may explain why real personal consumption is above 2% for the first time since October 2011.
At this point it is worth briefly turning our attention to the current state of inflation expectations. We believe that accommodative action is appropriate in the context of the Fed’s dual mandate given that medium and long-term inflation expectations remain subdued. Five-year breakeven rates lay just slightly above 2% and the Federal Reserve estimates Core PCE to remain between 1.8% and 2.3% through 2015.
Based on our analysis of the real economy, this publication believes the best way to strengthen the recovery is to reduce business uncertainty and encourage firms to hire and invest, while taking advantage of strength in housing to turn the industry into a source of growth. We therefore strongly support the Fed’s recent attempt to streamline QE3 by modifying its forward guidance. In our opinion, the Fed had not previously taken full advantage of this communication technique. Its use of a time-based target for lift-off had serious shortcomings.
First, under the old time-based target, any extension of the lift-off date might have been misconstrued as a pessimistic forecast. Every time the Fed moved its lift-off date, it was unclear whether this was the result of a downward revision in growth estimates or an increased commitment to accommodative policy. Rather than risk sending the markets mixed-signals or being misinterpreted, the elimination of a time-based target for lift-off alleviates this concern altogether.
Second, the previous mid-2015 date lacked credibility as anything more than a conditional expectation. If the economy recovered and inflation pressures began to mount before mid-2015, there was little reason to believe that the Fed would continue to keep interest rates low. This is exactly what happened to the Bank of Canada, which promised in April 2009 that conditional on inflation, it would keep rates low until mid-2010. However, as positive economic data was reported, expected interest rates rose, hindering the effectiveness of Canada’s forward guidance.
However, this is nothing more than a problem of implementation. Effective forward guidance has the power to be particularly potent. The clearest evidence for its effect comes from work done by Campbell and others at the Chicago Fed. Using a GSS model, they decompose the effect of Federal Reserve policy changes and isolate the effect of announcements as distinct from the effect of Fed Funds rate changes. They find that the announcement of an accommodative policy stance explains most of the downward movement in long-term treasury rates.
The impact is that Federal Reserve promises can have a direct effect on firms and on individual consumption decisions. As highlighted by Michael Woodford in his recent Jackson Hole Paper, central bank forward guidance can convince markets that short-term rates will remain low for an extended period of time. This ought to shift the yield curve so that the Fed can take advantage of the traditional effects of lower rates. Falling long-term rates make it cheaper to invest or consume. And, as articulated by Kuttner and Moser, a decrease in rates drives up equities and homes prices, increasing consumption through the wealth channel. Given the ability of forward guidance to cause a full shift in the yield curve, traditional economic correlations ought to still apply. These effects explain why the Fed’s FRB/US model predicts that a 20 bps fall in 10-year rates will cause a 20 bps reduction in unemployment.
The FOMC should not consider raising the Federal Funds rate until the unemployment rate falls below 6.5% or the medium-term outlook for inflation rises above 2.5%. Moreover, even after a reduction in headline unemployment to 6.5%, the Fed should not consider raising rates unless there are meaningful gains to full-time employment. Specifically, reductions in U3 caused by falls in labor force participation or increases in part-time employment should not warrant higher interest rates.
It seems clear to this publication that the 6.5% unemployment threshold was chosen based on a highly cautious estimate of current structural unemployment to avoid any serious inflationary risk. The evidence suggests 6.5% is safely above the natural rate. Research in 2012 by Sahin and others at the New York Fed estimate structural unemployment at 6%, as does the Survey of Professional Forecasters. This publication would have also preferred that a forward guidance strategy target 6%, rather than 6.5%, but we understand the Fed’s conservatism.
Nonetheless, as we previously mentioned, there must also be a strong focus by Fed officials on full-time employment, as the value of U3 as an indicator is not absolute. An indicator like U6, which counts discouraged and marginally attached workers, may better gauge the strength of the labor market. Because of the lack of public familiarity with this indicator, however, we understand why the Fed would not target U6 over U3. The goal here is to form clear expectations for the public, and the introduction of a complicated and less reported indicator might only hinder effective communication. A focus on full-time employment, without explicitly targeting U6, is an attempt to navigate a middle ground.
In essence, the Fed’s new change of policy is really only intended to define what it meant to “wait for a considerable time after the economic recovery strengthens,” the language the Fed formerly used. Instead of estimating when this would occur, the new policy will increase Federal Reserve credibility and reduce business uncertainty by providing greater clarity.
Now that we have discussed the Fed’s forward guidance policy, let us address the MBS purchases under QE3. Given that the housing market has the potential to drive U.S. economic growth in 2013, we believe the appropriate course of action is to help maintain momentum in housing by continuing MBS purchases. The Fed’s purchase of MBS from “To Be Announced” markets should have a more direct effect on lowering mortgage rates than the purchase of other assets, like treasuries. In a TBA market, mortgage originators can sell forward contracts on MBS at fixed rates, creating a strong incentive to originate more mortgages. And, according to Hancock and Passmore at the Board of Governors, QE1 lowered mortgage rates by 50 bps through conventional transmission channels. This publication believes we can recapture that effect with QE3.
It is important to note here that while we endorse further purchases, we have only moderate expectations for their effects. The efficacy of MBS purchases is still contingent on lower MBS yields translating into lower mortgage rates. And, as recently highlighted by President Dudley, frictions in the mortgage market like delays in mortgage processing and decreases in competition among mortgage originators may limit pass through. This helps to explain why MBS yields have fallen relative to treasury rates but the spread between MBS and mortgage rates has risen.
Assuming that MBS purchases have a moderate but meaningful effect, the question then becomes, when should the Fed end its purchases? After all, this targeted program ought to focus on different considerations than forward guidance in general. In helping the Fed answer that question, this publication would like to highlight one important variable. To the extent the Fed wants to stimulate the economy through refinancing and home purchases, it should focus on the spread between mortgage rates and treasuries. The current spread may encounter a lower bound at its historical long-run average of 150 bps. In recent months, spreads have hovered in this range. Consequently, further purchases may prevent a rise in rates and help maintain upward momentum in housing. Nonetheless, the Fed should not expect much additional stimulus from this program unless it is willing to drive down treasury rates first.
As a final way to stimulate the economy, this publication now advocates for a reduction of the interest paid on excess reserves (IOER) from 25 bps to 0 bps. This proposal is designed to take advantage of the 1.4 trillion dollars that currently lie in excess reserves. We believe that this action will positively impact the economy in two important ways. First, lowering IOER is likely to stimulate bank lending by generating an incentive for banks to move money out of reserves and seek returns elsewhere. In the worst case, if the money is not loaned out, but rather invested in other safe assets, it is likely to at least lower borrowing costs.
The second reason to lower IOER is to reduce the Fed Funds rate. IOER acts as a porous floor on the price of federal funds, as banks have little incentive to loan out money at a rate lower than the one they receive from the Fed risk-free. Therefore, when IOER falls to zero, the Fed Funds rate will fall as well. Lowering the Fed Funds rate has the important effect of driving down long-term yields. Cochrane and Piazzesi estimate that a surprise change in the Fed Funds rate of 100 bps changes 10-year rates by as many as 80 bps. If we conservatively project that a decrease in IOER would reduce the Fed Funds rate by 10 bps, we could see an 8 bps reduction in 10-year rates. For point of reference, the academic consensus is that the entirety of QE2 moved long-term rates 10 bps to 50 bps, and Fuhrer and Olivei at the Boston Fed estimated that QE2 created 700,000 jobs. The implication is that lowering IOER may create several hundred thousand jobs.
This publication understands that such an action is controversial, and so we would like to take a moment to address an important argument made against it. A major concern is that this policy will cause money market funds to break the buck and will therefore lead to a run on money markets. It is this publication’s belief, however, that this concern is overstated. A money market fund is only likely to incur a significant drop in net asset value if it holds paper in a corporation or government that collapses. After all, the Reserve Primary Fund only broke the buck because it held Lehman paper. Therefore, even if IOER reductions put downward pressure on short-term interest rates, they are unlikely to cause such an event. SEC Rule 2a-7 and the Investment Company Act of 1940 require that money market funds hold only the most highly rated corporate and government bonds. This means that the funds have little room to respond to changes in IOER with a reach for greater yields. The underlying risk of breaking the buck does not change.
Finally, as a philosophical matter, it is dubious at best to claim there is some positive externality associated with banks not lending money, which raises the question why the government would seek to subsidize such behavior. If anything, there are negative externalities associated with keeping large excess reserve balances, which would prompt the Fed in fact to set IOER at negative levels, not positive ones. Setting the IOER to 0 bps, where it remained for decades prior to 2008, is again the sensible policy at this time.
In totality, however, this publication would like to commend Mr. Bernanke’s wisdom and courage during this time of great economic malaise. We wish him continued success in the remainder of 2013.