While the market has been fixated lately on the question of when and how the Fed will taper its asset purchases, perhaps as important for the rates market (and the magic that levitates stocks) is the outlook for the Fed’s forward rate guidance. On this front, BofAML suggests that recent evidence shows the effectiveness of forward guidance is diminishing... already.
Via BofAML,
Is Forward Guidance Dead?
One measure of this is the ratio of realized 5y Treasury volatility versus 10y Treasury volatility.
The first major shift in the relationship between 5y and 10y realized vol occurred in late 2008, when the FOMC cut rates to 0-25bp and 5y vol dropped sharply relative to 10y vol. The second major decline in 5y vol occurred in August 2011, when the FOMC strengthened its forward guidance by replacing the “extended period” language with guidance that it would likely keep rates at zero “at least through mid-2013.” This initiated a low-vol, low-rate regime that persisted for nearly two years, until the June 2013 FOMC meeting. Below we discuss some factors that contributed to the surge in 5y vol beginning in mid June.
In the days prior to the June 19 FOMC meeting, markets had come to expect Bernanke’s press conference to reinforce or potentially strengthen the Fed’s low-rate guidance. Instead, Bernanke seemed to dismiss the recent rise in rates when asked multiple times about the issue, which prompted a sharp selloff and an unwinding of long positioning in the front end. This served to undercut the notion that the Fed’s forward guidance would protect investors in the 1-5y sector of the curve.
Another critical revelation at the June meeting was the FOMC’s decision to move forward its expectation of when the first rate hike would occur, as per the scatterplot diagram released with the Summary of Economic Projections. As we noted at the time, the average FOMC participant forecast for the appropriate fed funds target at the end of 2015 shifted up by 30bp to 1.0%.
This surprising shift occurred despite downward revisions to the central tendency inflation forecast and essentially no net changes to the GDP forecast. However, the 2015 unemployment rate forecast did fall by 0.25bp relative to the March projection. The fact that this drop was sufficient to prompt the Fed to forecast earlier hikes revealed critical information about the Fed’s reaction function, as it implied a fairly mechanical tightening of policy in response to lower unemployment. As we noted last week, there is also a risk that the 2016 SEP forecasts reveal a faster pace of tightening than the market has priced in, since the 2016 unemployment central tendency is likely to be near the FOMC’s estimate of the NAIRU.
The June FOMC meeting also sparked a sharp increase in speculation regarding Bernanke’s future as Fed Chairman and who might be appointed to replace him. In particular, Obama’s decision to state in an interview two days before Bernanke’s press conference that the Chairman had “already stayed a lot longer than he wanted or was supposed to” injected a considerable degree of uncertainty about who might be leading the Fed during the exit process, especially since Obama chose to wait until autumn to nominate a successor. The uncertainty about who will lead the Fed has obscured the market’s understanding of the Fed’s reaction function and undermined the efficacy of forward guidance, in our view, thereby resulting in higher risk premiums in the front end of the curve.
Can forward guidance credibly be strengthened?
Can put the genie back in the bottle? The situation in Europe offers a cautionary tale, as efforts by the BoE and ECB to keep short rates low have proven unsuccessful to date. Increased uncertainty about the path of US short rates in coming years has already prompted the swaption market to price in higher implied vol in 5y tails than 10y tails (across all expiries) for the first time since 2011.
In our view, a major obstacle standing between the Fed and lower front end rates, besides uncertainty about the next Fed chair, is the persistent decline in the unemployment rate that has occurred despite lackluster GDP and payroll growth. The unemployment rate now stands at 7.3%, down 0.8ppt over the past year and just 0.8ppt above the FOMC’s 6.5% forward guidance threshold.
Of course, Fed officials have gone to great lengths to explain that 6.5% is not a trigger, but it becomes harder for the market to price a prolonged period of nearzero rates the closer the unemployment rate gets to the Fed’s estimate of full employment (5.2% - 6.0%).
Fed officials have given some consideration to strengthening forward guidance, for example by lowering the unemployment threshold, according to the July FOMC minutes. However, the Committee does not appear very close to making a decision on that point, and some participants have expressed concerns that lowering the unemployment threshold now could introduce uncertainty about how the reaction function may be shifted (perhaps in a hawkish direction) in the future.
In addition, explaining to the market why the unemployment rate is now viewed by the FOMC as a misleading indicator of progress toward full employment (due to falling participation) will be challenging given the Fed’s past use of this indicator to guide market expectations for both the end of asset purchases (~7.0%) and the beginning of rate hikes (<6.5%). Said another way, moving the goal post may not necessarily help to clarify the reaction function.
Recommended positioning: sell rallies in 5s
To summarize, policy makers are finding it harder to convince markets that central bankers have more insight into the future course of the economy and policy than they actually do. Meanwhile, markets are learning that it can be painful to rely too heavily on forward guidance when the risk/reward of being long fixed income is asymmetrical when close to the zero lower bound. In our view, this should lead to a return to persistently higher front-end risk premiums than have prevailed over the last two years, barring a sharp deterioration in the economic outlook.
For these reasons, we shifted our stance on the 1-5y sector following the June FOMC meeting from a buy-dips stance, which we stuck to for the last three years, to a sell-on-rallies bias. Crowded front-end positioning also increases the risk of a short-covering selloff, as observed this week in the days leading up to the employment report.