We think it is fair to say that the meeting provided a modestly hawkish surprise last week, with the main element being the degree to which the dovish set of dots moved higher in 2015/16. That said, although key FOMC members seem more comfortable with a slightly earlier start (mid-2015) to normalization, they continue to project a very gradual hiking-cycle pace (roughly 1.25% in 2016 and 1% in 2017, based on the fourth dovish dot).
This has led to a far more supportive environment for longer-dated forward yields, which rallied on FOMC day and have continued to outperform since then; they now stand roughly 20bp lower than before the FOMC.
This price action has been exacerbated by the underperformance of risk assets across the board, somewhat reminiscent of what happened during the taper tantrum last summer. Figure 2 shows that emerging market currencies have underperformed, carry trades have been negatively affected (EM and mortgage ETFs) and spreads have generally widened.
The perception of the Fed’s reaction function is again turning hawkish, as is evident in the rise in real yields and tightening in breakevens (medium-term breakevens are close to the lows seen last summer). While the monetary policy outlook is gradually becoming less certain, it is not clear whether the Fed’s reaction function has turned more hawkish, as dovish Fed participants continue to project a very gradual path to normalization.
Later this week, the BLS is scheduled to release the September payroll report, which should give an updated view on the labor market. Our economists expect a strong report, with a 250K gain in payrolls and the unemployment rate to decline to 6% (see US employment report outlook, 25 September 2014). As Fed Chair Janet Yellen noted at the September FOMC press conference “If the economy proves to be stronger than anticipated by the Committee, resulting in a more rapid convergence of employment and inflation to the FOMC’s objectives, then increases in the federal funds rate are likely to occur sooner and to be more rapid than currently envisaged”.
Labor market conditions have improved over the past year across all indicators. In some areas, conditions are even better than they were in 2006 (eg, initial claims and the difficulty of filling jobs). Furthermore, in 7 out of 12 dimensions on the chart, labor market conditions are close to or better than the levels of early 2004; we are still presumably three quarters away from the first hike, suggesting that the Fed – appropriately, in our view – is not rushing toward normalization.
On aggregate, they are close to early 2004 levels. The spider chart also highlights that along some metrics, there has been less progress; for instance, the long duration unemployment rate, part-time employment and hiring rates are still worse than 2004 levels. These also happen to be key variables that Fed Chair Yellen has focused on recently.
Therefore, beyond the headline numbers on Friday, it will be critical to see if the recent solid progress in the U6 unemployment rate and long duration unemployment is maintained. Last, but not least, the pace of the hiking cycle is likely to be a function of wage/price inflation trends. While wage inflation for production and non-supervisory workers has risen from the lows (latest y/y at 2.5% vs post-crisis low of 1.3%), the series for all workers has been hovering close to 2%. Any upside surprises on this front will also have an increasing impact on rates markets, especially as we move closer to normalization.
Overall, we remain modestly bearish on US rates. We maintain our view that belly-long end curve flatteners (7s30s/10s30s) still offer better risk-reward than outright duration. With term premium negative in the belly of the curve, it should suffer the most on a positive payroll surprise (Consensus: 215K and 6.1% UR) and a negative surprise on wage inflation could still lead to a bull flattener. We also remain short the belly of 1yf 2s5s10s and 5s10s30s flys as an attractive risk-reward alternatives to position for a rise in term premium.
At the Jackson Hole symposium, she noted that the Fed was looking beyond just the headline payroll numbers to gauge the progress in the labor market. In Figure 3, we plot a spider chart that summarizes information across a range of labor market dimensions. The y axis has been calibrated, with the outermost band representing “ideal” labor market conditions (Q4-06, all values as 1); the innermost band represents the height of the recession, when labor market conditions were at their worst (Q4-09, all values at 0).
The other bands represent snapshots capturing the progress along each dimension. For instance, the blue band represents current conditions; the green one, conditions just prior to the last hiking cycle in 2004.