Yellen's NAIRU

Macro Man would like to thank all who extended such a warm welcome upon his return to the financial blogosphere yesterday.   It was nice to see a number of familiar names in the comments section; he can only hope that they bear with him as he settles back into the routine of writing for public consumption again.

Janet Yellen gave a speech yesterday in which she appeared to push back ever so slightly against the apparently hawkish message delivered a couple of weeks ago.  That's a nuanced interpretation, and one in which Macro Man may be trying to read too much to confirm prior views.  But let's give it a go anyways.

The money line, as far as he could make out, was Yellen's comment that her estimate of NAIRU is "now between 5.2 percent and 5.6 percent."  If you need any confirmation that Yellen's on the dovish side of the FOMC, this was it, as the center point of her range (5.4%) was below that of the committee at large (5.6%.)  But how realistic is her assessment?

Although the Fed no longer makes public its estimates of NAIRU, the Philly Fed has a plethora of historical data on its website, including historical estimates of NAIRU based on the prevailing models from a given Greenbook.  The data is, to be honest, a bit of a mess, and difficult to distill in an elegant fashion.  The chart below shows all model datapoints since 1960, with the prevailing unemployment rate on the X axis and the corresponding NAIRU estimate on the y axis.  The area approximating Yellen's current forecast is highlighted by the arrow.





There are a few interesting points to distill from this analysis:

1) Yellen's NAIRU is very much on the low side of those observations occurring with an unemployment similar to the current one.

2) The model that most closely fits with Yellen's current NAIRU estimate is from August 2006; in other words, she evidently believes that the structure of the labour market has not changed a whit since the crisis.  Readers are invited to decide for themselves whether this is a realistic conclusion

3) One thing that Macro Man can say for certain is that NAIRU is not 6.7%.   Did you know that the last quarter to average a 6.7% unemployment rate was in 1949?  (For reference, the average unemployment rate since 1948 has been 5.82%, above the central tendency of Fed NAIRU estimates.)

Point 2 above is particularly relevant to Monday's speech, as Yellen took pains to point out reasons why much of the current unemployment is cyclical rather than structural.   Among the arguments employed were the low level of wage growth.  While it is certainly true that wage growth is below long-term norms, it is somewhat disingenuous to suggest that it has been stagnant.  Indeed, private sector wage growth (the only type she has any power whatsoever to impact) has accelerated smartly over the past eighteen months, and now rests above the average level that prevailed when the past two Fed tightening cycles commenced.




At the same time, Yellen herself noted that payroll growth has risen a resounding 41 months in a row.  Indeed, smoothed private sector job growth looks, if anything, a bit stronger than it did during the last expansion:




So.....job growth has been solid and wages are accelerating.   Friday's payroll figure is widely expected to show robust job growth, partially as a catch-up to the miserable winter-affected results from December and January.  How are we left?

With a central banker that is very probably over-estimating the degree of controllable labour market slack in the economy (government jobs, like finance, ain't gonna grow like they used to for quite some time) and suggesting that rates will rise only gradually one they start.

Hmmmm....Macro Man loves the smell of policy error in the morning.  But how does one play it now with any degree of safety?   Front markets have already repriced to a degree after the March FOMC and have reached a critical layer of support.   One could easily make an argument for the next 20 ticks in EDM6 in either direction.



While the market currently has a case of "six month fever", Macro Man would suggest that the risks to the consensus timing of a June lift-off are heavily skewed towards later rather than earlier.  Ditching QE was a decision that was fairly easy to rally round, particularly in the context of the guidance that was then in place.  Even so, the FOMC looks set to spread it out over the better part of a year to mitigate the negative impact of weaning markets off of monetary methadone.

While the exact definition of "considerable period" is, ahem, inexact, we do of course have the precedent of the previous FOMC tightening cycle.   The Greenspan Fed inserted referred to a "considerable period" in August 2003, decided it could be "patient" in January 2004, assumed that policy accommodation could be removed at a "measured" pace in May, and tightened in June.  If you're keeping score at home, that's a 10 month "considerable period."

Assuming that QE wraps up in October, that would imply lift-off in August 2015. Somehow, Macro Man could see a Yellen Fed rounding up to September (and then hiking in October and December as well.)

Given the implication of an underestimated NAIRU, however, he finds the notion that the Fed would engage in an on-again/off-again, hike-every-other-meeting tap dance to be a load of old rubbish.   History strongly suggests that monetary policy is highly inertial, meaning that it takes a lot to change the existent policy path.

So a trade that Macro Man quite likes for this combination of circumstances (possible later lift-off in 2015, but a steady skein of hikes thereafter) is a simple EDM5/Z5/M6 butterfly, selling the wings and buying the belly.  In doing so, you're essentially betting that LIBOR rates will rise more in 1H16 than in 2H15.  At the current retail price of -8, you're not risking a lot, as it seems highly dubious that the market will move to price rates moving more next year than in the first half of 2016. 



Meanwhile, this specific fly was in the high teens in late December when EDZ5 was at current levels...and the generic (pictured above) was at -25!  Indeed, from current pricing you could probably risk 6 ticks to make close to 20- that's a pretty good ratio, particularly when you consider that it rolls positively about 3 ticks over the quarter.

Obviously things could get a bit spicy come Friday, so retaining some firepower to sell closer to -5 is probably in order.  Either way, something tells Macro Man that this trade is more likely to be a winner than Yellen's call on NAIRU.