The world as it is, not as we'd like it

It happened a day early, but we got our anniversary fixed income melt-up after all.  Although a weak retail sales figure (and Wal Mart earnings warning) was the ostensible catalyst, in reality Macro Man suspects that the market had been stewing on the flurry of dovish developments to have emerged from the Federal Reserve this week.  The poor number was merely the last straw that encouraged the market to go full-throttle on challenging the notion of lift-off by year end.   The result?  Dec 2016 eurodollars rallied more and closed higher than they did on payroll day.  Astonishingly, the contract has risen 40 ticks since the end of August.

The analytical and journalistic reaction to the dovish offensive from Brainerd and Tarullo this week has been revealing.  "The Fed cannot go with a body of dovish dissenters" seems to be the conclusion, even though the institution has managed to conduct policy just fine over the past few years with the occasional hawkish dissent- both verbal and voting.  In fairness, none of those dissents came from the Board of Governors, and while all voters are theoretically equal, it pays to remember George Orwell in Animal Farm: all animals are equal, but some animals are more equal than others.

Macro Man has found himself treading a fine line between seeing the world as it is, and seeing it as he would like it to be.   Being "right" intellectually is useless if the relevant policy-makers do not see things the same way, and at the end of the day it is their decisions that drive market pricing and P/L, which is the ultimate arbiter for market practioners.

Academic economists at universities and, yes, the Fed can wave away failed predictions with the sweep of a hand; failed trades, on the other hand, are not so forgiving.   Macro Man was struck by the contrast in reading the latest paper from the San Francisco Fed, which can only be described as a piece of dovish propaganda.

The paper deals with the so-called "natural rate of interest", a theoretical notion brought to the mainstream by the father of forward guidance, Michael Woodford.  The natural rate of interest is, to quote the paper, a construct that represents the real interest rate that is "consistent with an economy at full employment and with stable inflation."  OK, that seems simple enough.  The model output currently suggests a natural real rate of -2.1%.   Wow, that seems low, particularly with the unemployment rate at 5.1%.  The model does not project the real rate turning positive until late next year or 2017.

                                 Source:  SF Fed

The author helpfully provides historical real-time estimates of the natural rate, which is where things get weird.  As you can see, each snapshot starts with the natural rate estimate at very low levels, and the projects it to rise to turn positive within a year or two, much like the current forecast.   What makes this so bizarre is that this time frame coincides with the unemployment rate falling much more quickly than expected.....and yet the natural rate has risen much, much more slowly than expected?   By any reasonable definition, we have approached full employment more quickly than macro models have anticipated over the past few years (even if we are still a ways away, as many believe), and yet the model has continued to suggest a lower natural rate than it itself projected.  Macro Man has not delved into the nuts and bolts of the model's construction, but can only presume that there is an implicit Phillips curve in there that is much, much steeper than reality.

Now, in the hard sciences, any model that was so consistently wrong or contradictory would likely get chucked in the bin.  While economists and econometricians  certainly use a lot of mathematics in their work, what they do is clearly not a hard science, and as such is subject to the (substantial) biases of the analyst.  Of course, if the analyst has the ear of influential policymakers, his work is worth following, regardless of any biases that might be in play.

In this case, the contrast with the Taylor Rule is really quite striking.  Macro Man ran a standard Taylor Rule projection, using a NAIRU estimate of 5.0, an inflation target of 2.0, coefficients of 0.5 for inflation and employment, an Okun factor (representing the relationship between growth and employment) of 2.0, and a real rate of 2.0.   Unsurprisingly, the model thinks rates should be closer to 3% than zero.

The model clearly captures the "QE era," though in this iteration thinks rates should have gone positive by the end of 2011.   Of course, such a simple model can often jump around more quickly than it is practical for monetary policy to adjust, as you can clearly see from the chart.  For this and other reasons, so-called "inertial" policy rules have come into vogue, in which any prescription from the standard rule is essentially diluted by the current policy setting.  For example, if rates are zero, the standard rule prescribes a 3% policy rate, and the rho (i.e., how much the current policy setting is weighted relative to the policy rule) is 0.5, the inertial model would prescribe a policy rate of 1.5% (0% * 0.5 + 3% *0.5).

Unsurprisingly, an inertial model seems to do a much better job of capturing how the Fed actually behaves.   Many models, such as this one from 2008, use a relatively high rho, thus heavily weighting the current policy setting in the prescription.  For example, using a rho of 0.7 makes the fit look great....except it still prescribes the FF rate at 0.94%.   From the perspective of "absolute emergency setting are no longer needed", that doesn't sound like such a bad prescription.

OK, but since the crisis the argument can be made that the equilibrium real rate has gone down...what if we take that to 1% beginning in 2009 (which would render the current standard Taylor prescription as 1.85%)?  It starts to become interesting...the model suggests negative or zero rates until just before the announcement of the taper, and now prescribes a policy rate of 0.64%.  Either literally or figuratively, this is no doubt how a number of FOMC members are viewing the situation.

For fun, Macro Man set the unemployment and inflation policy goals at their extremes of the past 20 years- namely 3.8% and 2.5%, respectively.  Amusingly, the standard model still prescribes higher rates, albeit only at the 0.40% level.   The inertial model suggests a policy rate of 20 bps.

What does it all mean?   Macro Man will admit that he has probably underestimated the inertial considerations that are influencing the Fed's policy calculus.  Much as he would like them to take a step back and say "given where we are, zero rates are ridiculous", he has to concede that they are a model driven lot and simply don't operate that way.  Similarly, while he thinks that their inflation target is largely unattainable and therefore nonsense, it is there and the FOMC is taking it seriously.

The chances of seeing meaningful progress on the inflation front over the next couple of months seems fairly slim, and the FOMC therefore seems unlikely to move rates by the end of the year.   This is not exactly news to anyone with a set of eyes and as pulse, and Macro Man concedes that this feels an awful lot like stopping out at the lows.  Intellectually, it kind of is, though market-wise, not necessarily.

He closed the balance of his USD long on Tuesday night after yesterday's post was written, fortunately at levels well better than those prevailing at the current time of writing.  The breakout in gold is certainly suggestive that the FX market has not fully priced a Fed potentially on hold for many more months, as is the breakdown in USD/JPY.  Flat positioning seems the best idea for now until things clarify a bit.

As Macro Man has noted previously, he does not think that a continuation of ZIRP and an extension of lift-off uncertainty is necessarily a positive for equities, presuming that the Fed continues to direct markets that the next move is lift-off.   (If the Fed were to hint at QE4 or - God forbid!- negative rates, then all bets would be off.)

As for fixed income, he does think that the next policy move is a lift-off and does like a lot of ED curve trades to play for steepening.  The bull flattening has been remarkable, and there are some very nice technical set-ups that offer excellent risk-reward.  It's a relatively simple matter to adjust positioning a bit further out the curve to reflect the (likely) inflection point that the analytical community will shift their forecasts towards in the coming days/weeks.

As regular readers no doubt know, Macro Man thinks that the Fed should look through the current spell and end this nonsensical wait for the lift-off.   Who knows....maybe, come December, they still will.  But with markets deferring their rate hike pricing into next year and the Fed seemingly unwilling to strong-arm market expectations to what they, the Fed, would like to do, it seems as if the FOMC will play the prevent defense for a while longer.   At the end of the day,  we have to analyze and trade the world as it is, not how we'd like it to be...and through that prism, it looks like policy inertia will continue to win the day for a bit longer than Macro Man previously thought.....or would like.

Tomorrow:  some thoughts on financial conditions and calculating a downside to ZIRP.