Takeaways from Yellen and EM

Here are three takeaways from Yellen's speech:

* Somehow, the Fed's estimate of the long run trend in inflation has not budged from 2% since the mid-90's, despite a straight-line decline in durable goods prices, globalization, and the decline in...err.....the long term trend in actual reported inflation.

*  The reason for this is that the Fed's estimate of the long-term trend in inflation is based on the output of a macro model (y'know, the kind that failed to see the crisis and has completely misjudged the shape of the recovery) rather than on the actual trend of inflation in the US and globally.   During her speech, which focused on inflation dynamics, Yellen used the word "model" 13 times; she used the words "globalization", "China", "trade", "outsourcing", and "non-tradeables" a grand total of zero times.

*   Take whatever you want from her comment that she anticipates raising rates by the end of the year.  Unfortunately, when the inputs to your decision-making process are garbage, the output is likely to be garbage as well.

Switching gears, Macro Man would like to make a few observations about emerging markets, generally and specifically.  Chances are that unless you are over 50, most of your career has been spent with EM in a secular bull market (albeit with the occasional hiccup.)  If one believes that price ultimately dictates trend, then a pretty good argument can be made that the halcyon days are gone for the foreseeable future.   Over the last couple of weeks, the ADXY (i.e., the Asian currency index) has broken its long term support dating back to the beginning of the millennium.

What's particularly telling is that Chinese slowdown notwithstanding, Asia has been an oasis of relative calm in comparison with the debacles in Brazil, Turkey, etc.  A primary reason for that is a legacy of the Asian crisis- namely, that most Asian countries have not taken on the type of external debt that was their undoing in the 1990's.   (While HK and Singapore score poorly on the chart below, recall that they are international financial centers and the chart only shows liabilities, not assets.)  Brazil and Turkey, uncoincidentally, score very poorly indeed.

What's notable is that even cyclically, there are relatively few changes in trend for the ADXY; as we know with EM, serial correlation is very high, and downdrafts tend to be very steep (and overshoot.)  Macro Man crafted a very simple model to test the hypothesis using monthly data and the 2 year moving average.   Simply put, if the ADXY is above the MA, the model's long, and if it's below it, the model is short.

Note that this model is really just trying to gauge the efficacy of the 24 month moving average as a predictor of future price action.  The results above do not account for transaction costs or carry, both of which would detract from actual performance.  What the model does suggest, however, is that when the ADXY drops below that moving average, as it did last September, it tends to keep going.   To would-be buyers of EMFX: caveat emptor!

Finally, a bit more on Brazil.   USD/BRL cratered today on central bank selling, providing some hope that the worst is over.  Perhaps it is, and Bacen's magic beans will work.   Color Macro Man skeptical- does anyone else remember the 2014 emergency rate hike that "saved" the TRY 70 figures ago?

One of the unfortunate unintended consequences of ZIRP-world has been a tendency to focus on nominal rather than real rates.   However, as veteran EM watchers will attest, when it really hits the fan and inflation becomes an issue, real rates matter.  A lot.   With the help of a mate, Macro Man was able to get pricing for the Brazilian DI curve, which provides insight to what's priced for Bacen over the next several years.

Simply put, to Macro Man's jaundiced eye, it doesn't look like enough.  Observe the reaction function not only during the early 2000's currency/inflation scare, but also the mini-inflation scare a decade ago.   Unlike the US, Brazil is (more than) capable of having a domestically-generated inflation scare, and Macro Man wouldn't be surprised to see inflation tick another 3%-5% higher given currency pass-through and a negative feedback loop.

History (in Brazil and elsewhere) suggests that high inflation countries with large current account deficits and external funding problems need to jack up real rates to put a muzzle on currency crises.   Right now, the curve isn't pricing much in the way of higher real rates given the inflation outlook; understandable, perhaps, given the negative carry of being paid at 17% (while receiving a DI rate currently in the low 14's), but nevertheless wrong.

Macro Man would expect real rates to approach 10% before it's all said and done, with Bovespa P/E's in single digits (as they were a decade ago.)  As it is now, the forecast PE for Brazilian equities (per Bloomberg) is roughly the same as that for Germany.  Sure, Germany's got VW, but if you've seen a chart of Petrobras CDS recently you'll know that Brazil would love to have that kind of "first world" problem.

As it is, it's dealing with emerging world problems- and Macro Man cannot help but think that things will need to get a lot worse and overshoot before they get better.