China and the Fed, Part 2

For what seems like the thousandth time over the past few years, the market has read the tea leaves from Fed communication (in this case, the minutes of the July meeting) and apparently come to the conclusion that any sort of tightening is rather less likely it seemed beforehand.  Some of this is no doubt a Pavlovian response or, to put it somewhat more charitably, the triumph of empiricism over theory.

The trigger, of course, was the twin mentions of the risks posed by China and the rising US dollar.   When one considers the events of the other week- when China weakened the RMB against the dollar as a policy decision (cloaked however it may be in the guise of "market enhancement"), it seems eminently reasonable to suggest that anyone worried by these factors last month will be even more cautious now, thus rendering the chances of an imminent rate hike remote.

Of course, there was a line about the risks to domestic economic and labor market activity being "nearly balanced", but hey- the Fed has spent the last six years maintaining an Eeyore-ish fascination with doom and gloom, so it's hard to fault the market too much for assuming that the future will look like the past.

All of this, however, ignores a sublime irony.  China finds itself struggling to extricate itself from a misallocation of resources at least partially caused by its decision to essentially delegate its monetary policy to the Fed via the FX peg.   Sure, PBOC had its own policy rates, instruments, RRR, etc...but for a long period of time domestic policy in China was much easier than it should have been, partially to mitigate the capital account implications of the peg.  Now that the capital account is offsetting the current account, by and large, the significance of the issue has faded.

However, the market has apparently decided that the Fed will now delegate its monetary policy to the PBOC; in other words, the dollar's strengthening against the RMB thanks to the peg adjustment will now take the place of the putative first rate hike in September.  There is of course no guarantee that the Fed will do this, but the minutes certainty suggest that this type of mentality is held by at least a few committee members.

Needless to say, Macro Man doesn't think that delegating US policy to PBOC will work any better than China's delegating its policy to the Fed was.  He was intrigued, therefore, to see that equities- that most Panglossian of asset classes- shrug off the "good news" and finish on a downer.

A growing fear of a Chinese meltdown?

An absence of buybacks?

The weakness of the median stock finally filtering through to the average?

Algos playing "orca and baby seal" with the market on a hot August day?

Macro Man isn't really sure, but in normal circumstances the divergence of price action to newsflow would be telling.

None of this is to say that the Fed shouldn't think about China.  Of course they should.  However, the focus should be on the secular impact on inflation, rather than the cyclical impact on growth.  Macro Man has addressed this issue a number of times, both on wages and on prices.

Following on from the previous post, Macro Man built himself a simple model that attempts to project changes in the trend of CPI inflation in the United States.  There are only two explanatory variables in the model- the nominal GDP less Fed funds monetary policy proxy discussed last week, and the y/y change in nonfinancial private sector credit.  The model attempt to capture the delta of CPI over a two year horizon.   (For example, if at time t CPI is 1% and the model delivers a reading of +2%, the implication is that it expects CPI to be 3% in t + 2 years.)

It's fairly simple stuff, but the model worked surprisingly well from 1967 to 2001, as shown below. (Caveat: the entire 1967 - 2013 period was in-sample, so of course it should look good at some point.)

                                       R-squared = 0.57

At the end of 2001, something monumental happened: China joined the WTO and became the world's manufactory.  US firms in particular took advantage of the outsourcing opportunities to boost margins while flooding their domestic market with cheap Chinese goods.   Here's what the same model, with the same coefficients, looks like from 2002 to the present day:

                                       R-squared = 0.014

Put it all together and it looks like this.

                                       R-squared = 0.41

Interestingly, running the regression with core CPI actually reduces its accuracy by a small amount- so much for core inflation being a "superior measure!"  In any event this seems to be another brick in the empirical Great Wall of Evidence that China has had a meaningful downward impact on inflation in the US and other industrialized countries.   To ignore this impact would appear to be naive at best, yet we still see the Fed apparently operating under a domestic Philips curve model of inflation.

At the end of the day, Macro Man doesn't actually think that the Fed will outsource policy to China (though it's going to be awfully difficult to distinguish between doing that and standing pat from here on out.)  He would, however, like to see a more realistic assessment of the inflation target; given the global backdrop, hitting 2% on the core PCE deflator for any meaningful amount of time looks like dreaming the impossible dream.  Somewhat fittingly, that song is from the musical Man of La Mancha; expecting that monetary policy can drive inflation towards its target is little better than tilting at windmills.