The Problem With Small Surprises
The phrase “paradigm shift” is meant to sound dramatic. Like the sudden slipping of tectonic plates, a paradigm shift moves mountains (metaphorically). Deregulation of airlines caused a paradigm shift, transforming airplanes from airborne luxury cruises to cattle cars. Decimalization of equity bid/offers caused a paradigm shift, dropping bid/offer spreads about 90%.
Economic paradigm shifts can also be dramatic, as when the credit crisis of 2008 caused lending to contract almost overnight. But not all shifts are dramatic. The paradigm shift from horse-drawn carriages to ‘a car in every garage’ brought the world dramatically closer together. But it took a long time before it happened…before, anyway, it was obvious that it had happened. The paradigm shift was clear in retrospect, but not in prospect. It wasn’t as if people were waiting for the world to change: Henry Ford famously said that if he’d asked his customers what they wanted, they’d have said “faster horses.”
Core inflation, by its nature, rarely produces good surprises. Friday’s release was a case-in-point. Forecasters were looking for a +0.2%, and got a +0.2%, but they were actually looking for about +0.17% and got +0.23% instead. It doesn’t even look like a surprise, when the rounded data is announced, but it assuredly was one. The y/y core CPI, which decelerated last month for the first time in 16 months, rose back to 2.3% (although just barely, and shy of the high from January), after last month’s decline had produced a chorus of predictions that core CPI for the year would end up being in the low-to-mid 1% range.
It is hard to imagine that, following the 2008 collapse of Lehman Brothers, there were very many who didn’t see the paradigm shift. On the other hand, Henry Ford churned out millions of Model Ts before the wagon-wheel manufacturers went bust. Could we be in the midst of an inflationary paradigm shift without knowing it? Put another way, does the fact that many economists deny such a shift imply that there isn’t such a shift? Well, would the fact that many analysts still projected record profits for buggy-whip producers, which could plausibly have happened if our current research structures existed back then, have implied that there was no revolution happening in locomotion? I personally think it would have been very remarkable if an analyst covering such companies had deduced that the Ford development would change the demand/supply balance for automobiles in such a dramatic way.
This is not just abstract musing. In 2008, although the Fed was audibly speculating in May about how quickly the exceptional easing in the spring of that year (and late 2007) would need to be reversed, by October there was no mistake (although it took until December to lower the Fed Funds rate all the way to 0.25%). Missing the paradigm shift has implications, and thankfully the central bank didn’t miss this one although I would argue with what they did – I would not argue that central bank response of some kind was required. The accumulation of small misses, though, is more insidious. Traders know the phenomenon of getting nicked to death with small losses and only later realizing that these were signs that the game had changed. Economists are not good at this. Rejecting the null hypothesis requires a movement sufficient to refute the presumption that the result was merely noise. How do forecasters do that when the misses, as with core CPI, are small (but persistent)?
Part of the answer depends on your hypothesis. A truly ridiculous hypothesis – say, that the output gap drives inflation, so that we should have experienced a savage deflation in 2009-10-11 – should be simple to refute; however, we’ve seen even in this case that when the hypothesis is rejected, economists sometimes try to “fix the model” enough so that the hypothesis will not be rejected this time. This is bad economics, and it often comes even from otherwise good economists. It is hard to let go of a cherished old paradigm.
Better hypotheses, but still incorrect ones, create smaller misses, so that one point is almost never sufficient to reject the null. There are methods which can help identify when a series of misses is sufficient to (jointly) reject the null, although these have their own issues. More important, though, is the problem that economists are human and hate to be wrong, so they tend to try and reinterpret hypothesis rejections (I also hate to be wrong, but it happens so much that I am used to it; and, as a trader, I am at peace with fallibility because I must be).
Returning from the philosophical from the practical: at what point does the Federal Reserve admit at least internally that inflation is not going to self-regulate just because they say so? Already, we’ve seen the Fed move the goalposts from the presumptive 1.75%-2.00% target on CPI to an official 2.00% target on PCE, which is a 25-50bp nudge, and have heard many expressions from officials of the sentiment that “in the short run, inflation might exceed the target somewhat if we feel it is going to return to target.” That seems to me to be an approach designed to miss a paradigm shift as long as possible.
Year-on-year core PCE this month will likely reach the 2% target. Now what? I believe we may be in the middle of a paradigm shift from the 1%-2% low-and-stable inflation we have experienced in recent years to higher and less-stable inflation. The startling buoyancy of housing inflation, the possible shift in apparel inflation dynamics back to the 1970s-1980s paradigm, and the policy straitjacket all suggest that something very different is happening, or could be happening, at the moment. By the straitjacket, I mean that draining reserves, as necessary as it may become, will probably be more difficult to do technically than the Fed lets on, and would almost surely cause a large spike in interest rates while the Treasury continues to come to market with trillions more in new debt every year. Draining reserves while the Unemployment Rate is above 7%, especially during an election year, is almost out of the question. Trying to talk inflation down, while the only option at the moment, burns the credibility log (and like all logs, it can only be burned once).
I continue to believe the Fed is in a pickle, and the pickle will be worse the more time passes before the Committee recognizes that the inflation paradigm may have shifted. As I said, that will probably take some time.
The fact that Europe seems to be sliding back down from the comfortable but unstable equilibrium it had reached in March does not change my view. If anything, the probability that global central banks will continue to pour fuel on the fire in the form of banknotes increases, rather than decreases, the inflation risk. The news over the coming weeks and months will focus on the dramatic, while the important inflation paradigm shift takes place quietly, with small surprises, in the back pages of the financial papers.