In the first week of February, in the trading session just before the February 5 VIXtermination, the market tumbled as a result of a January average hourly earnings number that surged (even though as we explained at the time, the market had wildly misinterpreted the print), prompting speculation that the Fed was dangerously behind the curve and would need to accelerate its tightening, potentially hiking rates more than just 4 times in 2018, leading to an accelerating liquidation of risk assets which eventually culminated in the record VIX spike.
Since then, inflation fears moderated following several downward revisions (as expected) and more tame hourly earnings prints, with market concerns instead shifting to trader wars, the return of populism to Europe, the tech bubble, and the sustainability of record margins and net income.
But according to a recent analysis by Deutsche Bank's Aleksandar Kocic, traders are ignoring the risk of an imminent, "phase shift" spike in wages at their own risk. Specifically, Kocic looks at the current locus of the Philips curve - which many economists have left for dead due to its seeming failure to explain how the plunge in unemployment to record low levels has failed to boost wages - and notes that as the economy approaches the full employment, "wages tend to become more responsive." This, to the Deutsche Bank analyst, "is the inflection point that the Fed is monitoring."
Looking at the Phillips curve over the past 4 economic cycles, Kocic compares it to the Cheshire cat's smile from Alice in Wonderland, which is present even when the actual cat body is no longer there: "In each cycle, it falls apart, but after every annihilation, it re-composes itself and continues to play an important role."
Specifically, what Kocic highlights, is the sudden phase transitions between the end of one cycle and the start of another, in which one observes a "near vertical" spike in inflation to the smallest favorable change in underlying conditions. In the DB chart below, each cycle has a different color which implicitly marks their beginning and end.
In the current context, the most important message of this graph is the finale of each recovery. In the past, this stage always exhibited a dramatic (practically straight line) rise in wages in response to infinitesimal improvements in economic activity. These periods are highlighted with (almost) vertical lines in the chart.
Of course, as economists have long lamented, what has prompted many to speculate that the Phillips curve is broken or even dead, is the failure of the economy to respond with a sharp increase in wages to the already near record low unemployment rate, which is over 0.5% below what the Fed currently believes is the NAIRU (non-accelerating inflation rate of unemployment), or the unemployment rate which does not cause inflation to rise or decrease.
The NAIRU level corresponding to the 1990 is 6%, and 5% in 2000 and 2006. Currently, NAIRU is around 4.5%. This is the territory where the Fed is likely to become concerned. Allowing the economy to overheat would jeopardize Fed’s ability to control inflation if it rises too fast.
Another way of visualizing the potential risk facing both the Fed and traders is whosn in the next chart, which looks at the Phillips curve within the latest economic cycle, starting with the early stagflationary months of 2007, then progressing into several years of recession, before ultimately emerging into the recovery and, finally, "growth and inflation" phase. According to Kocic, "we are currently in the final stages of the recovery. If we enter the goldilocks region (upper left corner) too fast, the Fed could be caught behind the curve and might be forced to hike aggressively which could have a negative impact on growth while leaving only inflation behind"... a recipe for progressing straight into another recession.
As the Deutsche strategist warns, and as the events of early February so vividly demonstrated, "this is the most dangerous development which the Fed would want to avoid."
Underscoring why it is only a matter of time before the Phillips curve, which he believes is not dead, but merely waiting to erupt in "a near vertical spike" reveals some fireworks, Kocic notes that "in addition to the distance from the NAIRU and the inflation target, the actual non-linearity of the curve is the key development to watch, in particular the change in response of inflation to the decline in unemployment" and adds:
The economy is currently operating below NAIRU and just slightly below the target and has shown a sustained support for higher slope which has been persistent since the end of 2016. This has caught the market’s attention.
But it's not just the Phillips curve that has caught the attention of Kocic. As regular readers will recall, back in the summer of 2013 we first speculated that Okun's law is either broken, or there is a giant, and inexplicable output gap forming in the US economy, in which GDP was trending far below where the unemployment rate suggested it should be.
Now, five years later, that output gap is finally closing, and according to Kocic if past is prologue, it may do so in an "explosive" manner. This is how he frames it:
The figure below shows the Okun’s law which captures the interplay between the social and economic response to the crisis in terms of distance of unemployment rate from NAIRU (social) and output gap (economic). Historically, output gap closes roughly when unemployment reaches NAIRU (the long term intercept / beta is close to zero). Post-2007, there has been a structural break in this relationship: The economy was slower to recover – it required a more aggressive improvement of labor market in order for the growth to reach its potential. This is shown in the figure through lower intercept: unemployment has to decline about 0.8% below NAIRU for the gap to close.
So, according to Kocic, if the unemployment rate is already low enough to potentially trigger a "near vertical" spike in wages (Phillips), it is also low enough to launch a sharp reversal in the output gap (Okun) leading to a Fed that is behind the curve on the parameters, or as the DB analyst says, with "current unemployment rate already 0.5% below NAIRU it is only a quarter of a percent away from closing the output gap. This justifies possible Fed’s concerns."
Then again, if Kocic's historical analog for the current situation is correct, "nightmares" may be a better word than "concerns." The reason for that is that the appropriate historical period to compare the current regime to is the 1960s, when the output gap "exploded":
While we think that history might not be the best guide for the future at the moment, the lessons of the 1960s are difficult to be completely ignored. The figure below shows the unfolded Okun’s law. 1960s are a screaming example of the effect of “exploding” output gap as a consequence of injection of fiscal stimulus when the economy was already operating at full employment, a situation that bears keen similarity to the present.
If the DB analysis is accurate, the implications would be profound for the market, which would find itself not only observing a sudden spike higher in wages, but also an economy where the actual GDP is suddenly soaring far above the potential, resulting in a 1960s output gap rerun and a Fed that has never been so far behind the curve.
The outcome of both would be even more aggressive rate hikes by the Fed, which - if the scenario plays out as Kocic envisioned - would be forced to raise rates well more than the 7 or so hikes the latest dots currently envision for 2018 and 2019, resulting in another bout of exploding volatility as the market finds itself not only chasing the dots, but in a rerun of "February" where to catch up to inflation, first the Fed, and then the market would be forced to aggressively tighten financial conditions.