Markets may be underestimating inflation

The collapse of crude oil prices since mid-2014 has not only devastated the oil industry, it has also sowed confusion over the issue of inflation. Since crude prices started falling in the summer of 2014, headline inflation fell from 2% to essentially zero for most of this year. Yet if we exclude energy from the calculation, inflation has been fairly steady at 2% for a long time. This bears repeating: if not for the big decline in oil prices in the past 18 months, the underlying trend of inflation at the consumer level would have been 2% for the past 13 years.

Energy prices are not going to fall forever; once they simply stabilize, then headline inflation should bounce back to 2% fairly quickly. The Fed understands this, but—arguably—markets do not, since key measures of inflation expectations over the next 10 years range from 1.25% to 1.8%, and those expectations assume that oil prices will decline substantially further in the next year or so. We need to keep in mind that the active drilling rig count in the U.S. has declined by two-thirds in the past year (i.e., future oil supplies are going to be expanding at a far slower rate), while vehicle miles driven in the past year are up over 4% after being flat for the previous six years (i.e., the demand for oil is rising meaningfully, thanks to lower prices). Markets are adjusting rapidly to lower oil prices, so today's oil glut could easily disappear within a year or less.

The chart above shows the inflation-adjusted price of crude oil over the past several decades. Prices have been extremely volatile, ranging from a low of just over $10/bbl to a high of almost $150/bbl. Since mid-2014, oil prices have fallen by a staggering 65%. We've only seen one episode of similar magnitude, and that was in 1986. (I don't count the 2008 oil price collapse, since it was preceded by a spike in prices and everything was very turbulent around that time. But now, as was the case in 1986, conditions have been relatively stable.)

As a result of the shock of sharply declining oil prices in 1986, the CPI fell from 3.9% in early 1986 to 1.2% by the end of 1986, while the CPI ex-energy remained around 4% throughout. Six months after oil prices stopped falling, the CPI had returned to 4%. The chart above gives you a feel for just how much energy prices have contributed to the volatility of consumer price inflation, and it highlights the 1986 oil shock and the current oil shock. There is every reason to believe the current episode will play out in a similar fashion to the 1986-1987 episode. Once oil prices stop declining, the headline CPI will probably return to the underlying rate of non-energy inflation (2%), most likely by the middle of next year.

The chart above shows the level of the CPI ex-energy, plotted on a log scale. The dotted green line represents a 2% trend, which has been remarkably durable. For the past 13 years ex-energy inflation has averaged almost exactly 2% per year, and there is every reason to think this will continue for the next year. Seven years of near-zero interest rates did not "stimulate" inflation (nor the economy in general, I might add), so why should a modest increase in interest rates depress inflation, much less create deflationary conditions? The deflation threat is a myth, and since the risk of deflation is nonexistent there is no reason for the Fed to postpone liftoff.

Now let's look at the market's current inflation expectations, as embedded in the pricing of TIPS and Treasuries.

The chart above shows the nominal yield on 5-yr Treasuries, the real yield on 5-yr TIPS, and the difference between the two, which is the market's expectation for the annualized rate of CPI inflation over the next 5 years. Currently about 1.25%, it is significantly less than the prevailing rate of ex-energy CPI inflation. Either the market is correctly anticipating more and substantial declines in energy prices, or it is seriously underestimating future inflation. For example, if inflation were to average zero for the next two years (e.g., because of further declines in energy prices), and then rebound to 2.0% for the subsequent three years, the annualized rate of inflation over the next 5 years would be 1.2%. This scenario is pretty close to what the market is apparently assuming will happen.

The chart above shows the market's forward-looking inflation expectations (technically speaking, the 5-yr, 5-yr forward inflation expectation). This represents the market's expectation for the annualized rate of CPI inflation in years 6-10. Currently 1.8%, it is only marginally less than the underlying rate of ex-energy inflation. Over a multiple-year period, then, the market expects inflation to rebound to what it has been over the past 13 years.

If the market is making a mistake, it is in expecting the underlying rate of inflation to remain unusually low for the next few years, before rebounding within 5 years or so to the same 2% level we have seen over the past decade. The surprise, in other words, would be for inflation to end up closer to 2% than to zero over the next few years.

An inflation surprise next year might well unsettle markets, since today's benign expectations assume that the Fed will be very slow in raising interest rates. But at the same time, an upside inflation surprise would put a dagger through the heart of the market's continuing worries over deflation, and how deflation might (supposedly) act to slow the economy's growth rate. It's tough to know how this would all play out next year. But if inflation does surprise on the upside, then the Fed's decision to raise rates this week will prove completely justified, if not too little, too late.

The Fed has every reason to raise rates by 25 bps at this week's FOMC meeting.