Last week Federal Reserve Chair Janet Yellen gave a speech at the University of Massachusetts in Amherst in which she outlined her views on inflation and its role in the U.S. central bank's monetary policy.
The entire speech is well worth a read, but for the moment we will draw attention to footnote 26 where Yellen highlights an intractable problem faced by Fed officials when looking for measures of inflation expectations that will help inform their policy decisions. Expectations can be gauged either through market-based indicators or by looking at survey results. Before deciding which set of indicators to give most weight to, policy makers have to decide who they should listen to -- markets or the business and consumers.
26. Another complication is that we do not know whose expectations “matter” for determining inflation. Inflation expectations of professional forecasters (such as those collected in the Blue Chip Economic Indicators, the Survey of Professional Forecasters, or the Survey of Primary Dealers) or inflation expectations derived from asset prices probably capture the views of participants in financial markets but need not reflect the views of households and firms more broadly. As an empirical matter, the little information available on the longer-term inflation expectations of firms from the Atlanta Federal Reserve Business Inflation Expectations survey suggests that firms’ expectations more closely resemble expectations from the University of Michigan Surveys of Consumers than the expectations of professional forecasters.
The economic argument here is that household and business inflation expectations matter more in the real economy as current inflation expectations should influence investment decisions.
For market participants, it is useful to know that survey-based indicators may be better at gauging inflation expectations, but they do have some disadvantages to market-based expectations. Survey-based indicators are only released periodically, and usually with a substantial lag. When market participants get the information it may already be stale.
No wonder, therefore, that market-based indicators are still preferred as they can be accessed in real time. The European Central Bank added fuel to this when it recently cited the five-year, five-year forward inflation swap rate as its preferred gauge for inflation expectations.
It's unfortunate then, for markets, that in a note to clients Bespoke Investment Group has just poured cold water all over the use of forward rates from the bond market as an indicator of inflation expectations.
In an heroic piece of charting, they looked at every possible permutation of the forward curve form the 12 spot breakeven yields they had access to. If forward rates do indicate current inflation expectations, it would be reasonable to expect they would be relatively smooth and generally point in the same direction.
Instead they look like this:
Bespoke point out that during the peak of the financial crisis some forward rates were nearly 15 percent, while others were at negative 8 percent -- at the same time. Even today, if we say forwards are predicting inflation, then we get this scenario:
... inflation starting a year from now will be the strongest for any period over the next 30 years, while inflation will also run hot for the 10 years 20 years in the future. But the 10-year period 10 years in the future (the decade immediately preceding that with the second highest forward rate) has the lowest inflation forward of any period.
Which highlights the problem with these as indicators -- they are all over the place. It is probably better to just use one signal, than comparing several as the messages quickly become mixed, but which to pick may be little more than an arbitrary choice. Also, the forward rates are volatile, so while survey-based indicators are occasional and delayed, at least they are not nearly as noisy as these indicators.
Worst of all, Bespoke say forwards are not even useful when trying to calculate inflation expectations. They are, in as much as they do anything, only reflecting the amount of risk the market is willing to expose itself to.
Bespoke finishes by saying:
... in our view forecasting inflation is much, much more valid when using actual economic theory and input, rather than obscure relationships between various points of a yield curve built by subtracting two bond yields.
In the footnote to her speech Yellen asks "whose inflation expectations matter?"
In attempting to answer that, we should be clear as to what is and is not an inflation expectation. From Bespoke's note, it seems clear that bond market forward rates are the wrong place to start.