Hidden Message of the Fed's Dot Plot
Of all the maddening things about this month’s Federal Open Market Committee meeting, perhaps the single most annoying is the hoopla surrounding the so-called dot plot. It even has its own Twitter hashtag: #Dotplot.
The dot plot is a chart that shows the expectations of each FOMC member -- absent their names -- for where they believe the central bank's overnight lending rate will be in the future. Here's a recent example:
Source: Federal Reserve
That there were four members of the FOMC who thought at the end of 2014 that Fed rates would be at 2 percent by 2015 is, well, kind of adorable. Deeply misguided, totally wrong, but still cute in a wonky way.
The problem with this dot-plot chart is that it reflects a three-factor forecast:
-- Each FOMC member’s expectations of the state of the economy in both the near and distant future.
-- How each member believes they will vote on rates in light of that future economy.
-- And what the consensus of the FOMC will be on rates in the future.
As we have shown repeatedly over the years, human forecasters range from awful to terrible. The Federal Reserve’s economic predictions are no better than those of any other forecaster, and in some ways, they are worse. (Same is true of the Fed's economic models). Remember also the Fed’s institutional tendency to be a bit circumspect, so as not to roil markets. Given all of that, what are the odds that this dot plot will be accurate?
What the Fed is wrestling with now are a number of cross currents , making this a challenging period for understanding the economy. Economists might want to forget about forecasting when so few seem to fully grasp the present.
Consider the contradictory economic data this quarter: Job creation has been robust, but wage gains have been feeble. Initial reports of gross domestic product growth were strong, but have since been revised lower. Manufacturing has been slow, housing starts weak. Aside from auto loans, credit remains tight and has been crimping the residential real estate market. Oil, copper and most other commodities continue to fall on weak demand, though this might be a boon for consumer spending. The strong dollar, now at 12-year highs, attracts investment capital but threatens to make U.S. exports uncompetitive.
In light of all that, can anyone really forecast where rates will be in a year or two? If we look at the history of Fed forecasts, the answer is an emphatic no. But that actually misses the entire point of the exercise. The trick isn't to think of these forecasts as predictions, but rather, as a minor policy tool. The dot plot is yet another form of forward guidance, a bully pulpit to communicate the Fed’s intentions. If you focus on the dates, you will miss the hidden message contained within the dots.
What can we discern by looking at the most recent dot plot? This is, of course, a subjective exercise, but I am willing to make a reasonable guess: The Fed wants to begin raising rates, and sooner rather than later. But the data -- never forget that the FOMC is data-dependent -- isn't cooperating.
It is important to note that this isn't your standard-issue recovery from a recession; it's a recovery from a credit and financial crisis. As such, it has been halting, lumpy and unevenly distributed, with lots of soft patches. Compared with normal post-World War II recoveries, this one has been feeble. But as far as post-credit crisis recoveries go, it's actually been fairly robust.
You can see the Fed’s quandary when it comes to raising rates: Wait too long, and it risks an inflationary spiral; raise rates too quickly, and it risks a 1937-like recession that derails the recovery.
The dot plot probably can only be used to hint at the battle of views among the individual Fed members. But as a guide to the timing and size of future rate increases? Don’t make me laugh.
This column does not necessarily reflect the opinion of Bloomberg View's editorial board or Bloomberg LP, its owners and investors.
To contact the author on this story:
Barry L Ritholtz at email@example.com
To contact the editor on this story:
James Greiff at firstname.lastname@example.org