Why so much is read into the words of the Fed chairman.
Following Fed chairman Ben Bernanke's speech yesterday, the usual bout of trying to understand quite how much we can read in to his words has begun. Yet unlike the normally perjorative "Kremlinology" – attempting to infer things from the most minuscule turns of phrase – this Bernankeology is understandable and quite useful.
Central bankers have a strange job. They don't actually have many tools at their disposal; largely just the tripartite decision to raise, lower, or maintain interest rates. Yet many of the outcomes they create come, not from actually using this power, but from creating expectations as to their future use.
Suppose Bernanke knows he is likely to raise interest rates in the first quarter of 2013. Even though his actual power is relatively limited, he can create a wide spectrum of outcomes depending on how he announces this. The market reaction will be extremely different if Bernanke says now that he will raise rates in a years time, compared to if he maintains right up until the day that a rate rise would be inappropriate.
But this power to persuade brings with it its own problems. Just like a legislature, a central bank is fundamentally unable to constrain itself; it can make promises, but everyone knows that it is free to break them at any point.
All of this means that every speech Bernanke gives is likely to be very carefully aimed at creating just the right set of expectations. On the one hand, he can't ever gain a reputation for untrustworthiness, so they have to be scrupulously honest; on the other, actually saying what he believes may create the wrong impression.
Last week, Ryan Avent provided a detailed breakdown of exactly what the benefits of Bernankeology can be, focusing on the Fed's "forward guidance" where it hinted that it would keep interest rates low until at least 2013. He writes:
On the one hand, a pure focus on the language of the Fed's statement indicates that rates are likely to remain low through that period based on the state of the economy... On the other hand, the Fed may be hinting that it will be willing to keep rates low through late 2014 even if the trajectory of the economy warrants a rate increase.
In other words, the Fed might be attempting to commit itself to a deviation from its normal policy rules of the sort that might generate more rapid growth and inflation.
The problem the Fed has is that it needs to generate growth, but that growth is likely to come with relatively high inflation, of the sort which Bernanke has historically fought against. In order to help the economy, he needs to convince "the markets" that interest rates will be kept low even if inflation spirals out of control. The problem is that this, from an inflationary hawk like Bernanke, is unbelievable.
Avent points to a paper (pdf) which breaks down the distinction into two categories:
Delphic, corresponding to the first category above, and Odyssean, corresponding to the second, in which the central bank attempts to commit itself to deviations from typical rules.
Matt Yglesias offers a less refined version of the same strategy, breaking Bernanke's possible responses into an Eeyore response and a Tigger one. Either the Fed chief can "avoid optimistic forecasts as a way of signaling that rates will stay low for a long time," or he "can say we're climbing out of a steep hole so rates will stay low for the next 18 months come what may".
The test for Bernankeologists is to work out whether yesterday's gloomy speech is Odyssean-Eeyore, using gloominess as a mast to bind himself to, or simply Delphic, with the chairman making his most honest predictions and still being pessimistic.
More from New Statesman
- Online writers:
- Steven Baxter
- Rowenna Davis
- David Allen Green
- Mehdi Hasan
- Nelson Jones
- Gavin Kelly
- Helen Lewis
- Laurie Penny
- The V Spot
- Alex Hern
- Martha Gill
- Alan White
- Samira Shackle
- Alex Andreou
- Nicky Woolf in America
- Bim Adewunmi
- Kate Mossman on pop
- Ryan Gilbey on Film
- Martin Robbins
- Rafael Behr
- Eleanor Margolis