Bernanke Put?

The Times' Floyd Norris notes that Federal Reserve actions appear to be following the stock market:

Say this for the Fed. It pays attention to what Wall Street wants.

When Wall Street wanted the Fed to ignore all the wild gambling in the derivatives markets, the Fed did so. Alan Greenspan fought to keep regulation away from that market, and argued that it was assuring the safety of financial institutions by allowing risk to be transferred to others. It turns out the exact opposite was true.

Now that the crisis is upon us, and Wall Street is in trouble, the Fed seems to announce new actions whenever investors get worried enough.

I count six separate days in the first quarter, which ended today, when the Fed announced actions. They are:

Jan. 22, when it cut the discount rate.
Jan. 30, when it cut the discount and Fed funds rates.
March 7, when it announced plans to inject money into financial markets
March 11, when the Fed and other central banks announced plans to inject more credit into markets.
March 16, when it financed the rescue of Bear Stearns and cut the discount rate
March 18, when it cut the Fed funds and discount rates.

On a net basis, all the damage for the quarter was done in the days leading up to each Fed action. In the three trading days before each Fed move — a total of 15 days since there was some overlapping — the S.&P. 500 fell almost 145 points and the Dow Jones industrial average lost 1,003 points.

On the other 46 trading days during the quarter, a net nothing happened. The S&P was up about two points, and the Dow was down less than one point.

What we have here is a picture of a Fed that follows the market, and of a market that repeatedly rallies on the news of a Fed move, only to fall again as more bad news comes out.

This is reminiscent of the notion that the Fed, under Bernanke's predecessor, was effectively setting a floor under the stockmarket, sometimes called the "Greenspan put." To explain: a put option grants the holder the right to sell a stock at a certain price; the option is basically an insurance policy establishing a minimum value for the holder (for example, if I buy a share worth $45 and a put option with a "strike price" of $40, if the share falls to $20, I can exercise the option and sell the share to the writer of the option for $40, limiting my loss to $5 plus the price of the option). Reflecting on the 20th anniversary of the October 1987 "Black Monday" stock market crash, The Economist's "Buttonwood" column put it thus:

The third lesson [of the 1987 crash] is that central banks will quickly intervene if they fear the markets are in crisis. They did so again in 1998 [following the LTCM collapse] and in August and September this year.

Twenty years ago, the Fed feared a repeat of 1929: after that year's crash came the Depression. In fact, the economy shrugged off the meltdown with the help of a loosening of monetary policy and recession was postponed until the early 1990s. Indeed, Black Monday now looks like a blip on the long-term stockmarket graph.

Many investors came to believe that central banks would underwrite the markets (the so-called Greenspan, now Bernanke, put). Although central banks did not prevent the equity bear market of 2000-02, the current strength of the stockmarket suggests investors' faith in this put has not been eliminated.

But this could be the most dangerous lesson of all. In Japan in the 1990s, neither near-zero interest rates nor fiscal stimulus saved the market. One day, investors will realise central bankers are not magicians. That might be another Black Monday.

The risk of the "put" (like any insurance, whether actual or perceived) is essentially one of "moral hazard" - that protection from negative consequences leads to riskier behavior. The belief in the "Greenspan put" may have contributed to a stock market bubble (or at least some "frothiness") as people were more eager to buy stock when they felt that they would be protected from declines, and the share prices were therefore bid up excessively (In 2005, Brad deLong argued that the Greenspan put is a myth).

Of course, correlation does not imply causality. Norris' evidence establishes that stock market declines have preceded Fed moves (what an econometrician might call "Granger causality"), but it does not rule out some other event causing both. Over the past several months, it may be that bad news about the functioning of credit markets and the condition of financial institutions has led to both the market declines and the Fed actions. The Fed actions themselves, in turn, provide information to the stock markets about how the Fed will respond to credit market problems. The market movements mean that the Fed actions have not been fully anticipated: so far, the rallies indicate that they have been pleasantly surprised by the Fed's moves, but some ugliness may come when the markets start to expect the Fed to act in certain ways, and the Fed chooses to disappoint them.