Last week, William Dudley, president of the Federal Reserve Bank of New York, was widely quoted as saying that the case for a September interest rate increase has grown “less compelling,” apparently as a consequence of turmoil in stock markets around the world. Is Dudley saying that monetary policy is dependent on the stock market?

“Data dependence” has become a mantra of Federal Open Market Committee (FOMC) policy discussion. For example, Chairwoman Janet Yellen’s statement at her June press conference says, “…although policy will be data dependent.…”

It is long past time for the FOMC to make sense of what it means by “data dependence.” Dudley is not the first FOMC participant to react to events in a way that continues to muddle communication of policy strategy.

Early in my tenure as St. Louis Fed president, I gave a speech entitled “Data Dependence,” and across the years I tried to convey a sense of what this term ought to mean. The FOMC’s setting of the federal funds rate should depend in a predictable, systematic way on data available at the time of each FOMC meeting.

I followed events—especially economic news—closely. As I observed the daily flow of news and market reactions to it, I tried to think through whether the markets had it right. Markets continuously assess and reassess.

Day by day, I thought about what my policy position might be were the next FOMC meeting to be the next day. Eight times each year, the meeting was indeed the next day. And sometimes the chairman called a meeting between the regular meetings.

I was almost always opposed to policy action between regular meetings, because taking the market by surprise was a poor idea most of the time. A surprise meeting inevitably put more emphasis on the latest news than was justified. That is, a surprise meeting conveyed to the market that the news of that day had special importance to the FOMC. This is a risky way to conduct Fed policy.

An excellent example is the telephone conference call meeting on January 22, 2008, which was Martin Luther King Jr. Day. The FOMC cut its target fed funds rate that evening by 75 basis points. The U.S. markets had been closed, but European markets were down sharply.

As David Wessel explains in his excellent book, In Fed We Trust, “The timing, though, was terrible. Stock markets, it turned out, weren’t plunging on intensifying worries about the health of the global economy. Prices were falling because a big French bank, Société Generale, was secretly rushing to sell off a huge book of bets that one of its traders had made.”

Policy actions implicitly and inevitably define the Fed’s policy “rule” or regularity of behavior. No matter what the FOMC participants say, actions occur against a backdrop of history and recent events. That is, market participants try to divine the true reasons for policy actions, or inactions. Similarly, remarks about policy arise against a backdrop of recent events. Dudley must have understood that his comments would be interpreted as an effort to calm down the stock market.

But the comments come at a cost. Dudley changed the market’s assessment of a September liftoff of rates, which presumably was his intent, as shown by trading in the Fed funds futures market. The comments will still be fresh in everyone’s mind the day before the September FOMC meeting.

My questions for Bill Dudley are these: Do you or do you not want the market to believe that stock prices play an important role in FOMC decisions? How do you weight the stock market relative to other considerations such as employment growth and inflation?

Whenever I was pressed, in Q&A after a speech, to explain which data were most important, I explained that I kept an open mind to assess all information available at the time of an FOMC meeting. I emphasized that employment and other measures of economic activity were important as were measures of inflation. However, the most recent data on these were sometimes affected by temporary disturbances.

It was always important to be on the lookout for anomalies in the data, and keep an open mind about how to deal with unusual events such as the market upset caused by Long Term Capital Management. We might even be confronted by a cataclysmic event like the terrorist attacks on America of 9/11.

My effort, therefore, was to absorb all the information but to state a sensible policy strategy that would define a stable monetary environment over the medium term. With that environment, the market could respond day by day to the news and adjust sensibly to that news without the complication of trying to figure out how the Fed’s own posture might be changing.

I often ended an answer in a Q&A session with this comment. “My responsibility is not to predict monetary policy but to participate in setting it.” I wish that current FOMC participants would stop trying to predict policy actions and instead concentrate on defining a sensible and understandable long-run policy strategy.