Recently, this Journal published a paper that reported the first hard evidence of direct and systematic Administration influence on monetary policy (Havrilesky 1988a). An index was constructed from a simple weekly sum of the number of Wall Street Journal articles that indicated whether the Administration desired either easier (+ 1) or tighter (-1) monetary policy. The index, Signaling from the Administration to the Federal Reserve, or SAFER, was premised on the notion that the financial press efficiently reports the content of the mostly informal exchanges that are occumng between the executive branch and the Fed. One problem with using this index as a means of continually measuring an Administration's influence on monetary policy and whether its signals are heeded regularly or only episodically, say as Presidential elections approach,l is that periodically the level of signaling shifts rather abruptly. For example, a precipitous drop in signaling occurred after 1984. The yearly average of the absolute value of monthly signals fell by over one-half in the 1984-1988 period compared to the 1981-1984 period, the break coming in 1985 when the yearly total of the absolute value of monthly signals fell to six as compared to twenty-two in 1984.2 Unless such shifts can be explained, standard reaction function estimation of the causes and consequences of signaling is not feasible across all time periods. This paper will attempt to explain these and other changes in signaling over the period from October 1974 to December 1989 using data on the state of the economy and the partisan composition of the Federal Reserve Board of Governors. Except for the welldocumented shift that occurred during the Reagan Administration, it will be as-
Read full abstract