Academic journal article Economic Inquiry

The Biases of Federal Reserve Bank Presidents

Academic journal article Economic Inquiry

The Biases of Federal Reserve Bank Presidents

Article excerpt

I. INTRODUCTION

This paper tests the conjecture that there is a subgroup of Federal Reserve bank presidents who can be relied upon to vote in Federal Open Market Committee (FOMC) meetings in a manner that is consistent with the partisan orientation of the Administration that was in power when they were appointed. It seeks to identify the career characteristics of this reliable subgroup. Finally, it tests whether, while controlling for partisanship, Federal Reserve bank presidents, as a group, are less likely to vote for expansionary policies than are Federal Reserve board members (governors) as a group.

II. AN ALTERNATIVE TO MONETARY POLICY SIGNALING

Recent research by Havrilesky [1988; 1990; 1995] and Froyen, Havrilesky and Waud [1993] has presented evidence that the Administration, Congress and the banking industry systematically signal their desires for monetary policy to Federal Reserve leaders and that, under certain conditions, these leaders respond. Although they could legitimately protest the practice, Federal Reserve officials seldom object very loudly, even during the most blatant interludes of monetary policy signaling. They understand that politicians and bankers are aware of the private costs of too frequently appearing to pressure the Fed. Moreover, they are aware of the private costs of appearing to succumb. They know that excessive overt pressure on the monetary authority threatens the symbiosis that exists between the Federal Reserve, the Administration, Congress and the financial services sector.

The fine balancing of interests among the principle players in the monetary policy arena is essential to understanding the political economy of monetary policy. The Federal Reserve must be sufficiently autonomous for politicians to bash it whenever the economy goes awry, but it must not be so autonomous as to prevent politicians from having any say in monetary policy. Moreover, if political pressure became excessive, the Fed's clientele in the financial services sector would respond by reducing their support of politicians (Kane [1990]).

Given the costs of excessive overt political pressure it is not surprising that the Administration would try to influence monetary policy by controlling the composition of the Federal Open Market Committee.(1) In this way it could obtain the desired monetary policy, usually at a lower cost than that incurred by direct signaling, except in periods when new appointments to the Committee are highly visible. If it could control the Committee's membership, the degree of an Administration's influence on monetary policy would depend on whether its appointees were "reliable." An earlier study by Havrilesky and Gildea [1992] has shown that a class of reliable governors can be identified by their career and background characteristics as economists. (At the time this earlier study was prepared there were too few observations in the data set to treat bank presidents separately.) One purpose of the present paper is to determine whether the same can be said of Federal Reserve bank presidents viewed as a separate group. If it can be shown that bank presidents vote reliably in FOMC meetings, i.e., along partisan lines, then we wish to determine whether bank presidents and governors with the same partisan orientation are dissimilar in their voting behavior. This inquiry is motivated by the finding across a wide swath of literature that bank presidents, as a group, tend to favor less expansionary monetary policy than do governors, as a group.(2) This finding is reaffirmed periodically in the financial press and in recent years has become more pronounced whenever a group of bank presidents resists board members' desires for monetary ease.(3)

The only paper in the literature that takes issue with this near consensual is by Geoffrey Tootell [1991]. By including state-of-the-economy variables in separate multinominal logit functions (estimated for bank presidents and governors and for votes to tighten or ease monetary policy), Tootell rejects (at the 15 percent but not at the 10 percent level) the hypothesis that bank presidents' reactions are different from those of governors. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed

Oops!

An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.