The Operations of the Bank of England, 1890-1908: A Dynamic Probit Approach

Article excerpt

RECENT CONTRIBUTIONS Of Bloomfield (1992), Cooper (1992), and McKinnon (1993) in the Journal of Economic Literature confirm that interest in the various facets of the gold standard is as lively as ever. Previous empirical studies have used an implicit reaction function to evaluate the determinants of the British discount rate policy during the pre-World War I era. This work includes Goodhart (1972), Dutton (1984), and Pippenger (1984). Eichengreen (1992) contains a bibliography of the relevant literature as well as an insightful summary of the history of the gold standard.

Building on work by Eichengreen, Watson, and Grossman (1985), who analyze the interwar Bank rate, we extend research on the Bank of England's discount rate policy prior to World War I in several ways. Improvements include using weekly data, allowing for asymmetric responses, and applying a new estimation method, dynamic probit, to account for the discrete and dynamic nature of Bank rate policy.

Section 1 briefly reviews the operations of the Bank of England, and section 2 uses this review to develop our model of the policy-making process. The appropriate economettic method, dynamic probit, is discussed in section 3. (An appendix available from the authors provides a detailed description of the computational aspects of this technique.) Section 4 presents the econometric results and discusses their relevance for the operation of the gold standard, and section 5 is a summary.


Firsthand knowledge about the operation of the gold standard and the role played by the Bank of England is abundant. Sources include discussions in the contemporary financial press, articles in the Economic Journal, internal memoranda, private correspondence of bank officials, and records of the Bank. In his monumental three-volume study of the Bank, covering 1891 to 1944, Sayers (1976) analyzes most of this information. Goodhart (1988) provides a more general account of the history of the Bank of England and other central banks.

The picture of the Bank of England that emerges from this literature is that of a medium-sized gentlemanly commercial bank that was slowly evolving into a central bank and was concerned primarily with the convertibility of sterling deposits. It managed the government account on a more or less commercial basis, and the government's account was comparable in size to the accounts of its other customers, which include the governments of various countries forming the British Empire. This evolution from a commercial bank to a central bank had its counterpart in the reduction of the prerogatives of one of its major shareholders, the royal family.

During this period, defending convertibility required the Bank of England to exercise considerable judgment because its resources were limited relative to that goal. The automaticity ascribed to the workings of the gold standard, put forward in the report of the Cunliffe Committee in 1918 and given wide circulation by the time of Bloomfield's (1959) account, is perhaps a post hoc phenomenon. Flanders (1993) compares Sayers' accounts with statements of the various rules supposedly followed by the Bank during this era, ranging from the Bank Act of 1844 to the report of the Cunliffe Committee in 1918. She concludes that the difficulty in arriving at a single unambiguous formulation of the operational rules is a sign that the Bank in fact exercised a great deal of discretion in pursuing its goals.

Discretion was necessary because adjusting Bank rate to defend convertibility sometimes conflicted with, among other concerns, the proviso that the Bank earn a reasonable income for its shareholders. Although the need for income was very well understood at the time, it has been overlooked in most of the recent research, including Goodhart (1972) and Dutton (1984).(1) In fact, the Bank's efforts to deal with conflicts among its goals imparted important asymmetries onto its behavior. …