RDP 8811: Monetary Transmission in a Deregulated Financial System 5. Empirical Implications

The model of the transmission mechanism outlined above has empirical implications which differ substantially from both the textbook model and from more institutional descriptions of the monetary transmission process. In particular, traditional models tend to view central bank control over a money or credit aggregate as being the important exogenous force driving activity and prices. In such models the aggregates should, therefore, lead movements in activity and prices. In the deregulated world, where central banks operate directly on the price of financial assets, it will be interest rates and exchange rates which lead activity, with money and credit moving with, or after, changes in activity and prices. The above work therefore provides a strong alternative hypothesis to the conventional notion that money leads the economic cycle.

This paper will not attempt to provide new evidence on the question of money/income causality: that question is addressed directly in a number of other papers in this series.[14] It is, however, worth bringing together some of that evidence, as well as placing it within the context of the recent overseas work on the subject; particularly as the international evidence is starting to question the assumption that money and/or credit are leading indicators of activity.

A good summary of the US experience is provided in a series of papers by Benjamin Friedman (1988a and 1988b). He points out that the conventional wisdom of money leading output was established using data primarily from the pre-deregulation era.[15] As the deregulated data from the 1970s and 1980s began to be included in the data sets, it became increasingly difficult to make strong statements about the leading role of money and credit. The work of Stock and Watson (1987) was able to find, under certain restrictive specifications, that M1 still had some predictive power in explaining future industrial production with data up to 1985. However, Friedman (1988a) claims this appears to be breaking down once data up to 1987 is added to the model. To quote:

“As of 1979, the available evidence strongly supported the view that observed fluctuations of M1 in the United States did contain such information about future movements of U.S. income and prices. By contrast, the same experiments carried out with data for the most recent 18 years provide no support for the view that fluctuations in M1 carry information about future income and prices that is not already contained in fluctuations of income and prices themselves.”

In other countries the evidence is not necessarily as strong, although this is to be expected given that many countries have not yet proceeded as far in allowing financial market deregulation. In Japan, for example, the conventional wisdom remains that the monetary aggregates are good leading indicators of activity and prices, although it is already recognized that further steps aimed at deregulating the financial system may change this relationship (Suzuki 1987). Similarly, the stability of the monetary aggregates in a number of European countries (and the continued use of monetary targets) appears to be related to the gradual pace of deregulation in those countries. In countries where deregulation has proceeded further (the United Kingdom, Canada and New Zealand) there has generally been a breakdown in the relationship between conventional monetary aggregates and nominal spending. It is interesting that both the U.K. and New Zealand are now concentrating more closely on a definition of money that is much closer to the old textbook definition; i.e. currency in the hands of the public plus liquid reserves at the central bank. In all countries there is agreement that the effectiveness of monetary policy is not necessarily being reduced by the process of deregulation. Rather, as the model above would predict, the transmission mechanism is running directly from interest and exchange rates to real demand, activity and prices. The most recent OECD survey of the empirical literature concluded:

“The dismantling of administrative regulations and the adoption of market-orientated techniques of monetary control has resulted in financial market prices such as interest rates, and exchange rates, and in some cases equity prices, having to play a more important role in the transmission process of monetary policy.”(Chouraqui, Driscoll and Strauss-Kahn, 1988, p. 35)

These conclusions are also reinforced by the work of Ahktar (1983) at the Bank for International Settlements, by the Federal Reserve Bank of New York (1988) and by the International Monetary Fund (Watson, et al 1988).


See, for example, Bullock, Stevens and Thorp (1988) and Bullock, Morris and Stevens (1988). [14]

The foundation work in this case was Friedman and Schwartz (1963). This was backed up by Sims(1972) in the US and by numerous other studies for the other major industrial countries. [15]