International interest rate relationships: policy choices and constraints

Introduction

In many industrial countries the domestic impact of interest rate developments abroad has become an increasingly sensitive question. Outside the United States the issue has centred on the possible influence of high or rising interest rates on a still fragile economic recovery. When inflation expectations are unknown, interest rate levels are, of course, difficult to interpret, and partly for this reason interest rates have come to play a smaller role as explicit objectives of monetary policy. In some - though not all - countries nominal interest rates are nevertheless felt to be uncomfortably high in relation to rates of inflation. Moreover, while there are clearly domestic imbalances which may be contributing to unwelcome interest rate developments, it is evident that in recent years some countries have become more exposed to developments in financial markets abroad. The present paper considers how interest rates in different currencies might be related in different policy environments and examines some of the relationships that have actually been observed.

A glance at Graph 1 reveals that relationships among movements in nominal interest rates in different markets have undergone a change in recent years. Since the move to floating exchange rates in early 1973 interest rates in the industrial countries have gone through at least two major cycles, each associated with strong surges of inflation. After I979 short-term interest rates typically became more volatile on average than in the 1973-78 period, and their development in different markets diverged at least as much as before. Yet since 1979 long-term rates in different countries, though at different levels, seem to have moved more closely in step. The change can be seen not only in the case of rates in the European countries whose currencies are linked in the European Monetary System, but also in that of countries (including Germany and the United States) whose currencies have been floating in relation to one another. This synchronisation is the more noticeable given the increased variability in yields in the United States.

The new pattern of interest rate behaviour is probably mainly a reflection of various changes in the policy environment which took place in the late 1970s, including:

1. the removal of exchange controls and other restraints on international capital movements in individual countries, by domestic financial deregulation and by the growth of the international financial markets;
2. the establishment in early 1979 of the European Monetary System and, in a wider range of countries, an increase in the role played by exchange rate stabilisation objectives in the conduct of monetary policy;
3. the synchronised recourse in most of the major industrialised countries to monetary restraint policies in an effort to contain inflation following the second oil price shock; and
4. the introduction in October 1979 of new US monetary control procedures and the changes in interest rate behaviour which followed, given the overall policy mix and the way the economy responded.

The influence of this last change on interest rate developments elsewhere was not simple and direct but was conditioned by a number of other factors.

In the most general terms cross-country interest rate relationships are likely to depend mainly on how closely the financial markets are integrated, on exchange rate expectations and on whether financial assets denominated in different currencies are regarded as close substitutes by borrowers and lenders (after allowing for the impact on returns of expected exchange rate changes). This paper presents some empirical indications of the changing role of barriers to international capital movements and discusses the influence of asset substitutability and exchange rate expectations in a more impressionistic way.

It has long been recognised that the ability of countries other than the reserve centre to pursue independent money stock objectives depends ultimately on the freedom of exchange rates to move. At times, however, disturbing exchange rate developments have clearly complicated the execution of policy strategies based on targets for the money stock. There is also ample evidence that in cases where exchange rate objectives have been pursued it has been necessary at times to accept unwelcome movements in short-term interest rates. In the real-world context of money stock objectives and managed floating or of adjustable exchange rate pegs, an investigation of relationships among short-term interest rates and exchange rates should give some indication of the choices countries have actually faced and of the ways in which they have responded to them. It is also of interest to consider how the experience of the EMS has influenced the linkages between interest rates in European countries and the relationship between interest rates in these countries and those in the United States.

Long-term interest rates no doubt depend more on inflation expectations and savings investment relationships, as conditioned by fiscal and other policies. The uncertainty surrounding future developments might lead one to expect a priori that influences from abroad on domestic long-term yields could be relatively limited. Yet if financial markets were becoming more closely integrated internationally and domestic regulatory influences were declining, yields in many countries might increasingly come to incorporate expectations influenced by developments in major capital markets abroad, even when exchange rates were free to float. Evidence on longer-term interest rate relationships should indicate how important this latter effect has been in practice.

Relationships between yields in different currencies must be conditioned basically by longer-term exchange rate expectations and portfolio preferences. Both of these factors are subject to change over time and it would not be surprising to find yields in different countries at times moving closely together and at others behaving quite independently. Exchange rate expectations may depend not only on anticipated inflation differentials but also on changes in the pattern of exchange rate relationships which market participants think will ultimately take place - as a result, for instance, of a need to correct past deviations from purchasing power parity and/or to restore more balance in external current-account positions. To illustrate how cross-country bond yield relationships might have been conditioned by these factors, this paper considers ways in which real interest rates and real exchange rates may have interacted.

In support of much of the analysis in this paper, simple partial correlation and regression analysis is employed. Qualified use of the results seems adequate for the purpose in hand, which is not to test for stable underlying relationships but simply to present some stylised facts about recent developments. For some countries the results of such simple tests are far from conclusive but the aim is more to provide some simple international comparison than to explain fully developments in any particular country.

The plan of this paper is as follows: Section I briefly reviews the implications for monetary autonomy and for international interest rate relationships of various types of open economy analysis widely used in recent years. Section II recalls some recent changes in the policy environment which are likely to have affected countries' exposure to interest rate developments abroad und uses links between interest rates in national and international markets to indicate the degree of integration which has taken place. Section III discusses cross-country relationships among short-term interest rates as indicators of the ways in which central banks seem to have responded to developments in the external financial environment in seeking to meet their own intermediate or final objectives. Section IV seeks to cast some light on the international relationships among nominal longer-term interest rates observed in recent years. Section V discusses the role of exchange rate expectations in the international transmission of interest rate disturbances and some implications of this. Some tentative policy conclusions are set out in Section VI.