RDP 9406: Reserve Bank Operations in the Foreign Exchange Market: Effectiveness and Profitability 1. Introduction

When the Australian dollar was floated in December 1983, the Reserve Bank noted that it would retain discretion to intervene in the foreign exchange market. The Bank's operations in the market were very light for the first 18 months or so of the float but became more substantial towards the end of 1985. By the latter half of 1986, the Bank had significantly stepped up its intervention in the face of sharp downward pressure on the exchange rate.

Since then, the Bank has continued to intervene in the market. At times this has involved undertaking relatively small amounts of intervention to slow the movement in the exchange rate or to slowly rebuild reserves. At other times, when the exchange rate had moved a long way from what was perceived to be a sustainable rate over the longer term, the Bank has sought to prevent further movement or even reverse some of the earlier movement.^[1]

An obvious question arises: Has this intervention been successful in having a stabilising influence on the exchange rate? This question has no direct answer because we cannot know how the exchange rate would have behaved in the absence of the Bank's foreign exchange operations. As a consequence, we need to turn to various indirect measures. Probably the best known and appealing to a participant in the market is the profits test, first proposed by Milton Friedman in 1953. Other tests, which look at deviations of the exchange rate from a long-term trend, have been suggested by Wonnacott (1982) and by Mayer and Taguchi (1983).

This paper briefly reviews the literature on the effectiveness of foreign exchange intervention and then applies some of the tests discussed in this literature to the Reserve Bank's intervention over the floating rate period. While there can be shortcomings in the use of profits as a test of effectiveness of intervention, we argue that the existence of profits over long periods provides a strong case for the view that central bank intervention has been effective in stabilising the exchange rate. The paper shows that the Bank has made significant profits over the post-float period, though, as would be expected, profits have not been made in all sub-periods. Other tests carried out in this study support the finding that intervention has tended to stabilise the exchange rate over the period the currency has been floating.

It should be noted that the tests relate to what is often referred to in the literature as sterilised intervention. Academic discussion often makes much of the distinction between sterilised and unsterilised intervention (the latter leading to a change in financial conditions due to the effect of intervention on the monetary reserves of the banking system) but in reality central banks virtually always sterilise their intervention. To the extent that central banks want to undertake a change in monetary policy to influence the exchange rate they would do this through their usual means, which in most cases is operations in the domestic money market, rather than waiting for it to happen as a by-product of foreign exchange intervention.^[2]

The paper is organised as follows. Section 2 discusses whether profitability can be used to assess the effectiveness of intervention and examines some of the empirical results from studies for other countries. Section 3 gives a broad overview of the Bank's foreign exchange operations, discusses the data and details our methodology. Section 4 summarises the empirical results on profitability for Australia. Section 5 provides some additional evidence, based upon tests first proposed by Wonnacott, that the Bank's operations have stabilised the exchange rate. The final section provides some concluding thoughts.

Footnotes

For fuller explanations of the Bank's foreign exchange operations, see Fraser (1992) and Macfarlane (1993). [1]

One reason for this is that intervention only impacts on domestic monetary reserves with a lag, as foreign exchange transactions are settled two days after being undertaken. A central bank whose currency was under threat to such a degree that it would consider raising interest rates would not want to wait for two days to achieve this monetary tightening through its failure to sterilise its intervention. [2]