I would like to start by congratulating the Economic Society for putting together this policy forum. They certainly have been even-handed – one Central Bank Governor and one Secretary to the Treasury as speakers, a chairman who is a Secretary to the Treasury and a discussant who has been both a Central Bank Governor and a Secretary to the Treasury. I will not be able to get away with much in this company.

Introduction

Fortunately my aim is modest, and I do not want to introduce anything particularly new. Instead, I would like to look back over a subject that has been of great policy interest in Australia and elsewhere over the past decade. It is the issue of what is the optimal institutional structure for making monetary policy. What should be the respective roles of the Government and the central bank? This usually goes under the rubric of central bank independence, but it would be just as sensible to see it as a discussion of the optimal degree of delegation, including the circumstances in which the delegation could be withdrawn.

Since the beginning of central banking, the relationship between the Bank and the Government of the day has been an issue, with central banks routinely having a degree of independence that made them unlike government departments and more like the judiciary. The motivation for this separation was to provide some discipline on governments, by putting temptation out of harm's reach. Initially, the objective was to remove the temptation that governments might fund expenditure from the central bank, ie by money creation. See Keynes (1913) and Coombs (1964) for earlier statements of the issue.

Discussions of this subject were revived in Australia in the late 1980s, at about the same time as in a number of other countries. Almost from day one, however, it got caught up in politics, with one party putting a rather doctrinaire version of it straight into its platform, and the other party professing to see no value in it at all. This kept accusations of a lack of independence in the news, and it also made it difficult to have a calm and rational discussion of the subject. It made it particularly difficult for the Reserve Bank to play a constructive role in the debate. I have not participated publicly in this discussion until now, and I suppose it is a little ironic that my first contribution should come after most of the battle is over, and a reasonably bipartisan consensus has been established. However, this will not deter me, as I think there are still a few things that need to be said.

One thing I am not going to do is to argue the pros and cons of independence. I have the luxury of being able to put that to one side, because in Australia both Government and Opposition now accept the degree of independence given in the Reserve Bank Act. Incidentally, it is wrong to think that conservative parties favour central bank independence and labour or social democratic ones oppose it. In New Zealand, independence was introduced by the Labour Party, in France under a socialist administration, and in the United Kingdom the Labour Party is more disposed to it than the conservatives. All I have to do in this talk is to spell out what the situation in Australia was and now is, rather than argue for a new approach.

The Reserve Bank Act

The Reserve Bank Act only mentions monetary policy twice, and on each occasion it says that it is the responsibility of the Board of the Reserve Bank.^[1] More importantly, the Act does not give the Government any operational role in monetary policy decisions as is done in some other countries' central bank legislation. In the Bank of England Act, for example, they are told that they have to follow Treasury directives, and in the old (ie pre-1989) Reserve Bank of New Zealand Act, they were told that their job was ‘to carry out the government's monetary policy’.

In Australia, we were never in this situation; our Act always gave us a high degree of legislative independence. Of course, the Government always has the ultimate over-ride in extreme circumstances – that is how it should be. In our case, it can do this through Section 11(2) procedures, but these would be extremely politically costly and have never been used. Other countries also have an over-ride mechanism – for example, in the present New Zealand arrangements, the Government can exercise its over-ride by adjusting the inflation target if it feels that the monetary policy being pursued by the Reserve Bank of New Zealand is not appropriate. The important thing is the nature of the over-ride mechanism. If it is easy to use, concealed from view and used regularly, then little central bank independence exists; if it is transparent, politically costly and seldom used, then a high degree of independence exists.

The Act In Practice

Although we are now getting recognised as a reasonably independent central bank, this is only a very recent phenomenon. My predecessor, Bernie Fraser, was constantly having to defend the Reserve Bank against the charge that it was subject to political interference. The following extract from a speech in 1993 gives the flavour:

I have said many times that the Reserve Bank does, in fact, have a high degree of independence. We can and do pursue our statutory responsibilities without political interference. But we seek to do this in close consultation with the Government – to exercise independence with consultation. (See Fraser 1993.)

The question I want to address is why were we always on the back foot having to defend ourselves? Why were the charges of political interference so frequent until recently? And why was Australia regarded in international circles as a country whose central bank had little independence?

There are a number of answers to these questions. The first concerns the interpretation of the Act. There was a widespread assumption that the Act gave the Reserve Bank little independence. This assumption arose because, for virtually 30 years after it was enacted in the 1950s,^[2] monetary policy was implemented as though the Reserve Bank had no independence. Decisions were taken mainly in Canberra, with the Treasurer and the Treasury usually having a bigger say in any decision than the Reserve Bank. Generations of economists, politicians and journalists grew up with this and accepted it as the natural order of things; it was difficult for them to adjust quickly when things changed.

How could this situation persist for so long if the Act clearly said that decisions on monetary policy should be made by the Reserve Bank? The answer to this riddle is that the Reserve Bank did not have the instruments of monetary policy at its disposal – before deregulation, they were nearly all in Canberra. (See Phillips 1992.)

If we go back to the late 1970s or early 1980s, the main way of implementing monetary policy was by influencing the way the budget deficit was financed, i.e. by the setting of interest rates on government securities. This was entirely in the hands of the Treasurer. Bank lending and deposit rates were also centrally determined, ostensibly by the Reserve Bank, but changes required prior approval by the Treasurer.^[3] The exchange rate was determined by a group of three (or at times four) officials, one of whom was the Governor of the Reserve Bank. The general intent of monetary policy was contained in an M3 target; this was the Treasurer's target rather than the Bank's, and was announced in the Budget Speech. Major decisions on monetary policy were taken by the Monetary Policy Committee of Cabinet.

This state of affairs was regarded as normal, and to the best of my knowledge no-one criticised it on the grounds of lack of central bank independence. In the intellectual climate of the times, this was not an issue. It was only later in the late 1980s – by which time the Reserve Bank had gained a fair measure of independence – that this particular type of criticism started. This is not to say that we were not criticised for other reasons in earlier years; there was plenty of dissatisfaction with monetary policy, mainly on the grounds that it was doing too little to combat our high inflation.

The big change for the Reserve Bank was financial deregulation. It swept away the interest rate controls, freed up the exchange rate and made it possible to finance the budget deficit fully at market-determined interest rates. This left open-market operations, which effectively determined short-term money market rates, as the only instrument of monetary policy. This was entirely in the hands of the Reserve Bank, which put us operationally in the same position as the US Federal Reserve or the Bundesbank. For the first time, the intentions of the Act and the capacity of the Reserve Bank were in accord.

Most of the change occurred in the mid 1980s, with the last interest rate ceiling on banks not being removed until April 1986. It took time, however, for all parties to adjust fully to the new system, and it is not possible to point to an exact date when the Reserve Bank passed from being dependent to independent. In my view, it has clearly been independent in the 1990s, and a good case can be made to show that it was largely independent in the second half of the 1980s.

Measures of Independence

The economic literature on central bank independence and its relation to inflation began to attract attention in the mid to late eighties. Basically, it showed that there was a significant negative correlation between a country's degree of independence and its rate of inflation (i.e. the more independent the central bank, the lower the rate of inflation). It also showed no correlation between a country's degree of independence and its rate of economic growth, so the improvement in inflation was not bought at the cost of lower growth. This literature put forward a relatively simple proposition, and proved persuasive to a wide range of economists and policy makers from across the political spectrum. One of the best summaries is contained in an article co-authored by Lawrence Summers, who is now Deputy Secretary of the Treasury in the Clinton Administration. (See De Long and Summers 1993.)

Unfortunately for Australia, a lot of the literature on this subject quoted the earliest ranking of central bank independence developed by Parkin and Bade (1982). While those authors deserve great credit for starting or reviving) the literature on this subject, their particular index of central bank independence is an eccentric one. For example, it rates the Bank of Japan as being as independent as the Federal Reserve Board, and it ranks the Bank of England over the Reserve Bank of Australia. Australia's low ranking on this index was widely quoted until more accurate indices became available. The perception was reinforced by a poor inflation performance relative to other countries in the 1980s.

It is now recognised that the two most thorough and reliable guides are the GMT (1991) indices and the indices constructed by Cukierman (1992). These are shown in Table 1 below, and it can be seen that Australia is generally in the top half of the field. Of these two measures, the GMT aggregate index is probably the best, and now the most widely used. The GMT index also has the advantage of covering 18 OECD countries. Cukierman's questionnaire is hampered by lack of responses from a lot of OECD countries, and so has a high representation from developing countries.

Of course, it should also be recognised that all these measures are just attempts to simplify a very complex structure into one number or ranking. They are all imperfect, which is shown by the fact that they are all different. They take no account of personalities or of other policies which may impact on monetary policy. Even so, if they are going to be quoted – and they frequently are – it is better for us if the most representative are used.

The Statement on the Conduct of Monetary Policy

The Statement on the Conduct of Monetary Policy issued on 14 August at the time of my appointment was a means of clearing up any remaining ambiguity about the relationship between the Government and the Reserve Bank. In it, the Government stated its understanding of the high degree of independence given to the Reserve Bank in its Act and endorsed the Reserve Bank's inflation target of 2 to 3 per cent on average over the cycle. In one sense, this was the continuation of a direction that was becoming apparent under the previous Government and so shows the essentially bipartisan approach that now exists. The previous Treasurer had endorsed the inflation target and the Reserve Bank's independence, but not in a public document and not by formally relating it to the Reserve Bank Act. As a result, it had received limited recognition.

The Statement was well received in Australia, and we have received feedback that it was well received overseas, including by a couple of overseas central banks which showed considerable interest in it. While the Statement has been useful in achieving the objectives outlined above, it should also be seen as part of a more general trend towards establishing a framework for better economic policy.

We are confident that the combination of central bank independence and an inflation target endorsed by the Government will help to improve Australia's medium-term economic performance on both inflation and long-term output growth. Similarly, on the fiscal policy side, the focus on the underlying deficit and the commitment to balance it over the cycle contained in the recent Budget will improve Australia's savings and growth performance in future years.

They are two recent examples of the proposition that institutional change based on increased transparency is the best way to establish a framework for improved medium-term economic performance. There are others that have occurred over recent years and which are now firmly established. One that I think is particularly important is the agreement between Treasury and the Reserve Bank on the separation of monetary and debt management policy whereby the Reserve Bank is responsible for monetary policy and Treasury for debt management policy (formerly responsibilities were blurred). Within debt management policy, the commitment is to fully fund the budget deficit by borrowing from the public at market determined rates. (See Reserve Bank 1993.) Another example is the practice, introduced by the Bank in 1990, of announcing each change in monetary policy as it occurs and simultaneously documenting the reasons for the change.

I am sure we have not yet come to the end of the process. The art of good government is to constantly review practices to see where improvements can be made. The guiding principles should be to design systems that make it clear what the medium-term goals are, to choose goals that can be communicated easily to the public and accepted as being reasonable, and to ensure that the system is transparent so that people can judge whether policy changes are consistent with the goals.

Endnotes

The first is where it sets out the mandate for the Reserve Bank in Section 10(2), ‘It is the duty of the Board, within the limits of its powers, to ensure that the monetary and banking policy of the Bank is directed to the greatest advantage of the people of Australia …’. The second is in Section 11(1) where it says, ‘The Board shall, from time to time, inform the Government of the monetary and banking policy of the Bank.’ [1]

The current Reserve Bank Act dates from 1959, but its essentials were already in the Commonwealth Bank Act of 1951. [2]

The monetary policy instrument that was wholly in the Reserve Bank's hands was the Statutory Reserve Deposit (SRD) ratio. Unfortunately, this had very little effect on monetary conditions in circumstances where interest rates and the exchange rate were fixed, and so its possession did not provide the Reserve Bank with an effective monetary policy instrument. [3]

Bibliography

Coombs, H.C. (1971), ‘The Relationship of the Central Bank with the Government’, Other People's Money, Australian National University Press, Canberra, pp. 57–63.

Cukierman, A. (1992), ‘Central Bank Strategy, Credibility and Independence: Theory and Evidence’, MIT Press, Cambridge, Massachusetts.

De Long, J.B. and L. H. Summers (1992), ‘Macroeconomic Policy and Long-Run Growth’, Economic Review, Federal Reserve Bank of Kansas City, 77(4), pp. 5–29.

Fraser, B.W. (1993), ‘Some Aspects of Monetary Policy’, Reserve Bank of Australia Bulletin, April.

Grilli, V., D. Masciandaro and G. Tabellini (1991), ‘Political and Monetary Institutions and Public Financial Policies in the Industrial Countries’, Economic Policy, October, 13(1), pp. 342–392.

Keynes, J.M. (1913), in The Collected Writings of John Maynard Keynes, Volume XV: Activities 1906–1914 India and Cambridge, E. Johnson (ed.)(1973), Macmillan, London.

Parkin, M. and R. Bade (1982), ‘Central Bank Laws and Monetary Policy’, University of Western Ontario Working Paper.

Phillips, M.J. (1992), ‘Central Banking – A Parting View’, Reserve Bank of Australia Bulletin, April.

Reserve Bank of Australia (1993), ‘The Separation of Debt Management and Monetary Management’, Reserve Bank of Australia Bulletin, November.