1. The objectives of monetary policy

The main objective of monetary policy as applied in South Africa in recent years has been to fight inflation. This objective was pursued within the context of the broader concept of overall financial stability. The Reserve Bank has operated within a monetary policy framework anchored to predetermined guidelines for an acceptable rate of growth in the money supply.

The policy was, however, never dogmatically linked to a rigid money rule -- the Reserve Bank was indeed guided by a whole set of macro-economic financial developments such as, in addition to changes in the money supply, growth in the total amount of credit extended by the banking sector, the funding requirements of the Budget, the level of interest rates, changes in the net gold and foreign exchange reserves, movements in the exchange rate of the rand, and, of course, by the level of the rate of inflation. The Bank had certain objectives in respect of each one of these financial aggregates. The objectives were regularly disclosed and were widely debated in the media and elsewhere. They were, in our view, not only consistent with each other but also supportive of the overall objective of reducing inflation, and making South Africa more competitive in the world environment.

It is the task of the monetary authorities, that is of the Reserve Bank, to continue to operate within a consistent framework of monetary policy that must at all times take account of the interrelationship of these financial aggregates. It is very tempting for those involved only in certain sectors of the economy, for example in the export business, or in financial institutions, or in securities dealing, to judge monetary policy against the background of their sectoral needs. It is imperative for the central bank to apply monetary policy in the macro-interest, that is in the interest of the total economy. And for the total economy, inflation is bad. Therefore, the central bank has no alternative but to fight inflation, even if it should at times require unpopular restrictive measures.

The insistence on a financial approach for monetary policy based on changes in financial aggregates alone, may lead to the fallacy that monetary policy operates in a vacuum and has nothing to do with developments in the real economy, or with the implementation of the other branches of overall macro-economic policy. Monetary policy is, however, an integral part of overall economic policy, and is on a continuous basis influenced by developments in the markets for labour, for goods destined for domestic consumption and capital formation and for goods and services traded internationally. In the end, in economics, everything depends on everything else.

Over many years, excessive wage increases created relative price distortions in the South African economy that have led to the present situation where the country is in many areas no longer competitive vis-á-vis the out-side world. At the same time, South Africa is committed to introduce more liberal international trade arrangements in terms of the Uruguay Round of the World Trade Organisation. We must therefore now be even more cautious about wage adjustments, because we already have a situation where the present level of real wage rates are inconsistent with exported growth and more aggressive international competition, and with much needed increases in domestic investment and employment initiatives.

We therefore have to contend with two difficult situations in the labour market -- the one the historical result of excessive wage increases (relative to productivity changes), the other the continuing pressure for further excessive wage increases now and in the future. Both situations impede efforts to elevate the growth potential of the South African economy to a higher level on a sustainable basis, and to create more jobs for the absorption of the many unemployed people in the country.

Both the historical problem and the current pressures in the labour market cannot be ignored by the monetary policy authorities. It is an illusion, however, to believe that these problems can be solved by creating more money. An accommodative monetary policy in this situation will complicate the necessary correction and will contribute to the further erosion of the relative competitiveness of the South African economy. The decline in the country's competitiveness will, in the case of excessive money creation, only be redirected via the route of excessive inflation, and adjustment back to equilibrium will in the longer term be even more painful to achieve.

On the contrary, the present weak competitive pos- ition of the South African economy compels us to be even more restrictive with the monetary policies applied in this country. Keeping inflation down will focus the attention even more on the most effective way to solve the real problem, and that is by following the secure path of corrective wage and productivity adjustments that will lead us to higher output and employment.

It is likewise an illusion to believe that relative competitiveness in South Africa can be restored merely by devaluing the exchange rate of the rand. A devaluation of the currency can only bring lasting benefits to a country that is prepared and able to restrict the domestic consumption of goods and services and also to keep wage and price inflation under control. In other words, a devaluation of the currency must be accompanied by extremely restrictive monetary and fiscal policies, supplemented by an incomes policy that will prevent upward nominal wage adjustments intended to compensate for the increased prices of imported goods, such as oil. Real wages must, in other words, decline before a devaluation will improve corporate profitability and overall employment.

In contemporary South Africa, devaluation will be a risky route to follow. The odds are very much in favour of a continuous and eventual detrimental devaluation/wage in-crease/price rise/devaluation cycle. In the end, growth will be lower and unemployment more -- unless we can at the outset, and not halfway through the programme, agree to the implementation of the kind of restrictive monetary, fiscal and incomes policies that will make the cure of devaluation stick. The experience of the depreciation of the rand in 1993 and the first half of 1994, when its effect fell mainly on rises in producer and consumer prices, does not lend much support to this approach.

Monetary policy is likewise linked to fiscal policy. The Reserve Bank can only give advice to Government on fiscal policy, but cannot, for example, prescribe to Gov-ernment what the deficit on the Budget should be, or how it should be funded. Obviously, a larger deficit will increase the demand Government makes on the pool of the limited amount of saving available for the development of the economy. Making use of bank credit as an alternative source for financing purposes will increase the money supply and stimulate inflationary pressures in the econ-omy. Once the fiscal policy is a fait accompli for the central bank, monetary policy must be adjusted to continue to support the Reserve Bank's mission to protect the value of the rand. Co-ordination between monetary and fiscal policy often requires adjustment of monetary policy to a more restrictive mode, with the full support and the understanding of the Minister of Finance. Scope will only be created for an easier monetary policy once the deficit on the Budget has been reduced to a more palatable level, that is to a level more in line with total capital expenditure by Government.

Once again, in the present South African situation and given the stance of fiscal policy, there is not much scope for a more expansionary monetary policy. Recent trends in the South African money and capital market interest rates clearly indicated that the rising demand for funds emanating from expanding private sector economic activity, already created friction in the financial markets. It is a further illusion to believe that the central bank can prevent interest rates from rising in this situation by creating more money. More money will only lead to more inflation, and to even higher interest rates in the longer run.

2. Monetary policy achievements in 1994

The rates of increase established in most of the financial aggregates in 1994 lead to the conclusion that monetary policy last year was not restrictive enough. There were certain special circumstances such as the disruptions in the period preceding the election of April 1994, unforeseen large fluctuations within the financial year in Government deposits held with the banking sector, relatively wide fluctuations in the exchange rate of the rand because of volatile capital movements and speculation about the future of exchange control. This does not, however, distract from the fact that there was an over-expansion in the money supply which cannot be tolerated again in 1995.

During 1994:

the M3 money supply increased by more than R30 bil- lion which was equal to a 16 per cent increase on 1993. It is true that the money supply in 1993 increased by only 7,4 per cent and that the annual average rate of increase over the past three years, that is from the end of 1991 to the end of 1994, was only 10,5 per cent. To maintain financial stability, however, the rate of increase in M3 in 1995 will have to be constrained again to a much lower level than that of 1994;

the total amount of bank credit extended during the first eleven months of last year to the private and government sectors rose by no less than 21 per cent. The banking sector's total claims against the private sector alone increased by 17 per cent. This activity represented the main statistical cause for the excessive increase in the money supply;

the rate of inflation, measured by either the changes in producer prices or in consumer prices, reached lower turning points late in 1993 and early in 1994, before turning upwards again. Over the twelve months up to December 1994, the producer price inflation amounted to 9,7 per cent, whereas consumer price inflation escalated to 9,9 per cent. Although special factors also affected the inflationary trends in the country and although there were signs of some abate-ment in the rate of inflation towards the end of the year, the results give no reason for complacency;

the nominal effective exchange rate of the rand in 1994 again depreciated by more than the rate of inflation with the result that the real value of the rand declined by a further 2,6 per cent (in the first eleven months of 1994). From the end of 1992 to the end of November 1994 the real effective exchange rate of the rand decreased by 7 per cent. This has done little to improve South Africa's competitiveness in world markets, but has undoubtedly contributed to the escalation in inflation. This is confirmed by the quarterly changes in the producer prices of imported goods which declined by 5,4 per cent in the fourth quarter of 1993, before escalating to plus 3,2 per cent in the first quarter of 1994, and 16,6 and 14,8 per cent in the second and third quarters respect-ively. After the exchange rate was stabilised towards the middle of the year, the average price for imports at the producer level declined again by 3,6 per cent in the fourth quarter of last year.

Under these pressures, most interest rates increased last year. The Treasury bill tender rate, for example, rose from 10,2 per cent in December 1993 to 12,7 per cent in December 1994. Over the same period, the rate on three months' negotiable certificates of deposit rose from 10,4 to 12,8 per cent and the yield on long-term government stock from 12,3 to 16,8 per cent. The inertia of monetary policy is reflected in the increase of only 1 per cent in Bank rate and in related interest rates traditionally linked to Bank rate, for example the prime lending rates of banks and the mortgage rate for home loans.

The Reserve Bank can lean against the wind of rising interest rates for some time, but cannot keep interest rates low against market pressures, particularly not if it is committed to keep the rate of increase in the money supply under control at the same time.

3. Exchange controls and the exchange rate

The Reserve Bank's official attitude on exchange controls is well-known -- we must phase out this direct intervention in the South African financial markets as and when the South African balance of payments position will enable us to do so. Developments over the past year were encouraging. In the balance of payments:

the current account moved into a cyclical deficit. Over the year as a whole, the estimated deficit amounted to only about R2 billion, but this consisted of small surpluses in the first two quarters, followed by growing deficits in the last two quarters of the year. In the fourth quarter, the seasonally adjusted annual rate of the deficit amounted to about R7 billion;

improvements in the capital account, however, were more than sufficient to accommodate the current account deficits. From a net outflow of R3,6 billion in the first half of the year, the capital account switched to a net inflow of about R8 * billion in the second. What is even more encouraging is that long-term capital, that is capital with an original maturity of more than 12 months, accounted for a net inflow of about R4 * billion in the fourth quarter when the estimated total net inflow amounted to about R3,3 billion;

the country's net gold and foreign exchange reserves declined by R3,2 billion in the first half of last year, but then rose by R6,3 billion in the period from the end of June to the end of December 1994;

the exchange rate of the commercial rand in nominal terms depreciated by 12 per cent from the end of 1993 to the middle of July 1994, but then recovered slightly to register an overall decline of only 8,5 per cent for the year as a whole;

the financial rand rate of exchange appreciated from an all-time low of R5,58 per dollar shortly before the April election to R3,92 per dollar on 1994-10-21. The discount of the financial rand against the commercial rand accordingly declined from 35,7 per cent on 1994-04-11 to only 10,2 per cent on 1994-10-21. Since October it has remained relatively stable within the range of 10 to 15 per cent.

Good progress has therefore been made towards establishing a more stable underlying situation that will make the final decision on the abolition of the financial rand easier for Government. The one element of the pre-conditions that did not improve is the amount of financial rand deposits held with South African banking institutions. At the end of November 1994, these balances amounted to R6,4 billion, compared with R4,4 billion at 1993-12-31. These balances are, however, now being covered adequately by the amount of unutilised foreign exchange credit facilities available to the Reserve Bank.

It remains the objective of the authorities to reintegrate the two exchange rates and then to continue with the present system of a floating exchange rate regime where the price for the one and only unitary currency will be determined by overall demand and supply in the foreign exchange market. Regular Reserve Bank intervention will only be used to smooth out undue fluctuations of a reversible and short-term nature in the exchange rate.

This approach, however, brings with it a new challenge in a situation of persistent net capital inflows into the country. Demand and supply in the foreign exchange market operate on day-to-day developments and do not take account of any underlying trends in the balance of payments. It is the overall supply of and demand for foreign exchange that will determine the exchange rate, and not the balance on the current account or the net capital flows in their isolation. It can therefore easily happen, as experienced by South Africa during the past six months, that a deficit on the current account of the balance of payments could be consistent with an appre-ciating currency, at least for as long as net capital inflows exceed current deficits.

In the case of South Africa, this danger is not imminent. In the first phase of the new period of net capital inflows, the Reserve Bank has a dire need to increase the level of its foreign reserves -- it only needs to be a little more alert than over the past six months on the consequences for the money supply of its intervention in the foreign exchange market. In the second phase, there will be many taps that can be opened -- first by abolishing the financial rand system, and then by gradually relaxing the exchange controls on residents. South Africa therefore does not now have to face the threat of a persistent appreciation of the exchange rate because of capital inflows, but cannot escape the consequences of the global movement towards integrated financial markets in which capital flows, and not current account disequilibria, are becoming the dominant factor in setting the level of exchange rates.

The recent Mexican dilemma proved again that a country with an overall balance of payments deficit cannot, in the longer run, avoid a depreciation of its currency, irrespective of the amount of official foreign reserves it may have at its disposal. With a well-managed floating exchange rate regime as applied in South Africa (and not with a fixed exchange rate system, albeit within the context of a band, as Mexico applied) the exchange rate should be allowed to reflect at an early stage any persistent decline in the foreign reserves. This policy was followed in South Africa when the decline in our net foreign reserves from early 1993 up to the middle of 1994 was accompanied by a persistent depreciation in the nominal and real exchange rates of the rand.

But what about a persistent increase in the foreign reserves, and particularly in a situation of continuous net capital inflows into a country? In a symmetrical and consistent system, one would then expect the exchange rate to appreciate. Should this be allowed, even if it should further erode the competitiveness of local industries? And, if not, how can a country avoid an appreciation of its currency in a situation of an over-supply of foreign exchange in a market based on a floating system, with no exchange controls and with a commitment of the central bank to follow an independent domestic money supply and interest rate policy?

These intriguing questions are not only of academic interest for South Africa -- they are realities of the world in which we now operate. They form part of a new discipline in terms of which international capital markets can change allegiances at the touch of a computer button, and investors can desert with frightful speed any government (including its central bank) with suspect economic policies.

These are the realities South Africa will have to face on an increasing scale as we proceed with our programme of exchange control and financial market liberalisation. It is often not only a question of what we want to achieve, but also of how it can be achieved, if at all, in the environment in which we have to operate.

4. Inflation containment

It is and remains the main function of the Reserve Bank to curtail inflation. The laxity in switching to a more restrictive monetary policy already earlier in 1994 should not be interpreted as a sign of weakness. The increase in Bank rate from 12 to 13 per cent at the end of September last year signalled a new approach of a more restrictive policy that will be extended into 1995.

Now that the Reserve Bank is set on a course of a more restrictive monetary policy and will during the rest of this year have to introduce some unpopular measures, the critics of the Bank will undoubtedly retrieve many of the old clichés from their files. In pre-empting some of these criticisms, let me therefore defend the monetary policy approach by repeating already at this stage a few of the more well-known central banker's counter clichés:

Firstly, the level of interest rates is primarily determined by market forces -- the decision for the Reserve Bank is normally whether changes in the level of interest rates as dictated by market forces should be accepted and endorsed by the central bank, or whether it should oppose them. It will be irresponsible for the Bank to lean against the wind if interest rates want to rise in a situation where the rates of increase in bank credit extension and in the money supply already exceed 15 per cent per annum, as we now have in South Africa.

Secondly, in a situation of re-emerging inflation, it is not a question of inflation or growth, it is a question of inflation and growth. You cannot have more growth with more inflation, particularly not if the rate of inflation is already at the level of 10 per cent per annum. Sustainable economic growth in the longer run is only possible in an environment of low inflation.

Thirdly, timeous action by the central bank to fight inflation is not a spiteful action intended to stop economic recovery in the bud. If the Reserve Bank waits for inflation to go up before it acts, it will act too late. It is not too early to act now in South Africa. A more restrictive monetary policy is needed to make sure that the current economic upswing will not be of the boom-bust nature of earlier times, but will be more durable.

Fourthly, in the present South African situation, opponents of the Reserve Bank will try to discredit our anti-inflationary measures by intimating that our policies, how well they may be intended, frustrate the objectives of the Reconstruction and Development Programme. May I therefore also pre-empt this one by quoting from Section 3.9 of the White Paper on Recon- struction and Development:

"The Constitution asserts the need for the independence of the South African Reserve Bank, so as to ensure that it is insulated from partisan interference and is accountable to the broader goals of reconstruction and development. The main functions of the Reserve Bank are to maintain the value of the currency, to keep inflation relatively low, and to ensure the safety and soundness of the financial system."

The Reserve Bank remains determined to fight infla- tion and in the process to make the contribution demanded of it by the Reconstruction and Development Programme.