GLOBALISATION AND THE SKILLS BIAS OF
OCCUPATIONAL EMPLOYMENT IN SOUTH AFRICA
L. EDWARDS

This paper presents a critical analysis of existing research on trade and employment in South Africa.   In an advance upon previous approaches an input-output decomposition methodology is used to analyse the impact of trade and structural change on occupational employment in South Africa.   The analysis covers 1984-97, a period of increasing trade reform and trade flows.   There are two main findings.   First, the impact of net trade flows on employment is positive, but is insignificant in comparison to changes in final demand and technology.    Second, technological change is characterised by a striking skill bias that has reduced elementary skilled employment while raising high skilled employment.    The skill bias of net trade is less clear, although a bias is emerging within the manufacturing and service sectors.   This is particularly evident within the export market.

While the business sector accounts for most of South Africa’s R&D, its R&D performance is weak.    Government incentives designed to enhance business sector R&D in South Africa are confined to direct grants while R&D expenditures in South Africa attract substantially less favourable tax treatment than in many other countries. This article argues that tax incentives are increasingly international best practice in enhancing business sector R&D and both more effective and easier to administer than grants.    While some grant activity is justified, tax incentives are appropriate in the current South African context that seeks to enhance the overall performance of the business sector.

This paper examines whether developing countries are likely to gain from further moves to liberalize trade in services.   It finds that apart from the traditional gains from trade that stem from international division of labor, developing countries should realize improved productivity in all sectors through access to better (imported) producer services.    More developed economies will also encounter gains from intra-industry trade.    However, experience from reform suggests that the magnitude of the gains to developing countries largely depends on their capacity to impose a transparent and sound regulatory framework.  It is of particular importance to ensure competition in the liberalized producer service sectors.

A recent report by South Africa’s Department of Transport commented on the relative importance of shipping costs in the competitiveness of South African exports, and indicated that expectations are that shipping costs to and from South Africa is set to increase over the next decade.    In this light the present paper presents some empirical evidence on the relationship between South African exports and shipping costs.   It is shown that shipping costs, as represented by the import CIF/FOB band, is significantly higher in South Africa’s case than the world average.   Although shipping costs is found to have a statistically significant and negative effect on exports, the effect is relatively small and overshadowed by the effect of the real exchange rate. This reinforces the importance of trade and exchange rate policy, at least over the short term, to improve South Africa’s international competitiveness.

Inflation targeting provides a monetary policy framework under which policy actions are guided by expected future inflation relative to an announced inflation target (Green, 1996). Such a framework was implemented in South Africa when the South African government adopted a 3 to 6 per cent inflation target for the year 2002, as announced by the Minister of Finance in his Budget Speech on 23 February 2000.

This note briefly summarises the rationale for and practicalities involved in inflation targeting, with specific reference to South Africa. The rationale for its introduction will be presented first, followed by the key dimensions of the target decided upon and some observations on the impact of this target on the monetary policy process.

1. THE BACKGROUND TO AND RATIONALE FOR INFLATION TARGETING

During the final decades of the 20^th century, theoretical developments and practical experience led makers of economic policy to focus monetary policy on financial and price stability, rather than on attempting to fine-tune production or exploit the evasive Phillips curve trade-off. Mainstream economists and analysts became convinced that monetary policy could best serve economic development and growth by providing a stable financial environment within which the various economic processes could develop to the full. This found articulation in the Constitution of the Republic of South Africa and in the South African Reserve Bank Act, which instruct the Reserve Bank to protect the value of the currency in order to achieve balanced and sustainable economic growth.

Translating the quest for financial stability into practical monetary policy terms has involved various frameworks. Price stability always featured as an ultimate objective, but intermediate targets were formally incorporated with the adoption of annual money supply targets in 1986. In 1990 they were renamed money supply guidelines, to underscore their flexibility. In practice an eclectic monetary policy approach was used, taking a fairly wide range of variables into consideration when deciding on the appropriate monetary policy stance.

In March 1998 the then Governor of the Reserve Bank indicated that the Bank would strive "to bring the domestic inflation rate down to a level more in line with inflation in the economies of South Africa’s major trading partners which now fluctuates between 1 and 5 per cent per year. This cannot at this stage be accepted as an official or firm inflation target for South Africa, but it provides an important economic guideline when monetary policy decisions aimed at longer-term objectives are taken". So a form of informal inflation targeting was introduced by the Reserve Bank, simultaneously announcing guidelines for the rate of growth in the M3 money supply. It was not quite clear which objective would be deemed the most important if conflicting signals were given by M3 growth and by inflation.

In August 1999 the present Governor of the Reserve Bank devoted a section of his Governor’s Address to the annual meeting of shareholders of the Bank to the possible introduction of formal inflation targeting. This was followed by a number of initiatives: further research on the topic, development of the Bank’s inflation-forecasting and modelling capacity, technical discussions on the issue with the Department of Finance, consultation with other central banks that were already engaged in formal inflation targeting, and discussions with trade unions and business organisations. These initiatives culminated in the announcement of the 3 to 6 per cent target on 23 February 2000.

An inflation targeting framework has at its core the public announcement of numerical targets for inflation within a specific time horizon, and the full commitment of the monetary authorities to price stability as the primary goal of monetary policy, to which other goals are subordinated. It allows the authorities to use all information available and not just a single variable in deciding on the best settings for policy. Clearly it enhances the transparency of monetary policy and the accountability of the central bank, which is judged by whether or not it attains the inflation target.

The advantages of an inflation targeting framework that contributed to the decision to adopt it, include –

• transparency – the concept is easily understandable, with the ultimate policy objective translated into an explicit target value;

• enhanced clarity about the objective of monetary policy, which is conducive to sound planning in both the private and public sectors;

• improved accountability of the Reserve Bank;

• elimination of the need to rely on a stable relationship between the money stock and inflation, which has become increasingly difficult to identify;

• enhancement of economic policy co-ordination, with government and the central bank publicly committed to the same inflation target; and

• providing an anchor for inflation expectations and price and wage setting, thus reducing the friction which arises from widely divergent inflation expectations.

Some disadvantages of inflation targeting must be acknowledged. It is a complicated approach, relying heavily on forecasts in an uncertain economic environment. Where forecasts turn out to be wrong, even if for completely unforeseeable reasons, the central bank’s credibility could be impaired. The counter-argument is that it simply makes visible the uncertainty that would remain hidden in other monetary policy frameworks.

Inflation targeting, if pursued at any cost, runs the risk of inefficient output stabilisation. Significant supply shocks to the economy such as sharp oil price movements could require very large monetary policy adjustments to bring inflation back inside the target range within the stated time horizon. For such exceptional events, some discretion and patience in re-achieving the target range should be allowed for, as will be argued below.

2. KEY DIMENSIONS OF THE INFLATION TARGET

The measure of inflation chosen is of considerable importance. South Africa’s initial formal inflation target, announced in February 2000, specifies that the Reserve Bank is to achieve an average rate of increase in the CPIX measure of inflation of between 3 and 6 per cent for the year 2002. The CPIX measure includes the full consumer price index with the exclusion of mortgage interest costs. It was chosen to ensure as wide a coverage of consumer items as possible, but without the mortgage interest component that would perversely fall as the Reserve Bank relaxed monetary policy and vice versa. The CPIX measure chosen covers not only 14 metropolitan areas but also 39 other urban areas where prices are surveyed, ensuring as wide a geographical coverage of price information as possible. Wide coverage was deemed to be more important than non-susceptibility to non-monetary shocks, since the consumer eventually faces a cost of living in which those shocks cannot be wished away.

The time horizon was chosen after taking into consideration the fairly long and variable lags involved in monetary policy. Given the typical lag of some 1½ to 2 years between a change in short-term interest rates and inflation, an announcement of a target value for inflation should involve a target date at least some 1½ years into the future so as to be realistic in expecting the central bank to reach the target. Accordingly, the target set in February 2000 was for the average inflation rate during 2002.

The South African authorities have chosen a target range rather than a single point. A specific target value, like the 2 per cent by 1998 announced by the Bank of Spain in December 1996, is clear and straightforward. It focuses attention, expectations and policy actions on a single value. However, it implies a degree of precision which cannot realistically be expected of monetary policy. A degree of variation in inflation is to be expected, even under the calmest economic conditions. Most inflation targeting countries explicitly allow for such variation by setting a target range, such as New Zealand's 0 to 3 per cent. The United Kingdom's and Sweden's variation is to stipulate a specific point target but to indicate an allowable deviation therefrom – for example 2 per cent plus or minus 1 per cent per annum.

The width of the range was set at 3 percentage points; it allows for some variability of the inflation rate, conveying the message that some uncertainty is inherent in monetary policy and economic outcomes in general. At the same time it is not so wide as to reinforce widely divergent inflation expectations.

As far as the level of the target is concerned, low inflation reduces the stress and friction in the economy and provides a foundation for sustainable development and growth. At the same time, a large, rapid reduction in inflation through demand-management policy may be rather costly in terms of temporary reductions in production and employment. This is to a large extent because inflation expectations tend to be rigid, usually only changing slowly over time. Given these considerations and taking into account its budget framework, government has decided on 3 to 6 per cent as the appropriate target range for inflation.

3. THE PRACTICAL FORMULATION OF MONETARY POLICY
WITHIN AN INFLATION TARGETING FRAMEWORK

The South African Government has set the target at 3 to 6 per cent for the average inflation rate during 2002. A target for 2003 will only be announced in 2001, and so on every year. Technical consultation between the Reserve Bank and the National Treasury will inform this process.

Policy is formulated with a view to ensuring inflation outcomes within the target range. If it is felt that with unchanged short-term interest rates the inflation outcome in the target period will exceed the target, interest rates will be raised, and vice versa if undershooting of the target is expected. To aid this process, the staff of the Reserve Bank have developed a suite of econometric models aimed at understanding the inflation process and forecasting inflation. These models are used to investigate the outcome of various scenarios, and are indispensable for inflation targeting. However, in the end the monetary policy decisions are taken on the basis of professional judgement rather than mechanical acceptance of model results.

In practice, the new framework has undoubtedly reinforced the longer-term focus of monetary policy. The immediate effects of shocks like sudden oil price or exchange rate movements are discounted by considering their inflationary effects eighteen months into the future and responding accordingly, rather than attempting to neutralise the shocks immediately. This might reduce the amplitude of policy changes.

Regarding the matter of "escape clauses", it has been mentioned above that sudden shocks to the economy, especially close to or during the period when the quantitative target is to be achieved, can steer the inflation outcome off course. For instance, a doubling of the oil price in 2002 would make the set target extremely difficult to achieve. Some moderation and patience in monetary policy may be called for to allow the first-round effects of the shock to wear off; a temporary deviation from the target may be unavoidable. Indeed, trying immediately to neutralise the first-round effects on inflation of a large economic shock through monetary policy would require such a huge adjustment to interest rates that it would certainly be counterproductive. In some countries the conditions under which temporary deviations from the inflation target are allowable, are stipulated in a formal document. In South Africa this formalisation has not been deemed practical at this stage, not least because of the many potential sources of shocks. It is accepted that the framework adopted, involves an extremely serious but not totally inflexible target.

A reasonable degree of central bank independence, and in particular freedom to set policy instruments (mainly short-term interest rates), and freedom from budgetary demands, are deemed essential preconditions for successful inflation targeting. In South Africa the Reserve Bank has the necessary degree of independence, granted under the Constitution of the Republic of South Africa and the South African Reserve Bank Act. A tradition of independence has been developed in recent years. Independence should, however, not be seen as implying isolation; regular consultation with the Minister of Finance continues in the interest of policy co-ordination.

The nature of the inflation targeting framework, progress towards the target and policy steps taken have to be communicated to all relevant parties, including the general public, so that these will have an appropriate impact on inflation expectations and enhance public understanding of and support for monetary policy. To this end, numerous actions have been taken or are in the pipeline:

• the inflation target is publicly announced;

• policy is explained in a statement issued after each seven-weekly meeting of the Reserve Bank's Monetary Policy Committee;

• macroeconomic developments and monetary policy actions are also discussed in the Reserve Bank's Quarterly Bulletin, Annual Economic Review and also in the Governor's Address at the annual meeting of shareholders;

• a Monetary Policy Review is to be issued every six months;

• Monetary Policy Forum meetings, in which the Bank's governors explain developments in monetary policy to and obtain feedback from organised business, labour, politicians and academics, are held twice a year in ten centres, covering all provinces;

• the Governor regularly appears in parliament before the Portfolio Committee on Finance; and

• the Governors and senior staff give speeches and participate in panel discussions and other public events.

4. CONCLUSION

In a wider context, inflation targeting is part of a process in which economic policymaking is becoming more transparent and subject to more accountability and technical rules, and less susceptible to discretionary actions. Discarding the use of subtle intermediate variables and nuances is a great virtue, and the new framework may well prove to be more than a temporary fad. It is no panacea: central banks will still need to earn their credibility and work hard at maintaining it, through tight policies where necessary. However, to the extent that the inflation target focuses policies and expectations it can reduce the frictional costs associated with curbing inflation.

The key parameters of the inflation targeting framework implemented in South Africa from February 2000 were determined on the basis of thorough research and taking into account best international practice and advice. Monetary policy is firmly aimed at achieving the target in 2002 and beyond.

(i) Targeting inflation itself is a better policy goal than, say, growth in the money supply (which is not the only cause of inflation), or the exchange rate (which is greatly influenced by exogenous factors), or a range of competing considerations including inflation, currency stability and economic growth. Policy credibility is therefore greatly enhanced by such a move, and this is an important ingredient in attracting foreign capital inflows, increasing domestic and foreign investment and thus raising the economic growth potential over the medium term.

(ii) As inflation targets are announced by government the decision is squarely in the political realm and the monetary "technocrats" at the central bank therefore have an unambiguous and politically-backed mandate to fulfil. This reduces the likelihood of government allowing fiscal policy trends to impact negatively on inflation – co-ordination of monetary and fiscal policy has always been an important issue bearing on inflation.

(iii) Provided the inflation targets enjoy a broad level of societal support, they are a relatively painless way of reducing inflationary expectations and therefore the actual rate of inflation. This is clearly preferable to proceeding towards a defined target without general support - although even under such circumstances targets could still be announced and attained but with greater cost in the short term as a consequence of higher interest rates and lower growth in employment and the economy.

The commitment to inflation targeting from the South African Cabinet in the second half of 1999 effectively proscribed any debate about the merits of adopting such a policy in the period leading up to the actual announcement in February 2000 of a 3-6 per cent inflation target for 2002. Backing away from inflation targets then would, as now, almost certainly have given rise to a very negative reaction in financial markets.

2. WHAT MEASURE OF INFLATION SHOULD BE USED?

(a) "Headline" Inflation

"Headline" inflation is the official measurement of consumer inflation in South Africa’s metropolitan areas. It is published monthly. Weightings within the basket of goods and services measured are determined from household surveys conducted every 5 years.

The most important categories and their weightings currently are:

(b) "Core" Inflation

Because the policy instrument for combating inflation – interest rates – are included via the mortgage rate in the measurement of headline inflation, there is widespread agreement that this should be excluded from the measurement of inflation targeted (see appended Table 1). It can also be argued that other relatively volatile prices which are determined by outside influences (domestic and global) and are beyond the influence of monetary policy should also be excluded. In South Africa "core" inflation excludes:

1. Fresh and frozen meat and fish whose prices may be highly volatile, particularly during and following periods of drought.

2. Fresh and frozen vegetables and fresh fruit and nuts whose prices may be highly volatile from quarter to quarter due to their sensitivity to seasonal and climatic conditions.

3. Interest rates on mortgage bonds and overdrafts – changes in which reflect shifts in monetary policy. In this way the measurement of inflation avoids the perverse consequence whereby tighter monetary policy as a result of higher inflation will initially result in even higher inflation through higher interest costs.

4. VAT which is determined by fiscal policy.
   

5. Assessment rates which are determined by local government.

Core CPI Excluding Petrol. Some countries also exclude petrol and heating oil prices from their measurement of core inflation. South Africa chose not to do this, but for a while published a series of core inflation excluding petrol prices so that a comparable measurement could be made.

Headline CPI Excluding Mortgage Payments (CPIX). If the inflation targets are indeed to influence inflation expectations, it is crucial that the measurement used enjoys widespread acceptance and is easily understood. This cautions against excluding too many goods and services or items that feature significantly in household spending such as food and transport. For this reason South Africa’s chosen inflation measurement of CPIX measures headline CPI less only the mortgage rate. This measurement has the advantage that it is not complicated, it covers all those items to which poorer households (who often do not have mortgages) are most sensitive, and avoids the perverse impact that changing interest rates can have on the inflation measure. Moreover, the chosen measurement is for metropolitan and other urban areas (unlike headline and core inflation for which the measurements are mostly for metropolitan areas only) and thus ensures that it is applicable to as broad a geographical area as possible. This too enhances its credibility.

But the measurement runs the risk of measured inflation surging or falling as a result of weather-related changes in food prices, globally determined petrol prices, or changes in the VAT rate. It is therefore surprising that the inflation targets did not follow the example of some other countries and include acceptable escape clauses whereby sudden movements in any of these measures will provide an acceptable "excuse" for over/under-shooting the inflation target. This is partially offset by the decision to adopt a target range rather than a single inflation point. The target range gives the Reserve Bank some leeway in the face of unforeseen price shocks. At the same time the band is sufficiently narrow to reflect a strong commitment on the part of the monetary authorities to inflation reduction.

3. WHAT TARGET AND WITHIN WHAT PERIOD?

Any inflation target that would be considered too timid had clearly to be avoided as it would have been received very negatively in the financial markets. A broad guide was the oft-expressed desire to reduce South Africa’s inflation rate to the level of its major trading partners and thereby achieve a level of relative price stability that is widely accepted internationally as appropriate. This suggested that the inflation target could not have been more than 3-6 per cent (see appended Table 2).

However, a far more difficult issue in South Africa’s context was whether this should (and indeed could) have been achieved in one year or over two to three years. There was by no means unanimity amongst business analysts on the latter issue.

If it had been possible to be confident that inflation would trend rapidly downwards into the 3-6 per cent range in calendar 2000, this would not have been a serious issue. However many analysts were not certain that this would be achieved, inter alia because of their concern about the possible stickiness of wage settlements and the extent to which this could feed into price adjustments in a cyclical recovery in the economy. Clearly organised labour’s position with regards to inflation targeting was also a key factor in this respect. The harsh reality was that if the 3-6 per cent target had been immediately implemented and inflation had not trended down appropriately, interest rates would have had to rise irrespective of the short term implications for growth and employment. Failure to react appropriately would have been received extremely negatively. Indeed, it would have been better to not introduce an inflation target at all than to be seen to lack the will to achieve it.

It was these risks which prompted some analysts to argue in favour of a two or three year target date while others cautioned that such an approach will be viewed as being too timid. In the event government opted in February 2000 for a 3-6 per cent target to be achieved for CPIX (for metropolitan and other urban areas) as an average for calendar 2002. This caution proved justified as the actual increase for CPIX (metropolitan and other urban areas) for 2000 averaged 7.8 per cent and the measurement for December 2000 was 7.6 per cent. Most business analysts are confident that the 3-6 per cent target for 2002 will be achieved, though some still express reservations.

4. CONCLUSION

The introduction of inflation targeting and the likely backlash in the financial markets to any perceived departure from the set targets means Business has little choice other than to support this move. On balance the decision to introduce inflation targets was probably appropriate and the 3-6 per cent target realistic. The target bands are sufficiently lower than the actual inflation rate at the time of their introduction and narrow enough to demonstrate a firm commitment to reducing inflation. Given actual developments in the global oil market and continued stickiness in domestic wages and prices in important areas, it was also appropriate that the inflation target was phased in over three years.

APPENDIX

Table 1. Price Index Adopted by a Sample Group of Countries

Note: ¹ Australia switched from a core measure to the Headline CPI in 1998 when the latter was redefined by the statistical agency to exclude interest charges.

Source: Schaechter, A., Stone, M.R. and Zelmer, M., "Adopting inflation targeting: Practical Issues for Emerging Market Countries", Occasional Paper No. 202, International Monetary Fund, Washington D.C., 2000, pp. 10-11.

The objective was to bring inflation within this band by 2002. To ensure that the Bank would be targeting an appropriate measure, a new index of consumer prices – CPIX – had been introduced. This includes the full basket of goods and services that a typical household consumes – measured by the consumer price index (CPI) – but excludes interest costs as these are a direct outcome of monetary policy.

Recently, Fedderke and Schaling (FS)(2000) have modelled South African inflation using wages, aggregate demand and exchange rates. FS confirmed an earlier finding to the effect that inflation is a cost-push phenomenon (Greyling, 1987). More specifically, inflation is the outcome of wage costs plus profits, with inflation rising when either unions try to increase their share of the national cake (GDP), or when firms do so by increasing their mark-up over wage costs.

The way things seem to have worked in South Africa is as follows. Actual prices are a mark-up over unit labour costs. This means that when wages increase by, say, 10 per cent, firms increase their prices by about 13 per cent. So prices go up by more than would be justified by the increase in cost alone.

However, presently administrative prices are playing havoc with the South African inflation rate. Of the monthly increase of 0,9 per cent in the CPIX during the middle of last year, about 0,8 per cent came from administrative price hikes, mainly in the form of petrol, electricity and water rates.

Administrative prices are regulated by government in a monopolistic manner. These prices are hard to bring under control using interest rates, which is the main weapon that the South African Reserve Bank has in fighting inflation. They have a weighting of around a quarter in the inflation rate.

The Administrative Price Momentum Indicator (or APMI) again reached nearly 50 towards the middle of last year, indicating that the speed of administrative price rises was picking up again after two months of slowing down. The positive APMI implied that administrative prices were contributing nearly 50 per cent more to the inflation rate than their neutral weighting would justify.

Marginal increases in the CPIX last year were due to administrative price hikes; if these hikes had not happened, the CPIX would have actually fallen. Interestingly the SARB said in its policy statement on 18 August 2000 that it

believed inflation had now peaked and that "supply side" factors were influencing the inflation rate. All administrative prices can be considered supply side – in the sense of cost-push - factors. But if administrative prices should continue rising it seems unlikely that the rate of both core and CPIX inflation will fall much anytime soon.

In Eastern Europe the Czech Republic does not include administrative prices in the inflation measure that it uses for monetary policy decisions. Interestingly Hungary caps administrative prices just below the inflation target it has set itself; administrative prices such as bus tickets cannot be increased by more than the targeted rate of 5,8 per cent. Furthermore countries such as Brazil have used administrative prices to get inflation down during the currency crisis of 1999 when tariffs were lowered and government charges on port use were lowered to help keep inflation in check.

For South Africa to reach its inflation target of between 3 per cent and 6 per cent as an average is going to depend on administrative prices, as none are currently below the upper range of the inflation target. If administrative prices remain on average around 10 per cent then the rest of the inflation rate will have to be 3,7 per cent in order to meet the target. Until recently administrative prices were increasing at a rate of 11,7 per cent on the CPIX index, and they have not been below 10 per cent for some 15 months and have never been below 6 per cent since the early seventies.

The latest government policy framework on privatisation mentions the importance of competition to help keep administrative price hikes in check. Furthermore, the policy framework mentions that regulators need to be set up in many of the sectors earmarked for privatisation so that monopoly pricing can be contained when it happens. It seems that government has woken up to the fact that administrative prices could destroy its own set of inflation targets.

We have shown that administrative prices are still playing havoc with the South African inflation rate. However, government does seem to have recognised this fact and appears to be taking the necessary remedial actions.