UNIVERSITY OF GRONINGEN
Faculty of Economics and Management and
Organization
INFLATION TARGETING IN DEVELOPING COUNTRIES
AND ITS APPLICABILITY IN SOUTH AFRICA
MSc Economics
Student name:
Kiangi, Richard M.F.
Student number: S1655124
INFLATION TARGETING IN DEVELOPING COUNTRIES AND ITS APPLICABILITY IN SOUTH AFRICA
Abstract
This study examines inflation targeting (IT) in developing countries and its feasibility to the South African economy. Three prerequisites of IT are analysed: central bank independence, having a sole target, and the existence of a stable and predictable relationship between monetary policy instruments and inflation.
An empirical investigation is carried out using vector autoregressive (VAR) models to assess if South Africa satisfies the prerequisites of IT. Since the variables are nonstationary, integrated of the same order and cointegrated, Granger non-causality test is applied using the cointegration and the vector error correction methodology (VECM). Generalized impulse responses are also used to assess dynamic interactions among the variables used in this study. Generally, the outcomes of the study indicate that among the three prerequisites, two are not met. Yet, we still believe that South Africa can implement IT with success because of its ability to bring down inflation to low levels. Also the SARB’s legal framework is improved to establish full central bank independence.
Keywords: Inflation targeting, Central Bank independence, Vector AR systems, Granger
causality, South Africa.
KIANGI, RICHARD M.F. S.1655124
ii
Acknowledgements
iii
CONTENTS
1 Introduction... 1
2 Inflation targeting... 3
2.1 Definition of inflation targeting (IT)... 3
2.2 The advantages and disadvantages of an inflation targeting regime ... 6
2.2.1 The advantages of inflation targeting ... 6
2.2.2 The disadvantages of inflation targeting... 7
2.3 Prerequisites of inflation targeting... 8
2.4 Implementation of inflation targeting ... 10
2.4.1 Assignment of the target ... 11
2.4.2 Definition of the target variable... 11
2.4.2.1 Time horizon of the target... 11
2.4.2.2 Choice of the price index ... 12
2.4.2.3 Level of the target ... 13
2.4.2.4 Width of the target band ... 14
2.4.3 The transparency and accountability of monetary policy ... 15
3 Alternative policy regimes and inflation targeting in developing countries... 16
3.1 Alternative policy regimes... 16
3.1.1 Exchange rate targeting... 16
3.1.2 Monetary targeting... 19
3.2 Inflation targeting in developing countries ... 20
3.2.1 Problems with independent monetary policy in developing countries ... 20
3.2.2 Conflicts among policy objectives... 22
3.3 Literature review... 23
4 South African economic history and central bank independence ... 26
4.1 A brief history of the South African economy... 27
4.2 Rationale for adopting inflation targeting in South Africa ... 29
4.3 Central bank independence in South Africa ... 32
4.4 Having a sole target ... 37
5 An empirical analysis of the relationship between the monetary policy instruments and inflation targeting ... 38
5.1 Methodology ... 38
5.2 Data description ... 44
5.3 Empirical results ... 45
5.3.1 Unit root test ... 45
5.3.2 Johansen cointegration test ... 47
5.3.3 Vector error correction model (VECM) ... 50
5.3.4 Short-run Granger causality tests... 52
5.3.6 Generalized impulse responses ... 53
5.4 Summary ... 54
6 Conclusion ... 55
References... 57
1
1 Introduction
Inflation targeting (IT) is a monetary policy framework, characterized by public announcement of official target ranges or quantitative targets for price level increases and by explicit acknowledgement that low inflation is the most crucial long-run objective of the monetary authorities. There are three prerequisites for inflation targeting:
a) central bank independence, b) having a sole target and
c) the existence of a stable and predictable relationship between monetary policy instruments and inflation.
The objective of this study is to analyze the three above mentioned prerequisites of inflation targeting in developing countries and to evaluate its feasibility in South Africa. To analyse the applicability of this monetary policy framework we make an empirical exploration to reveal whether or not there is a stable relationship between the monetary policy instruments and inflation in South Africa. This is to fill the gap of other studies like that of Woglom (2000) that does not take into account the third prerequisite of the inflation targeting framework.
In many developing countries, the use of seigniorage revenues as an important source of financing public debts, the lack of commitment to low inflation as a primary goal by monetary authorities, considerable exchange rate flexibility, lack of substantial operational independence of the central bank or of powerful models to make domestic inflation forecasts hinder the satisfaction of these requirements.
2
bank seems to have not satisfied the independence criteria due to the problems associated with the appointment of the governors.
Having a sole inflation target was hindered by the existence of fixed exchange rate system throughout the years. However, in 1999 South Africa switched to a floating exchange rate regime, which is important for a successful inflation-targeting regime. Having a sole target within the system has also been supported by the new central bank law, which gives priority to price stability and supports any other objective as long as it is consistent with price stability.
In this study, an empirical investigation has been carried out in order to assess the statistical readiness of South Africa to satisfy the requirements of inflation-targeting by making use of vector autoregressive (VAR) models. We apply Granger non-causality tests in order to assess the short-run relationship between consumer price index (CPIX), broad money supply (M3), three-month treasury bill (R), nominal exchange rate (NER) and the real gross domestic product (GDP). Since the variables are nonstationary, integrated of the same order and cointegrated, Granger non-causality tests are applied using the cointegration and the vector error correction methodology (VECM). The empirical findings in the study suggest that, in the long-run, money growth has a significant positive impact on prices. Other policy instruments (exchange rates and interest rates) have no significant impact on prices in the long-run as indicated by the cointegrated equations and the impulse responses. In the short-run there is no evidence of a Granger-causal relationship between money and prices. Impulse responses support the outcomes of the cointegrating relationship, that is, money and prices are positively related in the longrun. In this empirical analysis, time-series data was used for the period 1988:1-2006:4.
3
government financing. Hence in addition to its instrumental independence, the SARB needs to have an operational independence. This means that monetary policy decisions should not be conditioned by other factors other than inflation. Furthermore, if IT is a desired monetary policy framework, the possible solutions available to the SARB are the pursuit of fully central bank independence and restricting the possibility of fiscal dominance.
This study is organized as follows: Section 2 discusses in detail the definition of IT and the prerequisites of this monetary policy framework. Section 3 examines the alternative policy regimes (exchange rate targeting and monetary targeting) and inflation targeting in developing countries. Section 4 examines the historical background of South African economy as well as central bank independence in South Africa while Section 5 conducts a series of tests to address the relationship between monetary policy instruments and inflation in South Africa. Section 6 concludes.
2 Inflation targeting
2.1 Definition of inflation targeting (IT)
Inflation targeting as a monetary policy framework has spread rapidly to developing countries as well as in developed countries since the 1990s. In 1990 New Zealand was the first country to implement this strategy followed soon by Australia, Brazil, Canada, Chile, Mexico, Sweden, the United Kingdom and others. In most cases the adoption of this policy framework helped these countries to respond practically to the difficulties they found in conducting monetary policy using an exchange rate peg or some monetary aggregate as the main intermediate target. The switch to this new policy framework also showed a deliberate attempt by these countries to improve their inflation records.
4
just monetary aggregates or exchange rate, are used for deciding the strategy the setting of policy instruments, 4) increased transparency of the monetary policy through communication with the public and the markets about the plans, objectives, and decisions of the monetary authorities and 5) increased accountability of the central bank for attaining its inflation objectives (Mishkin, 2001). This list of elements should clarify one crucial point that inflation targeting entails more than announcement of a numerical target over a specific time horizon.
Carare and Stone (2003) provide a broader definition of inflation targeting which includes three distinguished regimes: 1) Full-fledged inflation targeting- defined in accordance with the definition of inflation targeting by Mishkin (2001), 2) eclectic inflation-targeting and 3) inflation targeting lite.
Eclectic inflation-targeting countries holds to those countries that have so much credibility that they can maintain low and stable inflation without full transparency and accountability with respect to inflation target. Their low and stable inflation and high degree of financial stability affords them a flexibility to pursue the objective of output stabilization and price stability. Inflation targeting lite countries float their exchange rate and announce an inflation target but, owing to relatively low credibility are not able to maintain inflation target as the foremost policy objective (Carare and Stone, 2003).
South Africa is regarded as a full-fledged inflation targeter and thus this thesis will be basically restricted to this type inflation targeting regime. Carare and Stone (2003) distinguish seven industrial and eleven emerging market countries which practice this regime1. Although all these countries are regarded as full-fledged inflation targeters and they have some commonalities, there exist some differences in the target design and operation features among them.
1 The countries are Australia, Brazil, Canada, Chile, Israel, Hungary, Thailand, Iceland, Norway, Korea,
5
This thesis is particularly focused in three basic prerequisites for an inflation targeting regime. The first prerequisite for a country to apply inflation targeting is that the central bank be able to conduct monetary policy with a degree of independence. The second prerequisite is the absence of another targeted variable such as wages, level of employment, or the nominal exchange rate (Masson, Savastano and Sharma, 1997). The third prerequisite is the existence of stable and predictable relationship between the monetary policy instruments and inflation rate (Christoffersen, Slok and Wescott, 2001).
In the 1990s, a number of industrialized countries adopted inflation-targeting as their preferred framework for monetary policy. This replaced frameworks involving using targets for the exchange rate or monetary aggregates which had come to be judged unsatisfactory. The main purpose of this new framework is to establish accountability, and transparency of monetary policy, and improve inflation performance (Masson, et al. 1997). Having seen its beneficial results in the industrialized countries, such as New Zealand, Canada, the United Kingdom, Sweden, Finland, Australia, and Spain, some of the emerging market economies such as Brazil, Chile, Czech Republic, Poland, South Africa and Israel, started to implement an inflation targeting regime as well. In these latter cases, inflation targeting has also proved to be quite successful. It has brought about desirable long-run inflation outcomes, due in part to its features of transparency and accountability. In Europe, for example, there has been a wave of institutional reform of monetary policy, giving the central banks a clear mandate to pursue price stability, considerable operational independence and increased accountability (Svensson, 1997).
6
2.2 The advantages and disadvantages of an inflation targeting regime
An inflation targeting regime has been implemented by some of the developed and developing countries in the recent years because it removes some problems associated with intermediate targets by focusing mainly on the most important goal of monetary policy, which is price stability. However, this advantage may be offset in the absence of a stable relationship between the instruments of monetary policy and the inflation target (Debelle and Lim, 1998).
2.2.1 The advantages of inflation targeting
Jonsson (1999) shows that the main advantage of implementing an inflation-targeting regime is that it enables a country to attain and maintain a low and stable rate of inflation leading to some beneficial effects on economic growth. In addition, by means of the explicit mandate of the central bank to focus on achieving a low inflation rate, and also by means of the increased transparency and accountability of monetary policy, the uncertainties among wage and price setters about the future path of the inflation rate may decline, and thus, inflationary expectations may become more coordinated and accurate. Thus, the central bank’s concentration on an explicit inflation target may serve as a better focus for wage and price-setting than monetary or exchange rate targeting.
Second, the inflation target enables the central bank to enhance its credibility by providing a clear reference point about prices. It serves to confirm the central bank’s commitment to low inflation rate in the eyes of the public (Debelle and Lim, 1998). By announcing a quantitative target or range, the central bank makes known its commitment, which creates conditions for much more intensive influences of the inflationary expectations of market participants.
7
regime, in contrast, may not be compatible with stabilization of the aggregate demand and supply shocks (Jonsson, 1999).
Lastly, once IT is specified, it does not need to be adjusted frequently since it directly focuses on the final objective. In contrast, in the case of the monetary targeting framework, the monetary growth target may need to be adjusted periodically since there may be shifts in the money demand function resulting in changes in the relationship between monetary growth and the price stability goal (Debelle and Lim, 1998).
Due to these advantages, the inflation-targeting framework becomes more credible than other regimes. It makes the operation of monetary policy more transparent, provides accountability, and contributes to the improvement and stabilization of investor sentiment.
2.2.2 The disadvantages of inflation targeting
Despite the above discussed advantages, inflation-targeting regime has some disadvantages: First, there is no guarantee that the central bank will be successful in using its discretion to appropriately set monetary policy. Compared with monetary targeting and exchange rate targeting frameworks, inflation targeting is more complicated to implement (Jonsson, 1999).
8
discretion may allow policy makers to follow overly expansionary policies (Debelle and Lim, 1998).
Third, although the rigid structure of the inflation-targeting regime provides a better cyclical adjustment of the economy, this discretion has the potential to increase output instability, which will lower economic growth (Mishkin, 1999 and B ernake, Laubach, Mishkin and Pozen, 1999).
Fourth, in contrast to the exchange rate or monetary targeting framework, the inflation rate cannot be easily controlled by the central bank; inflation outcomes that incorporate the effects of changes in instruments settings are revealed only after a substantial lag. This requires that the central bank engage in what Svensson (1997) described as inflation forecast targeting in which the central bank seeks to make its inflation forecast equal to the inflation target over a relevant policy horizon. The developing countries are faced especially with this problem when inflation rates are brought down from high levels. In this case, large forecast errors and frequent target misses will be inevitable. As a result, the central bank will have some difficulties in explaining the reasons for the deviations from the target and in gaining credibility, which is crucial to the inflation-targeting regime (Kadioglu, Ozdemir and Yilmaz, 2000). Furthermore, the initial disinflation process resulting from the introduction of inflation targeting may lead to short-term output costs if private agents do not immediately find the policy framework credible (Jonsson, 1999). Finally, the inflation-targeting regime requires exchange rate flexibility, which may cause financial instability (Mishkin, 2000).
2.3 Prerequisites of inflation targeting
9
Firstly, the central bank must be given complete independence to adjust freely its instruments of monetary policy toward the attainment of the objective of low inflation. The independence does not mean the fully independence but implies at least instrumental independence which permits greater discretion in the conduct of monetary policy and which mainly implies that the central bank cannot finance the government budget.
In the same manner, the central bank should not be required to attain low interest rates on public debt or to maintain a particular nominal exchange rate. There should not be any political pressure on the central bank to raise the rate of economic growth in such a way that is inconsistent with the achievement of the inflation target (Debelle and Lim, 1998). The weights of public sector borrowing requirements on the financial system must be low and there should be no direct borrowing of the public sector from the central bank and heavy reliance on the seigniorage revenues by the public sector. In the case where these conditions are not met, the inflation will have fiscal roots. Fiscally driven inflation process will undermine the effectiveness of monetary policy to achieve any nominal target and forces the central bank to follow an increasingly accommodative monetary policy. Although there is no consensus about the threshold inflation rate at which monetary policy loses its effectiveness, it is generally accepted that a country which has experienced annual inflation rates in the 15-25 percent range for three to five consecutive years will not be able to rely on monetary policy alone to target any significant and lasting decline in the inflation rate (Kadioglu, Ozdemir and Yilmaz, 2000).
10
with an inflation target to the extent that the inflation target is given the priority when a conflict arises. But, in practice, such coexistence may be problematic because it is impossible for authorities to explain those priorities to the public in a credible manner before that conflict occurs. Under these circumstances, the public is going to make its own inferences about the policy actions and there is no assurance that the policy stance will give the appropriate signals to the public about the actions and cause true inferences. Therefore, the surest and safest way to avoid those problems is to have no other targeted nominal variable and to consider the inflation target as the main policy objective (Masson, et al. 1997).
Thirdly, there is a need for an existence of a stable relationship between the inflation outcomes and monetary policy instruments. Jonsson (1999) claims that in an inflation targeting regime, monetary authorities have to be able to model inflation dynamics in the country and to forecast the inflation to a reasonable degree. So, the monetary authorities should have access to policy instruments that are effective in influencing the macroeconomic variables. And also, there must be sufficiently developed money and capital markets to react quickly to the use of these policy instruments.
2.4 Implementation of inflation targeting
There are some practical issues regarding the implementation of inflation targets which makes a difference between inflation targeters. Van der Merwe (2004) shows important decisions that have to be made in the implementation of IT:
• Who sets the target?
• How will the target variable be defined? • How will the target variable be set?
• How will the monetary policy decisions be made?
11
2.4.1 Assignment of the target
The issue of who assigns the inflation target depends on the central bank’s instrumental independence and the announcement of the inflation target has differed across countries. For example, in Spain, Sweden, Mexico and Poland the inflation target is set and announced by the central bank. In countries such as Israel, Brazil, Korea, Peru and Thailand it is set by the government in consultation with the central bank whereas it is only in Britain the target is set by the government treasury. So even if the inflation target is originally announced by the central bank, the government should subsequently endorse it since this may promote the agreement between the two policy making institutions and increase the effectiveness and the credibility of the framework (Debelle, 1997).
2.4.2 Definition of the target variable
There are certain steps that should be followed to implement inflation targeting. These steps involve determination of the time horizon over which the inflation target is specified, choice of the price index upon which the inflation target will depend, the central point of the target, whether the target is defined in terms of a point or a band, and determination of possible escape clauses or exemptions to the inflation target under specific conditions.
2.4.2.1 Time horizon of the target
12
2.4.2.2 Choice of the price index
In all inflation targeting countries the target has been specified in terms of the consumer price index. However, the choice of the price level, which is used in calculating the targeted inflation rate, may differ from country to country because there are different methodologies in calculating the CPI across countries. In practice, the target has been generally specified in terms of the CPI rather than GDP deflator since it is the price index that is most familiar to the public, which is timely and does not need as much revision (Debelle, 1997). The first factor is important because the public needs to be familiar with the price level to be targeted for the transparency of an inflation-targeting regime (Kadioglu, et al, 2000). The GDP deflator on the other hand, has a broader coverage.
Many countries use an underlying measure of inflation, which is based on the CPI that excludes the volatile food and energy sector, and mortgage interest payments instead of the headline CPI inflation rate, which is based on all items index. The reason why underlying inflation is more preferred than the headline CPI inflation rate is that the former excludes the first-round effects of the shocks that are accommodated by monetary policy. However, it does not exclude the second-round effects of the shocks on wages and prices (Debelle, 1997).
When the monetary authorities focus on underlying inflation, they may not take into consideration the first-round effects on prices and may only take into account whether it brings about an increase in inflation expectations or not. Another problem with focusing on the underlying inflation is that if all the price and wage decisions in the economy are made on the basis of a headline index, their responses to the movements in the headline inflation rate will be captured by the underlying rate. A further problem with the underlying inflation is that it may not be transparent. However, this problem can be mitigated to some extent if the statistical office calculates it. In addition, the public should be well informed about the specification of the index (Kadioglu, et al. 2000).
13
effects of changes in mortgage costs (CPIX). However, South Africa opted to target the CPIX in metropolitan and other urban areas for it can be easier understood by the public than the core index and it excludes the direct effects that changes is the repurchase rate could have on prices (Van der Merwe, 2004).
2.4.2.3 Level of the target2
The level of the target is another important aspect of the implementation of inflation targeting. In most countries, the center point of the inflation target is referred to as their interpretation of the operational definition of price stability. While in theory, zero inflation appears to be equal to price stability, in practice, the concept of price stability is influenced by some other issues like price-level measurement and nominal rigidities.
Although the primary goal of monetary authorities is to establish price stability, all inflation-targeting countries have determined their target above zero due to the upward biases in the calculation of the consumer price index (CPI). These biases are caused by the introduction of new goods, the adjustment of the consumers to relative price changes by substituting similar goods with lower prices and quality bias.
Also, the precautionary behaviour of the central bank against some economic risks supports the non-zero inflation target. First, the possibility of downward rigidity in prices and wages may require a small positive inflation rate to provide the necessary relative price adjustment. Second, a zero inflation target may exclude the possibility of negative real interest rate since nominal interest rates are bounded below by zero. This may prevent the central bank from decreasing interest rates in the case of a recession.
In general, inflation targets have been set around 2 percent per annum in developed countries. In many of the developing countries on the other hand, there is no empirical evidence for the optimal inflation rate. However, it is commonly argued that developing countries should aim at achieving a medium term rate of inflation, which is somewhat higher than that of industrial countries, that is, between 4 and 8 percent per year and is
14
permitted to fluctuate within a somewhat wider band to accommodate larger supply shocks (Kadioglu, et al. 2000).
2.4.2.4 Width of the target band 3
The next step in implementing the inflation target is to decide whether the target will be a numerical number or a band. For example, Finland and Australia determined a particular point target for the inflation rate while Canada, the United Kingdom, Sweden and New Zealand specified a band for the inflation target. Spain, on the other hand, preferred a ceiling for the inflation rate. The reason why some countries construct an inflation band is the possibility of imperfect control of monetary policy over the inflation rate. Due to the long and variable lags of monetary policy and its imperfect ability to forecast future inflation, it is not possible to make a precise prediction about the future inflation rate. So, the inflation rate will display variability. Under these circumstances, the adoption of wider bandwidth will ensure some scope for output stabilization. However, as time passes and the public realizes the beneficial effects of the regime, the inflation-targeting regime may reduce the variability in inflation and the implementation of the band regime decreases the variability of the output (Kadioglu, et al, 2000). Specifying a band is also needed to maintain some flexibility in responding to short-term shocks.
Another issue related with the inflation band is the choice of bandwidth that reflects a tradeoff between announcing a tight band and breaching it occasionally, and announcing a wide band, which may be seen as softness on the part of the central bank. A narrower band indicates a stronger commitment to the inflation target but it is riskier than a wider band because due to the difficulties in remaining inside the band, frequent breaches may occur and these can undermine any credibility gain. In addition, with a narrower band, in the case of short-term shocks, the central banks are not as flexible as they would be with a wider band. However, it is much easier to observe the performance of central banks with a narrower band since it emphasizes the short run accountability of the central bank to achieve the inflation target. The central bank has to make an explanation for the reasons for any breach of the band.
15
An important consideration in determining the bandwidth of the target is that adopting a narrow band may induce instability in the instrument of monetary policy. He also states that to achieve a given movement in the inflation rate, the shorter the time horizon, the larger the change in the instrument of monetary policy. The change in interest rate may be higher within the narrow band than within the wider band. Such fluctuations in interest rates may lead to instability in the financial markets even though the inflation target is met.
Also, the necessary change in the inflation rate can be induced if the monetary authorities use the exchange rate to reach the inflation target. In that case, monetary authorities try to achieve exchange rate stability by fixing the exchange rate of the domestic currency to that of an anchor country in order to import low inflation. In the short term, the exchange rate may enable the monetary authority to reach the inflation target more easily. However, the systematic reliance on the exchange rate to achieve the inflation target may lead to a contradiction with the use of interest rates in the medium term. In the case of a wider band, monetary authority is more flexible in using the monetary instruments. Yet, the wider band causes the economic institutions to focus on the upper edge of the band thereby leading to higher inflationary expectations. This might be the reason why some countries prefer narrower ranges. The adoption of the point target introduces credibility to the implementation. However, because of the existence of the unpredicted events and the nature of the inflation, the achievement of the point target becomes difficult. Overall, implementation of a band decreases credibility but increases flexibility (Kadioglu, et al. 2000).
2.4.3 The transparency and accountability of monetary policy
16
more apparent whether any breach of the inflation target resulted from an error of the central bank or whether the breach was predictable at the time of the policy decision (Debelle, 1997).
Van der Merwe (2004)’s study shows that the inflation targeting framework improves the accountability of the central bank because it provides an explicit and publicly known benchmark that must be reached over a specific time frame.
3 Alternative policy regimes and inflation targeting in developing countries
3.1 Alternative policy regimes
This section provides the main features of exchange rate targeting and monetary targeting regimes. These regimes are the most frequently used in developing countries compared to inflation targeting, thus this study gives them a paramount importance. Their advantages and disadvantages are also discussed. Furthermore, the issues and problems about inflation targeting faced by developing countries are examined.
3.1.1 Exchange rate targeting
Exchange rate targeting is a monetary policy regime under which the central bank tries to establish exchange rate stability via interest rate changes and direct foreign exchange interventions designed to import low inflation from the anchor country. Maintaining the exchange rate has some prerequisites such as an appropriate macroeconomic policy mix that ensures a low inflation differential vis-à-vis the anchor currency, a sufficient level of international reserves, and maintaining the country’s competitiveness and overall credibility with regards to its institutional and legislative framework and political stability (The Czech National Bank, 1998).
17
involved fixing the exchange rate of domestic currency to that of a large, low inflation country whose inflation is lower than in the domestic economy and which has a substantial share in the small country’s trade. In this case, the exchange rate is pegged implying that the inflation rate will eventually gravitate to that of the anchor country. There are also other variants of exchange rate targeting. For example, a band can be specified for the nominal exchange rate. The rate is allowed to float freely within this band and the central bank intervenes when there are deviations from the band. In the case of an exchange rate band, speculative capital flows are restricted due to the increased uncertainties of the exchange rate, and this leads to an increase of the monetary policy’s autonomy. As an alternative, some countries have adopted a crawling peg in which the targeted nominal rate is shifted by being devalued in a controlled fashion by less than the inflation differential in the relevant period. Another modification of exchange rate targeting is called the currency board under which the domestic currency is issued only against growth in foreign exchange reserves and in a fixed ratio.
Exchange rate targeting also avoids the time-inconsistency problem by providing an automatic rule for the conduct of monetary policy. When there is a possibility of depreciation of domestic currency, tight monetary policy will be implemented. When there is a tendency for the domestic currency to appreciate, loose monetary policy will be implemented. Third, the nominal anchor of an exchange rate target is understandable to the public due to its simplicity and clarity. Finally, exchange rate targeting that results in a fixed exchange rate regime may lead to economic and political integration as in the case of Exchange Rate Mechanism (ERM), which was in place in the EMU states before the introduction of the euro.
18
First, they prevent the central bank from using monetary policy to respond to domestic shocks (Petursson, 2000). With liberalized capital flows, an exchange rate target causes domestic interest rates to be closely related to those of the anchor country. Therefore, the targeting country becomes unable to use monetary policy to respond to domestic shocks, which are independent of those occurring in the anchor country. In addition, shocks occurring in the anchor country are directly transferred into the target country since changes in interest rates in the anchor country bring about a corresponding change in interest rates in the targeting country (Mishkin, 1999).
Second, exchange rate targeting brings about financial fragility in developing countries if the exchange rate target fails. Financial fragility is a situation in which very small shocks can hit the economy over the edge into a full-blown crisis. Due to the uncertainty about the future value of the domestic currency, it is much easier for many nonfinancial institutions, banks and governments in those countries to issue debt in terms of foreign currency. Exchange rate targeting may further encourage this tendency. In that case, when there is a devaluation of the domestic currency, the debt burden of domestic firms rises because assets are denominated in terms of the domestic currency and there is no simultaneous rise in the value of firms’ assets. As a result, devaluation leads to a deterioration of the firms’ balance sheets, which causes a decline in economic growth (Mishkin, 1999).
19
Another disadvantage of an exchange rate target is that it may weaken the accountability of policymakers, especially in developing countries, because it removes an important signal that can help keeping monetary policy from becoming too expansionary. In many of these countries, the daily fluctuations of the exchange rate can provide an early warning signal when monetary policy is overly expansionary. Therefore, the foreign exchange market can prevent policy from being too expansionary and fear of exchange rate depreciations can make overly expansionary monetary policy less likely. Targeting the exchange rate eliminates this early warning signal and enables the central banks to pursue overly expansionary policies (Mishkin, 1999).
Although currency board application is problematic, it is often argued that it may be successful in developing countries, since strong commitment to fix the exchange rate contributes to increase the credibility of the central banks (Kadioglu, et al, 2000).
3.1.2 Monetary targeting
Monetary targeting is based on the fact that, in the long term, the price level is influenced by money supply growth. The primary aim of a monetary targeting policy is to ensure an appropriate growth rate of the chosen monetary aggregate. This monetary policy framework is associated with a number of advantages:
20
into the time-inconsistency trap because it is capable of promoting almost immediate accountability for monetary policy to keep inflation low (Mishkin, 1999).
Policy management through money targeting is suitable for an economy with a stable, reliable and predictable link between the targeted monetary aggregate and inflation. For instance, Orden and Fisher (1993) find evidence of existence of stable relationships among money, prices, and output under financial deregulation in Australia and New Zealand. However, due to financial innovations and liberalized capital flows, the stability of this link is decreasing in many countries. By the early 1980s, it became obvious that the relationship between monetary aggregates and inflation and nominal income had broken down in countries such as the United States and the United Kingdom. These countries abandoned monetary targeting (Kadioglu, et al, 2000).
3.2 Inflation targeting in developing countries
Inflation targeting has served as a monetary policy framework in many advanced as well as in the emerging economies. It has proved to be quite useful since it has improved policy transparency and accountability of monetary policy in many countries like Australia, Brazil, Canada, Chile, Columbia, Israel and others (Van der Merwe, 2004). The applicability of this policy framework in many developing countries started to be questionable, despite its significant successes. This was due to the failures in most of these countries to meet the main prerequisites of inflation targeting. These prerequisites are identified as central bank independence, including lack of fiscal dominance and commitment to another nominal anchor, and a stable relationship between monetary policy instruments and inflation rate.
3.2.1 Problems with independent monetary policy in developing countries4
Some of the developing countries do not meet the preconditions for inflation targeting. For instance, in the countries where the annual inflation is above 30-40 percent for a number of years, the nominal variables will tend to show a high degree of inertia and asynchronization, and the monetary policy will be largely accommodative. In these cases,
21
the effects of monetary policy on the inflation rate will be short-lived and unpredictable. The policymaker is required to implement a stabilization program, which aims at decreasing the inflation rate by diminishing the role of the central bank in financing the government deficits, to use nominal anchors consistent with inflation objectives to shape the inflationary expectations.
In many of the developing countries, it is difficult to assess whether or not the prerequisites of inflation targeting are met. Fiscal dominance does not always bring about high inflation rates. The extent to which monetary authorities accommodate other nominal anchors and shocks becomes obvious when the inflation rate is very high and is affected by many country-specific factors or when the adoption of the exchange rate regimes such as managed floats and crawling bands renders the monetary authorities discretionary in ranking their external and domestic objectives in a less-than-fully transparent manner.
Central bank independence is difficult to apply in developing countries because it is largely hindered by the presence of three factors: heavy reliance on seigniorage, shallow capital markets, and fragile banking systems.
The reliance on seigniorage that is the most common indicator of fiscal dominance is heavier in developing countries than in advanced countries due to the unstable sources of tax revenue, weak tax collection procedures, and political instability. Moreover, it is suggested that in these countries this source of revenue is abused during the times of crisis.
22
financial system may prevent the government from issuing domestic debt to finance transitory revenue shortfalls, leaving seigniorage and other types of financial repression as the only choices.
Fragile banking systems are also results of long financial repression periods. Especially after financial sector reforms, the banking sector has an independent effect on the conduct of monetary policies in developing countries. So, the conflict between the objectives of attaining price stability and preserving banking sector profitability becomes important within this context. As a result, the monetary authorities are required to rank clearly policy objectives in the early stages of financial liberalization.
In a nutshell, it can be suggested that an independent monetary policy in most of the developing countries is constrained by the existence of fiscal dominance and a poor financial infrastructure. However, the constraints on independent monetary policy in the 1990s became less severe in some of the high-middle income developing countries. For these countries, these constraints are imposed by the unwillingness of monetary authorities to give priority to inflation reduction and their inability to convey the policy objectives to the public in a credible and transparent way.
3.2.2 Conflicts among policy objectives5
In those developing countries where there are well-developed financial markets, low inflation rates and no signs of fiscal dominance, conducting an independent monetary policy depends on the exchange rate regime and on the extent of capital mobility. Due to the increase in the access to international capital markets, and subsequent rise in capital flows, the fixed exchange rate regime has become less popular. Consequently, the switch to the more flexible exchange rate regimes let the monetary authorities assign much lower weight to exchange rate objectives. Furthermore, the stabilization process and financial reform in these countries since the mid-1980s have lead to money demand instability hence decline in the informational content of monetary aggregates.
23
Due to these developments, the conduct and evaluation of monetary policy in these countries has become quiet challenging. There is still no common agreement whether the primary objective of the monetary policy should be the control of inflation or to get some type of balance between the competing objectives of external competitiveness and inflation reduction on a period-by-period basis.
Another difficulty arises from the absence of a coherent analytical framework for assessing the influences of monetary policy and forecasting inflation in these countries. This weakens both the central banks’ ability to formulate monetary policy and the external observers’ ability to assess monetary developments. Empirical studies for these countries mainly rely on three kinds of models. The first one is the monetarist model where the inflation is determined by disequilibria in the money market. The second one is the fiscalist model, which considers the budget deficits as an independent driving force of inflation. The third one is the Scandinavian model which links the inflation to wage pressures stemming from imported inflation and exchange rate changes. However, none of these models provides a useful tool for testing the links between the monetary policy instruments and monetary policy targets.
It is not easy to satisfy the preconditions of inflation targeting if nominal or real exchange rate stability is also an implicit objective of monetary authorities or if the understanding of the empirical links between the monetary policy instruments and targets is not well developed. Briefly, whenever monetary authorities have both an inflation target and other objectives at the same time, and the central bank does not have the means to inform the public about its primary objectives and its operating procedures in a credible and transparent way, there will always be some degree of tension between the inflation target and other policy objectives. In these cases, implementation of the inflation-targeting framework will not be feasible.
3.3 Literature review
24
inflation targeting in developing countries, conditions for successful inflation targeting, the reasons why these countries switch to inflation targeting regime, and to what extent these conditions are satisfied in developing countries (Masson, et al, 1997). Others evaluate inflation targeting as a monetary policy strategy in particular countries by offering empirical models to assess the feasibility of inflation targeting in these countries (Hoffmaister, 1999; Christoffersen, et al, 2001; Nelson-Douglas, 2004; etc).
Some studies analyzed the relevance of inflation targeting regime for developing countries. Masson, et al, (1997) argue that in most of the developing countries the requirements of inflation targeting are absent due to seigniorage being an important source of financing or due to the lack of consensus on low inflation as a primary objective. This is consistent with their main conclusion that the main obstacle of adoption of inflation targeting in developing countries does not lie so much on the countries’ inability to conduct independent monetary policy but on their tendency to overburden monetary policy with multiple and conflicting objectives. Out of a total of about 150 developing and transition economies, five countries were identified as satisfying the conditions for inflation targeting: Chile, Columbia, Indonesia, Mexico and Philippines.
Kadioglu, et al, (2000) analyze the general aspects of inflation targeting regime in developed countries and study the scope for inflation targeting in the developing countries by giving some examples of country experiences such as Mexico, Chile and Brazil. Their findings show that the preconditions for inflation targeting have not been satisfied by most developing countries and most of them do not have powerful models that explain the dynamics of the economy very well and give successful inflation forecasts. Despite these findings for most developing countries, observations from individual country experiences such as Mexico, Chile and Brazil revealed that adoption of inflation targeting has been relatively successful in spite of the countries’ inability to satisfy the prerequisites for inflation targeting.
25
capital mobility in the 1990s and of monetary targeting due to financial innovations, and the desire of some of the transition economies to join the European Monetary Union (EMU), which requires a clear target for disinflation.
Mishkin (2000) explains what inflation targeting involves for emerging market economies by discussing the advantages and disadvantages of this regime and making a reference to Chile’s experience of inflation targeting. He claims that inflation targeting may not be appropriate for many emerging countries because weak central bank accountability is a serious problem, which results from long lags from monetary policy instruments to the inflation outcome, and also, because financial instability caused by flexible exchange rate required by inflation targeting is a relevant fact for these countries. Moreover, he reveals that fiscal dominance and a high degree of dollarization, which may create severe problems for inflation targeting regime, are common features of emerging market economies.
There are many studies that judge whether or not inflation targeting is feasible in particular countries by offering theoretical or empirical evidence. Christoffersen, et al. (2001) claim that Poland appears to be ready for inflation targeting. They too analyze the statistical linkages between monetary policy instruments and inflation, and also between leading indicators of inflation and inflation itself by performing Granger causality tests. They observe that there are significant relationships between CPI and various leading indicators of inflation. They also reveal that although there is a predictable linkage between the exchange rate and inflation measures, the relationships between the changes in the short-term interest rates and changes in inflation are weak. However, they argue that as the Polish economy matures and stabilization is completed, the relationship between the policy interest rates and inflation will be more regular.
26
as policy variables, and also includes industrial production (IP) and the consumer price index (CPI) as output and price variables. In addition they include the German industrial output (GER_PROD) as a proxy for activity in Poland’s trading partners in order to separate domestic shocks from foreign shocks. They examine the effects of an exchange rate shock and interest rate shock on the price level. Their results show that although the exchange rate seems to be effective with respect to output and prices, the direct linkages between the interest rate and inflation do not appear to be very strong.
Hoffmaister (1999) makes an empirical exploration in order to assess the predictability of inflation in Korea. He uses the VAR methodology to calculate impulse responses to exogenous monetary policy. He examines the impulse responses to a negative M2 shock of inflation, output, real interest rate, real exchange rate, and capital flows. He finds that inflation in Korea is as predictable as it was in inflation targeting countries prior to their adoption of inflation targeting and concludes that the empirical evidence supports the feasibility of inflation targeting in Korea.
Nelson-Douglas (2004) explores the applicability of IT in Jamaica as a framework for monetary policy using the VAR system. The outcomes of his study show that the Bank of Jamaica’s (BOJ) instrumental independence and financial market depth is comparable to other developing countries that have implemented IT with success. The only remaining requirement for a successful IT regime for Jamaica is the reform of the legal framework governing the central bank in an effort to limit or prohibit government financing. He thus recommended the need for full central bank independence if IT is the desired monetary policy framework.
4 South African economic history and central bank independence
27
4.1 A brief history of the South African economy
South Africa has one of the most unique histories of all the countries in the world, characterized by years under Apartheid rule where a major part of the country’s population was marginalized. This history plays a major role in South Africa’s current economic situation, which can be characterized by high levels of inflation and a very high level of unemployment, both of which have persisted for the past several years.
Table 1: South African main Economic indicators
Year Inflation rate CPIX Real GDP (R bn) Real GDP Unemployment Exchange rate 1987 16.2% - 174.7 2.1% - 2.04 1988 12.8% - 209.6 4.2% - 2.27 1989 14.7% - 251.7 2.4% - 2.62 1990 14.3% - 289.8 -0.3% - 2.59 1991 15.3% - 332.1 -1% - 2.76 1992 14.0% - 372.2 -2.1% - 2.85 1993 9.7% - 426.1 1.2% - 3.27 1994 8.9% - 482.2 3.2% - 3.55 1995 8.7% - 548.1 3.1% 16.5% 3.63 1996 7.3% - 618 4.3% 20.3% 4.30 1997 8.6% - 685.7 2.6% 22% 4.61 1998 6.9% 7.0% 742.4 0.5% 26.1% 5.53 1999 5.3% 6.9% 813.6 2.4% 23.3% 6.11 2000 5.3% 7.8% 922.1 4.2% 25.4% 6.94 2001 5.7% 6.6% 1020.1 2.7% 27.9% 8.61 2002 9.1% 9.3% 1168.7 3.7% 30.1% 10.54 2003 6.0% 6.8% 1260.6 3.1% 29.6% 7.56 2004 1.4% 4.3% 1398.1 4.8% 27.1% 6.46 2005 3.4% 3.9% 1539 5.1% 26.6% 6.36 2006 4.6% 4.6% 1727.5 5% 25.6% 6.77
Source: International Monetary Fund: International Statistics and Statistics South Africa
Moreover, in recent years, South Africa has seen relatively high rates of inflation coupled with a depreciated currency. This has been coupled with growth rates in GDP and an improvement in the current account. South Africa’s exchange rate depreciated greatly between 1998 and 2002 (Table1).
28
facing the country6. The Reconstruction and Development Programme (RDP) focused primarily on job creation through public works programs such as water supply, sewage works, and road construction, while also attempting to improve housing availability, educational attainment, and other socio-economic indicators. The Growth, Employment and Redistribution Programme (GEAR), established in 1996, however, primarily concentrated on reducing government deficits and inflation while improving the rate of economic growth through privatization. This program was introduced in 1996 in response to the rand crisis.
In the face of major currency speculation and internal governmental division, the rand depreciated rapidly in 1996 and South Africa experienced an exchange rate crisis (Diamond, et al, 2003). The South African Reserve Bank (SARB) responded by selling off massive amount of its foreign reserves possibly because it believed that this depreciation was a temporary reaction to rumours of divisions within the government. South Africa was again faced with a currency crisis following contagion effects from the
Asian Crisis as investors believed that all emerging markets had a similar risk profileand decided against holding the rand. These crises were followed by increases in therate of depreciation, and then a gradual reduction in that increase.
In the February, 2000 Budget speech, the Minister of Finance announced a policy of IT, helping to bring consumer inflation, which had been running in the double digits for over 20 years, under control. Inflation fell from 6.9% in 1998 to less than 6.0% in 2000. The target was set to keep CPIX between 3% and 6% average per annum. Although initially successful, the rand's rapid depreciation in late 2001 led to greater inflationary pressure and the South African Reserve Bank missed the target during the course of 2002, with inflation coming in at an average of 9.3% for the year. Since September 2003, however, the CPIX inflation rate has remained consistently within the target range. The average annual rates of CPIX are shown in Table 1 above.
29
Success in keeping inflation at the desired level gave the government room to drastically bring down interest rates. During 2003 alone interest rates were cut by 550 basis points, while between 2002 and 2006 interest rates were cut by a total 650 basis points. The cut in interest rates increased consumer spending, the construction sector boom and the sale of new vehicles reach record levels. This in turn generated much needed GDP growth.
South Africa’s reserve bank also adheres to a floating exchange rate regime, but that policy is more properly referred to as a managed float. While they have no particular exchange rate targets, the bank intervened recently to reduce foreign currency holdings in the market. This intervention was primarily put forward to maintain inflation targets through open market operations. It is possible that these interventions occurred to work through the “signalling channel,” in which the bank demonstrates its desire to maintain the exchange rate to encourage investors to do the same. Despite these interventions, the exchange rate in South Africa has fluctuated since the currency crisis of 1996.
Given the above historical background of South African economy it is of interest to look into what led to the persistently high level of inflation in South Africa over several years. The main contributors are high food prices, increasing energy prices, and increases in the price of domestic services such as medical and education services and water rates. The South African government stated that the policy of maintaining price stability has not been incredibly successful in maintaining its own targets. This is why the South African government had to find an alternative way to bring down an inflation rate.
4.2 Rationale for adopting inflation targeting in South Africa
30
a high of 9.3 per cent before falling to 3.7 per cent, within the target range for the year to August 2004. The reduction in inflation to below 10 per cent was due in part to the adoption of money supply targets introduced in the mid-1980s to anchor monetary policy (Stals, 1999)7.
Given that the SARB was successful in bringing down inflation under an informal inflation targeting regime, Van der Merwe (2004) raises the question as to why the SARB then moved to formal inflation targeting. He gives four reasons for the SARB’s decision.
First, the informal inflation targeting framework was not all that clear in conveying reasons for the monetary policy stance adopted by the authorities. This was partly because of the breakdown in the relationship between money supply and inflation. The frequent above-guideline money supply figures did not always result in a rise in short-term interest rates because the SARB viewed this as a consequence of structural changes in the economy. Mboweni (1999)8 notes that he conveyed to the ordinary general meeting of shareholders on 24 August 1999 that it would be advisable to move away from the “eclectic” or informal inflation targeting monetary policy framework to formal inflation targeting. The reason was that the eclectic framework had at times created uncertainties about SARB decisions and actions which were “perceived as being in conflict with the stated guidelines for the growth in money supply and bank credit extension.”
The second reason Van der Merwe puts forward is that inflation targeting is more successful than other monetary policy frameworks at improving the co-ordination between monetary policy and other economic policies. It formally defines the co-ordinated effort necessary to implement the framework in order to achieve the broader objective of sustainable economic growth and employment creation.
Third, inflation targeting imposes discipline on monetary policy and increases central bank’s accountability. It does so by setting targets which have to be met. The framework
31
fulfils this criterion by detailing how deviations from the target will be addressed, chiefly in the form of escape clauses. In South Africa’s case, an escape clause detailed the circumstances (exogenous shocks) under which a deviation from the target would be allowed. These included an oil price shock, a drought, natural disaster or financial contagion affecting the currency (SARB, 2003).
According to Mboweni (1999), in order for South Africa to adopt the inflation targeting framework, a number of preconditions had to be met. The central bank should be free to pursue inflation targeting and to use any instrument to achieve this objective, despite the fact that the target is set jointly by the government and the central bank. This freedom to choose and set the policy instruments is referred to as instrument independence. However, given that the targets are set jointly with government, the SARB does not have goal independence. Further, there needs to be a commitment by all authorities to the objective of price stability and a harmonisation of monetary policy with other policies. It is debatable whether, in the initial years, there were coordinated policies. On many occasions Mboweni mentioned publicly of the high increase of administered prices and public-sector wage settlements. However, in 2004 the government announced that it would look into administered prices, showing, perhaps, a greater commitment to the goal of inflation targeting.
It is equally important that there should be well-developed financial markets to ensure that the financial markets react quickly to the central bank’s application of the instruments. On this point South Africa does well. The country is known to be an emerging-market economy in that it has a well-developed banking system and financial markets.
32
a feeling that the SARB wanted more time to test its forecasting and modelling procedures.
4.3 Central bank independence in South Africa
As noted in section 2, one of the important prerequisites for inflation targeting is that the central bank should have a considerable degree of independence, and in particular freedom to set policy instruments (mainly short-term interest rates), and freedom from budgetary demands that allow the authorities to gear monetary policy towards achieving the inflation target. To adhere to this prerequisite, the monetary policy should not be constrained by fiscal considerations and reliance on seigniorage or monetizing of fiscal deficits should be low to non-existent. Van den Heever (2001) notes that, the South African Reserve Bank has the necessary degree of independence granted under the Constitution of the Republic of South Africa and the SARB Act.
By the early 1990s, when South Africa’s Constitution was being drafted, central bank independence had already become a fairly widely accepted element of sound central banking in the international debate. Both the interim and the final Constitution of the Republic of South Africa therefore provide for an independent central bank. Section 224(2) of the Constitution states very clearly: “The South African Reserve Bank, in pursuit of its primary objective must perform its functions independently and without fear, favour or prejudice, but there must be regular consultation between the Bank and the Cabinet member responsible for national financial matters’’.
33
A second condition for greater autonomy is transparency. The South African Reserve Bank also fulfils this second condition for independence. Government and the public are provided with a stream of information on the monetary policy stance. The governor and some staffs of the SARB regularly appear before the Parliamentary Portfolio Committee on Finance. Comprehensive statements following each meeting of the Bank’s Monetary Policy Committee are also issued. The Reserve Bank’s Quarterly Bulletin, Annual Economic Report, and Governor’s Address at the annual meeting of shareholders provide comprehensive analysis of macroeconomic events.
The third condition for central bank autonomy is the creation of an efficient institutional framework within which decisions on monetary policy and on its implementation can be made, without undue interference by political functionaries. This involves decisions regarding: (i) functional independence, which means the right to decide on all matters regarding monetary policy and price stability; (ii) personnel independence, which covers the selection and appointment of Board members with a high professional competence and without an obligation to yield to political and other pressures; (iii) instrumental independence, which means control over the instruments that affect the inflation process, including in particular the prevention of any direct financing of government deficits; and (iv) financial independence, which requires the central bank to have access to adequate financial resources of its own and full control over its own budget.
The institutional framework in South Africa in the form of the South African Reserve Bank Act, Act No 90 of 1989, allows the Reserve Bank a great degree of autonomy in its operations. The Reserve Bank’s functional independence in monetary and related policies is clearly stated in Sections 10 and 35 of the South African Reserve Bank Act. Section 35 empowers the Board of the Bank to make rules for the good government of the Bank and the conduct of its business.
34
other Board members of the Reserve Bank. This section clearly precludes any person actively involved in politics, non-residents or officials of private banks from becoming a member of the Board of the Bank. Seven of the directors, including the Governor and three deputy governors, are appointed by the President of the Republic, and the other seven are elected by the shareholders. By means of these appointments the government, of course, does, have an effective say in the policies of the Bank.
The governors, however, after their appointment, normally operate independently without fear or favour. Certain conditions are given in the Regulations under the Reserve Bank Act which will disqualify a person for remaining a Board member, such as conviction for theft, fraud, forgery or perjury.
The instrumental independence of the Reserve Bank is also clearly spelled out in the Act and the Reserve Bank is precluded in Section 13(f) from making excessive direct purchases of government stock. This section states that the Bank may not hold in stocks of the Government of the Republic which have been acquired directly from the Treasury by subscription to new issues, the conversion of existing issues or otherwise, a sum exceeding its paid-up capital and reserve fund plus one third of its liabilities to the public in the Republic. This section therefore restricts the direct financing of government deficits.
At the same time, the Act also allows the Reserve Bank to provide unsecured loans and advances to government and companies in which it has acquired shares. Although there are limits on government borrowing, serious consideration in the revision of the Reserve Bank Act should be given to removing this potentially dangerous provision. The South African Reserve Bank is also financially independent from the government because of its adequate financial resources and full control over its own budget.
35
bank’s monetary policy as to protect the value of the currency (of the Republic) in the interest of balanced and sustainable growth. The SARB has set single policy objective for which it carries the responsibility and has independence to pursue.
The governor and the deputy governors of the SARB hold office for a (renewable) period of 5 years. They are appointed by the president after consultation with the Minister of Finance and the Board. In this regard therefore there is no question of full independence for the bank as the appointment of senior officials at the SARB resides with the government. In addition, since the government of South Africa appoints seven of the 14 directors of the SARB it has a clear and effective say on the top structure of the SARB. This does not ensure a fixed guarantee for independence though allowed to operate in an independent manner after their appointment.
Other weakness that limits the SARB’s independence is that the conflict between the bank and the government rarely come into the open, and there is no legal provision in either the constitution or the SARB Act prescribing such conflicts to be made transparent. In South Africa there is a perception that in the case of continuous conflict and dispute the governor and the president of the country, the former will have to resign.
For the case of economic independence the SARB can be viewed to have fully instrumental independence. It has fully discretion in this regard to change its repo rate, but does consult the government. The only problem comes with regard to extension of credits to the government. The SARB because of its function as the banker to the government can provide credit to the government. Granting financial assistance to the government diminishes the accountability and independence of the bank.
36
inflation forecasting framework of the South African Reserve Bank and further experience with the operational aspects of the repurchase system are needed before inflation targeting can be implemented.
Woglom (2000) provides empirical evidence to judge whether South Africa is a good candidate for inflation targeting. He uses vector autoregressions (VARs) to analyze the dynamic interaction of the variables of interest like a monetary policy instruments, the price level, the real GDP, and the nominal exchange rate. He makes comparisons between South Africa and pretarget periods of New Zealand and Canada and concludes that South Africa is not a good candidate for CPI inflation targeting due to the loss in benefits provided by a fully flexible exchange rate. His study also provides further evidence that the exchange rate in South Africa is very volatile and plays an important role of determining the CPI and thus provides an important automatic stabilization of IS shocks. Adoption of CPI inflation targeting thus causes the exchange rate to be less flexible in response to external shocks and partially removes some of the stabilization advantages of completely flexible exchange rates.
37
4.4 Having a sole target
One of the requirements for an inflation targeting is the absence of another targeted nominal variable within the system. However some other objectives might be achievable as long as they are consistent with inflation target. The price stability must be given a priority whenever a conflict arises.
Regarding the choice between setting the target as a single point or a range, the SARB chose a three-percentage-point target band. Mboweni9 (1999) argues that a narrower band might be interpreted as indicating a stronger commitment to the inflation target. However, repeated breaches of the target may damage the central bank’s credibility. Further, a single point is more difficult to reach than a band, which allows the central bank more discretion.
The central bank also needs to decide on the level of the target. Mboweni puts forward that high targets may lead the public to believe that the central bank is not serious about containing inflation. It is; therefore, better to lengthen the period over which the target will be reached. The time horizon chosen by the South African authorities was designed to take into consideration the time lags between a change in interest rates and their impact on inflation. Given these factors, the target was set in February 2000 for achievement in the year 2002.
Given the above considerations, when the adoption of inflation targeting was announced on 23 February 2000 the targets specified that the SARB must achieve an annual average for CPIX inflation of between 3 and 6 per cent for the year 2002. The same target was set for achievement in the year 2003 whereafter it was tightened to 3 to 5 per cent for 2004. However, in the October 2002 Medium Term Budget Policy Statement (MTBPS) the band was widened again to 3 to 6 per cent for 2004 and 2005, and in the November 2003 MTBPS the time period for this band was extended to 2006.
9
