Effect of exchange rate volatility on returns of investment portfolios in South Africa

(1)

Effect of exchange rate volatility on returns of investment

portfolios in South Africa

Marco Sgammini

STUDENT NUMBER: 22953817

Dissertation submitted in partial fulfillment of the requirements for the degree

Masters of Commerce (Economics)

in the

SCHOOL OF ECONOMIC SCIENCES at the

North-West University (Vaal Triangle Campus)

Supervisor:

Dr. Paul-Francois Muzindutsi

Co-Supervisor:

Dr. André Mellet

Vanderbijlpark

November 2016

(2)

Persistently persist.

(3)

DECLERATION

I declare that the dissertation entitled “Effect of exchange rate volatility on returns of investment

portfolios in South Africa”, which I hereby submit for the degree Masters of Commerce in

Economics, is my own work and that all the sources obtained have been correctly recorded and acknowledged. This dissertation was not previously submitted to any other institution of higher learning for marks.

(4)

ACKNOWLEDGEMENTS

First and foremost I would like to thank the Almighty for my given talents and ambitions. Onto Him I bestow all the glory.

Secondly, I would like to express my deepest gratitude to everyone who contributed towards the successful completion of this dissertation. The following people and institutions deserve special mention:

 My fiancé, Ruschelle du Plessis. Thank you for motivating me to embark on this journey. Without you, this dissertation would not have been. Your motivation, insight, care and understanding is deeply appreciated. I look up to you. I respect you. I treasure you. Most of all, I love you.

 To my Sgammini family, Lundi, Robbie, Jacques, Tertia and Chanté, thank you for your words of wisdom and support throughout this study.

 Andrew Laurens. “No person was ever honoured for what he received; honour has been the reward for what he gave”. Thank you for the example you set. I pray that one day; I can be half the man you are.

 Lynette Steyn-Laurens, words cannot describe how grateful I am for everything you have done for me. Thank you for your love, guidance, encouragement, and prayers.

 Thank you to the Safe Haven clan. Siblings, your passion for education and knowledge is infectious.

 My devoted study supervisors Dr. Paul-Francois Muzindutsi and Dr. André Mellet. Your excellence in the field is greatly appreciated. Thank you for all your inputs and patience.

 The North West University, thank you for the sound foundation you have provided me with along with the financial support you have given me throughout my postgraduate studies.

 The faculty members of the School of Economic Science. Thank you for your support, kindness, and friendship.

(5)

DECLARATION BY LANGUAGE EDITOR

Ms Linda Scott

English language editing

SATI membership number: 1002595 Tel: 083 654 4156

E-mail: lindascott1984@gmail.com

19 October 2016

To whom it may concern

This is to confirm that I, the undersigned, have language edited the dissertation of

Marco Sgammini

for the degree

Magister Commercii in Economics entitled:

Effect of exchange rate volatility on returns of investment portfolios in South Africa

The responsibility of implementing the recommended language changes rests with the author of the dissertation.

Yours truly,

(6)

ABSTRACT

The movement of globalisation has undoubtedly shaped the modern world of today. Investors have countless options available to them in their investment endeavours, including exploiting investment opportunities in foreign economies. Although international diversification of investment portfolios creates substantial financial gains opportunities, exchange rate risk can play a prominent role when investing in international markets. It is therefore essential to establish what the effect of exchange rate volatility is on investment portfolios.

Numerous studies have identified that there exists a relationship between exchange rates and investments. However, there seems to be no clear consensus on this relationship as a variety of studies produce mixed findings. One clear finding is that this relationship is largely dependent on the country, as each country is exposed to different shocks due to differing policies, political structures, and exposure to the international market and largely due to the exchange rate structure employed in the country.

This study examines how the exchange rate fluctuations influence South African investment portfolios. The empirical objectives are to establish whether exchange rate volatility influences portfolio prices of domestic-based investments relatively to portfolio prices of international diversified investment and to determine whether investment portfolios can be used to diversify exchange rate risk in the South African context.

An autoregressive distributed lag (ARDL) model with a causality analysis is used to analyse 125 monthly observations from April 2006 until August 2016. It is found that, in the majority of the cases, the exchange rate fluctuations have no long-run relationship with the investment portfolios. However, a few instances were identified where a long-run relationship exists. A uni-directional relationship from investment portfolio to exchange rate was noted, in most of the cases. The short-run analysis indicates that domestically diversified investment portfolios are influenced by the exchange rate through a lagged effect. Conversely, international investment portfolios tend to be influenced by the current exchange rates rather than by the lags of the exchange rate. An interesting finding was that domestically diversified investment portfolios tend to weaken as the South African rand depreciates in the short-run. Conversely, internationally diversified investment portfolios tend to strengthen in response to a depreciation of the South African rand in the short-run. It was thus concluded that exchange rate risk can be diversifiable in the South African context.

(7)

Keywords: Exchange rate volatility, exchange rate regime, investment portfolios, international

(8)

LIST OF TABLES

Table 5.1: South African real effective exchange rate descriptive statistics ... 72

Table 5.2: Overall descriptive statistics of domestic investment portfolios ... 73

Table 5.3: Domestically diversified investment portfolio descriptive statistics ... 74

Table 5.4: Overall descriptive statistics of international investment portfolios ... 76

Table 5.5: Internationally diversified investment portfolio descriptive statistics ... 78

Table 5.6: Investment portfolio and exchange rate correlation analysis ... 80

Table 5.7: Correlation between domestically diversified investment portfolios ... 82

Table 5.8: Correlation between internationally diversified investment portfolios ... 83

Table 5.9: Unit root with break test (p-values) ... 85

Table 5.10: Model selection and co-integration test (domestic funds) ... 86

Table 5.11: Model selection and co-integration test (international funds) ... 87

Table 5.12: ECM results for domestic investment portfolios 9 and 10 ... 89

Table 5.13: Short-run regression analysis results (domestic investment portfolios) ... 90

Table 5.14: Short-run regression analysis results (international investment portfolios) ... 93

Table 5.15: Causal relationship between the South African REER and domestic investment portfolios ... 97

Table 5.16: Causal relationship between the South African REER and international investment portfolios ... 98

(14)

LIST OF FIGURES

Figure 2.1: The supply and demand for the US dollar ... 17

Figure 2.2: The effect of increased export quality on the value of the US dollar ... 18

Figure 3.1: Integrated asset allocation ... 41

Figure 3.2: Strategic asset allocation ... 42

Figure 3.3: Tactical and insured asset allocation ... 44

Figure 5.1: Trend of the real effective exchange rate of South Africa ... 67

Figure 5.2: Trend of domestic South African investment portfolios ... 69

(15)

LIST OF ABBREVIATIONS

ADF : Augmented Dickey-Fuller AIC : Akaike information criteria ARDL : Autoregressive Distributed Lag CME : Chicago Mercantile Exchange CNB : Czech National Bank

CUSUM : Cumulative Sum

DF : Dickey-Fuller

ECM : Error Correction Model ECT : Error Correction Term ETF : Exchange Traded Fund EViews : Econometric Views FDI : Foreign Direct Investment

FINRA : Financial Industry Regulatory Authority

GARCH : General Autoregressive Conditional Heteroscedasticity GDP : Gross Domestic Product

ICAI : Institute of Charted Accountants of India IMF : International Monetary Fund

INET BFA : Net Bridge McGregor Bureau for Financial Analysis ITRISA : International Trade Institute of South Africa

MPT : Modern Portfolio Theory NAV : Net Asset Value

PIMCO : Pacific Investment Management Company REER : Real Effective Exchange Rate

SARB : South African Reserve Bank

(16)

SIC : Schwartz Information Criteria

UN : United Nations

US : United States

USD : United States Dollar ZAR : South African Rand

(17)

CHAPTER 1: INTRODUCTION AND BACKGROUND TO STUDY

1.1 INTRODUCTION

Over the past century, one movement that undoubtedly has shaped the modern world of today is globalisation (Selimi, 2012:363). Railo (2000:3) states that, due to the recent globalisation of economies, profound interest has been placed on international diversification of portfolios. In the 1990s when globalisation was on a steep rise, portfolio flows to emerging markets experienced a significant boom according to Schmukler (2004:2). Ray (2013:635) attributed the increased interest in international diversification of portfolios to three main sources. First, an internationally diversified portfolio reduces risk. Secondly, it provides substantial potential for growth and lastly, it creates many different alternatives for investors. Some might argue that this movement of globalisation has led to a degree of integration where most cultures have gone extinct, creating unsurpassed financial opportunities. Although the movement of funds tends to lead to a great deal of financial gains in opportunities, these gains in opportunities are associated with a number of risks (Kose et al., 2007:1) These risks include currency risks, which emerge from exchange rate fluctuations.

The definition of risk can take on many forms, as there seems to be no agreement on the conceptualisation of the term (Macmillan, 2000:19). Moles (2013:16) defined risk as “the chance or probability of a deviation from the anticipated outcome”. Another definition of the term merely reads that risk arises due to the possibility of undesirable results occurring (Macmillan, 2000:18). Based on the definitions it is clear that risks arise primarily due to factors beyond the control of the investor. This further implies that risk arises due to uncertainties faced by investors. Thus, based on the existing definitions of risk, risk can be defined as the situation where investors face a degree of uncertainty due to factors beyond the control of the investor, which can result in undesirable results occurring (Crouhy et al., 2014:5).

Although the movement of globalisation led to an increase in financial opportunities, which were accompanied by an increase in risks faced by investors, investors have not become reluctant to exploit these opportunities. This is mainly due to measures created to manage and reduce risk to ensure some protection for investors’ funds (Moles, 2013:18). Two major measures, which can be used to reduce uncertainty and risk, include the use of investment portfolios - also referred to as funds - and the diversification thereof (Reilly & Brown, 2012:172).

(18)

Investment portfolios generally are structured in such a way that the risk involved is minimised; while optimising the potential return of the portfolio (Reilly & Brown, 2012:21). This implies that one objective of a portfolio is to reduce the risk involved. Markowitz (1952:89), one of the pioneers of portfolio diversification theory, states that in order to reduce risk, a portfolio needs to be diversified in such a way that the assets involved share a low correlation with each other. Investors often diversify their portfolios internationally in an attempt to reduce risk and optimise their potential for return. Portfolios can thus include a percentage of international interest, which generally is done as potential for higher returns is created (Abid et al., 2014:46). The higher level of international exposure, however, does result in greater risk carried by an investor. Exchange rate risk in particular tends to increase substantially when a portfolio is diversified with higher international interest than domestic interest (Ziobrowski & Ziobrowski, 1995:65; Shehu, 2011:1305).

Portfolio diversification can assist investors in reducing their risks whilst optimising their potential return (Reilly & Brown, 2012:21). Vermeulen (2011) tested the gains of internationally diversified portfolios before and during the global financial crisis and found that internationally diversified portfolios provided large gains during the global financial crises. In an attempt to establish all risks associated with pension funds, Shehu (2011) argued on a theoretical basis that one of the factors that affect portfolio diversification is exchange rates. Shehu (2011:1305) and Doukas et al. (2003:291) argue that the degree to which pension funds are influenced by exchange rate risk are dependent on two variables. First, the degree of international exposure the fund receives and secondly whether firms incorporated in specific funds use foreign currency hedging instruments. Shehu (2011:1305) further states that funds with low international involvement would not be influenced by exchange rate risk as much as funds that have high international involvement. However, some pension funds tend to incorporate multinational firms and this gives rise to exchange rate risk in times of high exchange rate volatility (Shehu, 2011:1306). Abidin et al. (2004:51) also found that domestic-based portfolios and internationally diversified portfolios performed differently, when risk and some economic crises are considered. Ziobrowski and Ziobrowski (1995:65) also analysed the effect of exchange rate volatility on internationally diversified portfolios and found that portfolios, which keep the majority of their funds local, tended to be less prone to exchange rate volatility than funds that are highly internationally diversified. Backlund (2011:16) analysed currency risk in order to find means to manage risk created by exchange rate volatility. In doing so it was stated that organisations are

(19)

exposed to currency risk, regardless whether the organisation trades internationally or only domestically. This implies that investment portfolios with solely domestic-based interests also face currency risk.

Although numerous studies have been done to establish the motive for international diversification of portfolios, there seems to be mixed findings on how the effect of exchange rate volatility affect domestic-based and internationally diversified investment portfolios. Additionally, the effect of exchange rate volatility on domestic-based and internationally diversified portfolios has not been tested in the South African context.

1.2 PROBLEM STATEMENT

The annual report on exchange arrangements and exchange restrictions of 2014 indicated that 36 of the 191 International Monetary Fund (IMF) member countries employ a floating exchange rate regime (IMF, 2014a:5). It was stated further that 29 countries employ a free-floating exchange rate regime. South Africa shifted towards the use of an informal free-floating exchange rate regime in 1995 (National Treasury, 2009:1; Mtonga, 2011:3; Muzindutsi, 2011:58), which has resulted in more exposure to exchange rate fluctuations. The figures published by the IMF (2014a) indicates that more emerging economies around the world are shifting towards the use of a free-floating exchange rate regime. Yagci (2001:10) states that countries that employ any form of a floating exchange rate regime tend to produce high short-term exchange rate fluctuations. Thus, it is essential to establish how these fluctuations influence financial markets and investment portfolios in particular as a high number of countries employ some form of a floating exchange rate regime.

In South Africa, the adoption of a free-floating exchange rate system has increased the volatility of the local currency (Mtonga, 2011:6). The fluctuation of the South African rand is not a recent development but has been pronounced explicitly between 2009 and 2013 (De Lange, 2013). During this period, the South African rand/United States dollar (ZAR/USD) exchange rate fluctuated between as low as R6,60 per dollar and as high as R10,60 per dollar, which is close to a 61 percent fluctuation as stated by De Lange (2013). These fluctuations have also been highly pronounced in the last quarter of 2015 and the first quarter of 2016. Within this period, one dollar cost close to R13,10 in October 2015 and climbed as high as R16,85 (28,62 percent

(20)

increase) in January 2016 (Moneyweb, 2016). This greatly signifies the exchange rate risk faced by South African investors.

A number of studies have been done to establish the effect of the exchange rate on bond and stock markets around the world (Jarion, 1991; Zhu, 1998; Farooq & Keung, 2004; Mun, 2008; Muzindutsi, 2011; Lawal & Ijirshar, 2013; Mlambo, 2013). Due to the higher integration of the world, several studies placed emphasis on portfolio diversification (Markowitz, 1952; Aranyi, 1967; Eun & Resnick, 1988; Ziobrowski & Ziobrowski, 1995; Abidin et al., 2004; Shehu, 2011; Abid et al., 2014; Caporale et al., 2015) in order to reduce risks, some with specific focus on exchange rate risk. The studies focussing on diversification of portfolios in order to reduce exchange rate risk produced mixed findings with some (Jarion, 1991; Ziobrowski & Ziobrowski, 1995; Shehu, 2011; Mlambo, 2013; Caporale et al., 2015) finding that exchange rate risk can be diversified; while others (Choi & Rajan, 1997; Eun & Resnick, 1988) found that exchange rate risk is not diversifiable. In the South African context, the link between stock and bond market has been analysed but there seems to be a lack of research on the effect of exchange rate volatility on investment portfolios. More specifically, it is not clear whether South African investment portfolios are influenced by exchange rate volatility or whether domestic-based or internationally diversified funds yield the same return during the period of exchange rate volatility. Thus, this study aims to establish how the increased exchange rate volatility affects both domestic-based and internationally diversified investment portfolios in South Africa.

1.3 OBJECTIVES OF THE STUDY

The following objectives were formulated and identified for this study:

1.3.1 Primary objectives

The primary objective of this study is to establish how exchange rate volatility influences the return of investment portfolios in South Africa.

1.3.2 Theoretical objectives

The following theoretical objectives of the study were formulised in line with the primary objective:

(21)

 Track the development and transformation of the exchange rate regime in South Africa;

 Discuss theoretical concepts of investment portfolios;

 Establish a theoretical framework for portfolio diversification in line with exchange rate risk; and

 Review empirical studies on exchange rate risk and portfolio investment.

1.3.3 Empirical objectives

The following empirical objectives were formed in order to achieve the primary objectives of the study:

 Determine the degree of South African exchange rate volatility during the period of free-floating exchange rate;

 Empirically review the trend of different investment portfolios in South Africa;

 Establish whether exchange rate volatility influences portfolio prices of domestic-based investments;

 Establish whether exchange rate volatility influences portfolio prices of international diversified investment; and

 Determine whether investment portfolios can be used to diversify exchange rate risk in the South African context.

1.4 RESEARCH DESIGN AND METHODOLOGY

The study will comprise a literature review and an empirical study. Quantitative research, using secondary data will be used for the empirical portion of the study.

1.4.1 Literature review

Secondary data sources will be used to formulate the literature review of the study. These secondary sources will include journal articles, relevant textbooks, newspaper articles and Internet sources.

1.4.2 Empirical study

(22)

1.4.2.1 Target population and sampling frames

Primarily due to data availability, the sampling frame consists of all investment portfolios in South Africa of which 32 portfolios, 16 investment portfolios, which are designed with solely domestic interest in the South African economy and 16 investment portfolios, which encompass a certain degree of international interest. The two sets of investment portfolios that were used for this study can thus be classified into domestic investment portfolios and domestic investment portfolios with international interest (international investment portfolios). The selection of the respective investment portfolios was based on the following criteria:

 The fund should be in existence for the whole sample period, April 2006 until August 2016. The reason for the selected timeframe is based on the introduction of the informal free-floating exchange rate regime in 1995. The primary reason for the selected timeframe is due to data availability;

 The fund used to represent domestic investment portfolios should have majority interest (more than 50 percent) in South Africa; and

 The funds used to represent internationally diversified investment portfolios will be selected based upon the degree of international diversification. These funds should have majority interest (more than 50 percent) in countries other than South Africa. Most investment portfolios in South Africa publish a fund fact sheet, which clearly identifies how the funds within the investment portfolio are diversified. These fact sheets were used first, to establish whether the investment portfolio is internationally diversified and secondly, the degree of international diversification was taken into account to ensure all investment portfolios share more or less the same degree of international interest.

1.4.2.2 Data collection method

Accurate and frequent data was needed on the South African real effective exchange rate and prices of the selected South African investment portfolios to meet the objectives of the study. Secondary data consisting of 125 monthly time series data points, from April 2006 until August 2016, was collected. The reason for the selected timeframe is due to the introduction of the informal free-floating exchange rate regime in South Africa in 1995, which led to higher exposure to exchange rate fluctuations. Data was collected from sources, which include the

(23)

South African Reserve Bank (SARB, 2016b), Net Bridge McGregor Bureau for Financial Analysis (INET BFA, 2016), MorningStar (2016) and FundsData Online (2016).

1.4.2.3 Statistical analysis

In order to meet the objectives of the study, an autoregressive distributed lag (ARDL) model was implemented. The model was used to determine the co-integration amongst the variables which indicates whether a long-run relationship exists between the exchange rate and the investment portfolios. A long and short-run analysis was conducted to establish whether the investment portfolios are effected by the exchange rate. Furthermore, a causality test was done to establish whether a uni-directional or bi directional relationship exists between the two variables.

1.5 ETHICAL CONSIDERATIONS

The data needed to complete the study are solely of secondary nature and publicly available from the aforementioned databases. The North West University ethical considerations will be followed to attain ethical clearance in this regard.

1.6 CHAPTER CLASSIFICATION

The study will comprise of the following chapters:

Chapter 1: Introduction and background to study – this chapter provides a short overview of

exchange rate volatility, exchange rate risk and investment portfolios, which is followed by a description of the problem statement, theoretical and empirical objectives, the methodology and an overview of the chapter classification.

Chapter 2: A theoretical framework of exchange rate volatility – this chapter provides a

review of all existing literature of exchange rates with specific focus on the South African economy; predominantly the development of the South African exchange rate regime.

Chapter 3: A theoretical framework of investment portfolios – this chapter provides a brief

overview of investment portfolios. Specific focus is placed on the two major forms of investing − bonds and equities. Attention also is given to the other possible assets, which assist asset allocation in investment portfolios. A brief discussion of diversification follows, where emphasis is placed on domestic and international diversification. This chapter concludes with a review of

(24)

the empirical literature pertaining to the relationship between exchange rates and investment portfolios.

Chapter 4: Research design, data and methodology – this chapter describes the methodology

used in the empirical portion of the study, focusing on the data collection method, the sample size and the data used as well as the statistical analysis that is undertaken.

Chapter 5: Results and findings – this chapter provides a discussion on the volatility of the

South African exchange rate over a timeframe and is followed by an analysis of the impact of the volatility on investment portfolios with solely domestic interest and investment portfolios with a certain degree of international interest. These results and findings are used to determine whether exchange rate risk is diversifiable within South African investment portfolios.

Chapter 6: Summary, conclusion and recommendations – this chapter summarises and

concludes the study based on results and findings. Potential recommendations for future research are also outlined in this chapter.

(25)

CHAPTER 2: A THEORETICAL FRAMEWORK OF EXCHANGE RATE

VOLATILITY

2.1 INTRODUCTION

The aim of this chapter is to fulfil the objective of explaining a vast majority of the literature relating to exchange rates. More specifically, significant emphasis is placed on the development and transformation of the South African exchange rate regime, the volatile nature of exchange rates, as well as the factors that increase exchange rate volatility.

The chapter consists of five separate sections. First, exchange rate theory is discussed where emphasis is placed on the conceptualisation of exchange rates, the determinants of exchange rates and finally a discussion on the three main forms of exchange rate regimes. The second section provides a detailed discussion on the development and transformation of the South African exchange rate regime from 1945 until 2000. The third section emphasises the volatile nature of exchange rates. In this section, attention is given to exchange rates as a potential risk factor for investors, which is followed by a discussion of the factors that heighten exchange rate volatility.

2.2 EXCHANGE RATE THEORY

2.2.1 Conceptualisation of exchange rate

Theory suggests that no country in the world is completely self-sufficient and that companies often engage in trade beyond the borders of the domestic market (International Trade Institute of South Africa) (ITRISA, 2013:7). ITRISA (2013:19) further states that the reasons for the heightened interest in international trade are due to benefits such as expanded market opportunities, cost reduction, economies of scale, spread of risk, more balanced production, extension of product life cycle and improved operational efficiency, which are created by international trade.

Companies within the domestic market that want to exploit the benefits of international trade have to establish how much their domestic currency is worth in the foreign market. This can be done through the use of the exchange rates. Fourie and Burger (2011:130) defined exchange rate as the denotation of one currency expressed in terms of another. The exchange rate will thus indicate how much of one currency can be bought with another. Abel et al. (2008:477) state that

(26)

in order to explain exchange rates, it is important to consider the different forms of exchange rate - in particular that of the nominal and the real exchange rate.

2.2.1.1 Nominal exchange rate

The nominal as well as the real exchange rates can be used to describe the purchasing power of a certain currency (Abel et al., 2008:478). The main difference between these two forms is that the real exchange rate is adjusted in order to take price levels into account (Fourie & Burger, 2011:153). Thus, when considering the nominal exchange rate, inflationary pressures in each respective country is not removed or compensated for. The nominal exchange rate as stated by the Czech National Bank (CNB, 2016) indicates how many units of one currency can be bought with a single unit of the domestic currency. This implies that the nominal exchange rate is merely a ratio, which indicates the value of one currency in terms of another. For example, if the ZAR/USD is equal to R14 ($1 = R14), one USD can be bought with R14. Abel et al. (2008:78) state that the nominal exchange rate is not sufficient when explaining the purchasing power of a currency and hence the real exchange rate has to be considered.

2.2.1.2 Real exchange rate

The real exchange rate is calculated using the nominal exchange rate. The main reason why the real exchange rate has to be considered when measuring the purchasing power of a currency is due to role of differing price levels in different countries (Abel et al., 2008:478). Evrensel (2016) defines the real exchange rate as the comparison between the relative price levels of two different countries’ consumption baskets. Where the nominal exchange rate falls short is that the real exchange rate can be used to determine exactly how much can be bought with a certain currency compared to another (Catão, 2007:46). The nominal exchange rate includes inflation differentials, which could give a flawed indication as inflation rates differ from country to country. The Reserve Bank of Australia (2001:70) indicated Equation 2.1 as the formula to measure the real exchange rate.

(2.1)

In Equation 2.1, P represents the price level in the domestic economy and P’i represents the price

level in the foreign economy, represents the nominal exchange rate between the two countries and is expressed as the amount of foreign currency that can be bought with the

(27)

domestic currency. Thus, as the domestic currency appreciates, the value of will increase. This equation can be applied to determine the real exchange rate between two currencies. Given the inclusion of the price levels in the respective countries, it is clear that inflation is accommodated for as opposed to the case of the nominal exchange rate.

2.2.1.3 Real effective exchange rate

One of the main indicators of competitiveness of a country is the real effective exchange rate, which is a weighted average of the relative prices of a country with its main trade partners (Spilimbergo & Vamvakidis, 2000:3). Catão (2007:47) states that economists and policymakers often are more interested in the real effective exchange rate when determining the overall alignment of a currency. This is due to the real effective exchange rate being an important and effective measure to determine whether a currency is under- or overvalued. Fourie and Burger (2011:153) state that the real effective exchange rate expresses the value of the domestic currency relative to that of important foreign currencies. These foreign currencies usually include the most important trading partners of the domestic country. Fourie and Burger (2011:153) further state that the real effective exchange rate is a weighted average, which supports the definition of Spilimbergo and Vamvakidis (2000:3). As a number of currencies are used to determine the real effective exchange rate, it is less sensitive to currency disturbances within a single country. The real effective exchange rate is thus more stable than the nominal or real exchange rates. Although numerous policymakers would prefer to reason in real terms, Fourie and Burger (2011:154) state that it is often much safer to reason in terms of the nominal exchange rates than in real terms. This would make the analysis much more transparent and easier to follow.

2.2.2 Determinants of exchange rate

A number of studies (De Jager, 2012:7; Parveen et al., 2012:673; Patel et al., 2014:53 & Twarowska & Kakol, 2014:893) suggest a wide variety of fundamental factors that determine the exchange rate of a country. The majority of the factors identified by the studies are highly correspondent with one another and include economic and non-economic factors. Twarowska and Kakol (2014:892) clearly distinguish between the economic and non-economic factors. The economic factors identified includes economic growth, inflation rates, interest rates, balance of payments and capital speculations. Non-economic factors were listed as political risk, natural

(28)

disasters, policy approaches and psychological factors. The research of Parveen et al. (2012:673) argues that economic growth, inflation rates and the balance of payments influence the exchange rate, which corresponds to the arguments of Twarowska and Kakol (2014:892). Patel et al. (2014:53) also identified non-economic factors such as political stability and geopolitical events. It was further stated by Patel et al. (2014:53) that factors such as inflation rates, interest rates, speculation, gross domestic production (GDP), debt of a country and employment data determine the exchange rate.

From the existing literature, it is clear that exchange rates tend to be dependent on a large number of determinants. Within the South African context, De Jager (2012:7) argues that the South African exchange rate is influenced by a number factors within the financial sector, the fiscal sector, the real sector, the international sector as well as through commodity prices and terms of trade. The South African Reserve Bank (SARB, 2016a:3) states that the exchange rate of South Africa is also influenced by inflation, the international economy and natural disasters. These factors which were identified by De Jager (2012:7) and the SARB (2016a:3) highly correspond with the factors identified by Fourie and Burger (2011:156). These factors seem to be the widely accepted determinants of the exchange rate and include the inflow and outflow of funds, transactions, speculation and policy events, inflation differentials, international competitiveness and non-competitiveness and political-economic expectations. These factors are described in detail below.

 The inflow and outflow of funds – The inflow and outflow of funds is determined

largely by the state of the international economy as it entails the supply and demand of foreign exchange (Fourie & Burger, 2011:156). The exchange rate is most likely to appreciate in the event of excess demand for the domestic currency. Likewise, an excess supply of a currency with lower demand would most likely result in a depreciation of the domestic currency, according to Fourie and Burger (2011:156). South Africa is highly dependent on the international economy (Wesso, 2001:59). The SARB (2016a:5) also states that when the economy of China slows down, the South African economy feels the blow. The exchange rate of South Africa – and most developing countries – is thus dependent on the international economy to ensure a higher demand for the local currency that could result in a stronger currency.

(29)

 Transactions, speculations and policy events – The movement of globalisation has

led to countries being highly dependent on one another (Paechlke, 2009:190). The domestic exchange rate thus can also be influenced by events beyond the borders of the country. Fourie and Burger (2011:156) state that the exchange rate of the domestic country can depreciate due to the foreign country’s currency undergoing strong appreciation. This appreciation that is happening in the foreign country’s currency might have no relation to the domestic currency, but influences the domestic exchange rate nonetheless. It was stated further by Fourie and Burger (2011:156) that the balance of payments remains the underlying determinant of the direction in which the exchange rate moves.

 Inflation differentials – Most factors that determine the exchange rate are highly

linked to the international market and hence the balance of payments. Twarowska and Kakol (2014:893) state that as inflation increases the domestic exchange rate should depreciate. This is due to prices in the local market becoming more expensive and thus encourages imports to obtain better prices. Thus, as inflation increases, the balance of payments deficit could rise depending on the degree of price increases.

 International competitiveness and non-competitiveness – Fourie and Burger

(2011:157) state that international competitiveness is highly dependent on factors such as productivity, human resource skills, technological development, input cost tendencies and innovative management. These factors also indirectly link to the balance of payments, as highly competitive companies or countries would be able to have much more exports due to lower prices. Higher exports could result in appreciation of the exchange rate as demand increases. Dahlman (2007:36) identified a few measures to improve international competitiveness. These measures include technological development, strategic use of foreign investment, embracing globalisation and integration and developing state-directed technologies.

 Political-economic expectations – The general agreement seems to be that economic

and non-economic factors determine the exchange rate. Twarowska and Kakol (2014:892) argued that political factors also play a role in exchange rate determination. Patel et al. (2014:54) state that countries experience higher inflow of funds when the political environment is stable, which provides investors with certainty that their

(30)

investment will reap benefits. Fourie and Burger (2011:157) state that when a country is regarded as safe and a secure haven for funds and trading, significant capital inflows can occur. The advantage of providing a stable political environment is that exchange rates could experience a sustained appreciation. The movement of the exchange rate, however, largely is dependent on the balance of payments and the current account in particular.

2.2.3 Types of exchange rate regimes

One of the main objectives of the monetary policy of a country is to ensure price stability (Cecchetti, 2000:43). Over the last few decades countries have adopted and tested numerous exchange rate regimes (systems) in an attempt to ensure stable prices (Muzindutsi, 2011:71), which would attract foreign investments and hence fuel economic activity. Dornbusch et al. (2011:550) state that after the Asian financial crisis in 1997, a great deal of speculation was formed around which exchange rate regime was superior. This speculation is highly understandable given the large variations of exchange rate regimes.

The broad variations or exchange rate regime include hard pegs, soft pegs, intermediate regimes and floating regimes. Each of these separate categories then are divided into sub regimes. Hard pegs include full dollarisation and the use of a currency board. Soft pegs range from crawling pegs to fixed pegs. Intermediate regimes include managed-floating regimes and crawling broadband regimes. The last regime, which is floating exchange rate regimes, includes lightly floating and free-floating regimes (World Bank, 2001:4). Dornbusch et al. (2011:515) identified two broad regimes, which can be used by monetary authorities to intervene in the currency market. These regimes include the fixed and floating exchange rate regimes. The floating exchange rate regime can be managed-floating or free-floating whereas the fixed exchange rate regime can be pegged or fixed in nature (Tembo, 1999:31).

2.2.3.1 Fixed exchange rate

The fixed exchange rate regime was adopted shortly after the Second World War in an attempt to stabilise the global financial system (Muzindutsi, 2011:73). The fixed exchange rate system was based largely on the Bretton Woods system (1946 until 1971), according to Wang (2009:23), and is defined by ITRISA (2013:243) as a system where the exchange rate is fixed at a particular level. Stone et al. (2008:42) state that a fixed exchange rate system is formed when either there is

(31)

a legally mandated use of another country’s currency in the domestic market or when a legal mandate is formed that entails the domestic central bank should keep foreign assets equal to the local currency in circulation. Crouhy et al. (2014:126) state that the use of a fixed exchange rate regime began to break down in the late 1960s due to global economic forces. These economic forces include vast expansion of international trading and inflationary pressures. The fixed exchange rate regime also provided countries with advantages despite the breakdown in the late 1960s. ITRISA (2013:243) states that the fixed exchange rate system promotes foreign investment as investors can predict profits with relative certainty. Inflation also tends to be lower due to the control over prices of imports. These advantages, however, were outweighed by the disadvantages, which resulted in the regime breakdown. Abel et al. (2008:511) state that fixed exchange rate regimes hinder monetary policy flexibility to deal with economic downswings and recessions. In order to defend the fixed value of the currency, monetary authorities are required to keep large stocks of foreign reserves (ITRISA, 2013:243). Fixed exchange rate regimes also create obstacles for the implementation of domestic economic policies. Although use of the fixed exchange rate seems to be largely extinct, ITRISA (2013:243) states that countries such as Saudi Arabia, United Arab Emirates and Pakistan still follow this regime.

2.2.3.2 Managed-floating exchange rate

The main difference between a fixed and free-floating exchange rate regime is that market forces of demand and supply determine the exchange rate in the case of a free-floating regime, whereas supply and demand do not play a role in a fixed exchange rate regime (Abel et al., 2008:511). In the case of a managed-floating exchange rate regime, market forces still influence the exchange rate but only to a certain degree. ITRISA (2013:245) defined a managed-floating exchange rate regime as a situation where the exchange rate is determined largely by forces of demand and supply, but is managed when the exchange rate rises or falls too much. The advantage of a managed-floating exchange rate regime is that monetary authorities are able to smooth out short-term fluctuations. When an adverse effect on economic variables such as inflation, employment and international competitiveness is expected to result due to exchange rate fluctuations, the monetary authority is able to intervene in the currency market (ITRISA, 2013:245). Dornbusch

et al. (2011:550) state that economies with managed-floating exchange rate regime can be

adversely affected in times of high exchange rate fluctuations away from exchange rate equilibrium. This could result in the currency being under- or overvalued and could thus distort

(32)

trade flows and result in a financial crisis. ITRISA (2013:245) states that countries that employ a managed-floating exchange rate regime include Argentine, India, Singapore and Mauritius.

2.2.3.3 Free-floating exchange rate

The annual report on Exchange Arrangements and Exchange Restrictions of 2014 published by the IMF indicated that around 29 countries employ a free-floating exchange rate regime (IMF, 2014a:5). Since 1995, South Africa also employed an informal free-floating exchange rate regime (Muzindutsi, 2011:76) and later, in 2000, shifted to a formal floating regime. A free-floating exchange rate regime can be defined as one in which the price of the currency is determined by forces of demand and supply and the currency price is allowed to move freely with no borders or restrictions (ITRISA, 2013:241). In this case, the monetary authorities do not try to influence the price of the currency. The main disadvantage of the free-floating exchange rate regime is that currencies tend to be highly volatile and often undergo daily fluctuations (Crouhy et al., 2014:126). Due to the high volatility of the currency, investors are faced with a higher degree of uncertainty, which could result in less foreign investments. ITRISA (2013:242) states that when the currency rises too far or falls too low there could be dire consequences for the economy. As opposed to the fixed exchange rate regime, countries who employ a free-floating exchange rate regime do not have to keep large amounts of foreign reserves in order to defend the value of the currency. A free-floating exchange rate regime is also beneficial to the balance of payments of a country as the deficit could be brought back to equilibrium automatically (ITRISA, 2013:242). According to the IMF (2014a:7) countries that employ a free-floating exchange rate regime include Japan, Canada, Australia and the United States (US).

2.2.4 Determination of exchange rates: A supply and demand analysis

The different forms of exchange rate regimes clearly distinguishes between fixed and floating in the fact that forces of demand and supply determine free-floating exchange rates. Abel et al. (2008:487) state that under a floating exchange rate regime, the value of the currency changes constantly as compared to a fixed regime where the exchange rate remains constant for fairly prolonged periods. This section thus applies to floating exchange rate regimes where the exchange rate is determined largely by forces of demand and supply. Figure 2.1 illustrates the market for United States (US) dollars.

(33)

Figure 2.1: The supply and demand for the US dollar

Source: Abel et al. (2008:488)

Figure 2.1 illustrates the demand and supply for dollars. The vertical axis represents the nominal value of the currency in terms of another whereas the horizontal axis represents the quantity of US dollars supplied or demanded. The supply curve (S) represents the number of dollars the market is willing to supply at each given price level. The demand curve (D), on the other hand, represents the number of US dollars the market demands at each give price level. The equilibrium level (E) represents a case where the demand for and supply of the US dollar is equal. Knowing how the market of US dollars – or any flexible exchange rate regime for that matter – is represented, Figure 2.1 can be used to determine how the forces of demand and supply influence the value of a currency. It is now important to consider the determinants of an exchange rate as discussed in Section 2.2.2. Stalstedt (2006:6) argued that although each of the factors discussed tends to influence the exchange profoundly, foreign investing, speculations and, exports and imports are the three main factors, which determine an exchange rate. Figure 2.2 illustrates a situation where the value of the US dollar is influenced by exports.

E

Supply of dollars, S

Demand for dollars, D

Dollars traded Number of US dollars 𝑒𝑛𝑜𝑚1 V al ue o f U S d ol lar s, 𝒆𝒏 𝒐 𝒎

(34)

Figure 2.2: The effect of increased export quality on the value of the US dollar

Source: Abel et al. (2008:490)

In Figure 2.2, an increase in the value of the US dollar is noted. This increase has occurred due to an improvement in export quality. In this case, the demand for US dollars has increased, which is illustrated by the shift of the demand curve from level D1 to level D2. The higher

demand for the US dollar has also resulted in a new equilibrium level E2. At the new equilibrium

level, the market is willing to pay a higher price for US dollars and the market is willing also to supply a higher number of US dollars. The increase in the value of the US dollar is noted on the vertical axis where the nominal value of US dollars increased from level 1nom to 2nom. This

movement indicates that the US dollar has appreciated against another foreign currency due to increased exports from the US.

2.3 DEVELOPMENT AND TRANSFORMATION OF THE SOUTH AFRICAN

EXCHANGE REGIME (1945 UNTIL 2000)

In the modern global economy where countries are highly interdependent, foreign exchange markets of countries need to be managed effectively and efficiently. Van der Merwe and Mollentze (2012:134) state that a country’s foreign exchange market is of utmost importance as

E₁

Supply of dollars, S

Demand for dollars, D 1 Dollars traded Number of US dollars 𝑒1𝑛𝑜𝑚 V al ue o f U S d ol lar s, 𝒆𝒏 𝒐 𝒎 E₂ 𝑒𝑛𝑜𝑚2

Demand for dollars, D

(35)

bad exchange rate management can disrupt financial markets and the effective allocation of resources. Effective and efficient management of the exchange rate, on the other hand, promotes a stable economic environment that provides a sound foundation for healthy international trade, high levels of foreign investment and, ultimately, stable and sustained economic growth.

In the case of South Africa, it has been a relatively hard and long road to establish an effective foreign exchange market. South Africa has moved from a fixed, to managed-floating and eventually to a formal free-floating exchange rate regime in 2000 (Mlambo, 2013:8). There has also been a time when South Africa employed a dual exchange rate system during 1985 to 1995, according to Eun et al. (2012:1). Van der Merwe (1996:1) discusses the development and transformation of the South African exchange rate regime from 1945 until 1995 and states that South Africa has undergone numerous policy adjustments. Political instability – especially during the 1980s – in South Africa led to sanctions being placed on the country, which largely hindered the development of an effectively managed exchange rate regime in South Africa (Van der Merwe & Mollentze, 2012:142). This section will elaborate on the development and transformation of the South African exchange rate regime from around 1945 until 2015. The discussion will be broken down into four phases. The research of Van der Merwe (1996) discussed the development and transformation of the South African exchange regime between five key periods. However, Potgieter (2005), Zhang (2009) and De Villiers (2015) divided their discussions into four key phases. The timeframe of the discussion will roughly be from 1945 until 2000.

2.3.1 Phase 1: 1945 until 1971

Shortly after the Second World War (1939-1945), most countries around the world adopted a fixed exchange rate regime in an attempt to stabilise the global financial market (Muzindutsi, 2011:73). South Africa was no exception. According to Van der Merwe (1996:1), South Africa adopted a fixed exchange rate system and pegged the South African currency to the pound sterling and thus became part of the sterling area in 1945. Rossouw (2009:5) states that South Africa remained part of the sterling area at the outbreak of the Second World War and was thus required to accept the exchange control arrangements of the area. At the end of the Second World War, the Bretton Woods system was introduced (Wang, 2009:23) and South Africa became part of the international exchange rate system. During the Bretton Woods system,

(36)

currencies around the world were fixed against the USD, which in turn was linked to gold at a fixed price of 35 USD per fine ounce (Van der Merwe, 1996:2).

The South African pound of the time was fixed at 3,04 USD which in turn was valued at 3,58134 grams of fine gold (Van der Merwe, 1996:2) and was allowed a fluctuation range of 0,5 percent. Van der Merwe (1996:1) further states that in 1949 the pound sterling was devaluated and resulted in a reduced value of the South African pound. At this stage the South African pound was fixed at 2,80 USD or 2,48828 grams of gold. South Africa persisted with the use of the South African pound and only introduced the ZAR in 1961 when South Africa left the Commonwealth and became the independent Republic of South Africa (Rossouw, 2009:5). The new ZAR was valued at 50 percent of the old pound and was thus fixed at 1,40 USD or 1,24414 grams of gold.

The exchange control agreements of the sterling area were applied initially to residents of South Africa (Van der Merwe & Mollentze, 2012:142), but in 1962 the Blocked Rand System was introduced. This system resulted in the exchange controls being made applicable to non-residents as well (Van der Merwe, 1996:2). The last major change in the exchange rate regime during this period came in 1968. Van der Merwe (1996:2) states that during this year a limit was placed on borrowing facilities of South African subsidiaries of foreign companies. In summary of this period, South Africa employed a fixed exchange rate system. Mtonga (2011:3) states that during the period of February 1961 until July 1971, South Africa fixed its currency to the British pound and from August 1971 until November 1971 the currency was fixed to the USD.

2.3.2 Phase 2: 1971 until 1979

During the early phases of this period, the ZAR was still fixed to the USD. Van der Merwe (1996:2) argued that although major countries such as Canada, Germany and the US shifted towards the use of a floating exchange rate regime, South Africa was not yet able make the leap. The argument at the time was that South Africa’s foreign exchange market was underdeveloped and could not be successful as most of the country’s international transactions were made in dollars (Van der Merwe & Mollentze, 2012:138). As monetary authorities around the world began to shift their regimes in late 1971, South Africa’s ZAR was devalued by 12,3 percent. The balance of payments of South Africa was on a relatively stable recovery path and decided to fix the ZAR to the sterling again. Van der Merwe (1996:3) states, unfortunately, that the sterling

(37)

was on a downward path, which was followed by the ZAR. Primarily due to this reason, South African monetary authorities decided to shift back the fixing the ZAR to the USD. Mtonga (2011:3) states that the shift back towards the fixing of the ZAR to the USD began in October 1972 and lasted until May 1974. On 21 June 1974, monetary authorities announced that South Africa would no longer fix the currency to another major currency and would pursue an independent managed-floating regime (Van der Merwe, 1996:3). The new regime allowed the monetary authority of South Africa to intervene and iron out short-term exchange rate fluctuations. Van der Merwe (1996:3) states that this regime was employed from June 1974 until June 1975 during which time the ZAR was kept relatively stable through around seven upward and four downward adjustments to the ZAR/USD rate. The ZAR seemed to be performing relatively well, but only up until March 1975.

During this period, the USD gained substantial ground whereas the sterling seemed to be on a downward path once again. Potgieter (2005:27) noted that at the time, some speculations were formed that the ZAR would undergo substantial downward pressure that resulted in the monetary authority of South Africa deciding to fix the ZAR to the USD once again in June 1975. The general agreement was that the ZAR/USD rate would be kept stable for longer periods and would only be adjusted when it was deemed utterly necessary (Van der Merwe, 1996:3). Although the initial intention was to leave the ZAR/USD rate as stable as possible, the balance of payments recovery was seemingly losing steam and lags in foreign payments resulted in speculations that economic growth could take a blow. Thus, in September 1975, the monetary authorities of South Africa decided to devalue the ZAR by 17,9 percent. According to Van der Merwe (1996:3), this rate was kept until the beginning of 1979.

This period also saw a great instability in terms of exchange controls. Van der Merwe and Mollentze (2012:142) argue that exchange controls were frequently tightened and relaxed and were largely dependent on the domestic and international environment. It was mentioned further that decisions to change exchange controls were made increasingly difficult due to the political instability in South Africa at the time. In 1976, exchange controls, which focused on ensuring capital outflows in South Africa, were relaxed. Van der Merwe (1996:3) also recalled that blocked ZAR balances also became much easier to trade amongst non-residents. The period of 1971 until 1979 was largely unsuccessful in terms of ensuring stable prices, balance of payments equilibrium, job creation and economic growth (Potgieter, 2005:27). This could be due to the uncertainty created by the frequent changes made to the South African exchange rate regime.

(38)

2.3.3 Phase 3: 1979 until 1985

The period of 1979 until 1985 was likely one of the most interesting and pivotal periods in South Africa’s foreign exchange regime. The period started right at the back of a confusing time where the monetary authority of South Africa was uncertain of what steps to follow next. In 1977, a commission of inquiry (De Kock Commission) was appointed in an attempt to improve South Africa’s underdeveloped foreign exchange market (Van der Merwe, 1996:3). The commission identified several flaws on the foreign exchange regime of South Africa and recommended that money supply targets and more market-directed measures should be implemented to improve the underdeveloped regime (Potgieter, 2005:28).

The South African monetary authority decided to test a dual exchange rate regime from 1979 until 1983. De Villiers (2015:45) mentioned that the dual exchange rate regime implied the use of two separate exchange rates in South Africa, the financial ZAR and the commercial ZAR. The financial ZAR was a freer-floating rate, whereas the commercial ZAR was closer to an independently managed-floating rate where the monetary authority of South Africa controlled the fluctuations (Van der Merwe, 1996:4).

The South African foreign exchange regime was left largely unchanged between the period 1981 until 1982 primarily due to political and social unrest (Van der Merwe, 1996:6). This created an obstacle for South Africa to improve the underdeveloped foreign exchange market of South Africa. It was only in 1983 that South Africa made some changes to the exchange controls of the country. Potgieter (2005:29) identified these major changes as first, exchange controls over non-residents were lifted, secondly, the monetary authority decided to return to a unitary exchange rate regime and removed the financial ZAR and lastly, liquid asset requirements were relaxed and interest rate controls were dismantled. The unitary exchange rate regime, however, was only used for a short timeframe and soon after was followed by a second phase of the dual exchange rate regime. The final dismantling of the dual exchange rate regime only took place in March 1995 (Van der Merwe & Mollentze, 2012:143).

South Africa was still struggling with political instability and social unrest in 1985 that resulted in financial sanctions being placed on the country (Potgieter, 2005:29). These sanctions in turn led to a debt-standstill and one of the most detrimental crisis unfolded. South Africa was not able to make any new loans and was obliged to repay previous debt in line with the debt rescheduling

(39)

agreements. Potgieter (2005:29) states that it was due to this difficult time that the financial ZAR was reinstated in 1985. De Villiers (2015:45) argues that the exchange controls were reinstated primarily to protect the ZAR from downward pressure as foreign investors could easily convert their loans into equity shares and sell the shares through a unitary commercial ZAR exchange market, which would weaken the ZAR. It was stated further that South Africa’s ZAR soon experienced sharp depreciation due to a weakening gold price, increased political instability and an ensuing debt crisis.

In summary of the period 1979 until 1985, South Africa moved from a fixed exchange rate regime (fixed to the USD) to a dual exchange rate regime in 1979. This was followed by a removal of the financial ZAR and back to a unitary managed-floating exchange rate regime in 1983. The monetary authority of South Africa then decided to reinstate the financial ZAR in fear of a depreciation of the ZAR and thus moved back to a dual exchange rate regime in 1985. All these changes during this period clearly emphasise South Africa’s turbulence at the time and frustration as the foreign exchange regime was highly underdeveloped. Aron et al. (1997:3) state that political instability, increased violence and heightened capital outflows from South Africa largely characterised this period, which saw the ZAR depreciate substantially.

2.3.4 Phase 4: 1985 until 1994

The dual exchange rate regime was kept for the largest part of this period and only saw the introduction of the managed-floating ZAR in early 1995 (Mtonga, 2011:3). South Africa was going through an especially difficult time in the second half of 1984 as political instability, violence and an ensuing financial crisis was causing investors and international bodies to turn a cold shoulder to the country of South Africa. The United Nations (UN) and other organisations decided to impose sanctions on South Africa, which resulted in the majority of investors pulling their investments from the economy (De Villiers, 2015:46). As a result, Zhang (2009:24) identified that South Africa’s economic growth was at an all-time low, the balance of payments suffered severely, unemployment increased, labour disputes erupted and environmental factors such as droughts crippled the economy. Aron et al. (1997:5) emphasise South Africa’s injurious situation by pointing out that gold and commodity prices were declining. All these factors did not put South Africa in a healthy position and resulted in large capital outflows from South Africa in late 1984. Urgent change was needed to ensure South Africa gets back on the international scene.

Effect of exchange rate volatility on returns of investment portfolios in South Africa