3. Electronic Theses and Dissertations (ETDs) - All submissions
Permanent URI for this communityhttps://wiredspace.wits.ac.za/handle/10539/45
Browse
10 results
Search Results
Item Capital asset pricing model test on the Johannesburg stock exchange(2018) Mokgele, Kabelo KeosiThe Capital Asset Pricing Model (CAPM), jointly accredited to Markowitz (1952), Treynor (1961), Sharpe (1964), Lintner (1965) and Mossin (1966), provides that, at equilibrium, the return on all risky assets is attributable to their covariance with the market portfolio. This paper studies whether the CAPM holds for the South African market (represented by the Johannesburg Stock Exchange), by using the methodology developed by Fama and Macbeth (1973). Furthermore, the paper expands on other factors that influence asset returns and interrogates alternative asset pricing models. The findings of the study on individual assets rejects CAPM in the context of South Africa. This is consistent with other empirical studies. CAPM is also rejected for the Industrial Index as well as the Top 40 index. What is interesting to note however, is that for the Resources index, CAPM is validated.Item An assessment of the long-term commodity price assumptions on the Mineral Reserve estimates of South African gold and platinum mining companies from 200 to 2016(2018) Maseko, VulaniIn the 19th and 20th centuries mining underwent an evolution from the exploitation of small mining properties by individual claim holders to consolidation and the formation of multinational mining companies. With the rise of multinational mining companies came the advent of international finance in the mining industry. As companies developed and operated larger mining projects across several geographies, their capital requirements increased. As the number of investors in mining projects increased, through the raising of finance on international capital markets, so did the importance of mineral asset reporting. While governments and other organisations reported on the mineral inventories of countries, international mining companies had to report on only the economic portions of their mineral assets. The early days of international mineral asset reporting resulted in several reporting scandals where investors were misled by unscrupulous companies and individuals who reported false mineral deposit quantities and qualities in an attempt to manipulate investor sentiment for personal profit. As a result of these reporting scandals, the international mineral reporting codes were developed to increase the quality and transparency of mineral asset reporting and to protect investors. In the case of South Africa, the South African Code for the Reporting of Exploration Results, Mineral Resources and Mineral Reserves (SAMREC) Code, was first published in March 2000 and adopted by the Johannesburg Stock Exchange (JSE) later that year. The SAMREC Code, like the other international mineral reporting codes, aimed to ensure quality and transparent mineral asset reporting. Mineral Resources and Mineral Reserves are collectively a backbone asset for any company in the minerals industry This research study assessed the reporting of Mineral Reserves in South Africa in the period from the introduction of the SAMREC Code, in 2000, to 2016, with specific focus on reporting by South Africa’s gold and platinum mining companies. Mineral Reserve reporting was identified as a key focus, as Mineral Reserves are the economically mineable portions of Mineral Resources, and therefore give a better indication of the prospects of mining companies in the short to medium term. The literature review conducted as a part of this research study indicated that long-term commodity prices are the most important modifying factors used in Mineral Reserve estimation, hence there was a specific focus on the long-term commodity prices used in the Mineral Reserve estimates of South Africa’s gold and platinum mining companies. However, the relationships between the Mineral Reserve estimates and the other modifying factors were also assessed. The research demonstrated a general improvement in the quality of Mineral Reserve reporting by South Africa’s gold and platinum mining companies since the introduction of the SAMREC Code in 2000. However, the quality of Mineral Reserve reporting by the platinum mining companies was found be considerably less detailed to that of the gold mining companies. The long-term gold prices of the gold mining companies were found to be relatively conservative, often falling below prevailing spot prices, while the long-term prices of the platinum mining companies were often found to be more optimistic, with long-term platinum group metals (PGM) prices that were generally above prevailing spot prices. This research study also assessed the relationship between the Mineral Reserve estimates of South African gold and platinum mining companies and their reported modifying factors, with a specific focus on long-term commodity prices. The coefficient of determination (R2) was used to assess the relationship between Mineral Reserve estimates and their associated modifying factors. The statistical analysis conducted demonstrated that the Mineral Reserves of South African gold and platinum mining companies between 2000 and 2016 were often most sensitive to changes in long-term commodity prices, relative to the other modifying factors. The long-term commodity price assumptions of mining companies affect the value at which assets are recognised on their balance sheets. Sustained and fundamental differences between the long-term commodity price assumptions of mining companies and prevailing spot commodity prices may result in mining companies having to record impairments against their assets. In the period between 2000 and 2016, South African gold mining companies recognised a total of ZAR46bn (in nominal terms) in non-financial asset impairments, while platinum mining companies impaired their non-financial assets by a total of ZAR54bn. An assessment of the non-financial asset impairments of South Africa’s platinum and gold mining companies revealed that the least number of impairments were recorded when long-term commodity prices were within 5% of spot prices, suggesting that this should be the range within which long-term prices should be determined to limit future impairments.Item The fama-french five-factor model: evidence from the JSE(2018) Cox, ShaunThe desire to explain the returns of shares listed on stock exchanges has long driven research into asset pricing models. The formation of the Capital Asset Pricing Model (CAPM) served as the starting point for almost all models derived to describe share returns. Further research into returns discovered that there were various risk factors that impacted share returns; these could be captured by market value (size) and the book-to-market ratio (value). In 1993, Eugene Fama and Kenneth French used these to expand on the CAPM. The Fama-French three-factor model created a framework to model returns and allowed for other researchers to explore asset pricing further. In 2015, Fama and French augmented their profitability and investment factors onto their threefactor model. The author of this study postulated that these additional factors may proxy for quality, a formally undefined characteristic of shares that is used in selecting investments. This study tested the effectiveness of the five-factor model and the additional factors in explaining returns on the Johannesburg Securities Exchange (JSE). The five-factor model was compared to the traditional size-value three-factor model as well as other three-factor models that incorporated the additional factors. Furthermore, the study looked at what premiums are present in the returns on the JSE and captured by the various models tested. The results show that the size-value and size-profitability three-factor models best describe time-series returns when comparing models. The five-factor model best explains the crosssection of returns and overall, the results identify a strong inverse size effect and value and market premiums. Interestingly, the strength of the original size-value three-factor model is reinforced. The additional factors of profitability and investment contribute to explaining the returns on the JSE. Furthermore, profitability could be seen as a contributing factor in a “quality premium”. In addition to being a risk factor, quality could also be used as a filter through which the traditional premiums like size or value can be unlocked.Item Insuring resource revenues against commodity price volatility(2018) Thomas, AurelienSolutions to the so-called resource curse have been various and diverse. However, results have been at best inconclusive, if not counter-productive. In this research report, drawing from a thorough literature review on the resource curse, evidences are presented that a previously over-looked factor, volatility in commodity prices, is the cause of the resource curse. Building up on this conclusion, it is demonstrated that a simple insurance mechanism that would reduce volatility in commodity prices would help resource-rich countries to articulate sustainable macroeconomic and fiscal policies that would lead to sustainable growth. Back-testing this model, it is demonstrated that such an insurance mechanism would have withstood the financial crisis of 2008. Collectively, the results of this research report support the hypothesis that volatility carries an overwhelming share of responsibility in the process through which resource-rich nations lag behind resource-poor nations in terms of development, being economic or human. This research report concludes that this simplified model should serve as a foundation for future research and that this model should be expanded and tested on a greater scale to confirm its benefits. This research report provides an innovative approach to solving the resource curse which should certainly be used and expanded in future research.Item Testing the Fama-French five-factor model in selected emerging and developed markets(2017) Mosoeu, SelebogoThe focus of this research is on testing the adequacy of the Fama and French fivefactor model in explaining the patterns in average stock returns for selected developing markets, and developed markets. Further tests are conducted in evaluating the performance of the five-factor model in explaining returns for diversified portfolios. With the proposition that there is some form of relationship between the asset’s expected returns and market risk, the Capital Asset Pricing Model (CAPM) serves as a cornerstone for the asset pricing models. The evolution of the market over the years resulted in other factors being discovered that related to the asset’s expected returns. This led to the development of the three-factor model and later the five-factor model. The performance of the five-factor model depends on the region upon which it is being tested, especially for emerging markets, although the global five-factor model fails dismally as compared to the emerging market five-factor model. Across all the countries studied, the market premium (Mkt) is redundant, except for India and South Korea, together with some of the other factors depending on the country. For Indonesia, Mkt is the only redundant factor for explaining the patterns in average returns for the sample period.Item Essays on commodity prices and financial market variables evidence from Sub Saharan Africa(2017) Ndlovu, XolaniVolatility in international commodity prices is almost accepted as a stylised fact in modern financial markets. The drivers of commodity prices have evolved in addition to traditional global demand and supply factors. The literature suggests a number of other drivers, among them, activities of speculators – the so called “financialisation” of commodities postulate, the role of China and Fed policy. The question of whether exogenous shocks to commodity prices are transmitted through financial markets in Africa is investigated. In addition, the hypothesis of “wealth-transfer” from exporters to importers of commodities when prices fall is tested. Further, commodity prices are tested for their in sample and out of sample predictive ability. Finally recommendations are made for policy makers and financial market players in frontier markets in Africa. The three essays are organised in Chapters 3 to 5 of the study. In Chapter 3, an ARDL bounds testing approach is adopted and we find that a South Africa-specific commodity index significantly predicts (in-sample) the exchange rate in the short-run. While the long-run relationship is weak and the associated error correction process is slow existence of cointegration of commodity prices and the exchange rate suggests that commodities explain a significant part of terms of trade fluctuations for South Africa. With respect to the structural exchange rate models of the South African Rand, using the Dynamic Ordinary Least squares (DOLS) estimator, we find that commodity prices are significant and consistent explanatory variables of the changes in the nominal exchange rate. The commodity price variable improves the in-sample fit of the structural exchange rate models presented in this chapter and this evidence is robust to the other major Rand cross rates. Further, inclusion of the commodity price variable improves the out-of-sample short horizon forecasting ability of canonical exchange rate models. 2 In Chapter 4 we employ dynamic econometric modelling techniques to confirm the existence of a strong financial channel through which copper price shocks are transmitted to the Zambian economy. In the short run, changes in the copper price lead changes in all financial market variables. Financial market variables and the price of copper share a long-run equilibrium relationship. Importantly, if this system is out of its long-run equilibrium, short run corrections back to equilibrium are made by adjustments to the short term interest rate. Fittingly, the copper price-interest rate relationship appears strong in the long run. This result suggests that the policy makers “over-rely” on monetary policy to accommodate shocks from the international price of copper. The exchange rate and equity prices appear weakly exogenous to the system in-sample and out of sample. In Chapter 5, we confirm existence of a structural break in the price of oil in July 2008. We also show that the financial market time series for Kenya and Nigeria also exhibit a significant structural break around this period. We therefore partitioned our sample to investigate the effect of structural shocks to the financial markets of the two markets. On the whole, we find that the nexus between financial markets and oil prices is much stronger and statistically significant for an oil exporter (Nigeria) and weaker and statistically insignificant for a net oil importer (Kenya) after the 2008 financial crisis. Prior to the 2008 oil price shock, the results are roughly the opposite of the post oil shock period for both countries. Our results highlight that it is important to account for major structural shifts in modelling the impact of oil prices in developing countries (Le and Chang, 2011). The “wealth transfer” argument from net importer to net-exporters exists between Kenya and Nigeria in the short run although it is not robust to sample specification. Finally, we highlight the inherent flaws and limitations of ex-post stabilisation funds in Africa and make a case for market based oil-price hedging instruments. We argue for the adoption of market based hedging instruments given 3 the promising growth in financial markets of developing African countries in spite of several thorny implementation difficulties.Item Predicting exchange rates using Taylor rule fundamentals: evidence from a portfolio optimisation framework(2017) Jobo, Mathe NaleliThe paper studies exchange rate predictability using Taylor rule fundamentals in an optimal portfolio framework.The study seeks to link exchange rate dynamics with capital flows. Profit-seeking economic agents are assumed to repatriate funds across borders in response to differentials in rates of return from risky assets of portfolios held. We develop a uncovered portfolio return parity (UPRP) based exchange rate model in which changes in the short-term nominal exchange rate depend on the difference of optimal portfolio returns between two economies. In a two country economy where USA is taken as the foreign country we test the model in 5 countries namely South Africa, South Korea, Brazil, Mexico and Poland. The model is benchmarked against a UIP model and a Random walk model in order to establish whether the study’s extension enriches exchange rate prediction literature. We find that the main UPRP model outperforms the Random walk model in the 12 month horizon for 4 out of 5 countries using CW statistics. For the 1-month horizon the main model is outperformed by the Random walk model in 4 out 5 countries and for the 2-month and 3-month horizons the main model beats the Random walk using CW statistics. Theil’s U statistics also show that with the exception of South Korea, the main model beats the Random walk across all countries in the 3 and 12-month horizons. We conclude that out-of-sample exchange rate forecasting based on an optimal framework and Taylor rule fundamentals produces better nominal exchange rate forecasts relative to the Random walk model and UIP modelItem Higher moment asset pricing on the JSE(2016) Bester, JohanThe purpose of the study is to investigate the effects of relaxing the assumption of multivariate normality typically utilised within the traditional asset pricing framework. This is achieved in two ways. The first involves the introduction of higher moments into the linear Capital Asset Pricing Model while the second involves a Monte Carlo experiment to determine the impact of skewness and kurtosis on test statistics traditionally employed to assess the validity of asset pricing models. We commence by establishing non-normality for the majority of sample portfolios. A cross-sectional regression approach is employed to estimate factor risk premia and test higher moment Capital Asset Pricing Models. Unconditional coskewness and unconditional cokurtosis are found to be priced within the market equity (size) sorted and book equity/market equity (value) sorted portfolio sets over the period January 1993 to December 2013. Conditional coskewness and conditional cokurtosis are found to be priced for only the size sorted portfolios over the period January 1997 to December 2013. Factor risk premia estimated for coskewness are generally positive while risk premia estimated for cokurtosis are negative. This suggests a positive relationship between coskewness and expected return and a negative relationship between cokurtosis and expected return. The results of the asset pricing model tests are mixed. The pricing errors for higher moment Capital Asset Pricing Models are shown to be significantly different from zero for size sorted portfolios while pricing errors on the value sorted, dual size-value sorted and industry portfolios are found to be statistically insignificant. This suggest that none of the asset pricing models tested are the true model as it would explain variation in expected returns regardless of the data generating process. Finally we show that the Ordinary Least Square Wald test statistic has the most desirable size characteristics while the Generalised Least Squares J-test statistic has the most desirable power characteristics when dealing with non-normal data.Item An empirical evaluation of capital asset pricing models on the JSE(2014-03-07) Sacco, Gianluca Michelangelo;The Capital Asset Pricing Model (CAPM), as introduced by Markowitz (1952), Sharpe (1964), Lintner (1965), Black (1972) and Mossin (1966), offers powerful and intuitively pleasing predictions about the risk and return relationship that is expected when investing in equities. Studies on the empirical strength of the CAPM such as Fama and French (1992), however, indicate that the model does not reflect the share return actually obtained on the equity market. Attempting to improve the model, Fama and French (1993) enhanced the original CAPM by incorporating other factors which may be relevant in predicting the return on share investments, specifically, the book – to – market ratio and the market capitalisation of the entity. Carhart (1997) further attempted to improve the CAPM by incorporating momentum analysis together with the 3 factors identified by Fama and French (1993). This research report empirically evaluates the accuracy of the above three models in calculating the cost of equity on the Johannesburg Stock Exchange over the period 2002 to 2012. Portfolios of shares were constructed based on the three models for the purposes of this evaluation. The results indicate that the book-to-market ratio and market capitalisation are able to add some robustness to the CAPM, but that the results of formulating book – to – market and market capitalization portfolios is highly volatile and therefore may lead to inconsistent results going forward. By incorporating the short run momentum effect, the robustness of the CAPM is improved substantially, as the Carhart model comes closest to reflecting what, for the purposes of this study, represents the ideal performance of an effective asset pricing model. The Fama and French (1993) and Carhart (1997) models therefore present a step forward in formulating an asset pricing model that will hold up under empirical evaluation, where the expected cost of equity is representative of the total return that can be expected from investing in a portfolio of shares. It is however established that the additional factors indicated above are volatile, and this volatility may influence the results of a longer term study.Item A comparison of the forecasting accuracy of the downside beta and beta on the JSE top 40 for the period 2001-2011(2014-03-06) O'Malley, Brandon ShaunThe purpose of this research report is to determine whether the use of a Downside risk variable – the D-Beta – is more appropriate in the emerging market of South Africa than the regular Beta used in the CAPM model. The prior research upon which this report expands, performed by Estrada (1999; 2002; 2005), focuses on using Downside risk models mainly at an overall country (market) level. This report focuses exclusively on South Africa, but could be applicable to various other emerging markets. The reason for researching this topic is simple: Investors – not just in South Africa, but all across the world – think of risk differently to the way that it is defined in terms of modern portfolio theory. Beta measures risk by giving equal weight to both Upside and Downside volatility, while in reality, investors are a lot more sensitive to Downside fluctuations. The Downside Beta takes into account only returns which are below a certain benchmark, thereby allowing investors to determine a share’s Downside volatility. When the Downside Beta is included as the primary measure of systematic risk in an asset pricing model (such as the D-CAPM), the result is a model which can be used to determine cost of equity, and make forecasts about share returns. The results of this research indicate that using the D-CAPM to forecast returns results in improved accuracy when compared to using the CAPM. However, when comparing goodness of fit, the CAPM and the D-CAPM are not significantly different. Even with this conflicting result, this research shows that there is indeed value in using the D-Beta in South Africa, especially during times of economic downturn.