182 Afro-Asian J. Finance and Accounting, Vol. 4, No.

2, 2014

Analysing security performance in Morocco and

South Africa using CAPM

William Coffie
University of Wolverhampton,
Business School,
Wolverhampton, WV1 1AD, UK
E-mail: w.coffie@wlv.ac.uk

Abstract: This paper examines how well the capital asset pricing model
(CAPM) is able to describe the performance of individual securities listed on
Casablanca and Johannesburg Stock Exchanges, Morocco and South Africa
respectively. Jensen (1968) methodology is employed in the study. While there
is a reasonable amount of empirical studies on the performance of the CAPM in
Africa, the validity of the model has not previously been addressed in this
manner in Morocco and South Africa. The CAPM posits that the performance
of assets is solely explained by the market beta. The results of this study do not
support this assertion. Although it was found that beta contributes to the
variation of security returns in Morocco and South Africa, that contribution is
insufficient to fully explain security performance. Instead, we found positive
and significant alpha values, representing factors unexplained by market beta,
and hence deviations from the CAPM.

Keywords: security performance; stock exchange; capital asset pricing model;

CAPM; Johannesburg; Casablanca.

Reference to this paper should be made as follows: Coffie, W. (2014)

‘Analysing security performance in Morocco and South Africa using CAPM’,
Afro-Asian J. Finance and Accounting, Vol. 4, No. 2, pp.182–202.

Biographical notes: William Coffie has a cumulative work experience of

almost 15 years, which straddle four very important sectors in the public and
private sectors – the academia, local government, charity and industry. He
teaches at both undergraduate and postgraduate modules in finance,
econometrics and accounting. His research interest lies in asset pricing and
portfolio theory, volatility forecasting and correlations, time series and
cross-sectional analysis, stock market efficiency, Africa and emerging markets
finance and cost of capital.

1 Introduction

The capital asset pricing model (CAPM) of Sharpe (1964), Lintner (1965), and Mossin
(1966) has been the dominant orthodoxy in finance and has been empirically tested
extensively in USA, Europe, Japan, and many other developed capital markets over the
last four decades. In contrast, asset pricing research in emerging markets only recently
began in the mid 1990s. Claessens et al. (1995), Fama and French (1998), Patel and
Sarkar (1998), Rouwenhorst (1999), Lyn and Zychowicz (2004), and Ramcharran (2004)

Copyright © 2014 Inderscience Enterprises Ltd.

 Analysing security performance in Morocco and South Africa using CAPM 183

have investigated the relationship between asset returns and fundamental risk attributes in
emerging capital markets. Most of this international empirical evidence, in the main, has
ignored an important class of emerging African frontier markets, including Morocco and
South Africa. Emerging African market frontiers have relatively lower investable market
capitalisation, and are not integrated with world markets, and are also mostly illiquid.
In terms of applications, the CAPM is commonly used in the estimation of cost of
equity capital, and also frequently deployed in portfolio performance evaluation. This
may be so because the CAPM provides parsimonious and intuitively appealing definition
for risk as measured by the beta. In other words, the only risk investors care about and
will require compensation for is the systematic (or undiversifiable) risk created by the
market factor.
Given the paucity of empirical evidence on the CAPM in African frontier markets,
the wide applications of the Model in corporate finance and security analysis and
portfolio management, the contribution of this paper is to provide evidence on how well
it is able to describe the performance of individual securities trading on Casablanca and
Johannesburg Stock Exchanges (JSE).
One other important qualification or rider to make: it has been suggested by some
researchers that since global capital markets are becoming increasingly and significantly
integrated, the international capital asset pricing model (ICAPM) should be used to
estimate returns (O’Brien, 1999; Stulz, 1995, 1999; Schramm and Wang, 1999). This
implies that international investors can enter and leave any market anywhere in the world
with reasonable certainty and a minimum transaction costs. However, application of the
global version of the CAPM in emerging capital markets has proved impractical and
controversial. This is because these markets have remained highly segmented due to
country specific barriers that minimise their integration to the world markets (Bekaert,
1995; Harvey, 2000; Bekaert and Harvey, 2002; Chaieb and Errunza, 2007; etc.). The
ICAPM has not been used in this study.

2 Review of theoretical and empirical evidence

2.1 Theory
The CAPM was developed out of the modern portfolio theory (Markowitz, 1952, 1959)
and the capital market theory (CMT). According to Markowitz (1952), the portfolio
selection process begins with pertinent beliefs concerning future security performances
and ends with choice of portfolio. Expected return is considered by investors as a
favourable thing and variance of return as unfavourable.
The failure of both portfolio theory and CMT to define and measure risk in terms of
individual asset’s contribution to the portfolio risk led to the development of CAPM of
Sharpe (1964) and Lintner (1965). Fundamentally, the CAPM seeks to quantify the
relationship between expected return and non-diversifiable risk (known as beta). The
CAPM expresses the relationship between expected return of investment i and its
corresponding risk exposure as:

E ( Ri ) = R f + β i ⎣⎢ E ( RM ) − R f ⎦⎥ (1)
184 W. Coffie

σ i riM cov ( Ri , RM )
βi = = (2)
σM σ M2

where
Ri required return on asset i
Rf risk free return
RM the market return
βi the coefficient for the risk premium, E(RM) – Rf
σ M2 the variance of the market

Cov(Ri, RM) the covariance between the return of the market and the return of the asset.

2.2 Empirical literature

Over the last four decades, finance researchers have found substantial evidence in support
of the CAPM (Jensen et al., 1972; Black, 1972; Fama and McBeth, 1973; Blume, 1975).
However, the CAPM has come under attack from researchers who have identified
variations in asset and portfolio returns that are not captured by the classic CAPM’s beta
(Basu, 1977, 1983; Banz, 1981; Fama and French, 1992, 1993). Further, Ross (1976) and
Roll (1977) also criticised the Model for the dominant impact of the market portfolio.
Roll (1977) slated the CAPM’s recommendation of stock market portfolio as the efficient
portfolio and felt that the model has not yet been tested and almost certainly never will be
because the problem is that the market portfolio which is at the heart of the model is both
theoretically and empirically indefinable. Roll criticised the CAPM’s recommended
market proxy as not being mean-variance efficient and unlikely to be testable. However,
Levy and Roll (2010) examined the mean-variance efficiency of the market proxy
recommended by CAPM by adopting a reverse engineering approach, where they first
oblige that the return parameters ensure that the market proxy is efficient. They found
that their result is in conflict with earlier damaging and unsatisfactory results for the
CAPM theory. After 33 years Roll seems to suggest that the ex-ante mean-variance
CAPM’s market index proxy is after all consistent with empirical observed return
parameters and the market proxy portfolio weights. In effect this finding re-affirms the
fact that it will be premature to reject the CAPM as suggested by others regardless of
whether it is being applied in developed or emerging markets.
Banz (1981) and Reinganum (1981) provide evidence that, in addition to beta, size is
also a significant factor. This has become known as the ‘size-effect’ in assets returns. In
the same vein, Basu (1977, 1983) find that the same differences in returns applies to low
price-earnings ratios (P/E) stocks compared with high P/E ratio equities. Also, Fama and
French (1992, 1993) identified the joint significant role of size and BE/ME. Carhart
(1995, 1997) extended the Fama-French three factor model to include Jegadeesh and
Titman’s (1993) one-year momentum anomaly. The four variables are interpreted as
performance attributes, where the coefficients and the risk premia on the mimicking
portfolios indicate the proportion of mean return attributes to four fundamental strategies;
high versus low beta stocks, large versus small capitalisation stocks, value versus growth
stocks and one-year return momentum versus contrarian stocks. He finds that the
 Analysing security performance in Morocco and South Africa using CAPM 185

four-factor model substantially improves on the average pricing errors of the CAPM and
the three-factor model. These findings therefore challenge the explanatory power of
CAPM’s beta as the only priced or explanatory factor. However, this study is limited to
applying the one factor CAPM to security performance in Morocco and South Africa.
The emergence of new stock markets in the developing countries and the
re-emergence of previously dormant ones are now quite important for international
portfolio diversification. Since the mid 1990s, an extensive literature on risk-return
characteristics of such markets in the Asian and Eastern European has been documented
with little attention on Africa. This section reviews firstly, some studies of the CAPM in
non-African emerging markets, and then in Africa.
Claessens et al. (1995) provide evidence on the nature of asset returns by
investigating cross-sectional returns in 19 emerging markets. Using data from IFC
emerging markets database, they examine the effects of other risk factors on asset returns
in addition to beta. Following a regression similar to that of Fama and French (1992),
they find that in addition to beta, size and trading volume have significant influence on
asset returns in most of these markets. In a fewer of the markets, dividend yield and
earning-price ratios are contributory explanatory factors. Indeed, Bekaert and Harvey
(1995) show that ‘returns in emerging market are very different from returns in
developed markets’, and hence employing the standard asset pricing model or the global
version of the CAPM is unlikely to work due to the complex abnormal behaviour (such
as excessive volatility)of equity returns in these markets. Similarly, Harvey (2000) argues
that there would be serious problems in applying ICAPM to emerging capital markets
because of the model’s assumption of a perfect capital market.
Akdeniz et al. (2000) examined the impact of beta on monthly asset returns in Turkey
from 1992 to 1998. Following Fama and French (1992) regression approach, they
estimated the beta coefficients of 100 companies in their sample by regressing monthly
returns of assets on the contemporaneous and one-month-lagged returns on the
value-weighted Istanbul Stock Exchange (ISE) composite index. To adjust for
non-synchronous trading in the stock returns which can induce downward bias in the
‘true’ beta, they then calculated the beta estimate of each company as the sum of
contemporaneous and lagged values of the beta coefficients of the regression, which is in
line with Dimson (1979). Despite such adjustment, Akdeniz et al. found that beta was
insignificant in explaining realised asset returns of stock quoted on ISE Stocks. Similarly,
Michailidis et al. (2006) reported that their results ‘could not support the proposition that
higher risk (beta) is associated with higher returns in the Greek capital market, nor did it
support any alternative model’. Their study used the weekly and annual returns of 100
listed equities on Athens stock exchange from January 1998 to December 2002.
Outside European emerging markets, Pereira (2005) examined the challenges of
applying traditional valuation techniques and asset pricing model(s) adopted in
Argentina. He interviewed corporate executives, financial advisors, private equity funds,
banks and insurance companies using structured questionnaire technique. Pereira found
that although the CAPM is the model mostly reported for estimating the cost of capital
for discounting investment cash flows, however, quite often, the estimated figure was
adjusted upward to take account of country-specific risk factors such as asset
expropriation by regimes, volatile exchange rates, political instability etc. However, these
country-specific risks may be time-varying, and therefore using a constant additional risk
premium in the adjustment is inappropriate.
186 W. Coffie

Turning to African markets, Bundoo (2008) tests Fama and French three-factor model
on the Mauritius capital market taking into account time-variation in betas. The aim of
his methodology is to establish whether the size and book-to-market effects can disappear
or be reduced as time-varying risk premium is adjusted for temporal variation in
idiosyncratic risk, in this case time lag in beta. Bundoo found that his results were
consistent with Fama and French (1992, 1993); that is, in addition to beta, size and
book-to-market premia are present in the Stock Market of Mauritius. In other words, beta
alone does not explain assets’ returns in Mauritius as held by the CAMP.
Hearn and Piesse (2009) proposed and tested size and liquidity-augmented CAPM
focusing on emerging four African markets, namely, JSE, Nairobi Stock Exchange
(NSE), Swaziland and Mozambique. They measured “Illiquidity for a given stock on a
given day is measured as the ratio of the absolute value of the percentage price change
per US$ of trading volume”. Their results show that size-illiquidity augmented CAPM
performs better than the Sharpe-Lintner CAPM and Fama-French model. More
specifically, they found that size and illiquidity are priced factors in South Africa and
Kenya, but less significant in Swaziland and Mozambique. In another study, Hearn and
Piesse (2009) also estimated the cost of equity using the classic and augmented CAPM
for the major sectors in Morocco, Tunisia, Egypt, Kenya, Nigeria, Zambia, Botswana and
South Africa. The cost of equity was found to be highest in the financial sector of all
countries and lowest in blue chip stocks of Tunisia, Morocco, Namibia and South Africa.
In contrast to Bundoo (2008) and Hearn and Piesse (2009), Al-Rjoub et al. (2010)
found that beta has significant explanatory powers in predicting stock returns in Egypt,
Jordan, Morocco and Saudi Arabia, whereas other fundamentals – P/E, BE/ME and
M-CAP failed to account for variations in stock returns in these markets. Al-Rjoub et al’s
results are consistent with an earlier study in Morocco (Hearn et al., 2010), which also
found that the beta is significant in determining asset returns. This is also supported by
our own recent study of the Ghanaian capital market (Coffie and Chukwulobelu, 2012).
Reddy and Thomson (2011, 2013) investigated whether the CAPM can provide a
reasonable basis for actuarial modelling in South Africa. Unlike most previous studies
elsewhere and in Africa, they used quarterly and annual instead of monthly, weekly or
daily data. They separately first regressed the excess returns on sector indices against
their corresponding estimated betas, and then the excess returns on the market portfolio
against the estimated betas, both for the individual years and for the entire period of their
study. Their results show that, for sectorial regressions, the CAPM was rejected for the
entire period and in each sub-period. Again, similar results were also found with regard to
the regressions on the market portfolio.
In a departure from econometric studies, Nel (2011) conducted a field research on the
use of the CAPM by investment and accounting practitioners in South Africa. His
interviewees also included some academics. Surprisingly, he found that all the investment
practitioners interviewed indicated that they use the CAPM frequently, but not
surprisingly, less than 100% (74%) academics support its application. In general though,
both practitioners and academics agree that CAPM is the best approach to calculate cost
of equity.
From the forging, the evidence on the CAPM in African emerging markets seems to
be mixed. While some studies such as Bundoo (2008), Hearn and Piesse (2009), and
Reddy and Thomson (2011) found weak support for the model, others like (Hearn et al.,
2010), Al-Rjoub et al. (2010), Nel (2011), found that beta is very significant in explaining
return generating process in some African markets. In Morocco, Al-Rjoub et al. (2010)
 Analysing security performance in Morocco and South Africa using CAPM 187

tested the Fama-French three factor model based on P/E, BE/ME and size portfolios in
Morocco. Hearn and Piesse (2009) tested augmented CAPM based on size and illiquidity
portfolios in South Africa. Furthermore, Reddy and Thomson (2011, 2013) used the two
step approach to test whether CAPM can be accepted for stochastic modelling of
investment returns in typical actuarial applications, while Nel (2011) conducted field
research on the use of CAPM in South Africa.
These previous studies of CAPM in Africa and in particularly, Morocco and South
Africa have been based on portfolios; however, this paper evaluates the performance of
CAPM with individual securities rather than portfolios on Casablanca and JSE. Since the
CAPM was developed as an individual asset pricing model, accordingly, it is argued that
it is important to establish the performance of the model with regard to individual
securities initially before jumping to portfolios in frontier market studies. This study will
also help individual companies to understand and evaluate the relevance of the CAPM on
the basis of their firm specific risk profile rather than the risk characteristics of a
combination of portfolio of securities.

3 Methodology

3.1 Sample and data sources

JSE represents 85.7% of the total African market capitalisation in 1996 and this fell to
75% in 2007 according IFC emerging market database in 2007, as a result of other
markets being developed in the region. The JSE was ranked the 16th largest capital
market in the world in 1996 by market capitalisation (US $241.6 billion) and by the end
of 2007, JSE emerged as the largest emerging capital market in the world with
capitalisation of US $833.5 billion confirming a momentous flow of foreign capital flight
into South Africa. Morocco on the other hand saw a significant market capitalisation
growth over this period from US $8.7 billion in 1996 to US $18.5 billion, representing
113% increase. Likewise, market capitalisation as a proportion of GDP increased from
23.8% to 84.2% and turnover ratio from 5.9% to 27.1% from 1996 to 2007 respectively
(IFC International Financial Statistics, 2008). In comparison, Casablanca stock market is
less developed and fairly new market with active trading from 2000 but the only market
in the North African region with market data on DataStream database at the time of
study, while South Africa is the most capitalised and developed market in Africa and
fairly old dated back in 1887. Also, given their distinctiveness in terms of market
microstructure and differences in their industrial composition (i.e., Johannesburg is
dominated by mining and financial institutions while Casablanca has no one dominant
industry) it makes both markets to be of interest for asset pricing studies.
The sample for this study is carefully selected to provide adequate representation of
all industries in the Moroccan and South African capital markets, which are mainly
manufacturing, banking/financial services, real estate, mining, agriculture and trading.
This enhances comparability, and enables unbiased conclusion(s) to be drawn. However
to be included in the sample, a company also has to satisfy three criteria. Firstly, the
company must have at least three years of complete price data over the period covered by
the study 1995–2009. This is to enable sufficient price observations for each company in
the study. Secondly, to minimise the problem of infrequent trading, the company must
have traded at least once a month in each of those three year period. Some stocks do not
188 W. Coffie

trade in these markets for as long as six months and such companies are excluded from
this study. And thirdly, the share price must be denominated in local currency, Dirham
and Rand. Out of an initial total population of 44 companies, 29 companies listed on the
Moroccan exchange satisfied these criteria and only 55 listed companies on the South
African exchange satisfied these criteria out of initial 250 firms.
Table 1 Summary statistics for Morocco

Company Mean Median SD Skewness Kurtosis Jarque-Bera

Acred 9.4805 8.3520 13.5827 2.5143 13.5216 532.6273**
Afriquia Gaz 10.3655 9.3629 8.9401 0.8913 4.3432 19.5128**
Attijariwafa Bank 9.4989 8.4948 6.4281 0.6006 4.8680 19.3177**
Auto Hall 10.5369 9.9527 8.0785 –0.2181 4.7341 12.5233**
Auto Nejma 10.3138 8.5609 10.3113 1.8062 11.1155 309.0673**
BMCE Bank 10.0006 9.1854 7.0929 1.1313 6.7179 74.1948**
BQ. Maroc. Du Com. Etdl. 9.2712 9.2549 6.5543 0.3924 7.5946 85.0929**
Branoma 9.9288 8.9414 6.3380 0.5091 3.8339 6.7841*
Brasseries Du Maroc 10.0060 9.8203 8.2669 0.5423 4.3305 11.5412**
CDM Credit Du Maroc 8.8042 7.7453 7.5993 1.8907 11.8757 364.5535**
Centrale Laitiere 9.9443 9.3148 8.3372 0.0458 4.4731 8.5323*
Ciment Du Maroc 9.0819 9.3448 8.3908 0.1045 5.2901 20.7124**
Consumar 10.1663 10.2305 7.4851 –0.0857 3.0212 0.1167
Cr. Immobil. Et Hotelier 2.7431 0.9267 12.3056 0.8482 4.6629 22.1018**
Eqdom 9.0456 8.7418 7.2992 0.7092 6.4221 53.7468**
Holcim Maroc 9.3262 9.3108 8.0580 0.2732 4.6319 11.6007**
Lafarge Ciments 9.9803 9.7694 7.3234 0.0693 4.2122 5.8304*
Lesieur Cristal 8.6553 8.4983 7.7486 0.0749 3.9717 3.0640
Managem 7.2484 7.1832 12.8070 0.0433 3.7927 2.4905
Maroc Leasing 6.3338 6.0977 21.9299 0.3147 19.9261 1,123.6490**
Nexans Maroc 8.3020 8.2818 0.3275 –7.1314 63.8446 15,295.0700**
Rebab 9.3883 8.6770 12.4198 0.2365 8.2224 107.6969**
Samir 8.7066 8.1345 9.7089 1.3709 10.0099 221.9015**
Sc. Mtg. D’imiter 6.1788 6.7354 13.6378 –0.1393 4.3074 6.9985*
Sonasid 10.0944 11.1138 8.5370 0.0879 4.3894 7.6815*
Taslif 6.2937 5.5003 13.1142 2.8391 20.0709 1,267.6600**
Unimer 9.2313 8.5683 4.9889 0.2706 3.4205 1.8396
Wafa Assurance 7.6855 7.1660 9.6544 1.4778 7.2804 105.9736**
Zellidja 10.0413 8.4809 13.4118 0.2743 5.6044 27.7445**
Notes: The mean, median, standard deviation, skewness, kurtosis and Jarque-Bera
statistics are shown for each security in Morocco. The significance tests were set
at 1 and 5% levels.
** and *denote statistical significance at 1 and 5% levels respectively.
 Analysing security performance in Morocco and South Africa using CAPM 189

The stock price-data for the individual companies and for the value-weighted
all-share index of the Morocco and JSE, as well as the yield on Moroccan and South
African government’s Treasury bill are obtained from Thomson Reuters DataStream.
This stock price database is attractive because it has already been adjusted for all capital
changes such as rights issue, stock splits, and stock dividends, as well as the effects of
corporate restructuring as merger, acquisition and spin offs/demerger.

3.2 Descriptive statistics

As the descriptive statistics show the highest mean return in Morocco as registered by
Auto Hall is 10.3138%, while in South Africa the highest mean return is recorded by
Aspen Pharmaceuticals at 10.0469%. Furthermore, in Morocco, the lowest mean return is
2.7431% recorded by Cr. Immobil Et. Hotelier, while Sacoil recorded –2.7638% in South
Africa. Volatility as measured by standard deviation varies considerably among stocks
within and between the markets. In Morocco, the Maroc Leasing recorded the highest
standard deviation at 21.6024% while AG Industries registered the highest standard
deviation in South Africa at 35.3791%. The distribution for all stocks return data in both
markets is non-normal. They are either positively or negatively skewed with fat tails as
seen in the significant kurtosis well above the critical value of 3. The Jarque-Bera (JB)
which test the joint hypothesis that “the data used in this study are normally distributed,
with skewness of 0, and kurtosis of 3”, is rejected, as demonstrated by statistically
significant JB statistics at 1 or 5% levels for most firms in both markets.
Table 2 Summary Statistics for South Africa

Company Mean Median SD Skew Kurt J-B

ABSA Group 8.5819 7.9719 8.5505 –0.4320 9.4037 300.9721**
Acucap Properties 9.1857 9.4384 5.9429 –0.7145 4.6094 70.7553**
AECI 8.2729 9.0972 11.0569 –1.1651 8.5462 260.8742**
African Rainbow 7.9637 8.2125 13.1269 –0.5763 5.1399 42.5836**
African Oxygen 8.1175 8.5709 8.7013 –0.4378 4.3899 19.4522**
AG Industries 9.2032 6.6401 35.3791 9.0297 94.4942 44,936.2100**
Allied Technologies 0.3219 0.2259 10.7434 0.2512 4.9519 29.2824**
AngloGold Ashanti 8.4735 7.7119 11.9398 0.3505 4.0489 11.4724**
Anglo Platinum 9.1588 9.2797 12.7299 –0.7735 5.6786 68.9699**
Aspen Pharmaceutical 10.0469 8.8681 15.6703 2.7605 20.8814 2,524.5370**
Holdings
Aveng 8.6762 10.4513 10.4423 –1.3240 6.4019 96.0237**
Basil Read 8.6895 9.3631 18.0258 0.2143 4.9467 28.6398**
Ceramic Industries 9.3267 9.4260 9.0918 –0.6566 6.1662 80.2902**
City Lodge Hotels 8.6485 9.0069 10.9284 –0.0378 10.4215 397.0686**
COM AIR 7.8182 8.1179 14.8705 –0.4534 3.8541 8.7939**
Cullinan 2.4783 2.7660 20.8117 –0.6310 5.1632 45.2139**
Notes: The mean, median, standard deviation, skewness, kurtosis and Jarque-Bera
statistics are shown for each security in South Africa. The significance tests were
set at 1 and 5% levels.
** and *denote statistical significance at 1 and 5% levels respectively.
190 W. Coffie

Table 2 Summary Statistics for South Africa

Company Mean Median SD Skew Kurt J-B

Delta EMD 7.6996 7.7403 11.2818 –0.9582 8.4030 236.9024**
Discovery 3.3078 4.1837 8.3552 –0.3272 3.2827 2.5623
Distell Group 2.3350 2.5989 9.8565 –0.3069 3.3123 3.4184
DRD Gold –1.1873 –3.2341 19.0021 0.6139 3.9213 16.9865**
DS&WHSG Network 3.5165 4.2535 20.2259 0.4539 8.8256 250.5709**
First Rand Bank 9.1388 9.6468 10.2899 –0.6472 9.2936 297.5978**
Glenrand M I B 6.1790 8.1361 3.6935 –1.3400 2.8261 41.1738**
Gold Reef Resorts 7.6406 7.6799 14.0853 –0.3467 6.8249 102.6263**
Gold Fields 8.1477 7.8265 12.7826 0.1549 2.5587 2.0959
Group Five 8.7539 9.7238 13.4355 –0.1703 3.4364 2.1452
Growthpoint Properties 8.5659 9.0841 10.9926 –3.7232 32.3792 6,621.4700**
Harmony Gold Mining 5.3319 4.5465 16.5426 0.0897 3.5185 2.1695
Impala Platinum 9.3996 9.5607 13.0357 –0.3534 3.9038 9.4885**
Liberty Holdings 8.1356 7.9639 7.8622 –0.8193 5.3108 57.8475**
Masonite Africa 7.0945 7.6218 10.6249 –0.2621 5.4159 44.0547**
Merafe Resources 0.2919 –1.0430 19.0271 –0.2356 4.9860 30.0329**
MMI Holdings 8.5952 9.3387 10.5372 –0.6550 7.6477 168.0788**
MTN Group 9.4030 9.2778 12.6728 –0.6889 7.5787 158.1364**
Murray and Roberts 8.3725 8.8512 12.4262 –0.4940 3.5677 9.3606**
NED Bank Group 3.8349 4.6968 9.8486 –0.7073 6.0406 81.0678**
Octodec Investments 9.5094 9.1274 8.6071 –1.1293 7.5302 184.7244**
Omnia 8.3870 8.0625 11.2051 0.0615 5.5877 48.3790**
Pangbourne Properties 9.4516 9.8926 6.4022 –0.0480 2.7018 0.7074
Premium Properties 9.7690 9.2141 8.9289 0.2162 3.8763 6.8824*
Pretoria Port CMT 8.6943 9.7372 8.9855 –0.5445 3.9466 15.0067**
RMB Bank 8.9970 9.6540 10.1337. –0.7134 9.7114 339.3595**
SABLE 7.3732 7.6592 14.8890 –0.1909 7.0239 117.7671**
SACOIL Holdings –2.7638 –0.9960 32.3018 0.7237 4.6303 34.2583**
Saambou Bank 8.0095 8.5318 9.8252 –0.7734 11.4498 531.9097**
Sanlam 8.5307 8.2483 7.6512 –0.3250 3.8394 6.1999*
SASOL 9.0280 9.0730 10.5198 0.0095 3.2207 0.3539
Spanjaard 8.6083 7.7128 11.8790 1.9513 15.3505 1,209.3080**
Standard Bank Group 8.9100 0.6242 10.1349 –1.1122 14.4956 988.2434**
Sun International 8.2682 9.2204 9.5909 –0.4769 3.3773 7.5854*
TELKOM 9.2870 8.6695 8.8330 0.0552 3.4138 0.6115
VOX Telecom –0.9922 –0.7895 40.4853 1.4193 14.8232 825.4762**
White Water Resources –1.6962 –1.4596 24.7681 –0.3315 4.6872 23.6874**
WLSN Bayly 10.0145 9.9922 11.7940 0.1249 5.4711 44.4674**
Holmes-Ovcon
Zurich Insurance 8.5819 7.9719 8.5505 –0.4319 9.4037 300.9721**
Notes: The mean, median, standard deviation, skewness, kurtosis and Jarque-Bera
statistics are shown for each security in South Africa. The significance tests were
set at 1 and 5% levels.
** and *denote statistical significance at 1 and 5% levels respectively.
 Analysing security performance in Morocco and South Africa using CAPM 191

3.3 Empirical method

The study follow a methodology similar to Jensen (1968) time-series approach.
Parameters are estimated using ordinary least squares (OLS). Most research in capital
asset pricing in Africa has been conducted using cross sectional studies. This means that
the CAPM performance is measured at one particular point in time. However, this study
adopts time series approach and therefore is designed to measure asset performance
across time.
The testable unconditional CAPM is expressed as:

Rit = R ft = α it + β it ⎢⎣( RMt ) − R ft ⎥⎦ + ut (3)

Let
Rit − R ft = rit (4)

RMt − R ft = rMt (5)

(
Rit = ln Pit
Pit −1 ) (6)

RMt = ln PMt ( PMt −1 ) (7)

Rit – Rft = rit is the monthly excess stock return. RMt – Rft = rMt is the monthly market risk
premium. Rft is the one-month annualised yield on Moroccan and South African
government’s Treasury bill, observed at the beginning of the month t. Equations (6) and
(7) are the monthly natural log returns (i.e., compounding returns) of individual assets
and the market portfolio respectively. Also, lognormal return estimation methodology is
preferred in order to overcome the problem of non-normality of returns data used in this
study.
Therefore, equation (3) can be re-written as:

rit = αˆit + βˆit rMt + ut (8)

ut ≈ N ( 0, σi2 )

var ( ut ) = σ 2 < ∞ (9)

cov ( ui , u j ) = 0 for i ≠ j (10)

A testable restriction implied by equation (8) is that the intercept (denoted as, α) is equal
to zero and the beta (βi) must be must be positive or greater than zero to capture all
systematic risk. The error term ut represents the residual term with a mean value of zero
and a constant variance and assumed to be independent of all other variables in
equations (3) or (8).
In order to test the stationarity of the series, and determine the integrability order, the
presence of unit roots is tested using Augmented Dickey-Fuller (ADF). To make our data
stationary, and hence avoid spurious regressions that may arise from using non-stationary
time series returns data, non-stationary in the data is transformed by taking their first
differences. White test and Breuch-Godfrey test were used to detect heteroscedasticity
192 W. Coffie

and autocorrelation respectively. Furthermore, Newey-West heteroscedasticity

autocorrelation consistent covariance and standard error is used to correct for the
presence of both autocorrelation and heteroscedasticity in the estimator model [i.e.,
equation (8)].

4 Empirical results and analyses

4.1 Empirical results

The monthly security excess returns on the individual companies are regressed on the
contemporaneous monthly market risk premium using equation (8). The regression
results for all companies in the sample are presented in Table 3 and Table 4 as follows.

Table 3 Morocco results

Company β α R2
Acred 0.1517 9.3613 0.0034
(0.4014) (6.3752)**
Afriquia Gaz 0.9877 0.9876 0.3332
(6.3634)** (15.0106)**
Attijariwafa Bank 1.0126 8.7030 0.6775
(12.9662)** (32.3597)**
Auto Hall 2.26136 9.9277 0.2513
(6.10835)** (14.6520)**
Auto Nejma 0.2834 10.0910 0.0206
(1.7399) (9.1817)**
BMCE Bank 0.8371 9.3427 0.3802
(5.0528)** (14.1645)**
BQ. Maroc. Du Com. Etdl. 0.6494 8.7608 0.2680
(4.9042)** (17.3577)**
Branoma 0.2114 9.7627 0.0304
(1.4384) (15.4039)**
Brasseries Du Maroc 0.6459 9.4983 0.1667
(4.4445)** (11.7411)**
CDM Credit Du Maroc 0.8447 8.1403 0.3373
(3.7701)** (17.3443)**
Centrale Laitiere 0.5706 9.4959 0.1279
(3.2817)** (12.4560)**
Ciment Du Maroc 0.9317 8.3496 0.3366
(7.0149)** (11.5636)**
Consumar 0.2165 9.9961 0.0228
(1.4976) (13.8235)**
Cr. Immobil. Et Hotelier 0.9206 2.0196 0.1528
(5.0137)** (1.5186)
Eqdom 0.8369 8.7980 0.3589
(8.7979)** (15.4948)**
 Analysing security performance in Morocco and South Africa using CAPM 193

Table 3 Morocco results (continued)

Company β α R2
Holcim Maroc 1.1706 8.4062 0.5761
(11.9815)** (16.0050)**
Lafarge Ciments 0.9992 9.1949 0.5083
(8.9136)** (16.5508)**
Lesieur Cristal 0.4736 8.2830 0.1020
(2.8288)** (8.9467)**
Managem 1.5020 6.1679 0.3755
(5.5187)** (5.6317)**
Maroc Leasing 1.0967 5.4719 0.0683
(3.7003)** (3.2234)**
Nexans Maroc –0.0048 8.3058 0.0059
(–1.4630) (219.7072)**
Rebab 0.1696 9.2550 0.0051
(0.6918) (6.8845)**
Samir 0.8911 8.0063 0.2210
(3.0435)** (10.0494)**
Sc. Mtg. D’imiter 0.8785 5.4884 0.1133
(3.6343)** (4.2578)**
Sonasid 0.9307 9.3630 0.3244
(7.5912)** (14.6155)**
Taslif 0.3672 6.0051 0.0214
(0.9635) (4.0393)**
Unimer 0.0610 9.1834 0.0041
(0.5412) (18.2127)**
Wafa Assurance 0.9852 6.9112 0.2843
(5.3123)** (8.2053)**
Zellidja 0.1200 9.9469 0.0022
(0.4453) (7.6056)**

Table 4 South Africa results

Company β α R2
ABSA Group –0.0028 8.5823 0.0006
(–0.3194) (13.1675)**
Acucap Properties 0.1535 9.1434 0.0224
(1.2008) (14.1805)**
AECI –0.0002 8.2729 0.0000
(–0.0302) (8.8847)**
African Rainbow 0.0138 7.9617 0.0060
(1.8939) (7.1915)**
African Oxygen 0.0032 8.1170 0.0008
(0.3629) (12.2388)**
AG Industries 0.6212 8.9618 0.0106
(1.1449) (2.8181)**
194 W. Coffie

Table 4 South Africa results (continued)

Company β α R2
Allied Technologies 0.0091 0.3206 0.0039
(1.4636) (0.4338)
AngloGold Ashanti 0.0050 8.4727 0.0010
(0.8458) (10.8121)**
Anglo Platinum 0.0161 9.1565 0.0087
(1.8308) (9.5210)**
Aspen Pharmaceutical Holdings 0.0009 10.0468 0.0000
(0.1190) (7.6894)**
Aveng 0.7665 8.3784 0.1858
(4.5496)** (8.6009)**
Basil Read 0.0047 8.6888 0.0004
(0.8567) (5.3609)**
Ceramic Industries –0.0086 9.3272 0.0051
(–2.6776)** (10.7636)**
City Lodge Hotels 0.0038 8.6479 0.0007
(0.6634) (10.8529)**
COM AIR 0.8278 7.5604 0.1435
(4.7383)** (6.3762)**
Cullinan –0.0242 2.4816 0.0074
(–2.6257)** (1.3234)
Delta EMD 0.0165 7.6972 0.0116
(1.4148) (9.0000)**
Discovery 0.3493 3.1681 0.0615
(2.7931)** (4.2778)**
Distell Group 0.0124 –1.1890 0.0023
(2.5357)* (–0.7796)
DRD Gold 0.0124 –1.1890 0.0023
(0.6320) (–0.8216)
DS&WHSG Network 0.0144 3.5144 0.0028
(1.5783) (2.0668)*
First Rand Bank –0.0086 9.1399 0.0038
(–0.9872) (12.8689)**
Glenrand M I B 0.0610 6.1599 0.0126
(1.1058) (8.9767)**
Gold Reef Resorts 0.0165 7.6397 0.0080
(3.2072)** (6.3998)**
Gold Fields 0.0296 8.1435 0.0290
(5.0920)** (8.7318)**
Group Five 0.0066 8.7531 0.0014
(1.1115) (7.9644)**
Growthpoint Properties –0.0122 8.5675 0.0067
(–6.1910)** (9.1838)**
 Analysing security performance in Morocco and South Africa using CAPM 195

Table 4 South Africa results (continued)

Company β α R2
Harmony Gold Mining 0.0150 5.3297 0.0044
(1.3854) (3.7816)**
Impala Platinum 0.0176 9.3971 0.0099
(2.1411)* (9.6873)**
Liberty Holdings –0.0050 8.1363 0.0022
(–0.9299) (13.4469)**
Masonite Africa 0.0035 7.0940 0.0006
(2.0035)* (7.5568)**
Merafe Resources 0.0123 0.2901 0.0023
(0.9526) (0.1822)
MMI Holdings –0.0081 8.5963 0.0033
(–1.1130) (11.2869)**
MTN Group 0.0081 9.4026 0.0023
(0.9251) (8.7416)**
Murray and Roberts 0.0017 8.3722 0.0001
(0.3254) (7.6049)**
NED Bank Group 0.0038 3.8343 0.0008
(0.6170) (3.8410)**
Octodec Investments 0.0034 9.5089 0.0009
(0.8649) (15.6382)**
Omnia 0.0049 8.3863 0.0010
(0.8585) (10.3157)**
Pangbourne Properties –0.0052 9.4523 0.0037
(–2.5106)* (20.2173)**
Premium Properties –0.0032 9.7695 0.0007
(–1.0142) (14.7617)**
Pretoria Port CMT –0.0118 8.6959 0.0094
(–3.3341)** (13.3217)**
RMB Bank –0.004518 8.997639 0.0011
(–0.5599) (12.1575)**
SABLE 0.0061 7.3722 0.0009
(1.8694) (6.4856)**
SACOIL Holdings –0.0590 –2.7555 0.0180
(–3.2688)** (–1.4583)
Saambou Bank 0.0115 8.0079 0.0075
(2.1740)* (8.7840)**
Sanlam 0.518799 8.254811 0.1568
(5.0974)** (14.0726)**
SASOL 0.0096 9.0265 0.0046
(1.1906) (11.1217)**
Spanjaard 0.0016 8.6080 0.0001
(0.4862) (9.5818)**
Standard Bank Group –0.0019 8.9102 0.0002
(–0.2402) (13.4163)**
196 W. Coffie

Table 4 South Africa results (continued)

Company β α R2
Sun International 0.0012 8.2680 0.0001
(0.2291) (10.9313)**
TELKOM 0.4122 8.9186 0.0687
(1.9503) (9.7636)**
VOX Telecom 0.0074 –0.9965 0.0000
(0.0096) (–0.3486)
White Water Resources –0.0059 –1.6953 0.0003
(–2.0923)* (–1.1240)
WLSN Bayly Holmes-Ovcon –0.0122 10.0162 0.0058
(–2.4256)* (10.3671)**
Zurich Insurance –0.0028 8.5823 0.0006
(–0.5927) (13.6270)**

4.2 Empirical analysis

The two fundamental propositions of the CAPM are

1 security returns are positive (and linear) functions of beta

2 beta is the only determinant of security returns.

As can be seen from Table 1, with the exception of Nexans, which has a negative beta,
proposition (1) is supported by these results. The results in Table 2 also show that 40 of
the South African securities, which represents more than two thirds of the sample support
proposition (1). In other words, investors in both the Moroccan and South African
markets, like investors elsewhere, expect to be compensated more, the higher the
systematic risk on their investment. However, 16 securities which represents almost a one
third of the firms under study in South Africa exhibit negative linear relation between
stock returns and market risk premium, contradicting proposition (1). Furthermore, all the
securities trading on Casablanca bourse except five and all securities on Johannesburg
bourse have beta coefficients less than 1, and hence exhibit low variation in returns (less
risky) than the market portfolio. In other words, although by investing in such securities
investors will require lower returns in compensation for taking up higher systematic risk
than if they invested in an Index Fund (see Sharpe, 1964; Lintner, 1965), they are also
exposed to lesser loss in a falling and/or volatile market condition. The reverse is true for
the securities with equity betas of more than 1.
Positive betas also imply the CAPM is able to partly capture the pattern of returns
generating process in Morocco and South Africa. As Table 1 show, the beta estimates of
69% of the sampled securities in Morocco are statistically significant at 1% level and
only 32% of sampled securities in South Africa are statistically significant at 1% or 5%
level as in Table 2. This indicates that market risk has a significant effect in evaluating
security performance in Morocco, which is consistent with the evidence documented in
other emerging and African capital markets (Claessens et al., 1995; Bundoo, 2008;
 Analysing security performance in Morocco and South Africa using CAPM 197

Al-Rjoub et al., 2010; Coffie and Chukwulobelu, 2012) but less so in South Africa (see
also Michailidis et al., 2006; Chui and Wei, 1998).
The fundamental proposition of the CAPM namely, that only systematic risk
determines security returns since unsystematic can be eliminated through diversification
(Markowitz, 1952, 1959; Sharpe, 1964; Lintner, 1965), is rebutted by these
results. Significant deviations from the Model in the Moroccan and South African
markets are evident in the results, which can be seen from the statistically significant
alphas at 1% and/or 5% level. This is further buttressed by the R2 for the individual
regressions, which are very low. The highest total variation in security returns in
Morocco which can be explained by the CAPM, as measured by R2 , is 67.75% (for
Attijariwafa Bank), and only 18.58% (for Aveng) in South Africa, leaving more than
30% and 80% of the variations in the securities returns unexplained by the model in both
countries respectively. For a security like Zellidja in Morocco, with R2 of 0.22%, Vox
Telecom, Aspen Pharmaceutical and AECI in South Africa, with R2 of 0.00%, the
unexplained variation of 99.88% and 100% respectively renders the appropriateness of
CAPM even more questionable. These implies that there are other risk factors other than
systematic risk, including perhaps company-specific and industry/economy wide risk
factors, which equity investors seek compensations for in the Moroccan and South
African markets. This is consistent with Jensen et al. (1972), Ross (1976) and Fama and
French (1992). Jensen (1968) alpha is widely used to evaluate performance of security
and it essentially seeks to identify whether CAPM over or under estimate security’s
returns. The evidence gathered in Table 1 and Table 2 show that alpha values for all
securities in both markets except five in South Africa are positive, implying that these
securities generate returns in excess of CAPM’s prediction. In other words, these
securities outperform the market portfolio. The alpha values for White Water Resources,
Vox Telecom, Sacoil Holdings, DRD Gold and Distell Group are negative, implying that
these securities generate returns lower that CAPM’s prediction and underperforms the
market portfolio.
The abysmal performance of the CAPM in explaining security returns and the
inability of the beta to explain most variations in the return generating process in
South Africa turned out to be worrying to the researcher. This is because JSE is
the most developed capital market in Africa and it is expected that as capital market
develops its market microstructure also advances and become more correlated to the
world market (O’Brien, 1999; Stulz, 1995). As the impact of these market
microstructures such as illiquidity, thin trading, and marketability diminishes
systematic risk factor becomes more relevant in explaining the return generating
process (Amihud and Mendelson, 1986). As a consequence, the researcher revisited
his data used to estimate the parameters for South African securities to check for errors in
the data. In fact, there was none. All the series are first difference stationary since ADF
test revealed that some series were non-stationary at level. Newey-West standard error
was used to correct for both heteroscedasticity and auto correlation and this was
confirmed by acceptable range of Durbin-Watson (DW) statistic across firms. In order to
normalise returns used in the estimation lognormal returns of security prices and indices
were calculated. However, comparable empirical evidence on CAPM tests in South
Africa by Reddy and Thomson (2011) shows that the beta is unable to explain security
returns.
198 W. Coffie

5 Implications for practical application

The CAPM is a theoretical model with various practical applications. Theory suggests
that corporate managers should go ahead and invest in capital projects provided they
maximise or enhance corporate value. Subsequently, if some shareholders disagree with
management decisions, or wish to realise their investments for consumption purposes,
they can sell their shares at the best possible price. This underpins the theoretical
recommendation that managers invest only in those projects that yield positive net
present value (NPV).
The CAPM provides a method of assessing the riskiness of cash flows arising from a
project and also for estimating the relationship between that riskiness and the cost of
capital (or the risk premium for investing in that project). It also asserts that the important
measure of a project risk is systematic risk known as the project’s beta. According to the
CAPM, a project cost of capital is an exact linear combination of the rate on risk-free
project and the total market risk conditional upon the project beta of the project being
evaluated. This CAPM model has gained popularity as a tool used to compute company’s
cost of capital, which is then used to decide whether capital investment projects should
proceed or otherwise.
Investors and managers often use CAPM to estimate expected return on investment
and this expected return is interpreted as required rate of return on the investment which
is used as discount rate to compute the intrinsic value of the investment. CAPM expected
return can be interpreted as a fair return on investment if systematic risk as measured by
beta is the only risk factor that needs to attract compensation. As the CAPM expected
return can be interpreted as a fair return, public companies can use CAPM to determine
the prices they charge for their goods and services. Hence, public companies are therefore
expected to set prices that generate return equal to the CAPM expected return.
The CAPM model is also a well useful yardstick for measuring the performance of
portfolio/fund managers. For example, if an ex posts evaluation of security using CAPM
generates a positive alpha, and then conditional upon the CAPM, this suggests that the
portfolio manager has indentified underpriced securities and a negative alpha would
mean that those securities are overpriced according to CAPM’s prediction. in view of the
fact that positive alpha estimates the expected return in excess of the value predicted by
the CAPM, portfolio/fund managers with high positive alpha values are considered as
most successful and smart than those with low and negative values.
However, empirical results documented in this study appear to suggest that the risk
adjusted one factor CAPM’s beta is not sufficient to fully explain the security returns in
Morocco and South Africa. Instead, by and large, we found positive and significant alpha
values, representing factors unexplained by market beta, and hence deviations from the
CAPM. The evidence shows that all securities trading on Casablanca and JSE except five
in Johannesburg (that is, Distell group, DRD Gold, Sacoil Holdings, Vox telecom and
White Water Resources which show signs of negative alphas) exhibit positive alpha
values. This means that, conditional upon the CAPM, securities in Morocco and South
Africa are underpriced and portfolio/fund managers who have identified these securities
would considered successful and could earn excess returns for their clients. Furthermore,
by and large, investing in securities trading on Casablanca and Johannesburg bourses
would generate excess returns than could be predicted by CAPM expected return.
Portfolio and fund managers would welcome this as value for their clients and would take
advantage of higher returns these two countries presents.
 Analysing security performance in Morocco and South Africa using CAPM 199

Besides, positive and negative alpha values mean that there are other important risk
factors that are uncorrelated to the market beta but affect security returns due to country,
industry or firm specific characteristics. Thus, although statistically, beta is significant in
explaining security returns in Morocco and to some degree in South Africa, one should
interpret this with caution. As noted, these results have important implications for
corporate managers who use CAPM as a basis of estimating cost of equity for various
financial decisions, for example, when evaluating capital investment projects, and for
portfolio managers/investors who employ CAPM in investment analysis/portfolio
construction and/or rely on the ex-post-excess-return versions of the model such as
Jensen Alpha and Treynor Index for portfolio performance evaluation.

6 Conclusions

From the forgoing, the evidence on the CAPM in African emerging markets seems to be
mixed. While some studies such as Bundoo (2008), Hearn and Piesse (2009), and Reddy
and Thomson (2011) found weak support for the Model, others like (Hearn et al., 2010),
Al-Rjoub et al. (2010), Nel (2011), and Coffie and Chukwulobelu (2012) found that beta
is very significant in explaining return generating process in some African markets and/or
very commonly used in estimating the cost of equity capital by corporate entities and
investment communities alike. This study adds to the growing evidence on the current
state of the CAPM in Africa by focusing on the Moroccan and South African capital
markets using Jensen (1968) methodology, albeit tested and adjusted for the usual
violations of ordinary regression models. The adjustments not only serve to underpin the
robustness of the findings this study seeks, but also provide insights into how the CAPM
can better be used to explain security performance in Morocco and South Africa in
particular, and hopefully more widely in emerging markets that share similar country
characteristics and market microstructure as these two countries.
The main focus of this study was to analyse the performance of individual securities
trading on Moroccan and South African stock markets using the excess return version of
CAPM. Empirical evidence produced in this study by and large supports the explanatory
power of beta in Morocco but less so in South Africa. However, contribution made by
beta to variation in security returns in both countries is less than the CAPM’s prediction
as measured positive alpha values. The significant alpha values that are documented in
both countries mean that other risk factors, in addition to the market beta, are likely to be
present in Morocco and South Africa. Furthermore, the empirical literature documents
that size, BE/ME, P/E and liquidity may be contributing factors to the return generating
process. Therefore the direction for future studies in Morocco and South Africa will
extend the tests to include these company-specific fundamentals and/or to employ a
portfolio methodology approach in line with Fama and French (1993) and others.
Although other asset pricing models such as arbitrage pricing theory, Fama-French three
factor model, Carhart factor model, etc., exist, this study was limited to the single factor
CAPM and therefore future study is also intended to extend the tests to other pricing
models.
The CAPM is a theoretical model which has a variety of applications, including
computing cost of capital which can be used to decide whether capital investment
projects should proceed, determine prices equal to required rate of return by public
companies and evaluate security performance.
200 W. Coffie

The CAPM provides a mean to assess the risk of cash flows arising from a project
and also estimates the relationship between that risk and the cost of equity capital (or the
risk premium for investing in that project). The CAPM emphasises that the important
measure of a project risk is systematic or common risk known as the project’s beta.
According to the CAPM, a project cost of capital is an exact linear combination of the
rate on risk-free project and the total market risk conditional upon the project beta of the
project being evaluated. This CAPM model has gained popularity as a tool used to
compute company’s cost of capital, which is then used to decide whether capital
investment projects should proceed or otherwise.
Investors and managers often use CAPM to estimate expected return on investment
and this expected return is interpreted as required rate of return on the investment which
is used as discount rate to compute the intrinsic value of the investment. CAPM expected
return can be interpreted as a fair return on investment if systematic risk as measured by
beta is the only risk factor that needs to attract compensation. As the CAPM expected
return can be interpreted as a fair return, public companies can use CAPM to determine
the prices they charge for their goods and services. Hence, public companies are therefore
expected to set prices that generate return equal to the CAPM expected return.
The CAPM model is also a useful yardstick for measuring the performance of
portfolio/fund managers. For example, if an ex posts evaluation of security using CAPM
generates a positive alpha, and then conditional upon the CAPM, this suggests that the
portfolio manager has identified underpriced securities and a negative alpha would mean
that those securities are overpriced according to CAPM’s prediction. In view of the fact
that positive alpha estimates the expected return in excess of the value predicted by the
CAPM, portfolio/fund managers with high positive alpha values are considered as most
successful and smart than those with low and negative values.

References
Akdeniz, L., Atlay-Salih, A. and Aydogan, K. (2000) ‘Cross section of expected stock returns in
ISE’, Russian and Eastern European Finance and Trade, Vol. 36, No. 5, pp.6–26.
Al-Rjoub, S.A.M., Al Yousef, A. and Ananzeh, I.E.N. (2010) ‘Beta wins again: case of four
emerging markets’, Journal of Economic Co-operation and Development, Vol. 31, No. 1,
pp.1–16.
Amihud, Y. and Mendelson, H. (1986) ‘Asset pricing and the bid-ask spread’, Journal of Financial
Economics, Vol. 17, No. 2, pp.223–249.
Banz, R.W. (1981) ‘The relationship between return and market value of common stock’, Journal
of Financial Economics, Vol. 9, No. 1, pp.3–18.
Basu, S. (1977) ‘Investment performance of common stocks in relation to their price-earnings ratio:
a test of the efficient market hypothesis’, Journal of Finance, Vol. 32, No. 3, pp.663–682.
Basu, S. (1983) ‘The relationship between earning’s yield, market value and the returns for NYSE
common stocks: further evidence’, Journal of Financial Economics, Vol. 12, No. 1,
pp.129–156.
Bekaert, G. (1995) ‘Market integration and investment barriers in emerging equity markets’, World
Bank Economic Review, Vol. 9, No. 1, pp.75–107.
Bekaert, G. and Harvey, C.R. (2002) ‘Research in emerging markets finance: looking to the future’,
Emerging Markets Review, Vol. 3, No. 4, pp.429–448.
Bekaert, G. and. Harvey, C.R (1995) ‘Time-varying world market integration’, The Journal of
Finance, Vol. 50, No. 2, pp.403–445.
 Analysing security performance in Morocco and South Africa using CAPM 201

Black, F. (1972) ‘Capital market equilibrium with restricted borrowing’, The Journal of Business,
Vol. 45, No. 3, pp.444–455.
Blume, M.E. (1975) ‘Betas and their regression tendencies’, The Journal of Finance, Vol. 30,
No. 3, pp.785–795.
Bundoo, S.K. (2008) ‘An augmented Fama and French three-factor model: new evidence from an
emerging stock market, the stock exchange of Mauritius’, Applied Economics Letters, Vol. 15,
No. 15, pp.1213–1218.
Carhart, M.M. (1995) Survivor Bias and Mutual Fund Performance, Working Paper, School of
Business Administration, University of Southern Carolina, Los Angeles.
Carhart, M.M. (1997) ‘On persistence in mutual fund performance’, The Journal of Finance,
Vol. 52, No. 1, pp.57–82.
Chaieb, I. and Errunza, V. (2007) ‘International asset pricing under segmentation and PPP
deviations’, Journal of Financial Economics, Vol. 86, No. 2, pp.543–578.
Chui, A.C.W. and Wei, K.C.J. (1998) ‘Book-to-market, firm size and the turn-of-the-year effect:
evidence from Pacific-Basin emerging markets’, Pacific-Basin Finance Journal, Vol. 6,
Nos. 3–4, pp.275–293.
Claessens, S., Dasgupta, S. and Glen, J. (1995) The Cross-Section of Stock Returns: Evidence from
Emerging Markets, Research Working Paper, WP 1505, World Bank, New York.
Coffie, W. and Chukwulobelu, O. (2012) ‘The application of capital asset pricing model (CAPM)
to individual securities on Ghana stock exchange’, Research in Accounting in Emerging
Economies, Vol. 12B, pp.121–147.
Dickey, D.A. and Fuller, W.A. (1979) ‘Distribution of estimators in autoregressive time series with
a unit root’, Journal of American Statistical Association, Vol. 74, No. 366, pp.427–431.
Dimson, E. (1979) ‘Risk measurement when shares are subject to infrequent trading’, Journal of
Financial Economics, Vol. 7, No. 2, pp.197–226.
Fama, E.F. and French, K.R. (1992) ‘The cross section of expected stock returns’, Journal of
Finance, Vol. 47, No. 2, pp.427–465.
Fama, E.F. and French, K.R. (1998) ‘Value versus growth: the international evidence’, Journal of
Finance, Vol. 53, No. 6, pp.1975–1979.
Fama, E.F. and French, K.R. (1993) ‘Common risk factors in the returns on stocks and bonds’,
Journal of Financial Economics, Vol. 33, No. 1, pp.3–56.
Fama, E.F. and MacBeth, J.D. (1973) ‘Risk, return and equilibrium: empirical tests’, The Journal of
Political Economy, Vol. 81, No. 3, pp.607–636.
Harvey, C.R. (2000) ‘The drivers of expected returns in international markets’, Emerging Markets
Quarterly, Fall, pp.1–17.
Hearn, B. and Piesse, J. (2009) ‘An augmented capital asset pricing model: liquidity and stock size
in African emerging financial markets’, African Journal of Finance, Special Ed., pp.27–57.
Hearn, B., Piesse, J. and Strange, R. (2010) ‘Market liquidity and stock size premia in emerging
markets: implications for foreign investment’, International Business Review, Vol. 19, No. 5,
pp.489–501.
IFC International Financial Statistics (2008) Emerging Markets Database [online]
http://www.ifc.org.
Jegadeesh, N. and Titman, S. (1993) ‘Returns to buying winners and selling losers: implications for
stock market efficiency’, The Journal of Finance, Vol. 48, No. 1, pp.65–91.
Jensen, M.C. (1968) ‘The performance of mutual funds in the period 1945–1964, the Journal of
Finance’, Vol. 23, No. 2, pp.389–416.
Jensen, M.C., Black, F. and Scholes, M. (1972) ‘The capital asset pricing model: some empirical
tests’, in Jensen, M.C. (Ed.): Studies in the Theory of Capital Markets, pp.79–121, Praeger,
New York.
Levy, M. and Roll, R. (2010) ‘The market portfolio may be mean-variance efficient after all’,
Review of Financial Studies, Vol. 23, No. 6, pp.2464–2491.
202 W. Coffie

Lintner, J. (1965) ‘The valuation of risk assets and the selection of risky investments in stock
portfolios and capital budgets’, Review of Economics and Statistics, Vol. 47, No. 1, pp.13–37.
Lyn, E. and Zychowicz, E. (2004) ‘Predicting stock returns in the developing markets of Eastern
Europe’, The Journal of Investing, Vol. 13, No. 4, pp.63–71.
Markowitz, H. (1952) ‘Portfolio selection’, The Journal of Finance, Vol. 7, No. 1, pp.77–91.
Markowitz, H. (1959) Portfolio Selection: Efficient Diversification of Investments, Wiley,
New York.
Michailidis, G., Tsopoglou, S., Papanastasiou, D. and Mariola, E. (2006) ‘Testing the capital asset
pricing model (CAPM): the case of the emerging Greek securities market’, International
Research Journal of Finance and Economics, Vol. 4, No. 1, pp.78–91.
Mossin, J. (1966) ‘Equilibrium in a capital asset market’, Econometrica, Vol. 35, No. 4,
pp.768–783.
Nel, W.S. (2011) ‘The application of the capital asset pricing model (CAPM): a South African
perspective’, African Journal of Business Management, Vol. 5, No. 13, pp.5336–5347.
O’Brien, J. (1999) ‘The global CAPM and a firm’s cost of capital in different currencies’, Journal
of Applied Corporate Finance, Vol. 12, No. 3, pp.73–79.
Patel, S. and Sarkar, A. (1998) ‘Crises in developed and emerging stock markets’, Financial
Analyst Journal, Vol. 54, No. 6, pp.50–59.
Pereira, L.E. (2005) ‘The practice of investment valuation in emerging markets: evidence from
Argentina’, Journal of Multinational Financial Management, Vol. 16, No. 2, pp.160–183.
Ramcharran, H. (2004) ‘Returns and pricing in emerging markets’, The Journal of Investing,
Vol. 3, No. 1, pp.45–55.
Reddy, T.L. and Thomson, R.J. (2011) ‘The capital asset pricing model: the case of South Africa’,
South African Actuarial Journal, Vol. 11, pp.43–84.
Reddy, T.L. and Thomson, R.J. (2013) ‘The capital-asset pricing model reconsidered: tests in real
terms on a South African market portfolio comprising equities and bonds’, South African
Actuarial Journal, Vol. 13, pp.221–263.
Reinganum, M.R. (1981) ‘A new empirical perspective on the CAPM’, Journal of Financial and
Quantitative Analysis, Vol. 16, No. 4, pp.439–462.
Roll, R. (1977) ‘A critique of the asset pricing theory’s tests: part I’, Journal of Financial
Economics, Vol. 4, No. 2, pp.120–176.
Ross, S.A. (1976) ‘The arbitrage theory of capital asset pricing’, Journal of Economic Theory,
Vol. 13, No. 3, pp.341–360.
Rouwenhorst, K.G. (1999) ‘Local returns factors and turnover in emerging stock markets’, Journal
of Finance, Vol. 54, No. 4, pp.1439–1464.
Schramm, R.M. and Wang, H.N. (1999) ‘Measuring the cost of capital in an international CAPM
framework’, Journal of Applied Corporate Finance, Vol. 12, No. 3, pp.63–72.
Sharpe, W.F. (1964) ‘Capital asset prices: a theory of market equilibrium under conditions of risk’,
Journal of Finance, Vol. 19, No. 3, pp.425–442.
Stulz, R.M. (1995) ‘International portfolio choice and asset pricing: an integrative survey’, in
V. Maksimovic and W. Ziemba edition, The Handbook of Modern Finance, North Holland.
Stulz, R.M. (1999) ‘Globalization, corporate finance and the cost of capital’, Journal of Applied
Corporate Finance, Vol. 12, No. 3, pp.8–25.