WHAT DRIVES THE SHAREHOLDER VALUE?

Text-only Preview

AAMJAF, Vol. 1, 105–120, 2005
ASIAN ACADEMY of
MANAGEMENT JOURNAL
of ACCOUNTING
and FINANCE





WHAT DRIVES THE SHAREHOLDER VALUE?*

I. M. Pandey

Indian Institute of Management, Ahmedabad, India

Corresponding author: [email protected]


ABSTRACT

In the strategy literature a lot of emphasis is placed on growth as a dominant business
strategy. Is growth always desirable? The finance literature, on the other hand, focuses
more on economic profitability and value. This study empirically explores the
significance of profitability and growth as drivers of shareholder value, measured by the
market-to-book value (M/B) ratio. Profitability is defined as economic profitability; that
is, the spread between return on equity and the risk-adjusted cost of equity. Using panel
data and employing the Generalized Method of Moments (GMM) estimator, our findings
show a strong positive relationship between economic profitability and M/B ratio.
Growth, on the other hand, is negatively related to M/B ratio. However, the economic
profitability-growth interaction variable has a positive coefficient indicating that growth
associated with economic profitability influences shareholder value positively. This
finding is further supported when we analyse the relationships separately for the positive-
spread firms and negative-spread firms. Our results also indicate negative relationship
between M/B ratio and firm size and positive relation with business risk, financial risk
and capital intensity.

Keywords: shareholder value, economic profitability, EVA, spread, GMM





* Acknowledgement: This article is the revised version of the paper titled, Economic Profitability,
Sustainability Growth and Shareholder Value, presented in Annual Conference of the
International Digital Business Academy, September 25–28, 2005, Las Vegas, U.S.A. The author
expresses his gratitude to the Conference participants and an anonymous referee for their useful
suggestions.
105

I. M. Pandey
INTRODUCTION

In the recent years, there has been considerable managerial interest in shareholder
value and its management. One reason for this increased interest is the shift in
focus from accounting profitability to economic profitability. The most popular
measure of accounting profitability is return on equity (ROE). Economic
profitability, on the other hand, also considers the cost of equity (ke); it is a
spread between ROE and ke. The concept of economic profitability is equivalent
to the concept of economic value added (EVA). The spread between ROE and ke
multiplied by equity capital gives total EVA amount.

In strategic management literature growth is considered a most desirable
strategy. Growth is considered necessary as it makes a firm big, opens up
managerial opportunities for career advancements and gives pride to managers.
Growth for the sake of becoming big has been questioned by many in academics
and practice as well. In the quest for becoming big, firms may compromise
profitability. There are, therefore, practitioners who aim at making their
companies more profitable than bigger. The pursuit for growth perhaps implicitly
assumes that it drives a firm's profitability and value. But this is an empirical
question.

In this article, we explore two issues which are of considerable
theoretical and practical interest. Does growth enhance the value creating
potential of a firm? Does economic profitability help in creating shareholder
value? To explore these questions, we start with an approach similar in nature
with the study of Varaiya, Kerin and Weeks (1987) in the U.S. context. However,
our study differs in many major ways. First, we use panel data of 220 firms over
nine year period from an emerging market. Second, we employ an improved
model specification and introduce a number of control variables which influence
shareholder value. Third, our findings are based on the robust Generalized
Method of Moments (GMM) estimator.


SHAREHOLDER VALUE MODEL

A most common measure of the shareholder value creation is the comparison
between the market value and book value per share. When the market value
exceeds the book value, the shareholder value is created and when the book value
exceeds the market value, the shareholder value is destroyed.

A simple valuation model that can be used to make predictions about the
relationship between profitability and growth and shareholder value is the
106

What Drives the Shareholder Value?
constant-growth model. The market value of a share (M) is given as follows
(Brealey & Myers, 2003):

DPS
EPS(1− b)

M =
=

(1)
k −g
k −g
e
e


This model assumes that dividends grow at a constant rate in perpetuity.
Dividend per share (DPS) is equal to earnings per share (EPS) multiplied by one
minus retention ratio (b). EPS depends on the firm's return on equity (ROE) and
the equity investment, expressed as book value of per equity share (B). Eq. (1)
can be rewritten as follows:

B×ROE(1− b)
B(ROE − b×ROE)
M =
=
k − g
k − g
e
e

(2)
M
ROE − g
=

B
k − g
e

Eq. (2) implies that shareholder value will be created when market-to-
book (M/B) ratio is greater than 1, and value will be destroyed if it is less than 1.
We may further rewrite Eq. (2) as follows (Varaiya et al., 1987):

M
ROE − k

e
= 1 +
(3)
B
k − g
e

Eq. (3) indicates that M/B will be greater than 1 if ROE exceeds ke; that
is, the spread, ROE – ke, is positive. Both Eq. (2) and (3) assume that in
equilibrium ke is greater than g. However, this is not a necessary condition to
empirically test the effect of g on M/B ratio. The cost of equity is the risk-
adjusted return that shareholders require on their investment. Hence, a firm will
be creating value for its shareholders when it undertakes investments that
generate positive spread; which is, return on equity exceeding the cost of equity
(ROE > ke). It should be clear from this reasoning that a positive ROE alone is
not enough for creating shareholder value. A number of firms providing positive
ROE may in reality be destroying value if their cost of equity exceeds ROE. The
approach of focusing on the spread considers the quality of earnings – earnings
after adjusting for the risk-adjusted cost of equity.

Many researchers have argued that the appropriate measure of a firm's
profitability is the spread between ROE and ke, which may be referred to as
economic profitability (Hax & Majluf, 1984). We may notice from Eq. (3) that
107

I. M. Pandey
growth resulting from earnings reinvestment may affect shareholder value
depending on whether ROE is greater than or lower than ke. Thus, an interaction
between profitability and growth is indicated.

There are a number of studies on the relationship between growth,
profitability and value in the context of U.S.A. In a study of the U.S. companies,
Varaiya et al. (1987) find that both profitability and growth influence shareholder
value, but profitability has a greater impact. However, their results in case of
negative-spread earning firms are not very strong. This study omitted several
other variables that are expected to influence shareholder value, and hence it is
suspected that the results of the study may be biased. In an earlier study, Woo
(1984) finds similar results. More recently, Ramezani, Soenen and Jung (2002)
explore the relationship between growth (earnings or sales) and profitability
(measured as economic value added, EVA) and between profitability and
shareholder value. They use Jensen's alpha as a measure of shareholder value and
include several control variables in their estimation. They find that beyond a
point, growth adversely affects profitability and destroys shareholder value.


DATA AND METHODOLOGY

We use data of companies listed on Bursa Malaysia (formerly known as the
Kuala Lumpur Stock Exchange) for the period from 1994 to 2002. Companies
with missing data are excluded from the study. We also exclude the financial and
securities sector companies as their financial characteristics and use of leverage is
substantially different from other companies. We also drop companies with
negative equity. Our final sample size is 220 companies for each period. We
adjust data of those companies, which change their financial year. Such changes
result in one year with missing data and the subsequent year data of more than 12
months. We first annualise the subsequent year data, and then substitute missing
data by the mean value.

Our estimation model uses panel data. Panel data, unlike cross-section
data, allows controlling for unobservable heterogeneity through individual (firm)
effect – ηi (Baltagi, 2005). We also include dummies for time variable to measure
temporal effect (γt). This helps in controlling the effect of macro-economic
variables on the shareholder value. Thus, we use a two-way effects model as
follows:

(M/B) = β + β (ROE − k ) + β g + β (ROE − k ) g + β Ln TA

it
0
1
e it
2 it
3
e it
it
4
it (4)
+ β (TD/TA) + β Beta + β (FA/TA) + η + γ + ε
5
it
6
it
7
it
i
t
it

108

What Drives the Shareholder Value?
Given that the growth rate, g, is held constant, a higher spread,
(ROE – ke), implies a higher M/B ratio. Hence, the sign of β1 is expected to be
positive. However, the sign of β2 is unspecified.


VARIABLES

Three variables of direct interest in this study are market-to-book (M/B) ratio,
economic profitability or spread (ROE – ke) and growth (g). From Eq. (2) and
(3), we can predict that economic profitability, (ROE – ke), has a positive effect
on M/B ratio. It may be observed from Eq. (2) that g is subtracted from ROE in
numerator and from ke in denominator. If we consider numerator, M/B ratio
should decrease when g increases, other things remaining the same. On the other
hand, if we consider denominator, M/B ratio should increase if g increases. We
do not make any prediction about growth (g) variable. We can, however, predict
that, other things remaining the same, a combination of higher profitability and
growth should increase value. Hence, we may hypothesise that the interaction
between economic profitability and growth will have a positive effect on M/B
ratio. The dependent variable and three regressors in our estimation Eq. (4) are
defined below:

Market-to-book (M/B) ratio is the year-end closing market value of a
share divided by the closing book value of the share. This is the
dependable variable in our model.

Growth (g) is the sustainable growth based on a firm's financial policies.
The sustainable growth for each year is calculated ad follows:

g = [ROA + (ROA − i)D / E] 1
( − payout)

ROA = after-tax return on assets; i = after-tax interest rate on debt;
D/E = debt-equity ratio and (1 – payout) = retention ratio. Simplifying
the equation, the sustainable growth is equal to return on equity (ROE)
multiplied by retention ratio, that is: g = ROE × (1 – payout).

Hence, it is a proxy for the expected growth in the future. Sustainable
growth, rather than the growth based on past earnings data, is considered
a good proxy of the future growth potential of a firm.

Economic profitability (spread) is defined as the difference between
return on equity and cost of equity (ROE – ke). ROE is calculated as
profit net of all expenses and taxes and excluding all extra-ordinary items
109

I. M. Pandey
divided by the net worth (book value equity). Cost of equity (ke) is
calculated using the capital assets pricing model (CAPM). The estimation
of the equity beta is based on the daily closing share prices. The risk-free
rate is the monthly T-bill rate of the year under consideration. The risk
premium is calculated as the difference between average annual market
return and the average annual T-bill rate based on the monthly data from
the period January 1978 to December 2002.

Economic profitability-growth interaction variable is used in our
model to capture the joint effect of economic profitability and growth.

In practice, M/B ratio may be affected by several other factors. In a
number of studies of firm performance and value, a number of control variables
have been used and found significant (Ramezani et al., 2002; Perez-Quiros &
Timmerman, 2000; Opler, Pinkowitz, Stulz & Williamson, 1999; Philips, 1999;
Campbell & Shiller, 1998; Woo, 1984). Our model may suffer from
misspecification if we omit known observable regressors. The control variables
used in this study are defined as follows:

Size is measured as log of total assets (Ln TA). Total assets are used as
the measure of size as they reflect the firm's competitive strength and
staying power in the market.

Business risk measured by asset beta is used to account for the volatility
in a firm's earnings and value. Asset beta is calculated as the equity beta
of a firm multiplied by total assets, that is βE × TA/E. Total assets are
used as proxy for the firm's value.

Leverage is calculated as the book value of total debt divided by total
assets (which is equal to the book value of equity plus total debt
(TD/TA). It is used as a proxy for the financial risk.

Free cash flows ratio measured as free cash flows divided by total assets
(FCF/TA) captures the influence of cash flows on a firm's value. Free
cash flows include net profit (excluding extra-ordinary items) plus
depreciation.

Capital intensity (FA/TA) is defined as fixed assets divided by total
assets (FA/TA). This variable is used as a proxy for a firm's operational
flexibility.

110

What Drives the Shareholder Value?
MODEL SPECIFICATION

We have panel data; hence, our model can be estimated in several ways. These
methods are: (1) the ordinary least square (OLS); (2) the fixed effects models;
and (3) the random effects models. The fixed or random effects models may be
used with firm effects, or with time effects or with both firm and time effects.
The appropriate technique depends upon the structure of the error term, εit, and
the correlation between the components of the error term and the observed
determinants of the dependent variable (Baltagi, 2005), which is M/B ratio in our
study. If no firm- or no time-specific effects are anticipated, OLS is an
appropriate method of estimation. However, we may expect that both
unobservable firm-specific and unobservable time-specific factors may have an
effect on our dependent variable – M/B ratio. Both the fixed and random effects
models handle unobservable effects, but they adopt different assumptions and
approaches. The random effects estimation includes unobservable effects in the
error term. However, the results will be biased if the error term has correlation
with one or many of the regressors. To deal with this problem, a dummy variable
for each firm may be included in the estimation. This approach causes loss of
degrees of freedom, and hence, it is considered less efficient.

The fixed or random effects models with firm or time or both effects can
be estimated. We should first determine if there is any evidence of firm and/or
time effects. For this purpose, we can test the joint significance of the firm and/or
the time effects in the fixed effects models. If we find the evidence of these
effects, we should determine if firm and/or time dummy variables are correlated
with observable regressors. We can use the Hausman (1978) test for this purpose.
This test will indicate the choice between fixed or random effects model.

The fixed effect models control for unobservable heterogeneity, but they
give biased results if the models include endogenous variables. In order to
resolve this problem, we use the GMM, which controls the endogeneity problem
by using instrumental variables. In the model estimation, this study uses all
regressors plus an exogenous variable – sales-to-total assets ratio – as
instrumental variables. We use levels of all variables except the economic
profitability (ROE – ke) which is lagged t – 2. Further, to eliminate the individual
effects, the study uses the cross-section first differences of variables in the model
estimation.


RESULTS

Table 1 provides mean, median and standard deviation of the dependent and the
independent variables for each year from 1994 to 2002. The average M/B value
111

I. M. Pandey
TABLE 1
DESCRIPTIVE STATISTICS

Year
1994 1995 1996 1997 1998 1999 2000 2001 2002 ALL
Market-book ratio (M/B)

Mean
3.880 3.090 3.299 1.168 1.285 1.489 1.007 1.120 0.976 1.924

Median
2.975 2.312 2.459 0.655 0.787 0.986 0.712 0.728 0.629 1.187

Std.
Dev.
3.677 3.242 2.812 2.025 2.333 1.948 1.357 1.710 1.382 2.633
Spread (ROE – ke)

Mean
–0.089 –0.063 –0.103 –0.029 –0.106 –0.137 –0.101 –0.138 –0.121 –0.099

Median
–0.104 –0.058 –0.110 –0.035 –0.079 –0.097 –0.066 –0.090 –0.077 –0.075

Std.
Dev.
0.148 0.178 0.158 0.143 0.281 0.308 0.198 0.313 0.359 0.246
Growth (g)

Mean
0.059 0.071 0.068 0.049 –0.022 –0.023 –0.015 –0.052 –0.042 0.010

Median
0.053 0.069 0.059 0.052 0.017 0.018 0.012 0.004 0.011 0.034

Std.
Dev.
0.110 0.110 0.074 0.112 0.225 0.263 0.177 0.279 0.349 0.214
Size (log TA)

Mean
12.823 13.048 13.284 13.539 13.602 13.618 13.633 13.608 13.621 13.420

Median
12.812 13.058 13.310 13.532 13.625 13.628 13.715 13.629 13.680 13.417

Std.
Dev.
1.250 1.271 1.283 1.265 1.275 1.242 1.229 1.248 1.265 1.288
Leverage (TD/TA)

Mean
0.153 0.161 0.170 0.201 0.228 0.222 0.226 0.225 0.219 0.201

Median
0.116 0.129 0.158 0.185 0.238 0.205 0.223 0.198 0.204 0.176

Std.
Dev.
0.148 0.153 0.151 0.164 0.177 0.184 0.187 0.196 0.186 0.175
Asset beta (Beta)

Mean
0.978 0.874 1.093 0.637 0.646 0.706 0.551 0.610 0.596 0.743

Median
0.988 0.870 1.081 0.599 0.626 0.632 0.498 0.515 0.510 0.671

Std.
Dev.
0.418 0.417 0.602 0.343 0.344 0.378 0.320 0.423 0.431 0.452
Free cash flow ratio (FCF/TA)

Mean
0.079 0.084 0.081 0.068 0.027 0.044 0.050 0.024 0.034 0.055

Median
0.070 0.081 0.072 0.071 0.044 0.049 0.045 0.038 0.042 0.058

Std.
Dev.
0.132 0.054 0.051 0.075 0.109 0.104 0.079 0.155 0.092 0.102
Capital intensity (FA/TA)

Mean
0.360 0.361 0.358 0.351 0.365 0.355 0.355 0.364 0.363 0.359

Median
0.348 0.330 0.341 0.325 0.333 0.324 0.327 0.341 0.344 0.333

Std.
Dev.
0.240 0.242 0.236 0.231 0.232 0.232 0.234 0.244 0.246 0.237


for the year 1994 is 3.88 and for 2002, it is 0.98. M/B ratio has shown
continuous decline over years. It is noticeable that the average economic
profitability or (ROE – ke) spread is negative for all years from 1994 to 2002, and
this negative spread was larger post 1998. Growth rate (g) and ROE have shown
fluctuations during the 1994–2002 period. Like M/B ratio, they have generally
112

What Drives the Shareholder Value?
been lower after 1997, corresponding with the financial and stock market crisis in
Malaysia. Beta and cost of equity declined over years while leverage (TD/TA)
showed some increase. Average capital intensity (FA/TA) remained remarkably
more or less at the same level through out the period of 1994–2002. Like ROE,
free cash flow as a percentage of total assets showed a declining trend.

Table 1 also shows average value of variables over the 9-year period. The
average M/B ratio for the entire 9-year period was 1.92, indicating value creation
over and above the book value. The average spread or economic profitability has
been negative since average cost of equity (ke) was about 14% and average ROE
only 4%. The average sustainable growth was 1%, debt ratio (leverage) 20%,
asset beta 0.74, fixed assets-to-total assets ratio 0.36 and FCF-to-total assets ratio
of 5.5%.

Our concern in this study is on the relationship of spread and growth with
shareholder value as reflected by M/B ratio. Our data show a consistent pattern of
a high proportion of firms with negative economic profitability (spread) – almost
80% in each year from 1994 to 2002. Other studies have also found a high
proportion of negative spread firm (Varaiya et al., 1987; Fruhan, 1984). In Table
2, Panels (A) and (B), we report mean and median values for M/B ratio and
growth separately for the group of firms with positive spread and those with
negative spread. In Table 2, Panel (A), mean values of growth and M/B ratio
show a distinct pattern. For the group of positive-spread firms, the average annual
growth rates and M/B ratios are higher than the group of negative-spread firms.
Noticeably, average M/B ratios are 2 and more in all years. This implies that
high-growth coupled with high-spread leads to higher shareholder value. For
negative-spread firms, average annual growth rates are low and negative after
1998. The average M/B ratios in combination with negative spread but positive,
though low, growth rates are more than 2 for the period 1994–1997. For the
period from 1999 to 2002, the annual growth rates are negative. The negative
spread and negative growth rates cause erosion in shareholder value during this
period; the average annual M/B ratios are less than 1 during the 1999–2002
period. Similar conclusions emerge from median values of M/B ratios, growth
and spread in Table 2, Panel (B). It is also noticeable that the number of negative
spread firms has not varied significantly over years. Thus, it seems that it is quite
difficult for these firms to move from negative to positive spread group.







113

I. M. Pandey
TABLE 2
MEAN AND MEDIAN VALUES FOR NEGATIVE AND POSITIVE SPREAD
SAMPLES OF FIRMS

Year Obs. (R
<
ke)
g
M/B
Obs. (ROE > ke) g M/B
Mean value
1994 170 –0.144
0.039 3.641 50
0.096 0.125 4.690
1995 160 –0.122
0.045 2.659 60
0.094 0.141 4.241
1996 173 –0.157
0.052 2.887 47
0.098 0.126 4.815
1997 155 –0.157
0.052 2.887 65
0.098 0.126 4.815
1998 172 –0.081
0.020 0.740 48
0.095 0.116 2.188
1999 180 –0.186 –0.050 1.178 40
0.086 0.098 2.888
2000 177 –0.144 –0.044 0.736 43
0.077 0.102 2.119
2001 186 –0.183 –0.077 0.913 34
0.106 0.083 2.254
2002 175 –0.176 –0.079 0.701 45
0.096 0.100 2.047
Median value
1994 170 –0.130 0.038 2.734 50
0.060 0.141 3.826
1995 160 –0.104 0.048 2.020 60
0.064 0.131 3.186
1996 173 –0.150 0.049 2.207 47
0.062 0.112 3.467
1997 155 –0.062 0.035 0.525 65
0.052 0.109 1.191
1998 172 –0.107 0.006 0.703 48
0.032 0.097 1.341
1999 180 –0.129 0.006 0.918 40
0.038 0.108 1.835
2000 177 –0.093 0.005 0.593 43
0.034 0.093 1.396
2001 186 –0.106 0.000 0.643 34
0.049 0.090 1.396
2002 175 –0.104 0.002 0.525 45
0.043 0.085 1.324

Table 3 provides correlation matrix for the pooled sample of 1880
firms/years observations.1 We notice that none of the pairs of variables has a very
high correlation except economic profitability and growth. M/B ratio is
negatively correlated with size, leverage and capital intensity and positively with
economic profitability and beta; all pair correlation coefficients are small.
Economic profitability has positive correlation with growth, size, and capital
intensity; it has negative correlation with leverage and beta. Growth has positive
correlation with economic profitability and free cash flow ratio. The negative
relationship between risk (beta) and size implies that the large firms, being more
diversified, have lower risk. Generally low correlations among variables indicate
that the estimation equation may not have a serious problem of multi-colinearity.





1 Correlation coefficients, based on pooled OLS, do not control for firm and time effects.
114

Document Outline

  • WHAT DRIVES THE SHAREHOLDER VALUE?*
        • I. M. Pandey
              • Indian Institute of Management, Ahmedabad, India
        • ABSTRACT
    • INTRODUCTION
    • SHAREHOLDER VALUE MODEL
    • DATA AND METHODOLOGY
      • VARIABLES
    • MODEL SPECIFICATION
    • RESULTS
                  • Year
                • ALL
            • Mean value
            • Median value
          • M/B
            • Variable
    • CONCLUSION
    • REFERENCES