focuses on the analysis and resolution of managerial issues based on analytical

Research Article and empirical studies.

Linkage between Economic Value Added and Market Value:

An Analysis
Ashok Banerjee
Introduction
Maximizing shareholder value has become the new
corporate paradigm. Although shareholder wealth
maximization has traditionally been recognized by
managers and researchers as the ultimate corporate
goal, the maxim has gained new dimension in recent
years, thanks to the concept of Economic Value
Added (EVA) coined and registered by Stern Stewart
Maximizing shareholder value has become the & Co, New York. EVA is a residual income that
new corporate paradigm. Corporations in the subtracts the cost of capital from the operating profits
US have started disclosing EVA information generated by a business. Corporations in the US
from the beginning of 90s as a measure of started disclosing EVA information from the begin-
corporate performance. It is believed that ning of 90s. Since then, the number of companies
market value of a firm (hence shareholder adopting EVA has increased (Wallace, 1997). More
wealth) would increase with the increase in than 300 companies, with revenue approaching a
EVA. Various studies done in the US also trillion dollars a year, have implemented EVA
confirm this belief. EVA (a term coined and framework for financial management and incentive
registered by Stern Stewart & Co. New York) compensation (Ehrbar, 1998). Adopting EVA philo-
is a residual income that subtracts the cost of sophy forces a company to find ingenious ways to
capital from the operating profits generated by do more with less capital (Tully, 1993).
a business. The present study makes an at
This does not mean EVA concept retards growth.
tempt to find the relevance of Stewart's claim
It only suggests that so long as a company is earning
that market value of the firm is largely driven
a return on its investment in excess of the cost of
by its EVA generating capacity in the Indian
investment, there is no limit to growth. It is only
context. Based on a sample of 200 firms over
when the earning is insufficient to meet the cost of
a period of five years, the study shows that
funds tied up, there arises a need to unlock the fund
market value of a firm can be well predicted
and thereby avoid or minimize bad or uneconomic
by estimated future EVA streams. The study
investments. EVA is a modified version of share-
has also found that market value of most of
holder value theory. The shareholder value theory
the firms in the sample is explained more by
places shareholders at the top in analysing the
current operational value than future growth
economic performance of a business. The share-
value of firms.
holder value approach (Rappaport, 1986) estimates
the economic value of an investment by discounting
Ashok Banerjee is Associate Professor in the Accounting forecasted cash flows by the cost of capital. These
and Finance Area of the Indian Institute of Management, cash flows, in turn, serve as the foundation for
Lucknow. shareholder returns from dividend and share price
appreciation. EVA is different from shareholder
value theory in the sense that it deducts depreciation
to compute its residual income and also it makes
certain adjustments to convert accounting profit to
economic profit. It is believed that market value of
a firm, at any given moment, is the summation of

Vol. 25, No. 3, July-September 2000 23 Vikalpa

beginning invested capital and present value of maximize ROCE, it may reject the project. But,
future stream of expected EVAs. Stewart (1991) actually, the project would have added two per cent
emphasized that to get significant benefits, EVA economic surplus to the wealth of the firm. Consider
should be fully integrated into a company linking another example. Suppose the present ROCE of the
executive compensation to improvement in EVA. firm is ten per cent and cost of capital 16 per cent.
Stewart maintained that if executives' bonus and The firm receives a new investment proposal with
other incentives were linked to traditional para- an estimated ROCE of 12 per cent, with no change
meters [e.g., Earnings per share (EPS), turnover, in cost of capital. The firm would accept the proposal
Return on Net Worth (RONW), etc.], EVA would to maximize ROCE. But this decision would destroy
fail as a performance measure. This is because the firm's wealth. EVA compares ROCE with the
corporate executives, in that case, would have no cost of invested capital and a firm, with the objective
incentive to maximize EVA. EVA, thus, is not merely of EVA maximization, would accept all fresh invest-
a financial computation reported at the end of the ment proposals so long as the expected spread is
year but is a part of the fully integrated management positive.
system. Implementation of EVA system in an or-
ganization takes a long time as EVA does not mean
only laying down measurement or computational Definition of EVA
techniques. Stewart mentions in Ehrbar (1998) that
there are four Ms in the implementation process — EVA essentially seeks to measure a company's actual
Measurement, Management System, Motivation, and rate of return as against the required rate of return.
Mindset. Thus, Stewart (1991) argues that market To put it simply, EVA is the difference between Net
value of a firm is largely driven by its EVA generating Operating Profit after Tax (NOPAT) and the capital
capacity. The present study makes an attempt to find charge for both debt and equity (overall cost of
the relevance of Stewart's claim in the Indian capital). If NOPAT exceeds the capital charge, EVA
context. The relationship between EVA and market is positive and if NOPAT is less than capital charge,
value is tested on a sample of 200 companies. The EVA is negative.
results of the study confirm Stewart's claim. The definition of EVA can be mathematically shown
It is believed that EVA is a better performance as below:
measure than traditional measures like Earning per
Share (EPS), Return on Capital Employed (ROCE), EVA = NOPAT- Capital Charge ............. ( Eq. 1)
or Return on Net Worth (RONW). EPS depends = NOPAT- (WACC * Invested Capital)
largely on the vagaries of accounting policies fol- = (r * Invested Capital)-(c * Invested Capital)
lowed by a firm. Thus, EPS is as much reliable as EVA = (r-c) * Invested Capital
the accounting profit. Accounting profit (PAT) de- EVAt = (r-c) * Invested Capital (M) ................... (Eq.2)
pends, inter alia, on the firm's capital structure. A
Where,
lowly geared firm would return a higher PAT than
a highly geared firm, given the same level of WACC = Weighted Average Cost of
operating profit earned by both the firms. In Capital
computing accounting profit, only one part of cost Invested Capital = Invested Capital at the
of capital (i.e., borrowing cost) is deducted. As a Beginning of the Year
result, PAT does not reflect the true economic profit. r = NOPAT/Invested Capital
EVA, on the other hand, is the residual profit after c = WACC
deducting full cost of capital from operating profits. t = Time Period
ROCE or RONW considers only one side of the
performance. Exclusive reliance on ROCE or RONW
The spread (r-c) shows whether a company has
may lead to rejection of economically profitable
earned a return from its business that is more than
projects or acceptance of unviable projects. Both
its total cost of capital. If the spread is positive, EVA
would lead to destruction of shareholder value.
would also be positive. The logic for taking begin-
Consider a firm with a present ROCE of 22 per
ning invested capital for calculating periodic EVA
cent and an overall cost of capital of 18 per cent.
is that a company would at least take a year's time
The firm receives a new investment proposal with
to earn a return on investment. Given a particular
an estimated ROCE of 20 per cent, cost of capital
level of spread, EVA would depend on the beginning
remaining unchanged. If the firm's objective is to
invested capital. Given a particular level of invested

Vol. 25, No. 3, July-September 2000 24 Vikalpa

capital, EVA would depend on spread. Thus, there However, the actual number of adjustments
are two factors that drive EVA — the spread and the would depend on prevailing GAAP of a country.
invested capital. The spread denotes the relative In order to avoid complexity in the calculation of
profitability and invested capital denotes the size or NOPAT, Stewart (1991) suggested four common
growth. If a company has negative profitability (i.e., adjustments to be made — Adjustments for Deferred
spread), growth in size would reduce EVA. To reduce Tax Reserve, Last-in-First-Out (LIFO) Reserve, Goodwill
the impact of negative EVA, invested capital should Amortization and R&D Cost Amortization. These items
be economized. On the other hand, if the spread are called Equity Equivalents.* Equity Equivalents are
is positive, growth in firm size would indicate higher added to invested capital and periodic change is
EVA. However, it is true that for skill-based com- taken to NOPAT. These adjustments make NOPAT
panies (e.g., companies in the Information Technol- a realistic measure of yield generated for investors
ogy sector) growth does not involve commensurate for recurring business activities. It is believed that
increase in invested capital. This may prompt some these adjustments would truly convert accounting
people to conclude that EVA would not be a useful profit to economic profit.
variable to explain stock price movements of a
research-based or skill-driven company. But, Stewart However, for the purpose of this study, the
(1991) defended EVA on this count. Thus, the computation of NOPAT has been further modified.
message of the EVA formula (Eq. 2) is that if the NOPAT has been defined as below:
return (r) of a company is not adequate enough to
cover the cost of capital (c) in full, more investment NOPAT = PBIT (nnrt) * (1-T)
in the business would mean more negative EVA. In
such a situation, the company should try to either Where, PBIT (nnrt) = Profit Before Interest and
increase the 'r' or reduce the capital invested to Taxes (net of non-recur-
improve EVA. The idea behind EVA is that share- ring transactions)
holders must earn a return that compensates the risk
= Profit After Tax (PAT)+
taken. A zero EVA indicates that the return earned
Provision for Tax + Inter-
is just sufficient to compensate the risk. EVA holds
est Expense + Lease Rent-
a company accountable for the cost of capital it uses
Extraordinary Income+
to expand and operate its business and attempts to
show whether a company is creating real value for Extraordinary Expenses.
its shareholders. T = Effective Tax Rate (Provision for Tax/PBT).

Stewart (1991) defines NOPAT as the "profits Invested capital refers to total assets (net of
derived from the company's operations after taxes revaluation) net of non-interest bearing liabilities.
but before financing costs and non-cash-book keep- From an operating perspective, invested capital can
ing entries." But, in eliminating the impact of "non- be defined as Net Fixed Assets (i.e. net block), plus
cash-book keeping" entries, Stewart makes an ex- investments plus Net Current Assets. Net Current
ception. Depreciation is subtracted to arrive at Assets denote current assets net of Non-Interest-
NOPAT. Stewart argues that depreciation is sub- Bearing Current Liabilities (NIBCLS). From a financ-
tracted because it is "a true economic expense." In ing perspective, the same can be defined as Net
other words, NOPAT is equivalent to income avail- Worth plus total borrowings. Total borrowings denote
able to shareholders plus interest expenses (after tax). all interest bearing debts. Stewart (1991) mentioned
It may be noted that Stewart has considered regular that adjustments for four Equity Equivalents men-
non-operating income (e.g., interest/dividend on tioned above should be made. The adjustments for
investment) as part of NOPAT. This is a deviation Equity Equivalents are intended to arrive at the
from traditional definition of operating profit. Also, economic value of invested capital. Equity Equiva-
to compute NOPAT properly, Stewart identified 120 lents eliminate accounting distortions. Net worth is
adjustments (Ehrbar, 1998) to be made to accounting defined as paid up share capital plus reserves and
profit as reported in the profit and loss account. surplus (net of revaluation reserves) less miscellane-
These adjustments, it is argued, would eliminate ous expenditure less accumulated losses, if any. One
potential distortions in accounting results based on
Generally Accepted Accounting Principles (GAAP) 'For a detailed discussion on Equity Equivalents and their
of a country. treatment, interested readers may refer to Stewart, B III (1991),
The Quest for Value, Harper Business Publications.

Vol. 25, No. 3, July-September 2000 25 Vikalpa

may argue that this method of calculating invested the parameters in the computation of EVA, we now
capital is not free from depreciation distortions. look at the last factor i.e., WACC.
Since net block of depreciable assets is considered,
WACC has been defined in the study to include
different corporate depreciation policy would affect
three specific costs, viz., cost of equity shares, cost of
the invested capital and hence EVA. Stewart (1991)
preference shares, and cost of borrowings (debt).
tackles it by prescribing a uniform method of
charging depreciation. He mentions that a straight- Cost of Debt (Kd) is calculated -by multiplying the
line depreciation or annuity method of depreciation pre-tax debt cost with (1-t). It may be noted that
would minimize the distortions. Such adjusted in- 't' denotes the effective tax rate.
vested capital (after adjusting for Equity Equivalents Cost of Preference Shares (Kp)= (Preference dividend/
and depreciation) would be called economic capital. Beginning preference share capital) * 100. Corporate
However, invested capital for the purpose of the dividend tax has not been considered because it was
study is defined as follows: not in vogue during the period under study.

Invested Capital = Net Worth + Total Borrowings Cost of Equity (Ke) is an opportunity cost equal to
the total return that an investor in a company's equity
Where, Net Worth = Share Capital + Reserves and could expect to earn from alternative investments
Surplus - Revaluation Re- of comparable risk. Cost of equity is not an explicit
serve-Accumulated Losses - cost like cost of debt. The dividend-based approach
Miscellaneous Expenditure or earning-based approach of finding out cost of
Total Borrowings = Long-term Interest-bearing equity is not a valid way of calculating the return
Debt + Short-term Interest- expected by equity shareholders. These approaches
bearing Debt only measure the explicit cost of servicing equity.
But, the true measure of equity cost is not what a
company offers but what investors expect. The
Adjustments for Equity Equivalents are not opportunity cost of equity capital has been calculated
considered in the present study because these are by following Capital Asset Pricing Model (CAPM)*
largely non-existent or inapplicable in Indian con- (Sharpe, 1964).
ditions (Banerjee, 1999).
Thus, the computational methodology of EVA EVA and Net Present Value (NPV)
is not unique. Ehrbar (1998) talked about an EVA
spectrum. At one extreme is what is called "Basic It is widely tested that the value of a firm is given
EVA." This is a rudimentary form of EVA arrived by the present value of future stream of free cash
at without making any adjustments. Then follows flows. Cash flow is the value driver. Of course, cash
"Disclosed EVA." It is the EVA computed by Stern flow also depends on certain operating value drivers.
Stewart & Co to rank companies. "Disclosed EVA" is The NPV method of measuring firm value is used
computed by making about "a dozen of standard by Rappaport (1986) in defining shareholder value
adjustments to publicly available accounting data." of a firm. EVA proponents claim that the firm value
Next is "Tailored EVA." An insider can calculate this can be measured by discounting future EVAs instead
EVA by making tailor-made adjustments peculiar to of future cash flows. A question may naturally arise
the organization concerned. At the other extreme - will the firm value differ under EVA and cash flow
of the spectrum is "True EVA."This is the theoreti- approaches? As Table 1 illustrates, the life-time value
cally correct and accurate measure of EVA calculated of the firm would be the same in the EVA method
with all relevant adjustments to accounting data and of valuation as in the NPV method. We take a simple
using the precise cost of capital of each division of
an organization. It is extremely difficult to compute *CAPM recognizes the risks associated with equity instruments
"True EVA." and proposes that an investor in this instrument would expect
a risk premium over and above the risk-free rate of return
Truly speaking, "Tailored EVA" is the ideal EVA prevailing in the market. Such a risk premium woutd depend
measure. But, it is difficult for an outsider to use on the volatility of returns of the equity scrip vis-a-vis that of
this definition of EVA for sheer lack of information. market (usually represented by an index). Higher the volatility,
greater would be the risk premium. According to CAPM, the
Therefore, in the present study, EVA has been expected return on equity (i.e., opportunity cost of equity capital)
calculated in a manner that lies in between "Basic = Rf + P[E(RJ - R(], where f$ represents volatility, Rf the risk-
EVA" and "Disclosed EVA." Having defined most of free rate and E(R m ) the expected market return.

Vol. 25, No. 3, July-September 2000 26 Vikalpa

example to illustrate the similarity by considering figure. Thus, new economy firms may be better
a firm with a single line of business having five-year valued with EVA.
life and a cost of capital of 18 per cent (Table 1).
The advantage of EVA over cash flows' is also
The critical issue, therefore, is: Why should echoed by Sirower and O'Byrne (1998). They
we use EVA? EVA is better because it is an annual advocated the use of EVA in place of cash flows
measure as well as a life-time measure. NPV only to measure periodic operating performance. They
measures the life-time value of a firm. NPV or cash observed that the main shortcoming of free cash flow
flow-based method cannot return a reliable annual (FCF) as a measure of periodic operating perform-
performance measure. A firm with high growth ance is that "it subtracts the entire cost of an
potential would show negative annual cash flows investment in the year in which it occurs.... EVA
in the years of growth due to heavy investments. effectively capitalizes instead of expensing such
NPV method deducts the entire investments made corporate investment, and then holds management
for future growth in one year and thereby reports accountable for that capital by assigning a capital
a negative cash flow figure for high-growth firms. charge."
EVA, on the other hand, deducts only a capital
charge on such investments from NOPAT. A
dotcom firm, for example, would show huge EVA and Market Value Added:
negative cash flows in the first few years of its Relationship
existence in spite of high revenue. But, the share
price of the firm may still go up. In such a EVA theory simply emphasizes that earning a return
situation, the cash flow-based model may fail to greater than the cost of capital increases the value
explain the share price movements. Also, it may of a company and earning less than the cost of capital
be difficult to project future cash flows on the basis decreases the value. Stewart (1991) has introduced
of past negative cash flows. EVA measure would another measure of shareholder value called Market
deduct a capital charge on massive investments Value Added (MVA). MVA tells us how much value
made by the dotcom firm in initial years and hence the market adds over the book value of invested
would return a more reliable annual performance capital. MVA, therefore, denotes the confidence of
Table 1: A Comparison of NPV and EVA Methods

Period Invested Capital Operating Profit before Depreciation Depreciation NOPAT

0 200
1 75 90 15
2 120 49.5 70.5
3 130 27.23 102.77
4 145 14.97 130.03
5 (10.06)* 130 8.24 121.76
'Realizable value of assets at the end of five years.
NPV (Method) EVA (Method)
Period Cash Flows P V o f Cash NOPAT Invested Cost of EVA PV of EVA
Flows Capital Capital
0 -200 -200
1 75 63.56 -15 200 36 -51 -43.22
2 120 86.18 70.5 110 19.8 50.7 36.41
3 130 79.12 102.77 60.5 10.89 91.88 55.92
4 145 74.79 130.03 33.27 5.99 124.04 63.98
5 140.06 61.22 121.76 18.30 3.29 118.47 51.78
Value of the Firm 164.87 164.87
Vol. 25, No. 3, July-September 2000 27 Vikalpa
considered as independent variables and MVA has depends on the rate of return of a company. If a
been considered as dependent variables. The rela- company's rate of return exceeds its cost of capital,
tionship between independent and dependent var- the company will sell in the stock market with a
iables was tested in nine industries over a period premium compared to its book value of capital. On
of six years (1992-93 to 1997-98). In addition to the other hand, companies that have rate of return
conducting regression analysis on each industry smaller than their cost of capital will sell with
separately, the same analysis is carried on all discount compared to their book value of capital.
companies across the nine chosen industries (cross- This principle also applies to EVA. Thus, it is said
sectional analysis) for each year. The study failed that positive EVA also means positive MVA and vice
to conclude convincingly about the superiority of versa (Stewart, 1991). Maximizing MVA, therefore,
EVA over other independent variables in explaining should be the primary objective for any company
MVA. The study, however, singled out EVA as the that is concerned about its shareholders' welfare.
common significant explanatory variable across Thus, EVA is the internal measure of corporate
industries. The study could only show that of the performance and MVA is the external measure of
five independent variables, EVA is the better of the corporate performance. MVA reflects how much the
lot. capital market is putting value on the invested
capital.
The objective of the present study is to examine
whether the market value of a firm is best predicted Hence, market value is deemed to be predicted
by expected EVAs. In other words, Stewart's talked well with stream of future expected EVAs and not
about relationship of market value as equivalent to only with current year EVA. The sample size
the sum of present value of future stream of expected considered for the present study is 200 and the
EVAs is examined. The market value of a firm is relationship between EVA and market value is tested
always futuristic. It captures information about a firm over a five-year period data (1993-94 to 1997-98).
not yet operationally materialized and hence not
present in its income statement. Explaining change
in market value of a firm on the basis of one-year
Previous Studies
EVA value may not be quite correct. Stewart (1991)
also mentions that MVA is the present value of future The empirical research of academics to date on this
stream of EVAs. Stewart coined the term Market Value subject is limited. The results of these studies are
Added to measure shareholder wealth. MVA is mixed. Stewart (1991) has first studied the relation-
defined as absolute rupee spread between a com- ship with market data of 618 US companies. Stewart
pany's market value and its invested capital. In other observed that the relationship between EVA and
words, MVA is the difference between a company's MVA is highly correlated among US companies.
market value of invested capital and book value of Lehn and Makhija (1996) in their study of 241 US
invested capital. The market value of debt is not companies over two periods (1987-1988 and 1992-
readily available, as debts are mostly not traded. 1993) observed that both measures (EVA and MVA)
Therefore, the definition of MVA can be modified correlate positively with stock returns and that the
as below: correlation is slightly better with EVA than that with
traditional performance measures like return on
MVA = Market Capitalization - Equity Share Cap- assets (ROA), return on equity (ROE), etc. On the
ital + Reserves and Surplus - Revaluation predictive power of EVA in explaining MVA or
Reserves-Accumulated Losses - Miscella- shareholder wealth, several researchers (for example,
neous Expenditure. Uyemura, Kantor and Petit, 1996; McCormack and
................ (Eq-9) Vytheeswaran, 1998; O'Byrne, 1996; Milunovich
It can be observed from the above modified and Tsuei, 1996; Grant, 1996) observed that EVA
definition that MVA is almost similar to market price- is better correlated with MVA or shareholder wealth
book value (p/b) ratio. The only difference is that than other traditional parameters like ROCE, RONW,
MVA is an absolute measure and p/b ratio is a EPS, etc. However, there are adverse findings too.
relative measure. If MVA is positive, it implies that Dodd and Chen (1996) found that return on assets
p/b is greater than one. Negative MVA implies a (ROA). explained stock returns better than EVA.
less than one p/b ratio. Successful companies add Hamel (1997) was critical about the superiority of
their MVA and thus increase the value of capital EVA. He opined that EVA reveals little about a
invested in the business. It is argued that MVA company's share of new wealth creation.

Vol. 25, No. 3, July-September 2000 29 Vikalpa

Research Methodology of estimating cost of equity. Use of corporate beta
on the basis of daily share prices may not be
prudent. The basic limitation of taking daily values
It has already been mentioned that: is that two companies from the same industry may
not be traded on the same day. Also, the number
Cost of Borrowings (KJ = (Total Interest Expense/ of days traded would be different for different
Beginning Total Borrowings) * (1-T) * 100 companies. It would be different even for a particu-
............. (Eq. 10) lar company in different years. In the present study,
beta figures for companies have been estimated
It may so happen that Kd calculated according using monthly stock and index returns instead of
to Equation 10 may return an abnormally high figure. daily returns. Most estimation services in US use
This is because the total interest expenses may be five years of data. Our length of estimation period
too high compared to the beginning total interest- is also five years, i.e. sixty monthly return figures
bearing debts. This would give a distorted borrowing to calculate each beta.
cost figure. The borrowing cost number can show
Another reason for taking five years as the
an artificially high figure because of the following
length of estimation period is that the long-term risk
reasons: premium for the purpose of estimating cost of equity
• Loans taken during the current year have been has been estimated on the basis of five-yearly
repaid. average figure. Thus, beta for 1993-94 has been
Loans taken at the beginning of the current year calculated on the basis of past sixty months stock
so that it does not appear at the denominator return figures. It has been found that betas,
of Kd calculation. calculated on monthly returns, are less volatile
than betas calculated on daily return basis.
• Rescheduling of loan repayment.
Hence, a control variable "normal yearly bor- MVA has been calculated as the difference
rowing cost" has been considered. This is taken as between market capitalization and net worth (net of
the Prime Lending Rate (PLR) of the concerned year adjustments). The market capitalization figures are
of State Bank of India. The SBI's PLR for the period normally calculated as number of outstanding
under study is shown in Table 2 (RBI, 1996-97, 1997- shares multiplied by closing market price on the
98). last day of the year. This method of computing
market capitalization is subject to high volatility
If the computed K d in any year is more than because it is based on the share price of a single
the control variable (after-tax), the after-tax control day, whereas net worth is a cumulative figure and
variable has been considered as the borrowing cost the increment in net worth occurs throughout the
of that year. This has been done to see whether the year. Kondragunta (2000) pointed out that the
borrowing cost represents the prevailing lending average market value should be considered instead
rates of the banks. If the computed K d in any year of closing price. So, MVA has been computed by
is less than the control variable (after-tax), no suitably adjusting market capitalization as
adjustment is made to Kf This is to recognize the below:
innovative financing routes followed by companies
to minimize borrowing costs.
Market Capitalization = Mean Closing Adjusted
An earlier study on EVA (Banerjee, 1999) used Market Price (of Last 30 Trading Days) *
daily stock returns to compute beta for the purpose Number of Outstanding Shares
Table 2: SBI's Prime Lending Rate (PLR) Since MVA attempts to capture the value
Year SBI's PLR(%) added by the capital market at the end of the
year on the basis of a firm's performance, average
1992-93 19.0
adjusted market price of last 30 days (instead of
1993-94 19.0 the whole year) is taken. Adjusted market price
1994-95 15.0 refers to share price after adjustment for bonus
1995-96 16.5 issues and rights issues. It is believed that this
1996-97 14.5 method of computing market capitalization would
1997-98 14.0 reduce its volatility and hence MVA would be
more reliable.

Vol. 25, No. 3, July-September 2000 30 Vikalpa

Sample Selection EVA improvements (DEVA). Thus the above equa-
tion shows that if a firm has no future growth
prospect, its market value would be largely depend-
To test our hypothesis that EVA is a better predictor ent on current operational value. On the other hand,
of market value, we need a sizeable sample drawn a growth firm would derive its market value mostly
from different industries. We also need necessary from future growth value. Therefore, a look at the
data for a period covering one business cycle. We value of COV and FGV would tell us whether the
have selected 200 companies across industries. The firm in question has market-perceived growth po-
criterion for sample selection has been availability tential.
of necessary information for the period under study
(1993-94 to 1997-98). The companies are chosen in The computation methodology of COV and
such a way that there does not exist any industry FGV is shown in Table 3 with two companies from
bias. Companies represent industries like Automo- our sample — Abbott Laboratories (Drugs and
bile Ancillaries, Drugs and Pharmaceuticals, Cotton Pharmaceutical industry) and HINDALCO (Diver-
and Blended Yarn, Finished Steel, Paper and Paper sified Group).
Products, Tea and Coffee, Cement, Tyres, Heavy
Commercial Vehicles, Computer Hardware and The computation of COV and FGV (for Abbott
Software, Cosmetics and Toiletries, Paints and Var- Laboratories) is shown below:
nishes, Tobacco, and also a Diversified Group. Thus,
companies have been selected from mature as well COV = Invested Capital at the End of First Year
as growing industries. (1993-94) + Capitalized Value of Current
(1993-94) EVA
Computational Methodology = 12.85+1.16/0.1222 = 22.34 ...... (Eq.12)
FGV = Present Value of Incremental EVAs for the
The above relationship shows that market value of Period 1994-95 till 1997-98 + Present Value
a firm (given by market value of equity and book of Residual Value
value of interest-bearing debt) is a function of two = (0.39* 0.8911 -3.19*0.7941 + 3.02 *
components — Current Operational Value (COV) 0.7076 + 0.79* 0.6305) * (1+0.122)/0.122
and Future Growth Value (FGV). COV, in turn, = 4.15 .........(Eq.13)
depends on book value of beginning invested capital
and capitalized value of current year's EVA. FGV To put it simply, FGV is the capitalized value
is the summation of present value of future expected of present value of future stream of expected

Table 3: COV and FGV Computations

(Rs Crore)
Company Beginning Current Current Future Current Incremental Future
Invested EVA WACC Returns(r) Operational EVA Growth
Capital (94-93) (94-93)(%) (%) Value Value
Abbott Laboratories
1993-94 12.85(end 1.16 12.22 22.34
of 1993-94)
1994-95 12.85 24.25 1.55-1.16=0.39
1995-96 15.04 1.33 -1.64-1.55=-3.19
1996-97 23.93 18.01 1.39+1.64=3.02
1997-98 20.73 22.72 2.18-1.39=0.79 4.15
HINDALCO
1993-94 1188.96 49.95 15.79 1505.21
1994-95 1188.96 26.79 130.77-49.95=80.82
1995-96 1919.22 22.73 133.12-130.77=2.35
1996-97 2457.70 16.41 15.06- 133.12=-118.04
1997-98 2840.07 19.42 103.08-15.06=88.02 326.01
Vol. 25, No. 3, July-September 2000 31 Vikalpa
improvements in EVA. The present value of future Market Value = a + b, * COV + b2* FGV ....(Eq.16)
incremental EVAs is computed on the basis of
WACC of 1993-94. Hence, for Abbott Laboratories, Where,
the above mathematical relationship can be finally Market Value = Actual Market Value of a Company
shown as below: at the End of 1993-94,
= Market Capitalization (End of 1993-
Market Value = COV + FGV + e 94) + Preference Share Capital (End
(1993-94 end) of 1993-94) + Total Borrowings (end
60.18 = 22.34 -I- 4.15 + e .........(Eq.14) of 1993-94) .............. (Eq.17)
For HINDALCO, the relationship is Equation 17 attempts to show that estimated
2384.93 = 1505.21 + 326.01 -I- e ......(Eq.15) market value of a company is highly correlated with
actual market value. Estimated market value is
(Where e stands for the error term) largely a function of FGV of a company. The above
regression result would show how much of the
We can observe from Table 3 that Abbott variation in market value of different companies is
Laboratories experienced very high volatility in its explained by COV and FGV. It may be reiterated
returns. It was 24.25 per cent in 1994-95, dropped that EVA of 1993-94 has been considered as current
to a paltry 1.33 per cent in 1995-96 and again rose EVA and present value of incremental EVAs (1994-
to a respectable 18.01 per cent in 1996-97. Abbott's 95 to 1997-98), discounted by weighted average cost
profit before interest and tax (net of non-recurring of capital of 1993-94, is the FGV (end of 1993-94).
items) was only Rs 0.20 crore in 1995-96 as Therefore, a high correlation between independent
compared to Rs 5.4 crore in 1994-95 and Rs 4.31 variables and market value would indicate that
crore in 1996-97. The main reasons for such a dismal current and future EVAs significantly explain the
performance in 1995-96 were higher wage cost market value of a company. A company can increase
[14.07% of net sales as compared to 10.97%(1994- its market value by ensuring a sustained improve-
95) and 5.66%(1996-97)] and a high general and ment in EVA.
administrative overheads (16.86% as compared to
13.43% in 1994-95 and 9.44% in 1996-97). Such poor Regression Results and Interpretations
return in 1995-96 resulted in negative EVA and
hence a lower FGV.
The computed values of independent and dependent
A similar computational technique is followed variables are given in Appendix 1. Regression results
for other companies in the sample. Based on com- are encouraging. The results of the regression equa-
puted values of EVA, the following linear regression tion are summarized in the Box. The results show
has been drawn to establish the relationship between that independent variables (COV and FGV) signif-
EVA and Market Value: icantly explain the variations in market value

Box: Regression Results of EVA and MVA Relationship

1
Product-Moment Correlation (Sample Size: 200)
Market Value COV FGV
(MV)
Market Value (MV) 1.000 0.821 -0.279
COV 0.821 1.000 -0.592
FGV -0.279 -0.592 1.000
Regression Results
MV = 63.578 + 2.057 * COV + 0.986 * FGV (R 2 = 0.740, Adj.R 2 = 0. 737, F-statistic
(1.101) (22.380***) (7.047***) 279.991*** Durbin-Watson Statistic 2.047)
Note: Figures in parenthesis indicate t-statistic.
*** indicates significance at 1 per cent level.

Vol. 25, No. 3, July-September 2000 32 Vikalpa

(Adjusted R-squared '= 73.7%). This confirms our current state of affairs and future growth potential.
hypothesis that market value estimated on the basis It sends a clear message — it is important to earn
of current operational value and future growth value positive EVA but it is more important to achieve
is highly correlated with actual market value. It has a sustained improvement in EVA. The market value
already been noted that future growth value depends (end of 1993-94) of TISCO is significantly higher
on expected EVA improvements. In the present than its COV. This does not imply that the share
study, no projection has been made to estimate price of TISCO was increasing throughout. In fact,
future EVAs. The year 1993-94 has been considered share price of TISCO, which was Rs 193.75 as on
as the current year and actual data for next four March 31, 1994, had fallen to Rs 149.20 as on March
years (1994-95 to 1997-98) are considered to calculate 31, 1998. The negative FGV of TISCO captures
the future growth value. To calculate EVA for the this fall in share price.
four-year period (1994-95 to 1997-98), weighted
average cost of capital of 1993-94 has been consid- The t-statistics of COV and FGV is significant
ered. However, the same regression can be carried at one per cent level and the constant is insignificant.
on the basis of estimated EVAs for future periods. This implies that COV and FGV significantly explain
For that purpose, a number of assumptions are to the market value of the sample. The Durbin-Watson
be made to project growth. value also justifies the relationship. An important
feature to be noted in this result is that the coefficient
A look at the COV and FGV figures would tell
of EGV is lower than that of COV. It implies that
us the market implied growth potential of a company.
market value of most of the firms in our sample is
Consider a few examples as listed in Table 4.
a reflection of more of current operational value and
Table 4: COV, FGV, and Market Value of less of future growth value. It further implies that
Some Companies the firms in our sample have less market implied
Rs Crore growth potential.
COV FGV Market Value
Ashok Leyland 788.27 -598.65 2282.15
Conclusion
Hindustan Lever
Ltd. (HLL) 958.08 1976.42 7562.65 However, Appendix 1 also reveals a darker side.
Procter & Gamble There exists a huge gap in many cases between actual
India Ltd. (P&G) 41.73 133.02 717.71 market value and the sum total of COV and FGV.
TISCO 3866.62 -594.24 10296.28 For example, in case of HLL, the market value (end
Wipro Ltd. 112.22 218.68 82.46 of 1993-94) was Rs 7562.65 crore and the FGV was
only Rs 1976.42 crore. It implies that FGV has failed
HLL, P&G, and Wipro have future growth in these cases to capture the growth potential
potential. FGV is twice that of COV for HLL and factored in the market value of HLL. Another
W i p r o . F o r P & G , F G V i s mo r e t h a n t h r e e t i me s possible explanation for FGVs poor predictive power
of COV. The improvement in EVA is greater for could be that FGVs are calculated here on the basis
P&G. Ashok Leyland and TISCO, on the other of actual operating results of the firms during the
hand, have negative FGV figures. In case of Ashok period 1994-95 to 1997-98. A longer time horizon
Leyland, the negative FGV is more than 70 per cent and calculation of FGV on the basis of expected
of COV figure. It implies that whatever current EVA future EVAs might produce a better relationship
the company has been able to achieve, it could not between FGV and market value. The market cap-
generate positive EVAs for future years (1994-95 to italization factors in a longer time horizon and
1997-98, in this case). The company does not have capilatizes the growth potential of a firm during the
encouraging growth prospects. TISCO also shows future period. Computing FGV only on the basis
negative growth prospect. But, its current opera- of four-year data (1994-95 to 1997-98) might have
tional value (for 1993-94, in this case) is quite high. led to underestimation of future growth potential.
TISCO signifies a typical case of a steel company We have not attempted here to estimate future EVAs
i n I n d i a i n 9 0 s . E v e r y c o mp a n y i n t h i s i n d u s t r y simply because that exercise would involve more
is passing through a difficult phase — excess capac- assumptions. In the absence of published informa-
ity, inventory pile up, dumping by foreign players, tion about equity analysts' future projections of firms'
etc. The FGV figure tells that story. Appendix 1, performance, it would have been very difficult to
therefore, gives a better picture of a company's estimate economic parameters .of these 200 sample

Vol. 25, No. 3, July-September 2000 33 Vikalpa

firms drawn from diverse industries. Estimating might have been counter-productive. Therefore, the
future EVAs for these firms, then, would require above results could be considered keeping in mind
some generalization of assumptions. Such an effort these limitations.

Appendix 1: Computed Values ofIndependent and Dependent Variables

Company Name COV FGV MV Company Name COV FGV MV

(1994) (1994) (1994) (1994)

AFT Industries Ltd. 85.43 27.63 75.70 Ceat Ltd. 534.13 54.65 854.13
Abbott Laboratories (India) Ltd. 22.37 4.15 60.18 Century Textiles & Inds. Ltd. 1582.42 •1594.72 1140.23
Advani-Oerlikon Ltd. 51.36 22.86 76.29 Cheminor Drugs Ltd. 112.07 -37.38 170.26
Albert David Ltd. 22.16 -15.94 38.40 Cimmco Birla Ltd. 174.50 -84.20 138.06
Alfa Laval (India) Ltd. 126.10 -96.46 874.97 Cipla Ltd. 281.45 545.43 91.03
Amtek Auto Ltd. 20.53 9.86 17.74 Clutch Auto Ltd. 28.30 -3.45 32.91
Andhra Pradesh Paper Mills Ltd. 65.64 -20.51 102.88 Colgate-Palmolive (India) Ltd. 409.41 -2.08 6416.90
Apollo Tyres Ltd. 233.35 147.39 637.31 D C W Ltd. 175.33 -108.60 263.03
Arvind Mills Ltd. 1262.52 -847.74 2054.38 Dalmia Cement (Bharat) Ltd. 120.25 -46.91 296.33
Asea Brown Boveri Ltd. 277.95 -66.08 637.52 Deccan Cements Ltd. 29.41 10.05 42.16
Ashok Leyland Ltd. 788.27 -598.65 2282.15 Denso India Ltd. 23.29 10.06 105.68
Asian Coffee Ltd. 33.81 -10.02 77.05 Dewan Rubber Inds. Ltd. 124.55 -74.42 121.58
Asian Paints (India) Ltd. 288.42 160.10 528.20 Dr. Reddy'S Laboratories Ltd. 330.07 81.64 291.15
Associated Cement Cos. Ltd. 895.76 -645.90 824.83 Duphar-Interfran Ltd. 31.11 -14.66 59.28
Astra-Idl Ltd. 24.06 30.48 33.89 E I D-Parry (India) Ltd. 192.51 -186.60 370.14
Autolite (India) Ltd. 36.66 -46.43 51.12 E Merck (India) Ltd. 385.12 207.11 297.95
Automobile Corpn. Of Goa Ltd. 31.74 5.27 51.31 East Coast Steel Ltd. 34.67 -24.22 32.44
BASF India Ltd. 104.77 -28.21 284.83 Elgitread (India) Ltd. 38.86 18.24 36.20
Bajaj Auto Ltd. 1200.29 617.92 1660.82 Escorts Ltd. 267.25 -113.41 514.15
Banco Products (India) Ltd. 32.67 9.33 20.64 Essar Steel Ltd. 4095.27 3449.07 4075.24
Baroda Rayon Corpn. Ltd. 231.68 -235.60 123.13 Eurotex Industries &
Bayer (India) Ltd. 203.63 -45.28 711.27 Exports Ltd. 35.62 -1.30 - 72.97
Berger Paints India Ltd. 44.21 20.88 33.54 Falcon Tyres Ltd. 26.95 5.41 - 13.67
Bharat Bijlee Ltd. 26.48 -0.28 47.13 Finolex Cables Ltd. 269.29 273.62 254.20
Bharat Forge Ltd. 323.57 -148.08 695.31 Forbes Gokak Ltd. 139.64 -27.77 427.87
Bharat Gears Ltd. 24.06 29.59 43.62 Fulford (India) Ltd. 13.80 11.49 - 35.98
Bharat Seats Ltd. 14.70 -1.33 15.26 G K N Invel Transmissions 34.89 4.21 - 97.35
Bimetal Bearings Ltd. 44.88 -14.03 137.86 Ltd. 94.06 63.38 - 89.26
Birla Corporation Ltd. 284.85 -255.33 919.1 G S L (India) Ltd. 72.65 26.61 - 104.10
Birla Yamaha Ltd. 24.63 1.75 5 Gabriel India Ltd. 11.10 0.61 22.49
Blow Plast Ltd. 46.63 -33.53 47.24 Gajra Bevel Gears Ltd. 129.11 160.55 263.15
Blue Star Ltd. 64.10 15.44 50.26 German Remedies Ltd. 107.29 116.67 - 349.13
Bombay Burmah Trdg. Godfrey Phillips India Ltd. 87.40 10.43 - 127.99
Corpn. Ltd. 60.09 -43.07 44.46 Goetze (India) Ltd. 25.75 108.27 20.62
Bombay Dyeing & Mfg. Gontermann-Peipers (India) 88.65 -17.60 - 438.41
Co. Ltd. 710.94-490.02 1649.74 Ltd. 77.05 19.37 - 107.10
Britannia Industries Ltd. 102.94 49.16 511.03 Goodricke Group Ltd. 69.30 8.41 - 111.66
Burroughs Wellcome (India) Ltd. 71.91 8.98 193.40 Goodyear India Ltd. 2398.06 838.57 5573.31
Cadbury India Ltd. 42.67 59.52 233.36 Govind Rubber Ltd. 230.28 -44.93 - 365.35
Camlin Ltd. 21.22 16.42 29.61 Grasim Industries Ltd. 914.55 231.79 991.73
Caprihans India Ltd. 76.03 -118.79 88.51 Greaves Ltd. 79.60 -96.92 26.39
Carrier Aircon Ltd. 33.55 66.91 146.74 Gujarat Ambuja Cements Ltd.
Castrol India Ltd. 325.54 279.35 453.63 Gujarat Lyka Organics Ltd. 11.33 -5.41 15.99
Hardcastle & Waud Mfg.
Co. Ltd.
(Appendix 1 Contd.)

Vol. 25, No. 3, July-September 2000 34 Vikalpa

Company Name COV FGV MV Company Name COV FGV MV
(1994) (1994) (1994) (1994)
T I L Ltd. 54. 94 28 .21 74.14 VST Industries Ltd. 160.72 -88. 65 752.41
T T K Pharma Ltd. 66.67 1 .36 65.06 Vardhman Spinning &
TVS Srichakra Ltd. 15. 34 9 .31 17.48 General Mills Ltd. 265 .16 -30. 98 224.47
Talbros Automotive Videocon Appliances Ltd. 497.87 -114.68 669.74
Components Ltd. 10, ,7 5 .36 12.90 Vikrant Tyres Ltd. 86. ,56 51. 03 160.15
Tata Chemicals Ltd. 2204 .1 851 .43 4277.06 Voltas Ltd. 361. 08 -194.52 729.51
Tata Engineering & Warren Tea Ltd. 51. 95 28.15 327.02
Locomotive Co. Ltd. 1774. 63 -100 .01 5261.92
Tata Infotech Ltd. 80. 55 59 .54 103.26 West Coast Paper Mills Ltd. 46. 29 19. 39 13.11
Tata Iron & Steel Co. Ltd. 3866. 62 - .2410296.28 Wimco Ltd. 75. 25 -55. 22 181.29
Tata Tea Ltd. 552. 24 14 .58 1777.20 Wipro Ltd. 112. 22 218. 68 82.46
Tata-Yodogawa Ltd. 32 .1 -2 .87 62.94 Z F Steering Gear (India) Ltd. 8. 39 -9.37 18.44
Ucal Fuel Systems Ltd. 32. ,0 23 .37 50.24 Zuari Industries Ltd. 254. 44 168.14 328.22
Uttam Steel Ltd. 194 .1 -161 .11 198.44 Ambalal Sarabhai 103. ,56 - 55 176.72
V D O India Ltd. 46. 44 -21.51 33.08 Dee Pharma 41.57 - 52 26.24
VIP Industries Ltd. 97.43 9 .76 126.14 Gujarat Themis -6. ,82 3. 34 96.03

References
Banerjee, A (1999). "Economic Value Added and Shareholder Milunovich, Steven and Tsuei, Albertz (1996). "EVA in the
Wealth: An Empirical Study of Relationship," Paradigm, Computer Industry,"Journal of Applied Corporate Finance,
Vol 3, No 1, January-June. Vol 9, No 3, Fall.

Dodd, James L and Chen, Shimin (1996). "EVA: A New O'Byrne, Stephen F (1996). "EVA and Market Value.Vour-
Panacea?" Business and Economic Review, Vol 42 , July- nal of Applied Corporate Finance, Vol 9, No 3, Spring.
September. Rappaport, Alfred (1986). Creating Shareholder Value: The New
Standard for Business Performance, Chapter 5, First Edition,
Ehrbar, Al (1998). EVA: The Real Key to Creating Wealth, John New York: The Free Press.
Wiley & Sons, First Edition.
RBI (1996-97). Report on Currency and Finance, Vol II.
Grant, J L (1996). "Foundations of EVA for Investment
Managers," The Journal of Portfolio Management, Vol 23, Fall
RBI (1997-98). Report on Currency and Finance, Vol II.
Issue. Sharpe, W E (1964). "Capital Asset Prices: A Theory of
Market Equilibrium under Conditions of Risk," Journal of
Hamel, Gary (1997). "Killer Strategies that Make Sharehold- Finance, Vol 19.
ers Rich," Fortune, June 23.
Sirower, Mark L and O'Byrne, Stephen F (1998). "The
Kondragunta, Chaith S (2000). "Winning with EVA," Busi- Measurement of Post-Acquisition Performance: Toward
ness Today, February 22. a Value-based Benchmarking Methodology,"Journal of
Applied Corporate Finance, Vol 11, No 2, Summer.
Lehn, K and Makhija, A K (1996). "EVA and MVA as Per-
formance Measures and Signals for Strategic Change," Stewart, Bennett G III (1991). "The Quest for Value," Harper
Strategy and Leadership, Vol 24, May-June. Business, USA.
Tully, Shawn (1993). "The Real Key to Creating Wealth,"
Lev, Baruch (1999). "R&D and Capital Markets," Journal of Fortune, September 20.
Applied Corporate Finance, Vol 11, No 4, Winter.
Uyemura, Kantor and Petit, Justin (1996). "EVA for Banks:
McCormack, John L and Vytheeswaran, Jawanth (1998). Value Creation, Risk Management, and Profitability
"How to Use EVA in the Oil and Gas Industry,"/oama/ Measurement," Journal of Applied Corporate Finance, Vol 9,
of Applied Corporate Finance, Vol 11, No 3, Fall. No 2, Summer.

Miller, Merton and Modigliani, Franco (1961). "Dividend Wallace James S (1997). Adopting Residual Income Based Com-
Policy, Growth and the Valuation of Shares," Journal of pensation Plans: Evidence of Effects on Managerial Actions,
Business, Vol 34, No 4, October. Working Paper, University of California, USA.

Vol. 25, No. 3, July-September 2000 36 Vikalpa