Chapter 1

Introduction

1.1 General BackgroundThe pricing implication of common stocks has drawn considerable attention since

the publication of seminal work of Markowitz (1952) - the mean-variance

portfolio theory. Since then there is an ongoing debate on whether the market risk

factors explain better or there are some other anomalies influencing common

stock returns. Based on the mean-variance portfolio theory, Sharpe (1964), Linter

(1965), and Black (1972) then proposed extensively argued asset pricing theory-

the capital asset pricing model (CAPM). The central prediction of the CAPM is

that the rate of return associated with common stocks investment is determined by

the extent to which the common stock returns are correlated with market portfolio.

CAPM asserts that the market risk factors proxied by beta can capture significant

variation in common stock returns.

The empirical studies, such as Black, Jensen, and Scholes (1972), Miller and

Scholes (1972), Blume and Friend (1973), among others, have also documented

positive relationship between beta and stock returns. However, there are other

empirical evidences (for example, Basu (1977), Banz (1981), Fama and French

(1992), among others) which demonstrate the inability of market risk factor (beta)

in fully explaining common stock returns as opposed to that suggested by the

CAPM. As a result, these studies have evolved the attempts to identify firm

characteristics which explain differences in common stock returns. Among several

firm characteristics, the most prominent ones are earnings-to-price ratio (Basu

(1977)), firm size defined by market value of equity (Banz (1981)), and book-to-

market equity ratio (Stattman (1980); Rosenberg, Reid, and Lanstein (1985);

Chan, Hamao, and Lakonishok (1991)).

1

The joint role of beta, size, leverage, book-to-market equity and earnings-to-price

in the cross-section of average stock returns was evaluated by Fama and French

(1992). The study demonstrated that firm size and book-to-market equity tend to

absorb the significant role of leverage and earnings-to-price in average stock

returns. The empirical negative relationship between firm size and stock returns,

and a positive relationship between book-to-market equity and stock returns

encouraged Fama and French (1993) to propose a three-factor asset pricing model

comprising of the market risk factor, firm size factor and book-to-market equity

factor, that competes with one-factor CAPM to explain cross-section of average

stock returns. Thus, the contemporary debate on empirical performance of the

CAPM has broadened understanding of important asset-pricing factors. In

particular, the three-factor model, which includes the market factor, firm size and

book-to-market equity, is now widely considered to be state of art in cross-

sectional studies of common stock returns. Despite of the success of the model in

empirical studies of matured capital markets, little is known about the results of

applying the model to emerging and developing capital markets like Nepal.

Hence, there is a need to explore whether CAPM beta alone can predict stock

returns, or inclusion of firm size, book-to-market equity and earnings-to-price

ratio subsume the beta effect on stock returns in the context of stock market in

Nepal. Besides, there is a further need to examine the predictive power of the

three-factor model in the context of Nepal.

Additionally, the basic version of CAPM has one restrictive assumption as it over

specifies the unique role of market returns. As a result, multifactor model came

into existence in the form of Arbitrage Pricing Theory (APT) as initiated by Ross

(1976). The APT assumes that stock returns are determined by a number of

unnamed factors in the economy as opposed to single market risk factor.

Similarly, other studies on multifactor effect include King (1966) and Merton

(1973). Although these studies made contribution to asset pricing implications,

market factor was again the main pricing variable in their models besides the

other macroeconomic variables employed. By using the statistical factor analysis,

2

APT initiated the use of variables without the need of pre-specification of the

variables. But it did not take too long before the criticism to appear. One major

criticism was that APT could not specify the factors but just derive them

statistically. Chen, Roll and Ross (1986) then employed specific macroeconomic

variables as proxies for undefined variables in the APT. The study attempted to

express the stock returns as a function of macroeconomic variables. In an attempt

towards this, the study found that economic forces influence dividends and

discount rate, and therefore stock prices and stock returns are systematically

affected by macroeconomic variables. As postulated in the study, the discount rate

is expected to change with the level of interest rates, term-structure and risk

premium and expected dividend, and hence stock returns, may change because of

inflation rate, real activity and consumption. Many other studies, for example,

Chen (1991), Clare and Thomas (1994), Mukherjee and Naka (1995), Gjerde and

Saettem (1999), Flannery and Protopapadakis (2002), Adel (2004), and Gan, Lee,

Yong and Zhang (2006), have documented the relationship of stock returns with

macroeconomic variables in the context of developed stock markets around the

world. Some of these studies have observed that the rate of inflation, money

supply, interest rates and exchange rates have significant predictive power in

explaining stock returns while others have documented that real activity proxied

by real GDP and industrial production growth have significant explanatory power.

It is generally argued that if the value of corporate equity depends on the

economic movement in the country, then uncertainty in macroeconomic

environment would affect volatility in stock returns assuming constant discount

rates (Liljeblom and Stenius (1997)). There are several reasons why

macroeconomic volatility affects stock returns. One of them is the implication of

risk management. Adjasi (2009) stated that the use of information on

macroeconomic environment could help market analysts and other market

participants manage better the risk of their portfolios. The study further argues

that policymakers are also better placed to manage the economy and help develop

stock markets more efficiently by managing macroeconomic fundamentals that

3

affect stock market returns. However, there are no unanimous findings as to which

macroeconomic variable or a set of variables consistently predict the common

stock returns. As opposed to developed stock markets, this issue has been less

addressed and less explored in the context of emerging and developing markets.

The importance of stock markets as financial channels for saving and investment

is gaining significant role in Nepalese economy. As evidenced from

macroeconomic indicators of Nepal as of mid-July 2009, the ratio of stock market

capitalization to GDP is about 53.4 percent. Macroeconomic stability, therefore,

has become an important condition for financial development and economic

growth of the country. In addition, investment plans and financial sector returns

are driven largely by macroeconomic variables and hence influence the volatility

on the stock market returns. In the light of these facts associated with

macroeconomic environment of the country, it is necessary to examine how far

stock market in Nepal is being influenced by macroeconomic variables and how

far the stock market index serve as the leading indicator of macroeconomic

volatility. This study is also an attempt toward this direction using more recent

data on selected macroeconomic variables.

1.2 Statement of the ProblemThe capital asset pricing model (CAPM) of Sharpe (1964), Linter (1965), Mossin

(1966), and Black (1972) scripts the origin of asset pricing theory. The primary

implication of the CAPM is that the model is mean-variance efficiency. This

implies that differences in expected returns across stocks and portfolios are

entirely explained by differences in market beta. Put differently, there exists a

positive linear relation between expected returns and market betas, and variables

other than beta should not have power in explaining the cross-sectional variations

in common stock returns. The main attraction of the CAPM is that it offers

influential and naturally agreeable predictions about how to measure risk and the

relation between expected return and risk. However, the empirical documentation

of the model is poor enough to nullify the way it is used in application.

4

The empirical tests of the CAPM are based on three implications of the relation

between expected return and market beta implied by the model (Fama and French

(1993)). First, expected returns on all assets are linearly related to their betas, and

no other variable has marginal explanatory power. Second, the beta premium is

positive, meaning that the expected return on the market portfolio exceeds the

expected return on assets, whose returns are uncorrelated with the market return.

Third, assets uncorrelated with the market have expected returns equal to the risk-

free rate, and beta premium is the expected market return minus the risk-free rate.

The early empirical tests in US stock markets focused on the model’s predictions

about intercept and slope in the relation between expected return and market beta.

Many tests rejected the basic assumption of the CAPM. For example, Friend and

Blume (1970), Black, Jensen, and Scholes (1972), and Stambaugh (1982)

documented positive relation between beta and average stock returns, but it was

too flat. The CAPM also predicts that the intercept term is equal to risk-free rate

and the coefficient on beta is the expected market return in excess of risk-free

rate. On the contrary, the studies such as by Miller and Scholes (1972), Blume and

Friend (1973), Fama and MacBeth (1973), among others, found intercept term

greater than the average risk-free rate, and the coefficient on beta less than the

average excess market returns. However, there are few tests on empirical validity

of CAPM in the context of stock market in Nepal and studies find no unanimous

conclusion about this. Hence, the present study attempts to test, using more recent

data, whether the central prediction of CAPM holds true in Nepalese stock

market.

Contrary to the predictions of the CAPM model, empirical studies have found that

variables relating to firm characteristics have significant explanatory power for

average stock returns, while beta has little power. The most prominent variables

associated with firm characteristics are firm size, book-to-market equity, cash

flow yield and earnings-to-price ratio. Among the several contradictions, earlier

one was Basu’s (1977) evidence that when common stocks were sorted on

earnings-to-price ratios, future returns on high earnings-to-price stocks were

5

observed higher than that predicted by the CAPM. Similarly, Reinganum (1981)

reported excess returns on common stocks as a monotone increasing function of

earnings-to-price defined as the ratio of earnings per share to market price per

share. On the contrary, Chan, Hamao and Lakonishok (1991) observed earnings-

to-price ratio to loose its significance in predicting stock returns. Similarly, La

Porta (1996) demonstrated low earning growth stocks to have significantly lower

standard deviations and betas than higher earnings growth stocks. The study

concluded that not only did low earnings growth stocks yield higher average

returns than high earnings growth stocks, but they also did perform significantly

better than high earnings growth stocks in bear market. However, the studies have

failed to give unanimous conclusion regarding earnings-to-price effect on stock

returns. On the other hand, in relation to firm size effect, Banz (1981), Reinganum

(1981), and Keim (1983) observed that small firms have higher returns and larger

firms have lower returns than those predicted by the CAPM. Jagadeesh (1992)

also documented no explanatory power of beta in predicting cross-sectional

differences in average returns because when the test portfolios were constructed

the correlations between beta and firm size were found small.

Finally, Stattman (1980), and Rosenberg, Reid, and Lanstein (1985) demonstrated

high average returns for stocks with high book-to-market equity ratios that were

not captured by their betas. In later period, Chan, Hamao, and Lakonishok (1991)

revealed that the ratio of cash flow to price, in addition to book-to-market equity,

could explain stock returns in Japan. There is a theme in the contradictions of the

CAPM summarized in these studies. Ratios involving stock prices have

information about expected returns missed by market betas. However, most

empirical tests that have found those contradictions to the CAPM, involve an

error-in-variables problem, since true betas are unobservable and, thus, estimated

betas are used as proxy for the unobservable betas. Handa, Kothari, and Wasley

(1989), and Kim (1995) showed that the errors-in-variables problem could induce

an underestimation of price of beta risk and an over estimation of other cross-

sectional regression coefficients associated with firm characteristics variables

6

such as firm size, book-to-market equity, cash flow to price and earnings-to-price

that might be observed with error. As a mater of fact, a greater correlation

between the estimated betas and firm specific variables causes more downward

bias in the price of beta risk estimate and more exaggeration of the explanatory

power of the firm specific variables. Hence, this study also attempts to identify

whether higher correlation exists between betas and firm specific variables and

examine the joint role of beta, firm size and book-to-market equity in explaining

common stock returns in the context of Nepal.

Fama and French (1992) updated and synthesized the evidence on the empirical

failures of the CAPM. Based on the cross-section regression, the study confirmed

that size, earnings to price, debt-equity and book-to-market ratios could add to the

explanation of expected stock returns provided by market beta. Fama and French

(1996) reached the same conclusion using the time-series regression approach

applied to portfolios of stocks sorted on price ratios. The study also found that

different price ratios did have much the same information about expected returns.

As a result, Fama and French (FF) (1993, 1995, 1996) advocated a three factor

model in which a market portfolio return was attached by a portfolio long in high

book-to-market stocks and short in low book-to-market stocks (HML-high minus

low book-to-market equity) and a portfolio that is long in small firms and short in

large firms (SMB-small minus big size). Since then several studies have used the

FF three-factor model as an empirical asset pricing model.

However, there is controversy over why the firm specific attributes that are used

to form the FF three factors should predict stock returns. Some argue that such

variables may be used to find securities that are systematically mispriced by the

market (for example, Lakonishok, Shleifer, and Vishny (1994), Daniel and Titman

(1997)). Others argue that these measures are proxies for exposure to underlying

economic risk factors that are rationally priced in the market (for example, Fama

and French (1993, 1995, 1996)). A third view is that the observed predictive

relations are largely the result of data snooping and various biases in the data (for

example, Kothari, Shanken, and Sloan (1995), Chan, Jagadeesh, and Lakonishok

7

(1995)). In similar case, Berk (1995) emphasized that, because returns are related

mechanically to price by a present value relation, ratios that have price in the

denominator are related to returns by construction. As a matter of fact, if the

numerator of such a ratio can capture cross-sectional variation in the expected

cash flows, the ratio is likely to provide a proxy for the cross-section of expected

returns. Ratios like the book-to-market are therefore likely to be related to the

cross-section of stock returns whether they are related to rationally priced

economic risks or to mispricing effects. Ferson, Sarkissian, and Simin (1999)

illustrated that spread portfolios like SMB or HML could appear to explain the

cross section of stock returns even when the attributes used in the sort bear no

relationship to risk. Since the FF three factors are not derived from a theoretical

model, such concerns about their interpretation are natural. Given these

prominences of the FF three factor model, it is interesting to test its empirical

performance as an asset pricing model. Therefore, this study also attempts to

examine whether stock returns are largely associated with three factors as

suggested by FF in the context of small capital market in Nepal.

Besides firm specific variables, studies also suggest that there is significant

relationship between macroeconomic variables and stock returns. The underlying

theoretical constructs establish a link between macroeconomic volatility and stock

returns based on transmission mechanism between the key macroeconomic

variables, namely, inflation, money supply, interest rate, exchange rate, industrial

production growth and gross domestic product (GDP). Jaffe and Mandelkar

(1976), Nelson (1976), and Fama and Schwert (1977), among others, have argued

that stock returns are inversely related to inflation. This argument shows a

contrary opinion to the priori expectation of Fisher hypothesis which assumes that

stock returns are positively related to inflation, and hence stock investment can be

used as a hedge against inflation. The empirical evidences observed in 1980s (for

example, Fama (1981) and Solnik (1983)) documented the negative relationship

between stock returns and inflation. The evidences have suggested three dominant

8

hypotheses, namely, tax effect, proxy effect, and the reverse causality hypotheses,

explaining the negative effects of inflation on stock returns.

The tax effect hypothesis of Feldstein (1980) argues that inflation lowers stock

market returns due to the fact that tax assessment of depreciation and inventory

valuation are done in a non-neutral manner. Hence, inflation introduces a

corporate tax liability and reduces real after-tax earnings, thereby reducing

common stock returns. The proxy effect of Fama (1981) explains that real activity

is positively related to common stock returns, but negatively related to inflation

through the money demand effect. As a result, a negative relation between stock

returns and inflation is possible to observe.

According to reverse causality hypothesis of Geske and Roll (1983), the reaction

of stock markets to future economic activity is correlated with increased domestic

borrowing or increased supply of money through the central bank to balance the

budget. The increase in domestic borrowing or issuance of money has inflationary

effects that dampen real activity. In the end, stock market returns also fall due to

fall in real activity and the inflationary effect, and hence the negative relation

exists between stock market returns and the inflation. Dhakal, Kandil and Sharma

(1993) argued that money supply also influences stock returns through inflation.

Here, because of positive relationship between inflation and money supply, an

increase in money supply could reduce stock prices. Furthermore, portfolio theory

suggests that an increase in money supply results in portfolio shift from non-

interest bearing money assets to financial assets including common stocks. In the

opinion of Mukherjee and Naka (1995), the effect of money supply on stock price

is an empirical question. An increase in money supply would lead to inflation, and

may increase discount rate and reduce stock prices (Fama (1981)). Similarly,

Flannery and Protopapadakis (2002) reported the significant negative effect of

money supply on market value weighted returns indicating that higher than

anticipated inflation or money supply depressed equity values. However, contrary

opinion holds that negative effect might be countered by the economic stimulus

provided by money growth, which may increase future cash flows and stock

9

prices. For example, Mayasami and Koh (2000) observed a positive relationship

between money supply and stock returns. However, the studies have also

documented the effect of inflation and money supply on stock prices that are not

consistent. For example, in an attempt to establish a dynamic linkage between

stock prices and macroeconomic variables, Ibrahim and Aziz (2003) reported a

positive relationship between stock prices and inflation in the context of Malaysia.

The study demonstrated that the observed positive relation between stock prices

and inflation could provide better hedge against inflation for investors from stock

investment in Malaysia. However, the same study documented a long-term

negative relationship between stock prices and money supply indicating that

increase in money supply could contribute to the inflation uncertainty and as a

result it might exert the negative influence on stock prices. This study also

hypothesizes similarly that an increase in inflation is likely to result into tight

economic policies that increase interest rate level causing the stock price to

decline.

In relation to interest rate effect, several studies argue in favor of inverse

relationship between stock returns and level of interest rates. For example,

Thorbecke (1997) and Smal and Jager (2001) demonstrated that liquidity in the

economy could increase with reduction in interest rates. This extra liquidity could

be channeled to the stock market thus driving up the demand and prices of stocks.

Gan, Lee, Yong and Zhang (2006) observed that interest rate in the economy

could determine stock returns consistently. Similarly, Kandir (2008) demonstrated

a negative relationship between stock returns and interest rate. Such a negative

relation implies that investors tend to invest less in stocks when interest rates go

up causing stock price to fall. Though there are these evidences associated with

interest rate effects, the studies also reveal that interest rate changes may not be

enough to influence stock-price misalignments. For example, Bernanke and

Gertler (2001) argued that the volatile nature of stock prices makes them hard to

predict and that monetary authorities should only change interest rates in reaction

to stock price movements, when they expect such movements to affect inflation.

10

Goodfriend (2003) also noted no stable correlation between stock returns and

short-term interest rates, as a result it would be difficult for interest rates to target

stock price changes appropriately. Because of these controversies, this study

attempts to identify the interest rate effect on Nepalese stock market. For this

purpose, the study, however, hypothesizes that substantial investment in stocks

are made with borrowed funds and thus increase in interest rates make stock

transactions more costly, and lead to decline in demand and price of stocks.

The empirical evidences in relation to real sectors’ influence proxied by GDP and

industrial production growth on the stock returns also document mixed results. It

is argued that stock prices respond to the volatility in real macroeconomic

variables such as GDP and industrial production growth. In this context, Gjerde

and Saettem (1999) observed a significant positive association between the

industrial production and stock prices. The main reason for this relation to exist is

the fact that an increase in the real sector activity raises the future cash flows that

create a higher future dividend. With the expectation of higher dividend, investors

are always willing to buy common stocks at higher prices. Nasseh and Strauss

(2000), and McMillan (2005) also found similar results. Contrary to these

findings, in an attempt to examine effects of macroeconomic variables on stock

returns, Flannery and Protopapadakis (2002) reported no relation between stock

returns and two popular measures of aggregate economic activity, namely, real

GDP and industrial production. Similarly, a recent study by Kandir (2008)

revealed no significant effect of industrial production growth on the common

stock returns. However, this study hypothesizes a positive relation between real

GDP and stock market returns and attempts to identify whether there is significant

predictive power of GDP in the context of Nepalese stock market.

The empirical results have shown mixed evidences on macroeconomic variables

influencing common stock returns. This study does not consider the role of vast

majorities of macroeconomic variables. The effort simply confines to the

predictive power of real GDP, inflation and interest rates in explaining common

stock returns in Nepal. Besides, the study also focuses to examine whether there is

11

cointegration relationship between macroeconomic variables and stock market

returns. The main issue of this study is to analyze the variation in stock returns in

Nepal with respect to firm specific and macroeconomic variables. The study

basically deals with following issues:

a. Does CAPM hold true in explaining stock returns in Nepal?

b. Whether CAPM beta alone can predict stock returns, or inclusion of firm size

and book-to-market equity subsume the beta effect on stock returns?

c. Is there any relationship between earnings-to-price ratio and cross-section of

common stock returns?

d. Is there any consistency in explanatory power of firm size, book-to-market

equity, stock beta and earnings-to-price ratio when considered individually and

when considered together?

e. Are the stock returns related to three factors, namely market risk factor, size

factor, and book-to-market factor, as suggested by FF three factor model?

f. What is the direction and magnitude of causal relationship between stock

market returns and macroeconomic variables such as inflation, interest rate,

and gross domestic product?

g. Do stock prices in Nepal offer a hedge against inflation?

h. How do stock prices vary with interest rate?

i. Does the real GDP have significant power to predict common stock returns in

Nepal?

j. What are the views of market participants such as investors, executives and

security businesspersons in relation to preferences toward types of stock

market choice for trading, stock market efficiency, and factors affecting stock

returns in Nepal?

1.3 Objectives of the StudyThe main objective of this study is to analyze the cross-sectional variation in stock

returns in Nepal with respect to firm specific and macroeconomic variables.

However, the specific objectives of the study are as follows.

12

a. To evaluate the explanatory power of CAPM in explaining cross-section of

stock returns in Nepal.

b. To examine whether CAPM beta alone can predict stock returns, or inclusion

of firm size and book-to-market equity subsume the beta effect on stock

returns.

c. To analyze the relationship of cross-section of stock returns with earnings-to-

price ratio and evaluate whether inclusion of this variable subsume the effect

of beta, firm size and book-to-market equity.

d. To analyze the relationship of common stock returns with three factors,

namely market risk factor, size factors (SMB), and book-to-market factor

(HML).

e. To examine the causal relationship between stock market returns and

macroeconomic variables such as real GDP, inflation, and interest rate.

f. To analyze the views of market participants such as executives, investors, and

security businesspersons in relation to preferences toward type of stock market

choice, stock market efficiency, and factors affecting stock returns in Nepal.

1.4 Organization of the StudyThe study is organized into a total of five chapters. Chapter one contains general

background of the study including statement of the problem, objectives of the

study, and organization of the study. The chapter two consists of conceptual

review, review of literatures related to studies in global context as well as the

review of studies in Nepalese context. Besides, this chapter ends up with

concluding remarks associated with the findings and major ideas of the studies.

The chapter three covers the research design, nature and sources of data, selection

of enterprises, models used for data analysis and conclusion along with the

limitations of the study. The chapter four focuses on the systematic presentation

and analysis of data. This chapter is further divided into three sections, namely,

analysis of secondary data, analysis of primary data and concluding remarks

associated with the major findings of the study. The chapter five provides a

summary of overview on all works carried out in chapter one through four

13

including major conclusions derived from the study. This chapter also includes a

separate section for recommendations and scope for future research based on

major findings of the study.

14