5.1 FACTORS AFFECTING STOCK MARKET...
Transcript of 5.1 FACTORS AFFECTING STOCK MARKET...
CHAPTER – 5
DATA ANALYSIS – II 5.1 FACTORS AFFECTING STOCK MARKET VOLATILITY 5.1.1 Overview
The factors which result in volatility can be classified into two types micro and
macro. Micro factors are specific, like dividend decisions, major expansion plans, and
receiving of big contracts, earning per share, company size and book value per share
have significant impact upon the value of the stock of that company. Macro level factors
affect the whole economic structure of the economy, and thereby, the behaviour of the
stock market. The impact of these factors clearly gets reflected in the stock market in
terms of volatility. These factors include tax system, interest rate, inflation rate,
agriculture and industrial production, bank, GDP, Government expenditure, foreign
institutional investment, the exchange rate, union budget, imports growth rate, current
account deficit, money supply and foreign currency reserves. There are number of factors
which result in either rise or fall of stock prices or in other words lead to volatility. But
present study is confined to the following factors.
(i) Foreign Institutional Investment
(ii) Union Budget
(iii) Inflation
(iv) Interest Rate
(v) Corporate Fundamentals
5.1.2 FOREIGN INSTITUTIONAL INVESTMENT
5.1.2.1 Overview
Foreign investment refers to investments made by the residents of a country in
the financial assets and production processes of another country (Surumpudi, 2006). FIIs
are the entities which are established or incorporated outside India and invest in India
(Upadhyay, 2006). After the opening up of the borders for capital movement, these
investments have grown in leaps and bounds. The effect of foreign investment, however,
varies from country to country. It can affect the factor productivity of the recipient
country and can also affect the balance of payments. In developing countries like India
there has been a great need for foreign capital, not only to increase the productivity of
labour but also because foreign capital helps to build up the foreign exchange reserves
needed to meet trade deficits. Foreign investment provides a channel through which
developing countries can gain access to foreign capital. Such an investment can be of
two types: Foreign direct investment (FDI), and Foreign institutional investment (FII).
Foreign direct investment involves in direct production activities and is also of a medium
to long-term nature. But foreign institutional investment is a short-term investment,
mostly in the financial markets. FII, given its short-term nature, can have bidirectional
causation with the returns of other domestic financial markets such as money markets,
stock markets, and foreign exchange markets (Ray, 2009).
India has taken various measures to attract foreign investment since the beginning
of reforms in 1991. As a result, India succeeded in attracting a foreign institution
investment of around US $121559 million up to 2010-2011. It is evident that such a huge
foreign institutional investment can definitely play a significant role in the development
of a country. India is in the process of liberalizing its capital account, and this has a great
impact on foreign investment and particularly on FII, which affects short-term stability in
the financial markets.
FIIs have emerged as important players in the Indian equity market in the recent
past. Huge foreign capital flows affect not only wide range of economic variables such as
exchange rates, interest rates, foreign exchange reserves, domestic monetary conditions
but also savings and investments, and thus causing a snowballing effect on the Indian
stock market as well. There is a widespread belief that institutional investors particularly
the Foreign Institutional Investors (FIIs) play a major role in the movements of the
leading Indian stock indices (Bodla and Garg, 2007).
FIIs have both positive as well as negative aspects so far as the stock market is
concerned. On the positive side, FIIs investment can bring higher inflow of foreign
reserve which can help in improving the balance of payment situation of country; it can
provide liquidity in the system which can reduce the cost of borrowing, and hence
promote investment activities; and it can also improve the functioning of stock market as
it is believed to invest after a thorough analysis of stock valuation (Panda, 2005). On the
flipside, FIIs can cause heavy fluctuations in the stock market with slight market
disruption; speculators can artificially increase and decrease for short-term gains and
with the slightest hint they can pull back from stock market causing huge volatility in the
stock market, and in this process some make good profit by booking when the index is
high and some lose money that are not able to sell. Investment by foreign institutions
may not be that dangerous as high stock price indicates the gains of the stock holders.
Sudden withdrawal of investment is dangerous for the people who are holding the stocks
in anticipation of further increase. Hence, it is important for the investor to know when to
enter and when to exit (Sujit, 2010). In the last decade, capital flows attracted more
interest by policy-makers, central banks, international institutions and academia, mainly
because flow volume has grown at a phenomenal rate since the beginning of the 1990s.
5.1.2.2 Impact of Capital Inflows on Stock Market Volatility: Theory and Empirical
Evidence
There is a popular belief that institutional traders destabilize stock prices. Due to
their specific investment behaviour, institutions are supposed to move stock prices away
from fundamentals and thereby induce stock returns autocorrelation and increase returns
volatility. Herding, Positive feedback trading and Contagion are the main arguments put
forward for the destabilizing impact on stock prices induced by institutional investors
(Mazumdar, 2004).
Therefore, the first channel through which capital inflows cause volatility in the
stock market is that of ‘Positive feedback trading’. Positive feedback trading or ‘trend
behaviour’ refers to tendency among fund managers to buy ‘winner’ stocks and selling
‘loser’ stocks. It describes the strategy of rushing in when the markets are booming and
rushing out when the markets are on the decline. Batra (2003) finds strong evidence that
FIIs have been positive feedback investors at the aggregate level on daily basis. Investors
can be positive feedback traders for rational reasons or because of behavioural biases.
Investors with such strategies are often viewed to be destabilizing because their sales
lead the market to fall further and their purchases increase prices further. Besides
contributing to the volatility of stock returns, it is argued that such trading leads to
destabilizing capital flows. This is because equity investors rush into countries whose
stock markets are booming and flee from countries whose stock markets are falling (Bohl
and Brzeszczynski, 2005).
The second channel through which capital inflow causes volatility in the stock
market is herding mentality. Herding behaviour is defined as the tendency of fund
managers and investors to follow other fund managers trading behavior, ignoring their
own information. Batra (2003) found that in India foreign investors have a tendency to
herd on the equity market.
The third channel through which capital inflow causes stock market volatility is
through contagion. Contagion is then best defined as a significant increase in cross
linkages after a shock to an individual country (or a group of countries). Fund managers
often use contagion strategies, i.e., they sell assets in one country when a crisis hits
another country in the same region. The contagion channel is an extension of the herd
behaviour of fund managers. While herding attracts funds to one region of the world,
contagion precipitates exit of fund manager from the same region in times of crisis
(Dornbusch et al., 2000).
5.1.2.3 Trends in FII Investment
Foreign institutional investment (FII) is one of the main channels of foreign
investment in India. Foreign institutional investors (FIIs) were permitted to invest in
Indian securities market in 1992 to widen and broaden the Indian capital market. The
data showing FII investment trends in India during the period 1992-93 to 2010-2011 is
presented in the following table.
Table 5.1 reveals that cumulative net investment by FIIs which was only US $ 4
mn. in 1992-93 increased to US $ 121559 by 2010-11. Since then, the net investment by
FIIs in India has been positive every year except in 1998-99 and 2008-09 (Graph5.1). A
major factor which led to continuous outflow of funds during the middle and end of the
year 1998 was the worsening outlook on the emerging markets. Credit worthiness of
almost all the South-east Asian nations was severely damaged by the crisis which started
in July 1997. As a result, the FIIs were facing heavy redemption pressures from the
Emerging Markets Funds. The stock markets in all these countries fell continuously from
March 1998 to September 1998.
Table 5.1: Trends in FII Investment in India
Year Gross
Purchase
(Rs. Crore)
Gross
Sales
(Rs. Crore)
Net
Investment
(Rs. Crore)
Net
Investment
(US $ mn.)
Cumulative
Investment
(US $ mn.)
1992-93 18 4 13 4 4
1993-94 5593 467 5127 1634 1638
1994-95 7631 2835 4796 1528 3167
1995-96 9694 2752 6942 2036 5202
1996-97 15554 6980 8575 2432 7635
1997-98 18695 12737 5958 1650 9285
1998-99 16116 17699 -1584 -386 8898
1999-00 56857 46735 10122 2474 11372
2000-01 74051 64118 9933 2160 13531
2001-02 50071 411308 8763 1839 15371
2002-03 47062 44372 2689 566 15936
2003-04 144855 99091 45764 10005 25942
2004-05 216951 171071 45880 10352 36293
2005-06 346976 305509 41467 9363 45657
2006-07 520506 489665 30841 6821 52477
2007-08 948010 881839 66179 16442 68919
2008-09 614576 660386 -45811 -9837 59082
2009-10 846438 703780 142658 30252 89333
2010-11 992599 846161 146438 32226 121559
Source: Handbook of Statistics of Indian Securities Market, 2011, SEBI.
The integration of the Indian Capital Markets with the international markets thus
spilled over to Indian markets as well. But the low volatility of the Indian market as
compared to emerging market economies and an efficient market structure were some of
the factors responsible for continued confidence of FIIs in Indian Capital market. During
the last few years there has been phenomenal increase in portfolio investment by FIIs in
the Indian market (Graph 5.1).
Graph 5.1: Trends in Foreign Institutional Investment
-2000
0
2000
4000
6000
8000
10000
12000
14000
16000
18000
1992-93
1993-94
1994-95
1995-96
1996-97
1997-98
1998-99
1999-00
2000-01
2001-02
2002-03
2003-04
2004-05
2005-06
2006-07
2007-08
Year
0
10000
20000
30000
40000
50000
60000
70000
80000
Net Investment(US $ mn. Cumulative Investment(US $ mn.
Source: Compiled from Table 5.1
5.1.2.4 Analysis and Discussion
The causal relationship between the net FII investment and stock market
volatility, i.e., BSE Sensex and NSE Nifty has been analyzed by applying the Granger
Causality Test. For any time-series data analysis, all data series should be stationary. To
study the stationarity of data series, the Augmented Dicker-Fuller (ADF) test has been
conducted. The ADF test statistics for the FII investments is presented in Table 1. The
results show that ADF statistic is -4.408207 for FII. The ADF value for FII is lesser than
critical values leading to inference of rejection of hypothesis with unit root at 1%, 5%
and 10% level of significance. This proves that the time series data of net FII
investments are stationary.
Table 5.2: ADF Test Statistics for FII Investment
t-Statistic Probability
Augmented Dickey-Fuller test statistic -4.408207 0.0004
Test critical value 1% level -3.465014
5% level -2.876677
10% level -2.574917
Source: Calculated from the data of FII given in Appendix 1.
The ADF test statistics for the stock market volatility, i.e., BSE Sensex and NSE
Nifty are presented in Tables 5.3 and 5.4 respectively. The results show that ADF
statistic is - 4.588579 for volatility of BSE Sensex return and -5.136850 for volatility of
NSE Nifty return. Both the ADF values for volatility of stock market return of BSE
Sensex and NSE Nifty are lesser than critical values leading to inference of rejection of
hypothesis with unit root at 1%, 5% and 10% level of significance. This proves that the
time series data of stock market volatility BSE Sensex and NSE Nifty is stationary.
Table 5.3: ADF Test Statistics for Stock Market Volatility (BSE Sensex)
t-Statistic Probability
Augmented Dickey-Fuller test statistic -4.588579 0.0002
Test critical value 1% level -3.465014
5% level -2.876677
10% level -2.574917
Source: Calculated from the data of stock market volatility (BSE Sensex) given in
Appendix 1.
Table 5.4: ADF Test Statistics for Stock Market Volatility (NSE Nifty)
t-Statistic Probability
Augmented Dickey-Fuller test statistic -5.136850 0.0000
Test critical value 1% level -3.465014
5% level -2.876677
10% level -2.574917
Source: Calculated from the data of stock market volatility (NSE Nifty) given in
Appendix 1.
As the test confirmed that FII investment and stock market volatility i.e., BSE
Sensex and NSE Nifty are stationary in nature, the direction of causal relationship
between the stationary data series has been conducted by applying the Granger-Causality
test. The results of the Granger-Causality test are presented in Table 5.5.
Table 5.5: Granger-Causality Test Statistics
Null Hypothesis Observation F-statistics Probability
FII does not Granger Cause
Volatility(BSE Sensex) 190 1.38943 0.25180
Volatility (BSE Sensex) does not
Granger Cause FII 190 0.50388 0.60501
FII does not Granger Cause
Volatility(NSE Nifty) 190 1.18699 0.30745
Volatility (NSE Nifty) does not
Granger Cause FII 190 0.35368 0.70257
Source: Compiled from the data given in Appendix 1 using Eviews 5.
Results from Granger Causality suggest independence between FII and stock
market volatility, i.e., BSE Sensex and NSE Nifty. It can be clearly observed that the
probability for the first and second hypothesis is 0.25180 and 0.60501 respectively,
which is greater than critical value 0.05. Hence, null hypothesis is accepted, i.e., it proves
that foreign institutional investments do not cause stock market volatility (BSE Sensex)
and volatility in BSE Sensex does not cause foreign institutional investments. Similar
results appear for the third and fourth hypothesis thus, null hypothesis is accepted in
these cases also.
5.1.2.5 Results
All the series have been observed to be stationary since the calculated ADF
statistics are lesser than critical values. Though theoretically there seems to be a long run
relationship between stock market volatility and FII equity flows in India, yet the
relationship is not statistically significant. As far as causality relationship is concerned, it
suggests independence between FII and stock market volatility i.e., BSE Sensex and
NSE Nifty. Thus, it can be concluded that FII equity flows have neither increased nor
dampened volatility of stock market returns in India. There could be several reasons for
such insignificant relationship. One of the reasons could be that FIIs are allowed to
perform only delivery based trading and are not allowed to indulge in short selling. This
considerably restricts trading activity. Secondly, FIIs had to register with an Indian
broker to carry out their transactions (Mazumdar, 2004). They were required to pay
higher brokerage for trading as compared to domestic institutional investors. This made
their transactions more expensive than those of domestic institutions. Thirdly, FIIs have
expert knowledge, and therefore, can accurately make predictions about emerging
markets performance in advance, and hence, pro-active with regard to invest or withdraw
decision (Dhillon and Kaur, 2007). Despite the significant share that is allowed to
foreign institutional investments in India, there is no evidence of any strong statistical
relationship between FII equity flows and volatility of stock market returns in India.
5.1.3 UNION BUDGET
5.1.3.1 Overview The global financial crisis spared not even the strongest economies of the world.
In such a state of uncertainty, the fiscal measures taken by the government were meant to
revamp the faith of the investors in the financial system of the economy. The stock
markets are the barometer of fundamental strength of an economy as the development of
stock exchanges is based upon the potential future of the economy. Hence, any kind of
fiscal measures may have a great impact on the rational behaviour of the stock market. In
India, the contents of budget play a major role in developing the perception of investors
for the future prospect of the industrial development (Sharma and Mehta, 2010).
In Wikipedia, the budget is an annual financial statement containing the
estimated receipts and expenditure of the Government of India, which has to be laid
before Parliament in respect of every financial year, which runs from 1st April to 31st
March under article 112 of the constitution. A budget is a powerful tool in the hands of
the government to control the economic resources of the country. It contains proposals
regarding changes in direct and indirect taxes, industrial policy, trade policy, exchange
rate policy and financial sector reforms which may have favourable or adverse impact on
stock market (Gupta and Kundu, 2006). A budget is classified into revenue budget and
capital budget which is based on the nature of expenses. Overall, through budgets the
government is able to implement the fiscal policy. The finance bill actually determines
the rates of both the direct and indirect taxes affecting the individuals as well as all forms
of commercial enterprises. Thus, budget is one macroeconomic phenomenon, which
influences each and every company (Verma and Agarwal, 2005).
The Union Budget is perhaps the most watched event in economic policy making
in India. The core fiscal issues – taxation, expenditures, the fiscal deficit – are obviously
important for macro-economics. In addition, governments have often chosen to use the
Budget speech as a mechanism for announcing important new policy initiatives, and for
outlining some plans for economic policy in the coming months (Thomas and Shah,
2002).
The market participants expect many changes from the budget and revise their
expectations following the announcement of the same. Market starts reacting well in
advance due to the anticipated changes in policies. Budget proposals may prove better
than or fall short of expectations. They may include unexpected changes. Investors revise
stock prices in response to surprises or shocks and non-fulfilment of their expectations.
Union Budget and the stock market, both share a close relationship. Every time the
Finance Minister walks inside the Parliament, everyone at Dalal Street is glued to TV
sets to listen to what he says. And the index pointer dances to the tune. Stock Market
tends to be greatly influenced by the budget. The Stock Market response reflects the
quality of budget. If the government gives subsidy or tax rebate or if there are some
policy changes then the respective sectors get influenced positively or negatively which
affects the overall stock market (Patra, 2012).
5.1.3.2 Analysis and Discussion
Tables 5.6 and 5.7 depict the daily average returns and budget day returns
presented by Sensex and Nifty respectively during the previous and next 3, 15 and 30
days around the budget day.
Budget Day Effect on Sensex and Nifty: The first set of paired t-tests measure the on-
day (Z) influence of budget on Sensex and Nifty when compared to the previous 3, 15
and 30 days. A cursory glance at both the tables highlights that in most of the cases
budget day returns (ignoring sign) are more than the returns during the previous 30, 15
and 3 trading days. Therefore, when the budget day returns (Z) compared with the long-
term pre-budget return in the case of Sensex , it shows that budget day returns exceed in
all years, i.e., (19 out of 19 budgets),compared to medium-term (19 out of 19 budgets)
and short-term (13 out of 19 budgets); and in the case of Nifty it shows
Table 5.6: Daily Average Returns in Sensex
Year
X1(Last
30 days)
X2(Last
15 days)
X3(Last
3 days)
Z(Budget
day)
Y1(Next
3 days)
Y2(Next
15 days)
Y3(Next
30 days)
1996 0.52 0.56 0.18 -0.72 -0.89 -0.51 0.09
(interim)
1996 -0.09 -0.09 1.06 1.17 -2.25 -0.64 -0.27
1997 -0.21 0.05 0.68 6.33 2.57 0.14 -0.01
1998 -0.37 -0.66 -1.66 -1.19 -0.89 -1.12 -0.31
1999 -0.08 0.04 -0.51 4.99 1.93 0.67 -0.07
2000 0.16 0.32 0.58 -5.25 -0.42 -0.42 -0.17
2001 0.02 -0.48 -1.54 4.26 -2.01 -0.89 -0.89
2002 0.34 0.52 0.93 -3.94 0.73 -0.05 -0.10
2003 -0.08 0.04 -0.45 0.19 -0.59 -0.30 -0.30
2004
(interim) 0.17 -0.47 -1.70 -1.32 0.97 0.13 -0.05
2004 -0.08 0.23 0.58 -2.29 0.38 0.37 0.19
2005 0.25 -0.05 -0.10 2.17 0.35 -0.06 -0.13
2006 0.31 0.36 0.19 0.85 0.72 0.30 0.36
2007 -0.04 -0.49 -1.32 -4.09 -1.38 0.004 0.11
2008 -0.33 0.11 0.33 -1.39 -2.03 -0.59 -0.21
2009
(interim) -0.01 0.59 -0.04 -3.48 -0.73 0.35 -0.96
2009 0.24 -0.14 0.95 -6.01 0.60 0.22 -0.69
2010 -0.23 -0.10 0.04 1.07 0.39 0.24 1.08
2011 -0.27 -0.11 -1.10 0.69 0.06 0.33 1.22
Source: Calculated from the data taken from BSE website for the said period. Note: All returns are in percentage.
that budget day returns exceed (18 out of 19 budgets) when compared with the long-
term pre-budget returns, when the budget day returns (Z) compared with the medium-
term pre-budgets return, it shows that budget day returns exceed (18 out of 19 budgets)
and when the budget day returns (Z) compared with the short-term pre-budget returns, it
shows that budget day returns exceed (14 out of 19 budgets).
Table 5.7: Daily Average Returns in Nifty
Year
X1(Last
30 days)
X2(Last
15 days)
X3(Last
3 days)
Z(Budget
day)
Y1(Next
3 days)
Y2(Next
15 days)
Y3(Next
30 days)
1996
(interim) 0.53 0.57 0.08 -0.01 -0.94 -0.48 0.06
1996 -0.03 -0.05 0.78 0.57 -1.92 -0.58 -0.32
1997 -0.16 0.17 0.95 0.59 3.74 0.41 0.12
1998 -0.42 -0.59 -1.27 -0.89 -0.86 -1.20 -0.30
1999 -0.07 0.03 -0.30 4.17 1.98 0.52 0.04
2000 0.20 0.34 0.51 -4.01 0.02 -0.42 -0.29
2001 0.02 -0.46 -1.50 4.22 -2.03 -0.86 -0.86
2002 0.29 0.44 0.73 -4.05 1.05 0.01 -0.02
2003 -0.1 0.04 -0.54 0.99 -0.72 -0.32 -0.37
2004
(interim) 0.14 -0.56 -1.70 -2.28 1.20 0.19 -0.04
2004 -0.07 0.31 0.63 -3.15 0.46 0.43 0.19
2005 0.25 -0.05 0.04 2.03 0.40 -0.02 -0.13
2006 0.24 0.24 0.18 0.24 0.78 0.35 0.36
2007 -0.04 -0.53 -1.23 -3.89 -1.54 0.03 0.15
2008 -0.37 0.19 0.54 -1.17 -1.99 -0.52 -0.17
2009
(interim) -0.01 0.55 0.16 -3.45 - 0.7 -0.56 0.40
2009 0.14 -0.02 1.02 -6.02 -0.69 0.62 0.17
2010 -0.25 0.02 0.02 1.28 1.05 0.37 0.23
2011 -0.27 -0.11 -1.03 0.56 1.26 0.10 0.34
Source: Calculated from the data taken from NSE website for the said period. Note: All returns are in percentage.
The observations made in Tables 5.6 and 5.7 have been further statistically tested
by Paired t- test; and the results so obtained have been presented in Tables 5.8 and 5.9.
The budgets are found to take the markets by surprise in all the cases. In all the tests, the
actual values are found to exceed the table values leading to acceptance of alternative
hypothesis.
T-values from Paired T-test
Table 5.8: Effect of Budget Day Returns on Sensex
X1 and Z X2 and Z X3 and Z
Actual Value (5%) -5.35¬ -5.25¬ -4.29¬
Table Value (5%) -1.76
-1.76
-1.76
Source: Calculated from the data given in Table 5.6. Note: ¬ Signifies that Null Hypothesis (Ho) is rejected.
Table 5.9: Effect of Budget Day Returns on Nifty
X1 and Z X2 and Z X3 and Z
Actual Value (5%) -4.98¬ -4.99¬ -4.04¬
Table Value (5%) -1.76
-1.76
-1.76
Source: Calculated from the data given in Table 5.7. Note: ¬ Signifies that Null Hypothesis (Ho) is rejected.
T- values from Paired t-test
Table 5.10: Impact of Budgets on Sensex
Period Short-term Period Medium-term Period Long-term Period
X1 and
Y1
X2 and
Y1
X3 and
Y1
X1 and
Y2
X2 and
Y2
X3 and
Y2
X1 and
Y3
X2 and
Y3
X3 and
Y3
Actual
Value
(5%)
-5.83¬ -5.01¬ -2.41¬ -2.76¬ -1.78* 2.30 -0.51 0.89 4.05
Table
Value
(5%)
-1.76
-1.76
-1.76
-1.76
-1.76
-1.76
-1.76
-1.76
-1.76
Source: Calculated from the data given in Table 5.6. Note: ¬ Signifies that Null Hypothesis (Ho) is rejected.
The second set of tests provided in Tables 5.10 and 5.11 highlights that budgets
have maximum impact in the short-term (alternative hypotheses have been accepted in
all three cases). In medium-term, the alternative hypothesis has been accepted in two out
of three cases and in long-term, no alternative hypothesis has been accepted.
Table 5.11: Impact of Budgets on Nifty
Period Short-term Period Medium-term Period Long-term Period X1 and
Y1 X2 and Y1
X3 and Y1
X1 and Y2
X2 and Y2
X3 and Y2
X1 and Y3
X2 and Y3
X3 and Y3
Actual Value (5%)
-5.17¬ -4.62¬ -2.78¬ -3.15¬ -1.87* 2.23 -0.79 0.58 3.94
Table Value (5%)
-1.76
-1.76
-1.76
-1.76
-1.76
-1.76
-1.76
-1.76
-1.76
Source: Calculated from the data provided in Table 5.6. Note: ¬ Signifies that Null Hypothesis (Ho) is rejected.
In total, the actual values exceed the tabular values in five cases (3+2+0) out of
nine at the left tail. This proves that budgets, when taken together have the maximum
impact in the short term post-budget period, with some impact extending into the
medium-term and no significant impact at all on long-term average returns.
Table 5.12: Variance of Returns in Sensex
Year X1(Last 30 days)
X2(Last 15 days)
X3(Last 3 days)
Y1(Next 3 days)
Y2(Next 15 days)
Y3(Next 30 days
1996 (interim) 0.036 0.037 0.009 0.036 0.018 0.016 1996 0.019 0.022 0.016 0.026 0.023 0.024 1997 0.033 0.016 0.010 0.069 0.041 0.060 1998 0.032 0.030 0.007 0.019 0.108 0.083 1999 0.020 0.013 0.007 0.067 0.031 0.070 2000 0.038 0.056 0.113 0.118 0.050 0.081 2001 0.020 0.019 0.026 0.059 0.095 0.069 2002 0.011 0.010 0.024 0.049 0.021 0.013 2003 0.006 0.007 0.007 0.002 0.014 0.016 2004 (interim) 0.032 0.048 0.001 0.023 0.022 0.023 2004 0.024 0.014 0.013 0.024 0.008 0.009 2005 0.009 0.004 0.000 0.015 0.006 0.011 2006 0.009 0.008 0.004 0.012 0.010 0.014 2007 0.015 0.017 0.022 0.077 0.045 0.039 2008 0.095 0.050 0.002 0.105 0.108 0.072 2009 (interim) 0.071 0.044 0.029 0.031 0.031 0.058 2009 0.032 0.028 0.007 0.038 0.041 0.043 2010 0.014 0.013 0.001 0.013 0.005 0.006 2011 0.017 0.021 0.031 0.037 0.019 0.015
Source: Calculated from the data taken from BSE website for the said period.
Table 5.13: Variance of Returns in Nifty
Year
X1 (Last
30 days)
X2 (Last
15 days)
X3 (Last
3 days)
Y1 (Next
3 days)
Y2 (Next
15 days)
Y3 (Next
30 days
1996
(interim) 0.045 0.055 0.005 0.049 0.019 0.017
1996 0.013 0.015 0.046 0.015 0.019 0.017
1997 0.027 0.018 0.017 0.288 0.095 0.087
1998 0.017 0.022 0.0001 0.055 0.108 0.078
1999 0.031 0.023 0.012 0.067 0.027 0.031
2000 0.027 0.029 0.028 0.093 0.054 0.076
2001 0.017 0.016 0.010 0.051 0.101 0.069
2002 0.011 0.008 0.013 0.032 0.016 0.012
2003 0.008 0.008 0.007 0.002 0.013 0.018
2004
(interim) 0.037 0.054 0.004 0.040 0.026 0.025
2004 0.028 0.017 0.019 0.030 0.011 0.012
2005 0.010 0.004 0.0004 0.017 0.006 0.011
2006 0.008 0.006 0.003 0.007 0.010 0.014
2007 0.016 0.018 0.017 0.090 0.048 0.042
2008 0.121 0.055 0.005 0.105 0.098 0.064
2009
(interim) 0.060 0.039 0.024 0.033 0.029 0.049
2009 0.039 0.037 0.006 0.041 0.042 0.039
2010 0.014 0.011 0.001 0.012 0.006 0.006
2011 0.018 0.022 0.042 0.037 0.019 0.014
Source: Calculated from the data taken from NSE website for the said period.
Tables 5.14 and 5.15 present the F-test values for the tests that compare the
variance among the returns in Sensex (given in Table 5.12) and Nifty (given in Table
5.13) respectively during the short-term, medium-term, and long-term periods after the
budget with one another. In the case of Sensex, barring the year 2002, in no other year
the actual value exceeded the tabular value. However, in Nifty, there was no case where
the actual value exceeded the tabular value. This signifies that volatility does not
generally increase in post-budget situation as the time progresses.
Table 5.14: F-test Results Comparing Variance among the Returns (Post-budget)
with one another (BSE Sensex)
Year
Actual
Value
Table
Value
(5%)
Actual
Value
Table
Value
(5%)
Actual
Value
Table
Value
(5%)
Y1 and Y2 df = 14/2 Y2 and Y3 df= 29/14 Y3 and Y1 df= 29/2
1996
(interim) 2.00 3.74 1.13 2.03 2.25 3.33
1996 1.13 3.74 1.04 2.03 1.08 3.33
1997 1.68 3.74 1.46 2.31 1.15 3.33
1998 5.68 19.43 1.30 2.03 4.37 19.46
1999 2.16 3.74 2.26 2.31 1.05 19.46
2000 2.36 3.74 1.62 2.31 1.46 3.33
2001 1.61 19.43 1.38 2.03 1.17 19.46
2002 2.33 3.74 1.62 2.03 3.77¬ 3.33
2003 7.00 19.43 1.14 2.31 8.00 19.46
2004
(interim) 1.05 3.74 1.05 2.31 1.00 3.33
2004 3.00 3.74 1.13 2.31 2.67 3.33
2005 2.50 3.74 1.83 2.31 1.36 3.33
2006 1.20 3.74 1.40 2.31 1.17 19.46
2007 1.71 3.74 1.15 2.03 1.97 3.33
2008 1.03 19.43 1.50 2.03 1.46 3.33
2009
(interim) 1.00 3.74 1.88 2.31 1.87 19.46
2009 1.08 19.43 1.05 2.31 1.13 19.46
2010 2.60 3.74 1.20 2.31 2.17 3.33
2011 1.95 3.74 1.27 2.03 2.47 3.33
Source: Calculated from the data provided in Table 5.12. Note: ¬ Signifies that Null Hypothesis (Ho) is rejected.
Table 5.15: F-test Results Comparing Variance among the Returns (Post-budget)
with one another (NSE Nifty)
Year
Actual
Value
Table
Value
(5%)
Actual
Value
Table
Value
(5%)
Actual
Value
Table
Value
(5%)
Y1 and Y2 df = 14/2 Y2 and Y3 df= 29/14 Y3 and Y1 df= 29/2
1996
(interim) 2.58 3.74 1.12 2.03 2.88 3.33
1996 1.27 19.43 1.12 2.03 1.13 19.46
1997 3.03 3.74 1.09 2.03 3.31 3.33
1998 1.96 19.43 1.38 2.03 1.42 19.46
1999 2.48 3.74 1.15 2.31 2.16 3.33
2000 1.72 3.74 1.41 2.31 1.22 3.33
2001 1.98 19.43 1.46 2.03 1.35 19.46
2002 2.00 3.74 1.33 2.03 2.67 3.33
2003 6.50 19.43 1.38 2.31 9.00 19.46
2004
(interim) 1.54 3.74 1.04 2.03 1.60 3.33
2004 2.73 3.74 1.09 2.31 2.50 3.33
2005 2.83 3.74 1.83 2.31 1.55 3.33
2006 1.43 19.43 1.40 2.31 2.00 19.46
2007 1.88 3.74 1.14 2.03 2.14 3.33
2008 1.07 3.74 1.53 2.03 1.64 3.33
2009
(interim) 1.14 3.74 1.69 2.31 1.48 19.46
2009 1.02 19.43 1.08 2.03 1.05 3.33
2010 2.00 3.74 1.00 2.03 2.00 3.33
2011 1.95 3.74 1.36 2.03 2.64 3.33
Source: Calculated from the data provided in Table 5.13.
Table 5.16: F-test Results Comparing Variance among the Returns during Post-
budget Periods with Long-term Pre-Budget Period (BSE Sensex)
Year
Actual
Value
Table
Value
(5%)
Actual
Value
Table
Value
(5%)
Actual
Value
Table
Value
(5%)
X1 and Y1 df=29/2 X1 and Y2 df=29/14 X1and Y3 df=29/29
1996
(interim) 1.00 3.33 2.00 2.31 2.25¬ 1.85
1996 1.37 3.33 1.21 2.03 1.26 1.84
1997 2.09 3.33 1.24 2.03 1.81 1.84
1998 1.68 3.33 3.38¬ 2.03 2.59¬ 1.84
1999 3.35 3.33 1.55 2.31 3.50¬ 1.84
2000 3.11 3.33 1.32 2.03 2.13¬ 1.84
2001 2.95 3.33 4.75¬ 2.03 3.45¬ 1.84
2002 4.45¬ 3.33 1.91 2.03 1.18 1.84
2003 3.00 19.46 2.33¬ 2.03 2.67¬ 1.84
2004
(interim) 1.39 3.33 1.45 2.31 1.39 1.85
2004 1.00 3.33 3.00¬ 2.31 2.67¬ 1.85
2005 1.67 3.33 1.5 2.03 1.22 1.84
2006 1.34 19.46 1.11 2.03 1.56 1.84
2007 5.13¬ 3.33 3.00¬ 2.03 2.60¬ 1.84
2008 1.11 19.46 1.14 2.31 1.32 1.85
2009
(interim) 2.29 19.46 2.29 2.31 1.22 1.85
2009 1.19 3.33 1.28 2.03 1.34 1.84
2010 1.08 19.46 2.80* 2.31 2.33 1.85
2011 2.18 3.33 1.12 2.03 1.14 1.85
Source: Calculated from the data provided in Table 5.12.
Note: ¬ Signifies that Null Hypothesis (Ho) is rejected.
Table 5.17: F-test Results Comparing Variance among the Returns during Post-
budget Periods with Long-term Pre-Budget Period (NSE Nifty)
Actual
Value
Table Value
(5%)
Actual
Value
Table
Value
(5%)
Actual
Value
Table
Value
( 5%)
Year X1 and Y1 df=29/2 X1 and Y2 df=29/14 X1 and Y3 df=29/29
1996
(interim) 1.09 3.33 2.37* 2.31 2.65* 1.85
1996 1.15 3.33 1.46 2.03 1.31 1.84
1997 10.67* 3.33 3.52* 2.03 3.22* 1.84
1998 3.23 3.33 6.35* 2.03 4.59* 1.84
1999 2.16 3.33 1.15 2.31 1.00 1.84
2000 3.44 3.33 2.00 2.03 2.81* 1.84
2001 3.00 3.33 5.94* 2.03 4.06* 1.84
2002 2.90 3.33 1.45 2.03 1.09 1.84
2003 4.00 19.46 1.63 2.03 2.25* 1.84
2004
(interim) 1.08 3.33 1.42 2.31 1.48 1.85
2004 1.07 3.33 2.55* 2.31 2.33* 1.85
2005 1.70 3.33 1.67 2.03 1.10 1.84
2006 1.14 19.46 1.25 2.03 1.75 1.84
2007 5.63* 3.33 3.00* 2.03 2.63* 1.84
2008 1.15 19.46 1.23 2.31 1.89* 1.85
2009
(interim) 1.81 19.46 2.07 2.31 1.22 1.85
2009 1.05 3.33 1.07 2.03 1.00 1.84
2010 1.17 19.46 2.33 2.31 2.34 1.85
2011 2.06 3.33 1.06 2.03 1.29 1.85
Source: Calculated from the data provided in Table 5.13. Note: ¬ Signifies that Null Hypothesis (Ho) is rejected.
Tables 5.16 and 5.17 highlights the specifically F-test values for the tests that
compare the variance of returns in Sensex and Nifty during short-term, medium-term,
and long-term post-budget periods with that of long-term pre-budget period. In the case
of Sensex, maximum number of significant cases (in 9 out of 19 budgets) appeared
during the long-term period as compared to medium-term (in 6 out of 19 budgets) and
short-term (in 2 out of 19 budgets) periods. However, in the case of Nifty, maximum
number of significant cases (in 10 out of 19 budgets) emerged during the long-term
period as compared to medium-term (in 7 out of 19 budgets) and short-term (in 2 out of
19 budgets) periods. It indicates that the long-term period after the budget tends to be
more volatile than the medium-term and short-term periods as compared to similar long-
term period before the budget.
The test values, however, do not prove whether the market index will rise or fall
in the post-budget period because there has been an increase in the index values after the
presentation of budgets.
If one travels back in time to re-live the budget day moves made since the year
1996, the steep cut was witnessed in the year 2009 when the Sensex and Nifty tumbled
over 6% (as shown in Table 5.6 and Table 5.7). Silence on major policy reforms,
increase in MAT and soaring deficits were some of the irritants that led to the decline. In
2000, once again, the Sensex tumbled over 5 per cent and Nifty over 4%. The budget
caused disappointment among the investors as it did not live up to the `hype' created
ahead of its announcement. The increase in the tax on dividend outgo for companies and
subjecting export earnings to a 20 per cent tax per annum over the next five years were
seen as unfavourable by the market and also, the Budget failed to address macro-
economic issues such as fiscal deficit, Government spending and public sector
disinvestment.
In 2002, the stock markets fell by 3 per cent with a lukewarm budget. In 2007,
the stock market crashed by 4 per cent. This was the biggest fall on a Budget day in the
past few years. The fall was due to market unfriendly Union budget. It increased
dividend distribution tax from 12.5 per cent to 15 per cent, followed by an increase in
excise duty on cement prices, and an extension of minimum alternate tax (MAT) for the
IT sector.
On the other hand, there have been occasions when budget had helped the market
to move up in jiffy. In 1997, 1999, 2001 and 2005, popular measures taken by the
finance minister like relaxations in taxes helped the Sensex and Nifty to climb sharply.
Hence, if an investor wants to make any gain from budget swings, he will have to
predict the budget announcements that may cause a rise or fall in post- budget share
prices.
5.1.3.3 Results
The impact of various budgets on the stock market over a long period of 17 years
has been studied taking into account the returns and volatility in Sensex and Nifty. The
hypothesis tests at various levels of significance have provided some interesting results
not only from the of an investors’ and regulators point of view but also from the
government viewpoint.
With regard to returns, an investor has a chance to earn big profits by making
investments during the short-term and medium-term periods around the budget (up to15
trading days). However, there may be a risk of abnormal losses if his expectations are not
met from the budget. This is also true in the case of trading on the budget day. As one
moves away from the budget day (up to 30 trading days), the paired t-tests do not show
any significant change in average returns. Hence, budgets are seen to have effect only up
to 15 trading days from the budget day so far as return is concerned.
Volatility, on the other hand, does not generally increase in post-budget situation
as the time period increases. But the long-term period after the budget tends to be more
volatile than the medium-term and short-term periods as compared to similar long-term
period before the budget. In the case of Sensex, only 11% cases (2 out of 19 budgets)
post budget volatility during short term, in 32% cases (6 out of 19 budgets) post-budget
volatility during medium-term and in 47% cases (9 out of 19 budgets) post budget
volatility during long-term tends to increase in relation to the volatility during long-term
before the budget. However, in the case of Nifty, only 10% cases (2 out of 19 budgets)
post-budget volatility during short-term, in 37% cases (7 out of 19 budgets) post-budget
volatility during medium-term and in 53% cases (10 out of 19 budgets) post-budget
volatility during long-term tends to increase in relation to the volatility during long- term
before the budget. Hence, when volatility and return are considered together, the budget
has greater impact on return than volatility in short-term period, while in long-term
period the impact is reversed.
5.1.4 INFLATION
5.1.4.1 Overview
For a layman it means a rise in prices of commodities of daily use and
subsistence. However, the economists consider it as persistent rise in the general price
level over time. Here, two points are worth noticing; persistent rise and general price
level. One-time rise in prices cannot be termed as inflation; it should continue for some
time to be termed as inflation. Also, unless referring to specific index, inflation usually
rises in the basket of commodities (Rao et al., 2008). Inflation can also be described as a
decline in the real value of money - a loss of purchasing power. It is mostly referred to as
‘too much expenditure chasing too little goods’. It is also referred to as ‘a necessary evil’
for any economy. It is said of inflation that like an elephant, it is easy to spot but
somewhat tricky to measure (Agarwal and Barua, 1999). However, when the general
price level rises, each unit of currency buys fewer goods and services. Thus, inflation
reflects erosion in the purchasing power of money.
In India, the Consumer Price Index (CPI) and the Wholesale Price Index are the two
most important indices. While the CPI pertains to a set of items that a consumer
consumes, the WPI is a basket particular to the wholesale market. Also, CPI measures
changes in prices of goods at retail outlets, while the WPI measures changes in wholesale
prices.
Inflation is measured as the percentage rate of change of a price index over a
particular period of time (Rao and Bhole, 1990). Thus, if the inflation for a particular
week is, say 10%, it means the index is 10% higher than it was in the same week during
the previous year.
Inflation can be attributed to any of the following reasons:
• Inflation via Higher Demand: It is the case of more money chasing few goods. This
means that the demand for the goods is high which leads to higher prices of these goods.
The high demand is usually caused due to an increase in consumer and government
spending. In a developing country like India, inflation via higher demand is usually a
favourable situation for the economy because it is instrumental in continuing the growth
process; also, it can be better managed by the policy-makers.
• Inflation via Lower Supply: Every now and then we hear that poor monsoons have
led to lower harvests of foodgrains, pulses, vegetables, etc. Thus, lower supply in the
economy leads to higher prices. Even if the demand is the same the prices may continue
to rise due to lower supply. This form of inflation is particularly detrimental for a
growing economy like India and it is relatively difficult to manage.
There are different types of inflations like Creeping Inflation, Galloping Inflation,
Hyperinflation, Stagflation, Deflation (Agarwal and Barua, 1999).
Creeping Inflation: This is also known as mild inflation or moderate inflation. This type
of inflation occurs when the price level persistently rises over a period of time at a mild
rate. When the rate of inflation is less than 10 per cent annually, or it is a single digit
inflation rate, it is considered to be a moderate inflation.
Galloping Inflation: If mild inflation is not checked and if it is uncontrollable, it may
assume the character of galloping inflation. Inflation in the double or triple digit range of
20, 100 or 200 per cent a year is called galloping inflation. Many Latin American
countries such as Argentina, Brazil etc. had inflation rates of 50 to 700 per cent per year
in the 1970s and 1980s.
Hyperinflation: It is a stage of very high rate of inflation. While economies seem to
survive under galloping inflation, a third and deadly strain takes hold when the cancer of
hyperinflation strikes. Nothing good can be said about a market economy in which prices
are rising a million or even a trillion per cent per year. Hyperinflation occurs when the
prices go out of control and the monetary authorities are unable to impose any check on
it. Germany had witnessed this type of inflation in 1920s.
Stagflation: It is an economic situation in which inflation and economic stagnation or
recession occur simultaneously and remain unchecked for a period of time. Stagflation
was witnessed by developed countries in 1970s, when world oil prices rose dramatically.
Deflation: Deflation is the reverse of inflation. It refers to a sustained decline in the price
level of goods and services. It occurs when the annual inflation rate falls below zero per
cent (a negative inflation rate), resulting in an increase in the real value of money. Japan
suffered from deflation for almost a decade in 1990s.
Very low inflation rates are not good especially for the growing economies
because it can happen due to
(a) Lower Demand, and/or
(b) Oversupply.
Both these situations are detrimental to the growth process of the economy, and hence,
not desirable (Henry, 2002).
While seeking for a wage hike from our bosses we usually quote high inflation
rates as one of the reasons. But when there is a dip in inflation rates we don’t go running
for wage cuts!! Generally, it is not so easy to reduce the wage rates in an economy. So,
the company usually resorts to reduction in work force. On a macro level, it leads to
higher unemployment in the economy. At this point, the RBI takes action by lowering
interest rate to boost domestic investment but this also leads to lower net inflows. Thus,
zero or very low inflation is not conducive for the growth process (Tribedy, 1991).
The preferred solution, therefore, is a middle path. Moderate inflation rate of 4%-
6% is usually considered not bad for a developing economy like India. Moderate
inflation is necessary to act as a lubricant for the wheels of economy. It signifies healthy
demand in the economy and also avoids RBI intervention which affects consumer and
investment spending (Knif et al.,2003).
How does inflation affect stock market?
Generally, stock markets and inflation are believed to be inversely related. For
businesses, higher inflation means that raw materials become expensive which leads to
higher costs of production. The companies usually pass on a fraction of the price rise to
its final consumers; however invariably they have to absorb a fraction themselves. Thus,
it adversely affects the earnings of the companies. Lower earnings by companies can
make them less attractive for investors, and hence, the stock prices may fall (Cohn and
Lessard, 1981).
Also, during a high inflation period investors look for better rate of return on their
investments in order to maintain or improve the purchasing power of their income. The
investors will find better returns only when the P/E ratios are relatively lower. Thus, in
high inflationary periods the P/E ratio should be ideally lower and similarly, at lower
inflation rates P/E ratios would be higher. Summers (1981) analyzed that increased
inflation raises the expected return on alternative assets such as real physical assets (e.g.
owner-occupied housing). Investors make changes in their portfolios by shifting out of
equity holdings and investing the funds released in the process in other assets. Share
prices decline in response to these portfolio adjustments by investors. Rise in inflation
leads to decrease in after tax profits and hence share prices which are present discounted
values (PDVs) of future after tax earnings of firms - decline (Feldstein,1980).
Unexpected increases in inflation may provoke government or central bank reaction in
the form of changes in fiscal or monetary policy or both. For example, government may
impose price controls or change in tariff rates which might adversely affect the
profitability of firms. The central bank may resort to open market operations to contain
the expansion of money supply pushing up the interest rates in the process. The rise in
interest rates may increase the interest cost of working capital in the short run and
adversely affect the cash flow in the long run as firms resort to cuts in interest sensitive
capital expenditures. This expectation of government or central bank’s reaction may be
the reason for the response of stock prices which is referred to as information effect
(Jaffe and Mandelker, 1976).
5.1.4.2 Analysis and Discussion
The causal relationship between inflation and stock market volatility, i.e., BSE
Sensex and NSE Nifty has been analyzed by applying the Granger Causality Test. For
any time-series data analysis, all data series should be stationary. Augmented Dicker
Fuller (ADF) test has been conducted to study the stationarity of data series. The ADF
test statistics for the inflation is presented in Table 5.18. The results show that ADF
statistic is -4.956150 for inflation. The ADF value for inflation is lesser than critical
values leading to inference of rejection of hypothesis with unit root at 1%, 5% and 10%
levels of significance. This proves that the time series data of inflation is stationary.
Table 5.18: ADF Test Statistics for Inflation
t-Statistic Probability
Augmented Dickey-Fuller test statistic -4.956150 0.0000
Test critical value 1% level -3.465014
5% level -2.876677
10% level -2.574917
Source: Calculated from the data of Inflation given in Appendix 2.
The ADF test statistics for the stock market volatility, i.e., BSE Sensex and NSE
Nifty are presented in Tables 5.19 and 5.20 respectively. The results show that ADF
statistic is - 4.588579 for volatility of BSE Sensex return and -5.136850 for volatility of
NSE Nifty return. Both the ADF values for volatility of stock market return of BSE
Sensex and NSE Nifty are lesser than critical values leading to inference of rejection of
hypothesis with unit root at 1%, 5% and 10% levels of significance. This proves that the
time series data of volatility of stock market in the case of both BSE Sensex and NSE
Nifty is stationary.
Table 5.19: ADF Test Statistics for Stock Market Volatility (BSE Sensex)
t-Statistic Probability
Augmented Dickey-Fuller test statistic -4.588579 0.0002
Test critical value 1% level -3.465014
5% level -2.876677
10% level -2.574917
Source: Calculated from the data of stock market volatility (BSE Sensex) given in
Appendix 2.
Table 5.20: ADF Test Statistics for Stock Market Volatility (NSE Nifty)
t-Statistic Probability
Augmented Dickey-Fuller test statistic -5.136850 0.0000
Test critical value 1% level -3.465014
5% level -2.876677
10% level -2.574917
Source: Calculated from the data of stock market volatility (NSE Nifty) given in Appendix 2.
As the test confirmed that inflation and stock market volatility of BSE Sensex
and NSE Nifty are stationary in nature, the direction of causal relationship between the
stationary data series has been conducted by applying the Granger-Causality test. The
results of the Granger-Causality test are presented in Table 5.21.
The Granger-Causality results suggest independence between inflation and stock
market volatility, i.e., BSE Sensex and NSE Nifty. It can be clearly observed that for the
first and second hypotheses, the probability values are 0.91553 and 0.12785 respectively,
which are greater than critical value 0.05, hence, null hypothesis is accepted.
Table 5.21: Granger-Causality Test Statistics
Null Hypothesis Observation F-statistics Probability Inflation does not Granger Cause Volatility(BSE Sensex) 190 0.08830 0.91553
Volatility (BSE Sensex) does not Granger Cause Inflation 190 2.07997 0.12785
Inflation does not Granger Cause Volatility(NSE Nifty) 190 0.01326 0.98683 Volatility (NSE Nifty) does not Granger Cause Inflation 190 1.72908 0.18030
Source: Compiled from the data given in Appendix 2 using Eviews 5.
It proves that inflation does not cause stock market volatility (BSE Sensex) and
volatility in BSE Sensex does not cause inflation. A similar position exists in the case of
third and fourth hypotheses and hence, null hypothesis is accepted in these cases also.
5.1.4.3 Results
All the series have been found to be stationary since the calculated ADF statistics
are less than critical values. Though theoretically, there seems to be a long run
relationship between stock market volatility and inflation, yet the relationship is not
statistically significant. As far as causality relationship is concerned, it suggests
independence between inflation and stock market volatility i.e., BSE Sensex and NSE
Nifty. Thus, it can be concluded that inflation neither increases nor dampens volatility of
stock market returns in India. There could be several reasons for such insignificant
relationship. However, it could be due to net inflows brought by the FIIs and the FDIs.
Thus, in a growing economy like India, even at double digit inflation rates
investors might find it worthwhile investing in the stock market. So, while investing in
stock market, an investor needs to consider the trend of macro-economic indicator like
inflation which affects the whole economy, and hence, the stock market.
5.1.5 INTEREST RATE
5.1.5.1 Overview
The stock market has become an important indicator of the performance of the
Indian economy over the years. With this, the working of the stock market has become a
vital and facilitating subject for academics, investment professionals, and monetary
policy-makers. The stock market works with the sentiments of participants, which
depend on several factors, making it a very sensitive segment of the economy.
Globalisation and financial sector reforms have added to the sensitivity by increasing
determinants of the stock market movement manifold. Among the several other
important determinants, interest rate is the one which has a relationship with the stock
market and it needs to be carefully studied.
Interest rate, in simple words, means the cost of borrowing funds. It is the
payment a borrower makes to the lender for the facility of using his money. Generally,
interest rate is considered as the cost of capital, means the price paid for the use of
money for a period of time. From the point of view of a borrower, interest rate is the cost
of borrowing money (borrowing rate). From a lender’s point of view, interest rate is the
fee charged for lending money (lending rate) (Amadeo, 2012).
The interest rate in India is not market determined but controlled by the RBI through its
monetary policy. Primarily, there are two main reasons for the fluctuation in interest rate:
• Government alters interest rate to affect the investments: When an economy goes
into recession it does not come back on its own. The US government during the
recent downturn of 2008-09 lowered its interest rate to around 1-2% in order to
propel private investments and for consumers to borrow and spend more.
• Inflation- Interest Rate Linkage: When the inflation starts rising, the central bank
or the RBI responds by increasing the interest rate to reduce money supply in the
economy.
If price is the heart of a commodity market, the rate of interest, which is the price
for the use of money, is the heart of the money and capital markets. Even as in respect of
any commodity there is no single price - there being a complex of prices for the different
varieties of the commodity - there is no single rate of interest. But this is not to say that
the various rates are not related to one another; in fact, they are generally integrated into
a harmonious pattern.
Broadly speaking, the different rates of interest reflect three factors, namely, the
element of risk, the period of the loan and marketability or liquidity of the security.
Greater the probability and magnitude of risk, higher would be the rate. Likewise, longer
the period of the loan, the higher the rate, not so much because of the risk of default as
that of depreciation in the capital value, on account of a possible rise in the rate of
interest during the period of the loan (Tessaromatis, 1989). Greater the marketability,
lower would be the rate of interest. Marketability is also influenced by statutory
provisions and practices with regard to the transferability of assets and the investment
pattern of institutional investor in particular. If the long-term rate is steadily rising, for
instance, the short-term rate should also rise since there would be tendency for the
borrowers to switch from long to short-term loans, and for lenders to shift funds into the
long-term market. The short-term rates proper are those which rule in the money market,
namely, inter-bank call rate, the Treasury bill rate, the commercial bill rate , the deposit
and lending rates of commercial and co-operative banks and finally the bank rate, which
is the lending rate of the central bank. The long-term rates or the capital market rates
refer to the Government and corporate bond rates, yield on ordinary and preference share
etc. (Apergis and Eleftheriou, 2002).
Bank Rate is the rate at which central bank of the country allows finance to
commercial banks. It is a tool, which central bank uses for short-term purposes. Any
upward revision in Bank Rate by central bank is an indication that banks should also
increase deposit rates as well as Base Rate / Benchmark Prime Lending Rate. Thus, any
revision in the Bank Rate indicates that it is likely that interest rates on our deposits are
likely to either go up or go down, and it can also indicate an increase or decrease in our
EMI.
In theory, the relationship between interest rates and stock prices is negative. This
is due to the cash flow discounting model according to which, present values of stocks
are calculated by discounting the future cash flows at a discount rate. If the discount rate
increases, then present values of stocks decline and vice versa. This discount rate is a risk
adjusted required rate of return and equal to the level of interest rates in the economy.
Therefore, an increase in interest rates lowers present values of stocks directly. Even a
relatively small rise in interest rates can have a major effect on present values if it is
spread out over several years (Durre and Giot, 2005). In addition, rising interest rates
reduce cash flows by reducing the profitability of firms. Due to these two reasons,
present values of stocks decline and so do current stock prices. The inverse holds true as
well. Apart from the above theoretical reasons, there are some other reasons also which
account for the negative relationship between interest rates and stock prices. First,
interest rates are risk free returns on bonds and as interest rates on bonds rise, bonds
become more attractive and stocks less attractive. Consequently, there is a change in the
asset allocation in favour of bonds rather than stocks. This moves funds from the stock
market to the bond market, which invariably increases the demand for bonds and reduces
the demand for stocks. As a result, the prices of stocks fall. The opposite is true when
interest rates fall and funds are shifted from the bond market to the stock market (Shiller,
1987).
Second, corporate profitability is hit because of increase in interest rates. Interest
rates affect profits in two ways. Firstly, almost all companies borrow money to finance
capital equipment and inventory, so the cost of money, that is, the interest rate they pay,
is of great importance. Secondly a substantial number of sales are in turn financed by
borrowing. The level of interest rates, therefore, has a great deal of influence on the
ability and willingness of customers to make additional purchases. One of the most
outstanding examples is the automobile industry, in which both producers and consumers
are highly financed. The capital-intensive utility and transportation industries are also
large borrowers, as are all the highly leveraged construction and housing industries
(Flannery and James, 1984).
Perhaps the most important effect of interest rate changes on equity prices comes
from the fact that tight monetary policy associated with rising rates adversely affects
business conditions, whereas falling rates stimulate the economy. Given time, most
businesses can adjust to higher rates, but when they change quickly and unexpectedly
most have to curtail expansion plans, cut inventories, and so on. This has a debilitating
effect on the economy, and therefore, corporate profits. Higher rates and smaller profits
mean lower price/earnings multiples, and therefore, lower stock prices. When the
authorities become concerned about the economy, they lower short-term rates and a
reverse effect takes hold (Bordo et al., 2008).
Third, margin debt is money loaned by brokers for which securities are pledged
as collateral. Normally, this money is used for the acquisition of equities, but sometimes
margin debt is used for purchases of consumer items, such as automobiles. The effect of
rising interest rates on both forms of margin debt is similar, in that rising rates increase
the carrying cost. There is, therefore, reluctance on the part of investors to take on
additional debt as its cost rises. When the service charges become excessive, stocks are
liquidated and the debt is paid off. Rising interest rates have the effect of increasing the
supply of stock put up for sale with consequent downward pressure on prices (Binder
and Merges, 2000). The link between interest rate and stock market is evident from the
following diagram also.
The diagram reflects that there are two sides of the economy (a) The Business
Side, and (b) The Consumer Side. A continuous rise in interest rate affects the stock
price from both the investment and consumption side. Combining the effects from both
the sides, it spells a gloomy situation for the economy. The overall future revenue growth
of the companies would be adversely affected which would lead to negative sentiments
in the market leading to deflated stock prices. Also, at higher interest rate, people tend to
invest more in fixed deposits and bonds as it would be offering higher returns at very low
risk. Hence, it moves funds out of stock market affecting the stock prices adversely (Ray
and Vani, 2007).
Impact of Interest Rate on Different Sectors
• In the short term: The immediate impact of rise in interest rate is on companies
with high debt in their balance sheet .The interest payment made by them rises
which reduces their EPS. Thus there would be negative sentiments for such stock;
resulting into depleted stock price.
• Over a longer term, high interest rates would have more sector specific impact.
The impact of high interest rate can be witnessed the most in real estate,
automobile and all the capital intensive industries. Hence, any investment made
in these sectors during a situation when the interest rates are high can put an
investor into a big loss.
• Banking sector is likely to benefit most due to high interest rates. The Net Interest
Margin (difference between the interest earned by the banks on the money they
lend and the interest they paid to the depositors) of the banks is likely to increase
leading to growth in profits & the stock prices.
• The sectors like pharmaceuticals, FMCG, IT, etc. are least affected by the interest
rates; and investment can be made in these sectors without much risk.
If interest rate continues to rise for a longer duration, it causes negative impact on
the economy leading to recession.
The effects of interest rates change on a stock’s intrinsic value are more complex than
outlined earlier because of the existence of other economic variables that interact with
interest rates in determining a stock’s value. In addition, if the inflation rate is quite high
and real interest rates do not exist, then the investors are unlikely to move their funds
from the stock market to the bond market in response to an increase in interest rate
(Chakradhara, 2008). Hence, the negative relationship between interest rates and stock
prices is not necessarily true. The relationship can also be positive due to the following
reasons. First, if interest rates increase in response to the economy growing too rapidly
then corporate earnings should also be growing rapidly and so should stock prices.
Second, higher interest rates suggest higher anticipated inflation. This leads to a likely
increase in corporate pricing power because of which higher growth rates of earnings per
share are witnessed by firms. Therefore, when the discount factor is increased in the
stock valuation formula, the earnings per share are affected and increased. This implies
that lower stock prices are not necessarily warranted (Durre and Giot, 2005). Third, a
positive relationship can be explained in terms of a changing risk premium. For example,
a drop in interest rates could be the result of increased risk or/and precautionary saving
as investors move away from risky assets such as stocks towards less risky assets like
bonds (Barsky, 1989). However, it is important to note that although the negative
relationship between interest rates and stock prices is not automatic or perfect, in the
long run, it is unavoidable. The above discussion reveals that the relationship between
interest rates and stock prices can either be positive or negative. Therefore, the study tries
to examine the nature of relationship between interest rates and stock prices in the Indian
context.
5.1.5.2 Analysis and Discussion
Tables 5.22 and 5.23 depict the daily average returns and announcement of
interest rate day returns given by Sensex and Nifty during the previous and next 3, 15
and 30 days around the announcement of interest rate day.
On Day Effect of Announcement of Interest Rate on Sensex and Nifty: The first set
of paired t-tests measure the on-day (Z) influence of announcement of interest rate on
Sensex and Nifty when compared to the previous 3, 15 and 30 days. A cursory glance at
Tables 5.22 and 5.23 highlights that in most of the cases announcement of interest rate
day returns (ignoring sign) are more than the returns during the previous 30, 15 and 3
trading days. Therefore, when the announcement of interest rate day returns (Z)
compared with the long- term pre-announcement of interest rate return in the case of
Sensex, it shows that announcement of interest rate day returns exceed in all years ( i.e.,
15 out of 15 ), compared to medium-term (12 out of 15) and short-term (11 out of 15),
while in the case of Nifty it shows that when the announcement of interest rate day
returns (Z) compared with the long-term pre-announcement of interest rate return, it
shows that announcement of interest rate day returns exceed (14 out of 15 ),compared to
medium-term (12 out of 15) and short-term (12 out of 15) that budget day returns
exceed.
Table 5.22: Daily Average Returns in Sensex
Year
X1 (Last
30 days)
X2 (Last
15 days)
X3 (Last
3 days)
Z (Int.
rate)
Y1 (Next
3 days)
Y2 (Next
15 days)
Y3(Next
30 days)
15 April, 97 0.15 -0.26 -0.48 1.60 1.37 0.16 0.14
25 June, 97 0.36 0.47 0.25 -0.63 1.30 0.14 0.33
21 October, 97 0.10 0.38 0.98 -0.92 -1.32 -0.95 -0.57
16 January, 98 -0.19 -0.46 -0.29 0.49 0.35 -0.13 0.31
18 March, 98 0.56 0.65 0.35 0.89 1.19 0.49 0.27
2 April, 98 0.54 0.29 0.52 0.58 1.64 0.15 -0.08
29 April, 98 0.28 0.15 -0.08 -2.80 1.45 -0.11 -0.77
1 March, 99 -0.08 0.04 -0.51 8.60 0.73 0.29 -0.08
1 April, 00 -0.45 -0.65 -1.02 1.03 -1.25 -0.59 -0.62
21 July, 00 -0.02 -0.30 -1.23 -2.48 -2.10 -0.42 0.12
16 February, 01 0.30 0.17 0.24 -2.46 -0.22 -0.93 -0.65
1 March, 01 0.14 0.10 0.99 0.58 -1.80 -1.08 -0.85
22 October, 01 -0.20 0.70 0.27 -0.50 -0.20 0.17 0.45
29 October, 02 -0.33 -0.22 -1.32 1.78 0.75 0.48 0.47
29 April, 03 -0.24 -0.47 -0.12 0.46 0.28 0.24 0.39
Source: Calculated from the data taken from BSE website for the said period. Note: All returns are in percentage.
Table 5.23: Daily Average Returns in Nifty
Year
X1 (Last
30 Days)
X2(Last
15 Days)
X3(Last
3 Days)
Z (Int.
Rate)
Y1 (Next
3 Days)
Y2( Next
15 Days)
Y3 (Next
30 Days)
15 April, 97 0.12 -0.32 -0.89 1.55 1.44 0.25 0.12
25 June, 97 0.38 0.49 -0.13 0.57 0.86 0.12 0.30
21 October, 97 0.24 0.61 1.08 -1.12 -1.10 -0.78 -0.50
16 January, 98 -0.14 -0.42 -0.21 0.82 0.48 -0.15 0.21
18 March, 98 0.39 0.47 0.01 1.83 0.62 0.61 0.24
2 April, 98 0.49 0.38 0.28 0.63 0.86 0.04 -0.05
29 April, 98 0.21 -0.15 -1.01 0.37 1.05 -0.14 -0.57
1 March, 99 0.09 0.46 0.81 3.46 1.25 0.30 -0.16
1 April, 00 -0.38 -0.58 -0.86 0.41 -1.83 -0.44 -0.57
21 July, 00 -0.09 -0.31 -1.27 -1.91 -1.94 -0.43 0.07
16 February, 01 0.31 0.24 0.34 -2.53 -0.27 -0.95 -0.65
1 March, 01 0.24 0.34 -0.77 0.49 -1.70 -1.04 -0.80
22 October, 01 -0.20 0.62 0.18 -0.03 0.23 0.20 0.43
29 October, 02 -0.27 -0.18 -1.23 1.53 0.51 0.49 0.46
29 April, 03 -0.25 -0.55 -0.17 0.30 0.47 0.25 0.38
Source: Calculated from the data taken from NSE website for the said period. Note: All returns are in percentage.
T-values from Paired t-test
Table 5.24: Effect of Announcement of Interest Rate Day Returns on Sensex
X1 and Z X2 and Z X3 and Z
Actual Value (5%) -2.65¬ -2.41¬ -2.10¬
Table Value (5%) -1.76
-1.76
-1.76
Source: Calculated from the data based on Table 5.22. Note: ¬ Signifies that Null Hypothesis (Ho) is rejected.
The above inferences (Tables 5.22 and 5.23) have been further statistically tested
by Paired t- test (Tables 5.24 and 5.25).
Table 5.25: Effect of Announcement of Interest Rate Day Returns on Nifty
X1 and Z X2 and Z X3 and Z
Actual Value (5%) -3.54¬ -2.92¬ -2.24¬
Table Value (5%) -1.76 -1.76 -1.76
Source: Calculated from the data based on Table 5.23. Note: ¬ Signifies that Null Hypothesis (Ho) is rejected.
The announcements of interest rate are found to take the markets by surprise in
all cases. In all the tests, the actual values are found to exceed the table values leading to
acceptance of alternative hypothesis.
T-values from Paired t-test
Table 5.26: Impact of Announcement of Interest Rate on Sensex
Period Short-term Period Medium-term Period Long-term Period
X1 and
Y1
X2 and
Y1
X3 and
Y1
X1 and
Y2
X2 and
Y2
X3 and
Y2
X1 and
Y3
X2 and
Y3
X3 and
Y3
Actual Value (5%)
-4.91¬ -4.00¬ -3.45¬ -1.79* -.617 1.57 -1.88* -.607 1.39
Table Value (5%)
-1.76
-1.76
-1.76
-1.76
-1.76
-1.76
-1.76
-1.76
-1.76
Source: Calculated from the data based on Table 5.22. Note: ¬ Signifies that Null Hypothesis (Ho) is rejected.
Table 5.27: Impact of Announcement of Interest Rate on Nifty
Period Short-term Period Medium-term Period Long-term Period
X1 and
Y1
X2 and
Y1
X3 and
Y1
X1 and
Y2
X2 and
Y2
X3 and
Y2
X1 and
Y3
X2 and
Y3
X3 and
Y3
Actual
Value
(5%)
-4.65¬ -3.72¬ -3.12¬ -1.88* -0.05 1.65 -1.84* -0.56 2.06
Table
Value
(5%)
-1.76
-1.76
-1.76
-1.76
-1.76
-1.76
-1.76
-1.76
-1.76
Source: Calculated from the data based on Table 5.23. Note: ¬ Signifies that Null Hypothesis (Ho) is rejected.
Table 5.24 provides the second set of tests, which proves that announcement of
interest rate has maximum impact in the short-term period (alternative hypotheses have
been accepted in all the three cases). However, in medium-term and long-term periods,
the alternative hypotheses have been accepted in one out of three cases. In total, the
actual values exceed the tabular values in five cases (3+1+1) out of nine at the left tail.
This proves that announcement of interest rate, when taken together, has maximum
impact in the short-term post- announcement of interest rate period, with some impact
extending into the medium-term and long- term average returns.
Table 5.28: Variance of Returns in Sensex
Year
X1(Last
30 Days)
X2(Last
15 Days)
X3(Last 3
Days)
Y1(Next
3 Days)
Y2(Next
15 Days)
Y3(Next
30 Days
15 April, 97 0.072 0.080 0.018 0.019 0.011 0.008
25 June, 97 0.005 0.006 0.002 0.020 0.017 0.015
21 October, 97 0.012 0.009 0.001 0.002 0.012 0.022
16 January, 98 0.023 0.025 0.014 0.048 0.029 0.028
18 March, 98 0.023 0.025 0.02 0.030 0.020 0.025
2 April, 98 0.020 0.016 0.016 0.009 0.028 0.034
29 April, 98 0.021 0.028 0.008 0.023 0.037 0.044
1 March, 99 0.020 0.013 0.007 0.047 0.290 0.052
1 April, 00 0.053 0.025 0.001 0.288 0.152 0.125
21 July, 00 0.022 0.018 0.014 0.255 0.057 0.036
16 February, 01 0.016 0.020 0.018 0.009 0.053 0.080
1 March, 01 0.026 0.082 0.995 0.076 0.094 0.071
22 October, 01 0.064 0.027 0.042 0.023 0.015 0.017
29 October, 02 0.005 0.006 0.000 0.013 0.006 0.008
29 April, 03 0.016 0.015 0.002 0.001 0.004 0.006
Source: Calculated from the data taken from BSE website for the said period.
Table 5.29: Variance of Returns in Nifty
Year
X1(Last
30 Days)
X2(Last
15 Days)
X3(Last
3 Days)
Y1(Next 3
Days)
Y2(Next
15 Days)
Y3(Next
30 Days
15 April, 97 0.087 0.080 0.018 0.018 0.016 0.011
25 June, 97 0.007 0.008 0.001 0.009 0.021 0.018
21 October, 97 0.012 0.011 0.002 0.004 0.090 0.060
16 January, 98 0.023 0.025 0.003 0.017 0.019 0.023
18 March, 98 0.019 0.029 0.011 0.003 0.016 0.024
2 April, 98 0.023 0.018 0.053 0.005 0.029 0.027
29 April, 98 0.024 0.029 0.002 0.010 0.026 0.029
1 March, 99 0.037 0.030 0.087 0.050 0.020 0.029
1 April, 00 0.040 0.035 0.002 0.218 0.148 0.107
21 July, 00 0.015 0.013 0.003 0.169 0.040 0.027
16 February, 01 0.013 0.017 0.015 0.080 0.050 0.069
1 March, 01 0.016 0.015 0.09 0.080 0.101 0.710
22 October, 01 0.054 0.018 0.023 0.017 0.013 0.015
29 October, 02 0.005 0.006 0.001 0.006 0.004 0.009
29 April, 03 0.018 0.021 0.004 0.001 0.004 0.006
Source: Calculated from the data taken from NSE website for the said period.
Tables 5.30 and 5.31 shows F-test values for the tests that compare the variance
among the returns in Sensex (given in Table 5.28) and Nifty (given in Table 5.29)
respectively during short-term, medium-term, and long-term periods after the
announcement of interest rate with one another. In case of Sensex and Nifty, except in
two years, in no other year the actual value exceed the tabular value .This signifies that
volatility does not generally increase in post-announcement of interest rate situation as
time period increases.
Table 5.30: F-test Results Comparing Variance among the Returns (Post-
announcement of Interest Rate) (BSE)
Year
Actual
Value
Table
Value
(5%)
Actual
Value
Table
Value
(5%)
Actual
Value
Table
Value
(5%)
Y1 and Y2 df = 14/2 Y2 and Y3 df= 29/14 Y3 and Y1 df= 29/2
15 April, 97 1.73 3.74 1.38 2.03 2.38 3.33
25 June, 97 1.18 3.74 1.13 2.03 1.33 3.33
21 October, 97 6.00 19.43 1.83 2.31 11.00 19.46
16 January, 98 1.66 3.74 1.04 2.03 1.71 3.33
18 March, 98 1.50 3.74 1.25 2.31 1.20 3.33
2 April, 98 3.11 19.43 1.21 2.31 3.78 19.46
29 April, 98 1.61 19.43 1.19 2.31 1.91 19.46
1 March, 99 6.17 19.43 5.58* 2.03 1.11 19.46
1 April, 00 1.89 3.74 1.22 2.03 2.30 3.33
21 July, 00 4.48* 3.74 1.58 2.03 7.08* 3.33
16 February, 01 5.89 19.43 1.51 2.31 8.89 19.46
1 March, 01 1.24 19.43 1.32 2.03 1.07 3.33
22 October, 01 1.53 3.74 1.13 2.31 1.35 3.33
29 October, 02 2.17 3.74 1.33 2.31 1.63 3.33
29 April, 03 4.00 19.43 1.50 2.31 6.00 19.46
Source: Calculated from the data based on Table 5.28. Note: ¬ Signifies that Null Hypothesis (Ho) is rejected.
Table 5.31: F-test Results Comparing Variance among the Returns (Post-
announcement of Interest Rate) (NSE)
Year
Actual
Value
Table
Value
(5%)
Actual
Value
Table
Value
(5%)
Actual
Value
Table
Value (5%)
Y1 and Y2 df = 14/2 Y2 and Y3 df= 29/14 Y3 and Y1 df= 29/2
15 April, 97 1.13 3.74 1.45 2.03 1.64 3.33
25 June, 97 2.33 19.43 1.17 2.03 2.00 19.46
21 October, 97 22.5* 19.43 1.50 2.03 15.00 19.46
16 January, 98 1.12 19.43 1.21 2.31 1.35 19.46
18 March, 98 5.33 19.43 1.5 2.31 8.00 19.46
2 April, 98 5.8 19.43 1.07 2.03 5.40 19.46
29 April, 98 2.6 19.43 1.12 2.31 2.90 19.46
1 March, 99 2.5 3.74 1.45 2.31 1.72 3.33
1 April, 00 1.47 3.74 1.38 2.03 2.04 3.33
21 July, 00 4.23* 3.74 1.48 2.03 6.23* 3.33
16 February, 01 1.60 3.74 1.38 2.31 1.16 3.33
1 March, 01 1.26 19.43 1.42 2.03 1.13 3.33
22 October, 01 1.31 3.74 1.15 2.31 1.13 3.33
29 October, 02 1.50 3.74 2.25 2.31 1.50 19.46
29 April, 03 4.00 19.43 1.50 2.31 6.00 19.46
Source: Calculated from the data based on Table 5.29. Note: ¬ Signifies that Null Hypothesis (Ho) is rejected.
Table 5.32: F-test Results Comparing Variance among the Returns during Post-
announcement of Interest Rate Periods with Long-term Pre-
announcement of Interest Rate Period (BSE)
Year
Actual
Value
Table
Value
(5%)
Actual
Value
Table
Value
(5%)
Actual
Value
Table
Value
(5%)
X1 and Y1 df=29/2 X1 and Y2 df=29/14 X1and Y3 df=29/29
15 April, 97 3.79 19.46 6.55* 2.31 9.00* 1.85
25 June, 97 4.00* 3.33 3.40* 2.03 3.00* 1.84
21 October, 97 6.00 19.46 1.00 2.31 1.83 1.84
16 January, 98 3.39* 3.33 1.26 2.03 1.22 1.84
18 March, 98 1.30 3.33 1.15 2.31 1.09 1.84
2 April, 98 2.22 19.46 1.40 2.03 1.70 1.84
29 April, 98 1.09 3.33 1.76 2.03 2.09* 1.84
1 March, 99 2.35 3.33 14.5* 2.03 14.5* 1.84
1 April, 00 5.43* 3.33 2.87* 2.03 2.36* 1.84
21 July, 00 11.59* 3.33 2.00* 2.03 1.64 1.84
16 February, 01 1.78 19.46 3.31* 2.03 5.00* 1.84
1 March, 01 2.92 3.33 3.61* 2.03 2.73* 1.84
22 October, 01 2.78 19.46 4.27* 2.31 3.76* 1.85
29 October, 02 2.60 3.33 1.20 2.03 1.60 1.84
29 April, 03 16.00 19.46 4.00* 2.31 2.67* 1.85
Source: Calculated from the data based on Table 5.28. Note: ¬ Signifies that Null Hypothesis (Ho) is rejected.
Tables 5.32 and 5.33 depict specifically F-test values for the tests that compare
the variance of returns in Sensex and Nifty during short-term, medium-term and long-
term post- announcement of interest rate periods with that of long-term pre-
announcement of interest rate period. In the case of both Sensex and Nifty, there have
been 9 out of 15 significant cases in the long-term period, whereas the medium-term and
short-term periods are represented by 8 out of 15 and 4 out of 15 such cases respectively.
It indicates that the long-term period and medium-term period after the announcement of
interest rate tend to be more volatile than the short-term period when compared to similar
long-term period before the announcement of interest rate.
Hence, if an investor wants to have any gain from the swing arising after the
announcement of interest rate, he needs to predict whether such an announcement will
cause a rise or fall in share prices.
Table 5.33: F-test Results Comparing Variance among the Returns during Post-
announcement of Interest Rate Periods with Long-term Pre-
announcement of Interest Rate Period (NSE)
Year
Actual
Value
Table
Value
(5%)
Actual
Value
Table
Value
(5%)
Actual
Value
Table
Value
(5%)
X1 and Y1 df=29/2 X1 and Y2 df=29/14 X1and Y3 df=29/29
15 April, 97 4.83 19.46 5.44* 2.31 7.90* 1.85
25 June, 97 1.28 3.33 3.00* 2.03 2.57* 1.84
21 October, 97 3.00 19.46 7.50* 2.03 5.00* 1.84
16 January, 98 1.35 19.46 1.21 2.31 1.00 1.85
18 March, 98 6.33 19.46 1.19 2.31 1.26 1.84
2 April, 98 4.60 19.46 1.26 2.03 1.17 1.84
29 April, 98 2.40 19.46 1.08 2.03 1.21 1.84
1 March, 99 1.35 3.33 1.85 2.31 1.28 1.85
1 April, 00 5.45* 3.33 3.70* 2.03 2.68* 1.84
21 July, 00 11.27* 3.33 2.67* 2.03 1.89* 1.84
16 February, 01 6.15* 3.33 3.85* 2.03 5.31* 1.84
1 March, 01 5.00* 3.33 0.16 2.03 4.43* 1.84
22 October, 01 3.18 19.46 4.15* 2.03 3.60* 1.84
29 October, 02 1.20 3.33 1.25 2.31 1.80 1.84
29 April, 03 18.00 19.46 4.50* 2.31 3.00* 1.85
Source: Calculated from the data based on Table 5.29. Note: ¬ Signifies that Null Hypothesis (Ho) is rejected.
5.1.5.3 Results
An attempt has been made to measure the impact of 15 announcements made
about the interest rate over a period of 17 years on the stock market considering the
returns and volatility in Sensex and Nifty. The hypothesis tests conducted at various
significance levels have provided some interesting results from the investors, the
regulators and the Government point of view.
With regard to returns, an investor has a chance to earn greater profits by
investing during the short-term and medium-term periods (up to15 trading days).
However, he also faces the risk of abnormal losses, if his expectations from such
announcements do not come true. This is also true in the case of trading on the
announcement day. As one moves away from the announcement day (up to 30 trading
days), the paired t-tests do not show any significant change in average returns. Hence,
such announcements have an effect only up to 15 trading days from the budget day so far
as returns are concerned.
Volatility, on the other hand, does not generally increase in a post-announcement
situation as the time period increases. But the long-term period after the announcement
tends to be more volatile than the medium-term and the short-term periods when
compared to similar long-term period before the announcement. In the case of both
Sensex and Nifty, only 26% ( 4 out of 15) post-interest rate announcements caused
volatility during the short-term period, 53% (8 out of 15) post-interest rate
announcements caused volatility during the medium-term period, while and 60% ( 9 out
of 15) post-interest rate announcements caused volatility during the long-term period in
comparison to volatility during the long-term period before such announcements.
Hence, when volatility and return are considered together, it is found that an
announcement of interest rate has a greater impact on return than volatility in short-term
period, but in long-term period such an announcement of interest rate has greater impact
on the volatility than return.
5.1.6 CORPORATE FUNDAMENTALS
5.1.6.1 Overview
Share price is the most important indicator readily available to the investors for
their decision to invest in a particular share. The market price of shares tends to fluctuate
every moment. It may boost up to new heights or even may drop down tremendously. No
one can exactly predict what could be the next price of the share. For predicting share
prices there are different approaches. Fundamental approach predicts share price on the
basis of financial, environmental and managerial factors, whereas technical approach
takes the help of past trends in predicting future share price. Generally, the price depends
on the demand and supply of the shares. If there are more buyers than the sellers, the
prices of the shares are very likely to rise and vice versa. This is called the demand-
supply mechanism of the share price determination. Another vital factor is the
information regarding any company, which falls under the wide concept of Efficient
Market Hypothesis. If there is any good news regarding the company, the share prices of
that company are likely to boost and viceversa. Understanding of the effect of various
fundamental variables on share price is very much helpful to various parties such as
investors, management, government, etc., as it will help them in taking various important
decisions. Though investors are aware of these facts, but what drives the demand of the
shares? What kind of information causes the rise or fall in share prices? Which are the
factors that contribute to the determination of equity prices? In other words what are the
determinants of equity prices? Various factors undertaken for the study are:
MARKET PRICE (MP):
The market price of the share is mainly determined by the forces of demand and
supply of a particular security in the market (Malhotra, 1987; Piatroski, 2004; Zahir and
Khanna, 1982). The market price reflects the collective wisdom and knowledge of the
market. Daily price fluctuations arise because of changes in the buying and selling
pressure. Due to these fluctuations it becomes difficult to decide as to which market price
should be regressed as a measure of dependent variable. However, the present study
takes into consideration the arithmetic means of high and low market price of share
during the financial year of a firm. Mathematically, it is calculated as:
MP =
Where, PH is the highest market price, PL is the lowest market price during the year
which relates to the t’ period.
EARNING PER SHARE (EPS)
Since equityholders are residual claimants to the earnings of the corporation, the
net profits to be taken for computation of EPS are usually arrived at after subtracting the
preference dividends. The significance of this ratio flow from the fact that higher the
earnings per share, the more is the scope for a higher rate of dividend and also of retained
earnings, to build up the inner strength of the company. Therefore, a higher EPS would
increase the market price and viceversa.
EPS =
DIVIDEND PER SHARE (DPS)
Dividend per share is the actual amount of dividend (gross) declared per share.
The net profit after tax belongs to shareholders. But the income, which they really
receive, is the amount of earnings distributed and paid as cash dividends. Therefore, a
large number of present and potential investors are more interested in the dividend per
share, rather than the earning per share. The amount of dividend paid to the shareholders
depends upon the dividend policy pursued by a company. The stable dividend policy
helps in resolving uncertainty from the minds of the investors and also plays an
important role in creating a healthy investment climate. The dividend rate of a company
has a significant influence on the market price of a share.
Dividend Per Share =
BOOK VALUE (BV)
It is also known as net asset value per share because it measures the amount of
assets, which the corporation has on behalf of each equity share. BV shows the
investment per share made in the business by the shareholders. A high book value
usually indicates that the company has a good record of past performance, i.e., high
reserves, therefore, high market price (Grewal,1986).
Book value per share =
PRICE EARNING RATIO (P/E )
P/E ratio expresses the relationship between the market price of a company’s
share and its earnings per share. It indicates the extent to which the earnings of each
share are covered by its price. The ratio helps an investor to make an approximate
calculation of the time required to recover his investment in a company share. The price-
earning ratio has a positive relationship with market price.
P/E =
DIVIDEND YIELD
This is the return earned by an equity shareholder by way of dividends. Dividend
yield is computed as:
Dividend yield = *100 5.1.6.2 ANALYSIS AND DISCUSSION
The multivariate regression analysis has been carried out with Market Price (MP)
as dependent variable and Earning per Share (EPS), Dividend Per share (DPS), Book
Value (BV), Price-Earning Ratio (P/E) and Dividend Yield (DY) as independent
variables. The multivariate analysis technique was selected because it is the most
appropriate tool for evaluating the individual and combined effect of a set of independent
variables on dependent variable.
5.1.6.2.1 ANALYSIS OF HDFC BANK
The test for significance of overall multiple regression model was made through
F-test. This test explains whether there is a linear regression relationship between the
dependent variable and any of the independent variables. The results of the test as shown
in Table 5.34 reveal that in both the cases, i.e., BSE and NSE the F-values are significant
for being less than the significance level of 0.05. Hence, null hypothesis that there is no
relationship between the market price and independent variables in the case of both BSE
(Hoq) and NSE (Hor) stands rejected. As the time series analysis of economic variables
hold true under the assumption of no autocorrelation, Durbin-Watson test has been used.
This assumption is likely to be met if the Durbin-Watson statistic is close to 2. The
computed values of Durbin-Watson statistics are 1.78 and 1.75 in the case of BSE and
NSE respectively. Hence, no autocorrelation exists.
The coefficient of multiple correlation, R, indicates the relationship between
predictors and dependent variable. The value of R in both the cases is 0.985 which shows
a stronger relationship. Similarly, R2, the coefficient of multiple determination, measures
the percentage of the variation in the dependent variable which is explained by variations
in the independent variables taken together. For regression equation, R2 is 0.970 in both
the cases, indicating that 97.0 per cent of the variation in market price is explained by the
variations in Earning Price share and Price/Earning ratio.
Table 5.34: Regression Model Summary and ANOVA of HDFC BANK
Model R R2 F- change Sig. F-change Durbin-Watson
BSE(Sensex) 0.985 0.970 70.221 0.000 1.78
NSE(Nifty) 0.985 0.970 72.210 0.000 1.75
Source: Compiled from the data given in Appendix 6 using SPSS 16.
Regression Equation:
For BSE Sensex
Market Price (MP) = b0 + b1 Earning Per Share (EPS) +b4 Price/Earning Ratio (P/E)
= -296.345+27.38EPS+10.12P/E
For S&P CNX Nifty
Market Price (MP) = b0 + b1 Earning Per Share (EPS) +b4 Price/Earning Ratio (P/E)
= -294.45 + 27.35EPS+10.09P/E
The b-values denote about the relationship between dependent variable and each
predictor. If the value is positive it explains that there is a positive relationship between
the predictor and the outcome, whereas a negative coefficient represents a negative
relationship. In both the regression equations two of the predictors have positive b-values
indicating positive relationships. Hence, in both the cases market price increases with an
increase in Earning Per share as well as Price Earning Ratio. The b-values also explain
what degree each predictor affects the outcome if the effects of all other predictors are
held constant.
Table 5.35: Coefficients of HDFC Bank (BSE Sensex)
Model b-value t-value Sig.
(Constant) -296.345 -2.588 0.023
EPS 27.376 20.440 0.000
P/E 10.117 2.570 0.023
Source: Compiled from the data given in Appendix 6 using SPSS 16.
Earning Per Share (EPS) (b = 27.37): This value indicates that as EPS increases by
Re.1, MP increases by Rs.27.37. This interpretation stands true only if the effect of P/E
is held constant.
Price/Earning Ratio (P/E) (b =10.11): This value indicates that as P/E increases by Re.
1, MP increases by Rs.10.11. This interpretation is true only if the effect of EPS is held
constant.
Table 5.36: Coefficients of HDFC Bank (S&P CNX Nifty)
Model b-value t-value Sig.
(Constant) -294.450 -2.588 0.023
EPS 27.351 20.418 0.000
P/E 10.090 2.563 0.024
Source: Compiled from the data given in Appendix 6 using SPSS 16.
Earning Per Share (EPS) (b = 27.35): This value indicates that as EPS increases by
Re.1, MP increases by Rs.27.35. This interpretation stands true only if the effect of P/E
is held constant.
Price/Earning Ratio (P/E) (b =10.09): This value indicates that as P/E increases by Re.
1, MP increase by Rs. 10.09. This interpretation is true only if the effect of EPS is held
constant.
In simple regression, a significant value of t indicates that the slope of the
regression line is significantly different from horizontal, but in multiple regression, it is
not so easy to visualize what the value explains. Well, it is easiest to conceptualize the t-
tests as measures to find whether the predictor is making a significant contribution to the
model. Therefore, if the t-test associated with a b-value is significant (if the value in
column labelled significant is less than the significance level of 0.05) then predictor is
making a significant contribution to the model. The smaller the value of significant and
the larger the value of t, the greater the contribution of the predictor. For both the
models, the EPS and P/E are the significant predictors of market price as their values
stand below the significance level of 0.05. The magnitude of t-statistics reflects that EPS
has greater impact than P/E in both the cases.
5.1.6.2.2 ANALYSIS OF STATE BANK OF INDIA
The test for significance of overall multiple regression model was made through
F-test. The results of the test as shown in Table 5.37 reveal that in both the cases, the F-
values are significant for being less than the significance level of 0.05. Hence, null
hypothesis that there is no relationship between the market price and independent
variables in the case of both BSE (Hoq) and NSE (Hor) stands rejected. The computed
values of Durbin-Watson statistics are 1.77 and 1.19 in the case of BSE and NSE
respectively. Hence, no autocorrelation exists.
The values of ‘R’ are 0.983 (BSE) and 0.977 (NSE) which reflect a stronger
relationship. For regression equation, R2 values are 0.966 (BSE) and 0.954 (NSE)
indicating that 96.6 per cent and 95.4 per cent respectively of the variation in market
price is explained by the variations in Earning Per Share, Book Value, and
Price/Earning Ratio.
Table 5.37: Regression Model Summary and ANOVA of State Bank of India
Model R R2 F-change Sig. F-change Durbin -Watson
BSE (Sensex) 0.983 0.966 58.185 0.000 1.77
NSE (Nifty) 0.977 0.954 48.592 0.000 1.19
Source: Compiled from the data given in Appendix 6 using SPSS 16.
Regression Equation:
For BSE Sensex
Market Price (MP) = b0 + b1Earning Per Share (EPS) +b3 Book Value (BV) +b4
Price/Earning Ratio (P/E)
= -744.74- 2.21EPS+1.64BV +36.54P/E
For S&P CNX Nifty
Market Price (MP) = b0 + b1Earning Per Share (EPS) +b3 Book Value (BV) +b4
Price/Earning Ratio (P/E)
= -931.92- 10.60 EPS+1.84BV +97.49P/E
The b-values denote about the relationship between dependent variable and each
predictor. In both the regression equations two of the predictors have positive b-values
indicating positive relationships and one has a negative b-value showing a negative
relationship. Hence market price increases with an increase in Price Earning Ratio as
well as Book value. However, a decrease in Earning Per Share causes an increase in
market price. The b-values also explain what degree each predictor affects the outcome if
the effects of all other predictors are held constant.
Earning Per Share (EPS) (b =-2.21) : This value indicates that as EPS decreases by Re.
1, MP increases by Rs. 2.21 .This interpretation stands true only if the effects of BV and
P/E are held constant.
Table 5.38 Coefficients of State Bank of India (BSE Sensex)
Model b-value t-value Sig.
(Constant) -744.743 -7.477 0.000
EPS -2.213 -.359 0.000
BV 1.643 10.215 0.000
P/E 36.538 11.756 0.000
Source: Compiled from the data given in Appendix 7 using SPSS 16.
Book Value (BV) (b =1.64): This value indicates that as BV increases by Re. 1, MP
increases by 1.64. This interpretation stands true only if the effects of EPS and P/E are
held constant.
Price/Earning Ratio (P/E) (b = 36.53): This value indicates that as P/E increases by Rs.
1, MP increases by Rs.36.53. This interpretation is true only if the effects of EPS and BV
are held constant.
Table 5.39 Coefficients of State Bank of India (S&P CNX Nifty)
Model b-value t-value Sig.
(Constant) -931.923 -7.823 0.000
EPS -10.602 -8.575 0.000
BV 1.843 11.220 0.000
P/E 97.494 68.121 0.000
Source: Compiled from the data given in Appendix 7 using SPSS 16.
Earning Per Share (EPS) (b = -10.60) : This value indicates that as EPS decreases by
Re. 1, MP increases by Rs. 10.60 .This interpretation stands true only if the effects of BV
and P/E are held constant.
Book Value (BV) (b =1.84): This value indicates that as BV increases by Re. 1, MP
increase by 1.84. This interpretation stands true only if the effects of EPS and P/E are
held constant.
Price/Earning Ratio (P/E) (b = 97.49): This value indicates that as P/E increases by Re.
1, MP increase by Rs. 97.49. This interpretation stands true only if the effects of EPS and
BV are held constant.
The t-test measures whether the predictor is making a significant contribution to
the model. Therefore, if the t-test associated with a b-value is significant (if the value in
column labelled significant is less than the significance level of 0.05) then predictor is
making a significant contribution to the model. The smaller the value of significant and
the larger the value of t, the greater the contribution of the predictor. For both the
models, the EPS, BV, and P/E are the significant predictors of market price as their
values stand below values the significance level of 0.05. The magnitude of t-statistics
reflects that P/E has greater impact than BV and EPS in both the cases.
5.1.6.2.3 ANALYSIS OF TATA MOTORS
The test for significance of overall multiple regression model was made through
F-test. The results of the test as exhibited in Table 5.40 reveal that in both the cases the
F-values are significant for being less than the significance level of 0.05. Hence, null
hypothesis that there is no relationship between the market price and independent
variables in the case of both BSE (Hoq) and NSE (Hor) stands rejected. The computed
values of Durbin-Watson statistics are 1.57 and 1.92 in the case of BSE and NSE
respectively. Hence, no autocorrelation exists.
The values of ‘R’ are 0.947 (BSE) and 0.982 (NSE) which prove a stronger
relationship. For regression equation, R2 values are 0.897 (BSE) and 0.964 (NSE),
indicating that 89.7 per cent and 96.4 per cent respectively of the variation in market
price is explained by variations in Dividend Per Share, Book Value and Dividend Yield.
Table 5.40: Regression Model Summary and ANOVA of Tata Motors
Model R R2 F-change Sig. F-change Durbin -Watson
BSE (Sensex) 0.947 0.897 14.253 0.001 1.575
NSE (Nifty) 0.982 0.964 38.950 0.000 1.925
Source: Compiled from the data given in Appendix 8 using SPSS 16.
Regression Equation:
For BSE Sensex
Market Price (MP) = b0 + b2Dividend Per Share (DPS) +b3 Book Value (BV) +b5
Dividend Yield (DY)
= 133.52 + 50.76DPS +1.26BV - 127.8DY
For S&P CNX Nifty
Market Price (MP) = b0 + b2Dividend Per Share (DPS) +b3 Book Value (BV) +b5
Dividend Yield (DY)
= 169.81 + 58.12 DPS +.97 BV – 145.46 DY
The b-values denote about the relationship between dependent variable and each
predictor. In both the regression equations two of the predictors have positive b-values
indicating positive relationships and one has a negative b-value showing a negative
relationship. Hence, with the increase in Dividend Per Share, market price increases as
other variables like Book value and Dividend Yield remain constant. Similarly, with
increase in Book Value and decrease in Dividend Yield individually (if other variables
remain constant), market price increases. The b-values also explain what degree each
predictor affects the outcome if the effects of all other predictors are held constant.
Table 5.41: Coefficients of Tata Motors (BSE Sensex)
Model b-value t-value Sig.
(Constant) 133.101 1.793 0.103
DPS 50.779 6.458 0.000
BV 1.2550 2.611 0.026
DY -127.851 -6.517 0.000
Source: Compiled from the data given in Appendix 8 using SPSS 16
Dividend Per Share (DPS) (b = 50.77) : This value indicates that as DPS increases by
Re. 1, MP increases by Rs. 50.77 .This interpretation stands true only if the effects of BV
and DY are held constant.
Book Value (BV) (b=1.25): This value indicates that as BV increases by Re. 1, MP
increases by Rs. 1.25. This interpretation stands true only if the effects of DPS and DY
are held constant.
Dividend Yield (DY) (b = -127.85): This value indicates that as DY increases by Re. 1,
MP decreases by Rs. 127.85. This interpretation stands true only if the effects of DPS
and BV are held constant.
Table 5.42: Coefficients of Tata Motors (S&P CNX Nifty)
Model b-value t-value Sig.
(Constant) 169.813 1.763 0.112
DPS 58.121 14.853 0.000
BV 0.974 1.981 0.079
DY -145.468 -7.123 0.000
Source: Compiled from the data given in Appendix 8 using SPSS 16
Dividend Per Share (DPS) (b = 58.12) : This value explains that as DPS increases by
Re. 1, MP increases by Rs. 58.12 .This interpretation stands true only if the effects of BV
and DY are held constant.
Book Value (BV) (b =0.97): This value depicts that as BV increases by Re. 1, MP
increases by 97 paise only. This interpretation stands true only if the effects of DPS and
DY are held constant.
Dividend Yield (DY) (b= -145.46): This value reflects that as DY increases by Re. 1,
MP decreases by Rs. 145.46. This interpretation stands true only if the effects of DPS
and BV are held constant.
The t-test measures whether the predictor is making a significant contribution to
the model. Therefore, if the t-test associated with a b-value is significant (if the value in
column labelled significant is less than the significance level of 0.05) then predictor is
making a significant contribution to the model. Smaller the value of significant and
larger the value of t, the greater the contribution of the predictor. For both the models,
the DPS, BV, and DY are the all significant predictors of market price as their values
appear below 0.05. The magnitude of t-statistics reflects that DPS has deeper impact than
DY and BV in both the cases.
5.1.6.2.4 ANALYSIS OF MAHINDRA AND MAHINDRA
The test for significance of overall multiple regression model was made through
F-test. The results of the test as shown in Table 5.43 describe that in both the cases, i.e.,
BSE and NSE the F-values are significant for being less than the significance level of
0.05. Hence, null hypothesis that there is no relationship between the market price and
independent variables in the case of both BSE (Hoq) and NSE (Hor) stands rejected. The
computed values of Durbin-Watson test are 1.28 and 1.46 in the case of BSE and NSE
respectively. Hence, no autocorrelation exists.
The values of ‘R’ are 0.991 (BSE) and 0.959 (NSE) which show a stronger
relationship. For regression equation, R2 values are 0.983(BSE) and 0.920 (NSE),
indicating that 98.3 per cent and 92.0 per cent respectively of the variation in market
price is explained by variations in Earning Price Share, Price/Earning Ratio and
Dividend Yield.
Table 5.43: Regression Model Summary and ANOVA of Mahindra and Mahindra
Model R R2 F change Sig. F-change Durbin –Watson
BSE (Sensex) 0.991 0.983 90.428 0.000 1.289
NSE (Nifty) 0.959 0.920 25.008 0.000 1.460
Source: Compiled from the data given in Appendix 9 using SPSS 16.
Regression Equation is:
For BSE Sensex
Market Price (MP) = b0 + b1 Earning Per Share (EPS) +b4 Price/Earning Ratio (P/E) +b5
Dividend Yield (DY)
= -379.35 + 13.99 EPS + 28.03P/E + 3.94DY
For S&P CNX Nifty
Market Price (MP) = b0 + b1 Earning Per Share (EPS) +b4 Price/Earning Ratio (P/E) +b5
Dividend Yield (DY)
= -266.43 + 15.12 EPS + 19.29 P/E + 5.65DY
The b-values denote about the relationship between dependent variable and each
predictor. In both the regression equations three of the predictors have positive b-values
indicating positive relationships. Hence, with the increase in Earning Per Share market
price increases as other variables like Price Earning Ratio and Dividend Yield remain
constant. Similarly, with increase in P/E and DY individually (if other variables remain
constant), market price increases. The b-values also explain what degree each predictor
affects the outcome if the effects of all other predictors are held constant.
Table 5.44: Coefficients of Mahindra and Mahindra (BSE Sensex)
Model b-value t-value Sig.
(Constant) -379.352 -11.149 0.000
EPS 13.986 18.351 0.000
P/E 28.026 11.959 0.000
DY 3.943 11.921 0.000
Source: Compiled from the data given in Appendix 9 using SPSS 16.
Earning Per Share (EPS) (b = 13.98): This value reflects that as EPS increases by Re.1,
MP increases by Rs 13.98. This interpretation stands true only if the effects of P/E and
DY are held constant.
Price/Earning Ratio (b = 28.02): This value indicates that as P/E increases by Re. 1,
MP increases by Rs. 28.02. This interpretation stands true only if the effects of EPS and
DY are held constant.
Dividend Yield (DY) (b = 3.94): This value indicates that as DY increases by Re. 1, MP
increases by Rs. 3.94. This interpretation stands true only if the effects of EPS and P/E
are held constant.
Table 5.45: Coefficients of Mahindra and Mahindra (S&P CNX Nifty)
Model b-value t-value Sig.
(Constant) -266.432 -4.089 0.001
EPS 15.116 9.444 0.000
P/E 19.290 4.445 0.001
DY 5.652 4.421 0.000
Source: Compiled from the data given in Appendix 9 using SPSS 16
Earning Per Share (EPS) (b = 15.11): This value explains that as EPS increases by
Re.1, MP increases by Rs. 15.11. This interpretation stands true only if the effects of P/E
and DY are held constant.
Price/Earning Ratio (b =19.29): This value means that as P/E increases by Re. 1, MP
increases by Rs. 19.29. This interpretation stands true only if the effects of EPS and DY
are held constant.
Dividend Yield (DY) (b = 5.65): This value indicates that as DY increases by Re. 1, MP
increases by Rs. 5.65. This interpretation stands true only if the effects of EPS and P/E
are held constant.
The t-test measures whether the predictor is making a significant contribution to
the model. Therefore, if the t-test associated with a b-value is significant (if the value in
column labelled significant is less than the significance level of 0.05) then predictor is
making a significant contribution to the model. The smaller the value of significant and
the larger the value of t, the greater the contribution of the predictor. For both the
models, the EPS, P/E, and DY are the significant predictors of market price as their
values appear less than the significance level of 0.05. The magnitude of t-statistics
reflects that EPS has greater impact than P/E and DY in both the cases.
5.1.6.2.5 ANALYSIS OF HERO MOTORS
The test for significance of overall multiple regression model was made through
F-test. The results of the test as exhibited in Table 5.46 reveal in both the cases the F-
values are significant for being less than the significance level of 0.05. Hence, null
hypothesis that there is no relationship between the market price and independent
variables in the case of both BSE (Hoq) and NSE (Hor) stands rejected. The computed
values of Durbin-Watson test are 1.78 and 1.45 in the case of BSE and NSE respectively.
Hence, no autocorrelation exists.
The values of ‘R’ is 0.997(BSE) and 0.996(NSE) which prove a stronger
relationship. For regression equation, R2 values are 0.994 (BSE) and 0.996 (NSE) ,
indicating that 99.4 per cent and 99.6 per cent respectively of the variation in market
price is explained by variations in Earning Price Share, Price/Earning Ratio, Dividend
Yield.
Table 5.46: Regression Model Summary and ANOVA of Hero Motors
Model R R2 F-change Sig. F-change Durbin -Watson
BSE (Sensex) 0.997 0.994 198.68 0.000 1.788
NSE (Nifty) 0.996 0.992 124.028 0.000 1.450
Source: Compiled from the data given in Appendix 10 using SPSS 16.
Regression Equation:
For BSE Sensex
Market Price (MP) = b0 + b1 Earning Per Share (EPS) +b4 Price/Earning Ratio (P/E) +b5
Dividend Yield (DY)
= -406.13 +17.24 EPS + 25.77P/E -18.15 DY
For S&P CNX Nifty
Market Price (MP) = b0 + b1 Earning Per Share (EPS) +b4 Price/Earning Ratio (P/E) +b5
Dividend Yield (DY)
= -497.43 +17.79 EPS + 29.15P/E -11.62DY
The b-values denote about the relationship between dependent variable and each
predictor. In both the regression equations two of the predictors have positive b-values
indicating positive relationships and one has a negative b-value showing a negative
relationship. Hence, market price increases with an increase in Earning Per Share as well
as Price Earning Ratio. However, a decrease in Dividend Yield causes an increase in
market price. The b-values also explain what degree each predictor affects the outcome if
the effects of all other predictors are held constant.
Table 5.47: Coefficients of Hero Motors (BSE Sensex)
Model b-value t-value Sig.
(Constant) -406.132 -7.822 0.000
EPS 17.241 38.273 0.000
P/E 25.768 9.736 0.000
DY -18.152 -2.795 0.016
Source: Compiled from the data given in Appendix 10 using SPSS 16.
Earning Per Share (EPS) (b = 17.24): This value indicates that as EPS increases by
Re.1, MP increases by Rs. 17.24. This interpretation stands true only if the effects of P/E
and DY are held constant.
Price/Earning Ratio (b = 25.76): This value indicates that as P/E increases by Re. 1,
MP increases by Rs. 25.76. This interpretation comes true only if the effects of EPS and
DY are held constant.
Dividend Yield (DY) (b = -18.15): This value explains that as DY increases by Re. 1,
MP decreases by Rs. 18.15. This interpretation appears true only if the effects of EPS
and P/E are held constant.
Table 5.48: Coefficients of Hero Motors (S&P CNX Nifty)
Model b-value t-value Sig.
(Constant) -497.427 -10.835 0.000
EPS 17.787 27.760 0.000
P/E 29.155 11.556 0.000
DY -11.619 -2.374 0.035
Source: Compiled from the data given in Appendix 10 using SPSS 16.
Earning Per Share (EPS) (b = 17.78): This value denotes that as EPS increases by Re.1,
MP increases by Rs 17.78. This interpretation comes true only if the effects of P/E and
DY are held constant.
Price/Earning Ratio (b = 29.15): This value reflects that as P/E increases by Re. 1, MP
increases by Rs. 29.15. This interpretation stands true only if the effects of EPS and DY
are held constant.
Dividend Yield (DY) (b = -11.61): This value indicates that as DY increases by Re. 1,
MP decreases by Rs. 11.61. This interpretation comes true only if the effects of EPS and
P/E are held constant.
The t-test measures whether the predictor is making a significant contribution to
the model. Therefore, if the t-test associated with a b-value is significant (if the value in
column labelled significant is less than the significance level of 0.05) then predictor is
making a significant contribution to the model. The smaller the value of significant and
the larger the value of t, the greater the contribution of the predictor. For both the model,
the EPS, P/E, and DY are the significant predictors of market price as their values appear
less than the significance level of 0.05. The magnitude of t-statistics reflects that EPS has
greater impact than P/E and DY in both the cases.
5.1.6.3 Results
The results of the test reveal that in all the selected companies (HDFC Bank,
State Bank of India, Tata Motors, Mahindra and Mahindra, and Hero Motors) and in both
the cases (i.e. BSE and NSE) the F-values are significant for being less than the
significance level of 0.05. Hence, the null hypothesis that there is no relationship
between the market price and independent variables in the case of both BSE (Hoq) and
NSE (Hor) stands rejected. The computed values of Durbin-Watson test in the case of all
the selected companies are less than 2 which reflect that no autocorrelation exists.
The values of R and R2 are greater than 0.90 in all companies under study as well
as in BSE and NSE which establish a stronger relationship between Market Price and
independent variables while R2 indicates that 90.0 per cent of the variation in market
price is explained by variations in independent variables respectively.
The regression analysis further provides that in all the companies under study as
well as in BSE and NSE, the variables such as EPS, P/E, and DY have contributed most
in determining share prices. It has also been found that EPS, and P/E are the most
important determinants which have influenced the share price in banking sector, while in
the automobile sector DY, P/E and EPS are the most significant variables.
5.2 RELATIONSHIP OF VOLATILITY AND STOCK PRICE
RETURNS
5.2.1 Overview
The relationship between the return on an asset and its variance (or volatility) as a
proxy for risk has been an important topic in financial research. The theoretical asset
pricing models (e.g., Sharpe, 1964; Linter, 1965; Mossin, 1966; Merton, 1973, 1980)
typically link the return (or the price change) of an asset to its own return variance, or to
the covariance between its return and the return on the market portfolio. However, it has
been controversial whether such a relationship is positive or negative. As summarized by
Ballie and DeGennaro (1990)in their research work, most asset-pricing models (e.g.,
Sharpe, 1964; Linter, 1965; Mossin, 1966; Merton, 1973) postulate a positive
relationship between a stock portfolio’s expected returns and volatility. On the other
hand, there is also a long tradition in finance that models stock return volatility as
negatively correlated with stock returns (Black, 1976; Cox and Ross, 1976; Bekaert and
Wu, 2000; Whitelaw, 2000). For example, Bekaert and Wu (2000) claimed that it
appears that volatility in equity markets is asymmetric; returns and conditional volatility
are negatively correlated.
Standard GARCH models assume that positive and negative error terms have a
symmetric effect on the volatility. In other words, good and bad news have the same
effect on the volatility in this model. In practice, this assumption is frequently violated,
in particular by stock returns, in that the volatility increases more ‘after bad news’ than
‘after good news’. This so called Leverage Effect appeared firstly in Black (1976), who
noted that:
“a drop in the value of the firm will cause a negative return on its stock, and will usually
increase the leverage of the stock. [...] That rise in the debt-equity ratio will surely mean
a rise in the volatility of the stock''.
A very simple but plausible explanation for the leverage effect is that negative
returns imply a larger proportion of debt through a reduced market value of the firm,
which leads to a higher volatility. The risk, i.e., the volatility reacts first to larger changes
of the market value, nevertheless it is empirically shown that there is a high volatility
after smaller changes. On the other hand, Black said nothing about the effect of positive
returns on the volatility. Although the positive returns cause smaller increases, they do
cause an increase in the volatility. From an empirical point of view the volatility reacts
asymmetrically to the sign of the shocks, and therefore. a number of parameterized
extensions of the standard GARCH model, i.e., EGARCH, QGARCH, TGARCH have
been suggested recently.
EGARCH model has several advantages over the pure GARCH specification.
First, since the log (σ2t ) is modelled, then even if the parameters are negative, σ2
t will be
positive. There is thus no need to artificially impose non-negativity constraints on the
model parameters. Second, asymmetries are allowed for under the EGARCH
formulation, if the relationship between volatility and returns is negative, γ, will be
negative (Brooks, 2008).
EGARCH Model
The EGARCH model was proposed by Nelson (1991). The main purpose of this
model is to explain the asymmetrical response of the market under the positive and
negative shocks. The specification for the conditional variance is:
−+++=
−
−
−
−− πσσ
γσβωσ 2)()( 21
1
21
121
2
t
t
t
ttt
uauInIn
The left-hand side is the log of the conditional variance. This implies that the
leverage effect is exponential, rather than quadratic, and that forecasts of the conditional
variance are guaranteed to be non-negative. The presence of leverage effect can be tested
by the hypothesis that γ < 0. The impact is asymmetric if γ≠0.
5.2.2 Analysis and Discussion
Fitting EGARCH Model
The most popular member of the GARCH class of models, i.e., EGARCH model
has been used to examine the relationship between stock returns (BSE Sensex and NSE
Nifty) and volatility. EViews 5 software has been used examine the relationship. The
results are reported in Table 5.49 and 5.50.
Table 5.49: Coefficient of EGARCH Model (BSE Sensex)
Coefficient Std. Error z-Statistic Prob.
C 0.060662 0.019763 3.069469 0.0021
Variance Equation
C(2) -0.155475 0.009579 -16.23074 0.0000
C(3) 0.247142 0.013459 18.36192 0.0000
C(4) -0.077313 0.007431 -10.40423 0.0000
C(5) 0.958650 0.004080 234.9542 0.0000
R2 -0.000140 Mean dependent variance 0.040432
Adjusted R2 -0.001150 S.D. dependent variance 1.711505
S.E. of regression 1.712489 Akaike info criterion 3.662615
Sum squared residual 11613.16 Schwarz criterion 3.670541
Log likelihood -7256.134 Durbin-Watson statistic 1.860334
Source: Compiled from the data taken from NSE website for selected period using
Eviews 5.
The following conditional variance equations for BSE Sensex and NSE Nifty get while
running EGARCH process.
For BSE Sensex
−+++=
−
−
−
−− πσσ
γσβωσ 2)()( 21
1
21
121
2
t
t
t
ttt
uauInIn
Where, Constant C(2), i.e. ω = - 0.155
GARCH coefficient C(5), i.e. α = 0.958
ARCH coefficient C(3), i.e. β = 0.247
Leverage coefficient C(4), i.e. γ = -0.077
Table 5.50: Coefficient of EGARCH Model (NSE Nifty)
Coefficient Std. Error z-Statistic Prob.
C 0.061056 0.020487 2.980304 0.0029
Variance Equation
C(2) -0.144746 0.007440 -19.45437 0.0000
C(3) 0.239162 0.009918 24.11423 0.0000
C(4) -0.083336 0.007381 -11.29087 0.0000
C(5) 0.955600 0.003803 251.2848 0.0000
R-squared -0.000143 Mean dependent variance 0.040753
Adjusted R-squared -0.001146 S.D. dependent variance 1.700851
S.E. of regression 1.701826 Akaike info criterion 3.668464
Sum squared residual 11550.09 Schwarz criterion 3.676343
Log likelihood -7319.088 Durbin-Watson statistic 1.874163
Source: Compiled from the data taken from NSE website for selected period using Eviews 5.
For NSE Nifty
−+++=
−
−
−
−− πσσ
γσβωσ 2)()( 21
1
21
121
2
t
t
t
ttt
uauInIn
Where, Constant C(2), i.e. ω = - 0.144
GARCH coefficient C(5), i.e. α = 0.955
ARCH coefficient C(3), i.e. β = 0.239
Leverage coefficient C(4), i.e. γ = -0.083
The presence of leverage effects can be tested as follows:
If γ = 0, then a positive surprise has the same effect on volatility as a negative surprise of
the same magnitude.
If -1< γ <0, a positive surprise increases volatility less than a negative surprise.
If γ < -1, a positive surprise actually reduces volatility while a negative surprise increases
volatility.
The impact is asymmetric, if γ≠0.
5.2.3 Results
It has been found that all the coefficients are significant. In the case of both BSE
Sensex and NSE Nifty, ARCH and GARCH coefficients are positive and leverage term
is negative, i.e. BSE Sensex (-0.077) and NSE Nifty (-0.083) and statistically different
from zero indicating the existence of the leverage effect for the stock market returns
during the period under study. Therefore return is negatively correlated with volatility.
This implies that returns tend to be more volatile in response to bad news and less
volatile in response to good news i.e. evidence of asymmetry in stock price behaviour.
The observations of the present study just document as evidence of the findings
developed by Schwert (1989), French et al. (1987), Christie (1982) and Black (1976).
They also found out that returns are negatively correlated with volatility. The main
reason of negative correlation between stock and volatility is leverage effect, i.e., when
stock price declines, the value of equity relative to corporate debt also reduces. Thus, it
leads to increase the risk of holding stocks.
5.3 FORECASTING STOCK MARKET VOLATILITY USING
GARCH MODEL 5.3.1 Overview
The study of volatility is always a serious concern for analysts and researchers
because high degree of volatility can affect the smooth functioning of any stock market.
It may also adversely affect the economic growth and business environment (Srivastava,
2008). An increase in stock market volatility can be interpreted as an increase in assets
classes i.e. debt. This shift can result an increase in cost of capital to firms and hence
new firms might abide this effect as investors will turn to purchase of stock in superior,
well-known firms. Whereas there is a general agreement on what constitutes stock
market volatility, and to a smaller extent, on how to quantify it, there is far less
conventionality on the reasons of changes in stock market volatility. A number of
researchers investigated the causes of volatility in the arrival of new, unexpected
information that affect expected returns on a stock. Thus, changes in market volatility
would just reproduce changes in the domestic or global economic environment. Others
maintain that volatility is caused largely by changes in trading volume, practices or
trends, which in turn are resolute by factors such as changes in macroeconomic policies,
shifts in investor's risk appetite and growing uncertainty (Srinivasan et al., 2010).
Detection of volatility trends would provide insight for designing investment
strategies and for portfolio management. Investors seeking to avoid risk, for example,
may choose to adjust their portfolios by reducing their commitments to assets whose
volatilities are predicted to increase or by using more sophisticated dynamic
diversification approaches to hedge predicted volatility increases. Accurate forecasts of
stock market volatility may improve the performance of option pricing models. In order
to value an option precisely, it is important to accurately forecast the future standard
deviation of returns over the remaining life of the option. This would be useful for
holders and writers of options on the underlying assets. Moreover, the stock market
volatility forecast is an important input for dynamic portfolio insurance strategies
(Srinivasan and Ibrahim, 2010).
As such, financial market volatility has wider ramifications for the entire
economy. The Black Monday on May 17, 2004 had virtually shaken the entire Indian
economy, and regulators, self-regulators, market players as well as government
functionaries were on tenterhooks. The Ketan Parekh fiasco, Harshad Mehta Scam, etc.
are the cases that concern even a common man in the country. Similarly, the stock
market crash of 1987, mini crash of 1989, Southeast Asian currency turmoil, etc. are the
events that have demonstrated the impact of volatility not only on the domestic market
but also the world market abroad. Therefore, volatility forecasting is an issue that has lot
of practical significance in the present day environment wherein events taking place in
one part of the world may affect other parts within seconds (Srivastava and Jain, 2006).
Hence, there is no gainsaying with the statement that volatility estimation is an
essential part in most finance decisions be it asset allocation, derivative pricing or risk
management. However, the question as to what model should be used to calculate
volatility, there is no unique answer as different volatility models are proposed in
literature and are being used by practitioners and these varying models lead to different
volatility estimates.
Financial market volatility is predictable; a prediction of high volatility is really
just a high variance - a prediction that the potential size of a price move is great. Thus,
even perfect predictability of variance does not mean perfect predictability of the size of
market moves or of their direction. Volatility forecasting is a little like predicting
whether it will rain: You can be correct in predicting the probability of rain, but still have
no rain (Engle, 1993).
The technical term given to volatility clustering behaviour is autoregressive
conditional heteroskedasticity (ARCH). Conditional Heteroscedasticity (ARCH) became
a very popular method in the modelling of stock market volatility. As compared to
traditional time series models, ARCH models allowed the conditional variances to
change during time as functions of precedent errors. First approach was to improve the
univariate ARCH model with a different requirement of the variance function. One
development was introduced by Bollerslev (1986) where the Generalized Autoregressive
Conditional Heteroscedasticity (GARCH) method was presented. Then after, the
integrated GARCH (IGARCH) (Engle et al. 1994) and the exponential GARCH
(EGARCH) (Nelson, 1991) were significant one wherever re-specification of variance
equation was considered. Nevertheless, the extent of empirical research on stock return
volatility in emerging markets like India was not plentiful. While Roy and Karmakar
(1995) focused on the measurement of the average level of sample standard deviation to
investigate whether volatility has gone up; Goyal (1995) used conditional volatility
estimates, as recommended by Schwert (1989), to spot the trend in volatility. He also
analyzed the impact of carry forward system on the intensity of volatility.
ARCH/GARCH models have been worn by Pattanaik and Chatterjee (2000) to model the
volatility in Indian financial market.
5.3.2 Analysis and Discussion
A descriptive investigation of the plot of daily returns on Sensex and Nifty
(Figures 5.1 and 5.2) reveals that returns continuously fluctuated around the mean value
that was close to zero. The return measures were both in positive and negative area.
More fluctuations be tending to cluster together and were separated by periods of relative
calm. This was in agreement with Fama's (1965) observation of "volatility clustering"
(Volatility Clustering means that large changes in time series tend to be followed by
large changes and small changes by small changes.).From the time series graph of the
returns for both the markets, it is analyzed that high volatilities are followed by high
volatilities and low volatilities are followed by low volatilities. That means time series
have important time varying variances. Additionally, it is appropriate to put conditional
variance into the function to clarify the impact of risk on the returns. Hence, GARCH
model is the excellent tool for the study.
Figure 5.1: Volatility Clustering of Daily Returns of BSE Sensex
Source: Based on the data taken from BSE website for selected period.
Figure 5.2: Volatility Clustering of Daily Returns of S&P CNX Nifty
Source: Based on the data taken from NSE website for selected period.
Table 5.51: Descriptive Statistics of Daily Returns
Sensex S&P CNX Nifty
Observation Period April 1996-March 2011 April 1996-March 2011
Number of
Observations 3699 3747
Mean 0.000471 0.000472
Median 0.001128 0.001162
Maximum 0.159900 0.163343
Minimum -0.118092 -0.130539
Std. Dev. 0.017296 0.017168
Skewness -0.103023 -0.184996
Kurtosis 8.102657 9.297716
Jarque-Bera 4018.426 (2-tailed p =0.00) 6211.823(2-tailed p =0.00)
Q(1) 19.720 (2-tailed p =0.00) 12.42 (2-tailed p =0.00)
Q2(1) 122.89 (2-tailed p =0.00) 150.61 (2-tailed p =0.00)
ARCH LM
statistics(at lag = 1) 127.95 158.40
Source: Calculated from the data taken from BSE and NSE website for selected period. Notes:
(a) Skewness is a measure of asymmetry of the distribution of the series around its
mean.
(b) Kurtosis measures the peakedness or flatness of the distribution of the series.
(c) Jarque-Bera is a test statistic for testing whether the series is normally distributed.
(d) Q(K) is the Ljung Box statistic identifying the presence of first order
autocorrelation in the returns. Under the null hypothesis of no autocorrelation, it
is distributed as chi-square (K).
(e) Q2(K) is the Ljung Box statistic identifying the presence of first order
autocorrelation in the squared returns. Under the null hypothesis of no
autocorrelation, it is distributed as chi-square (K).
(f) ARCH LM statistic is the Lagrange Multiplier test statistic for the presence of
ARCH .Under the null hypothesis of no heteroskedasticity, it is distributed as a
chi-square (K). Critical value at 1 per cent level of significance is 6.63 at 1 degree
of freedom. Values for other higher lag are also significant.
Descriptive statistics for both Sensex and Nifty returns are summarized in Table
5.51. For Sensex and Nifty, the skewness statistic for daily returns is found to be
different from zero indicating that the return distribution is not symmetric. Furthermore,
the relatively large excess kurtosis suggests that the underlying data is leptokurtic or
heavily tailed and sharply peaked about the mean when compared with the normal
distribution. The Jarque-Bera statistic calculated to test the null hypothesis of normality
rejects the normality assumption. The results confirm the well-known fact that daily
stock returns are not normally distributed but are leptokurtic and skewed.
Unit Root Tests
Stationarity of the return series were tested by conducting Dickey-Fuller test. The
calculated p-values of ADF for both markets were less than the significance level of 0.05
which leads to an inference that the data of time series under study is stationary. The
results of test confirm that both the return series are stationary. Table 5.52 and 5.53
present the results of these tests.
Table 5.52: ADF Test Statistics for Daily Returns (Sensex)
t-Statistic Probability
Augmented Dickey-Fuller test statistic -56.49521 0.0001
Test critical value 1% level -3.465014
5% level -2.876677
10% level -2.574917
Source: Calculated from the data taken from BSE website for selected period.
Table 5.53: ADF Test Statistics for Daily Returns (Nifty)
t-Statistic Probability
Augmented Dickey-Fuller test statistic -57.75137 0.0001
Test critical value 1% level -3.465014
5% level -2.876677
10% level -2.574917
Source: Calculated from the data taken from NSE website for selected period.
Application of Box-Jenkins Methodology
Volatility clustering implies a strong autocorrelation in squared returns; therefore,
a simple method for detecting volatility clustering is to calculate the first-order
autocorrelation coefficient in squared returns. To test this hypothesis, one can use the
modified Box-Pierce (Ljung-Box- Pierce or simply Ljung-Box ) statistic (Q), developed
by Ljung and Box, which is defined as Q = n (n + 2) Σ r2 k / (n-k), where n = sample size
and k = lag length and Σ r2 k = autocorrelation function (Ljung and Box, 1978). In an
application, if the computed Q exceeds the critical Q value from the Chi-square table at
the chosen level of significance, one can reject the null hypothesis. The values of Q2(1)
test statistic (reported in Table 5.51), reject the null hypothesis which confirmed the
presence of first-order autocorrelation in the series. The existence of a leptokurtic
distribution, presence of volatility clustering suggested an ARCH or GARCH process,
which was confirmed by computing the value of Lagrange Multiplier (LM) which rejects
the null hypothesis. To sum up, the analysis indicates that daily return series of the index
are not normal and exhibit ‘ARCH effect’.
Fitting GARCH (1, 1) Model
The most popular member of the ARCH class of models, i.e. GARCH (p,q)
model has been used to model volatility of Sensex and Nifty returns. EViews 5 software
for model estimation has been used. The results from the model estimated are reported in
Table 5.54 and 5.45.
The following estimated conditional variance equations for Sensex and Nifty are
obtained while running GARCH (1,1) process.
For Sensex
ht = ω +α1 ε t-1 2 + β1h t-1
ht = 6.5E-06+.120 ε t-1 2 + 0.863 h t-1
For Nifty
ht = ω +α1 ε t-1 2 + β1h t-1
ht = 7.55E-06+.130 ε t-1 2 + 0.851 h t-1
Table 5.54: Coefficients of GARCH Model (Sensex)
Coefficient Std. Error z-Statistic Prob.
C 0.001148 0.000208 5.509233 0.0000
RET(-1) 0.087753 0.017837 4.919702 0.0000
Variance Equation
C 6.56E-06 8.29E-07 7.909030 0.0000
RESID(-1)^2 0.120143 0.007744 15.51357 0.0000
GARCH(-1) 0.862982 0.007805 110.5744 0.0000
R2 0.003396 Mean dependent variance 0.000471
Adjusted R2 0.002317 S.D. dependent variance 0.017298
S.E. of regression 0.017278 Akaike info criterion -5.520149
Sum squared residual 1.102212 Schwarz criterion -5.511743
Log likelihood 10209.00 F-statistic 3.145587
Durbin-Watson statistic 2.018649 Probability (F-statistic) 0.013613
Source: Compiled from the data taken from BSE website for selected period using Eviews 5.
Table 5.45: Coefficients of GARCH Model (Nifty) Coefficient Std. Error z-Statistic Prob.
C 0.001092 0.000220 4.962655 0.0000
RET(-1) 0.083970 0.017566 4.780146 0.0000
Variance Equation
C 7.55E-06 7.73E-07 9.767712 0.0000
RESID(-1)^2 0.130346 0.006860 19.00021 0.0000
GARCH(-1) 0.851039 0.006154 138.2827 0.0000
R2 0.001139 Mean dependent variance 0.000470
Adjusted R2 0.000071 S.D. dependent variance 0.017169
S.E. of regression 0.017169 Akaike info criterion -5.524704
Sum squared residual 1.102431 Schwarz criterion -5.516389
Log likelihood 10350.01 F-statistic 1.066612
Durbin-Watson statistic 2.042994 Probability (F-statistic) 0.371309
Source: Compiled from the data taken from NSE website for selected period using Eviews 5.
5.3.3 Results
For time series analysis, it is desirable to have stationary series. Stationarity of
the series can be found by summation of α1+ β1 and the value of summation should be
less than one. As for the stationarity of the variance process, it was observed that α1+ β1
is 0.983 for Sensex. (value of α1 is + 0.120 and that of β1 is +0.863, reported in Table
5.54) and for S&P CNX Nifty it was observed that α1+ β1 is 0.981 (value of α1 is + 0.130
and that of β1 is +0.851, reported in Table 5.55) . Hence, stationarity condition (α1+ β1 <
1) is satisfied in both markets. However, the sum was rather close to one which indicated
a long persistence of shocks in volatility. It implies a ‘long memory’. A large value of
GARCH lag coefficients β1 (+ 0.863 for Sensex and +0.851 for Nifty) indicates that
shocks to conditional variance take a long time to die out, so the volatility is ‘persistent’.
Low value of error coefficient α1 i.e., (+ 0.120 for Sensex and +0.130 for Nifty) suggests
that market surprises induce relatively small revisions in future volatility.
The present study has attempted to devise a volatility forecast model for the BSE
Sensex and S&P CNX Nifty, and concluded the GARCH(1,1) specification fits the
Sensex return and Nifty return time series quite well.
Forecasts of Market Volatility
Once the model has been fitted to market return series, (i.e. Sensex and Nifty, it
can be used to forecast volatility. The model has used to forecasts volatility for one-day-
ahead.
For Sensex ht = ω +α1 ε t-1 2 + β1h t-1
ht = 6.56 E-06+.120 ε t-1 2 + 0.863h t-1
h t+1 = 6.56E-06+.120 (-0.00608)2 + 0.863 (.000163)
=.00008118
For Nifty ht = ω +α1 ε t-1 2 + β1h t-1
ht = 7.55E-06+.130 ε t-1 2 + 0.851 h t-1
h t+1 = 7.55E-06+.130 (-0.00609)2 + 0.851 (.000158)
=.00008663
Where, h t+1 is the one-day-ahead volatility forecast. The values of ε t-1 2 and ht-1 are
calculated by using EViews 5. The value of one-day-ahead volatility is .00008118 for
Sensex and .00008663 for Nifty. Hence, one-day-ahead volatility can be forecast from
constructed model.
REFERENCES Agarwal, Manju; and Barua, Biswajit (1999), “Inflation: A Study with respect to India”,
Finance India, Vol. XIII, No. 2, pp. 381-401.
Amadeo, Kimberly (2012), “What are Interest Rates and how do they work?”,
http://useconomy.about.com/od/glossary/g/Interest_Rate.htm.
Apergis, N.; and Eleftheriou, S. (2002), “Interest Rates, Inflation, and Stock Prices: The
Case of the Athens Stock Exchange”, Journal of Policy Modelling, Vol.24,
pp. 231-236.
Ballie, R.T.; and De Gennaro, R.P. (1990), “Stock Returns and Volatility,” Journal of
Financial and Quantitative Analysis, Vol. 25, No. 2, pp. 203-214.
Barsky, R. (1989), “Why Don’t the Prices of Stocks and Bonds Move Together?”,
American Economic Review, Vol.79, pp.1132-45.
Batra, Amita (2003), “The Dynamics of Foreign Portfolio Inflows and Equity Returns in
India”, ICRIER Working Paper No.109.
Bekaert, G.; and Wu, G. (2000) Asymmetric Volatility and Risk in Equity Markets”,
Review of Financial Studies, Vol. 13, pp. 1- 42.
Black, F. (1976), “Studies of Stock Price Volatility Changes,” Proceedings of the 1976
Meetings of the American Statistical Association, Business and Economics
Statistics Section, Washington, DC, American Statistical Association, pp.
177-181.
Bodla, B.S.; and Garg, Ashish (2007), “Determinants of Foreign Institutional Investment
in India”, Amity Management Analyst, Vol. 2, No. 2, pp. 17-30.
Bohl, Martin; and Brzeszczynski, Janusz (2005), “Do Institutional Investors Destabilize
Prices? Evidence from an Emerging Market”, CERT Discussion Paper
2005/01.
Bollerslev, T.(1986),“Generalized Autoregressive Conditional Heteroskedasticity”,
Journal of Econometrics, Vol. 31, No. 3, pp. 307-327.
Bordo, Michael; Dueker, Michael; and Wheelock, David (2008), “Inflation, Monetary
Policy and Stock Market Conditions”, FRB of St. Louis Working Paper No.
2008-012A, http://research.stlouisfed.org/wp/2008/ 2008-012.pdf.
Binder, John; and Merges, Matthias (2000), "Stock Market Volatility and Economic
Factors", Review of Quantitative Finance and Accounting, http://ssrn.
com/abstract=265272.
Brooks, Chris (2008), Introductory Econometric for Finance, Cambridge University
Press, Second Edition.
Chakradhara, Panda (2008), “Do Interest Rates Matter for Stock Markets?” Economic &
Political Weekly, Vol. 43, No. 17, pp. 107-115.
Christie, A.A. (1982), “The Stochastic Behaviour of Common Stock Variance - Value,
Leverage and Interest Rate Effects,” Journal of Financial Economics,
Vol.10, No. 4, pp. 407-432.
Cohn, Richard; and Lessard, Donald (1981), “The Effect of Inflation on Stock Prices:
International Evidence”, The Journal of Finance, Vol. 36, No. 2, pp. 277-
289.
Cox, J.; and Ross, S. (1976), “The Valuation of Options for Alternative Stochastic
Process”, Journal of Financial Economics, Vol. 3, pp. 145-166.
Dhillon, Sharanjit; and Kaur, Majinder (2007), “Foreign Institutional Investment and
Returns on Indian Stock Market Indices: A Causality Analysis”, The Indian
Journal of Commerce, Vol. 60, No.2, pp. 78-85.
Dornbusch, Rudiger; Park, Y.C.; and Stijn, Claessens (2000), “Contagion: How it
Spreads and How it can be Stopped”, World Bank Research Observer.
Durre, A.; and Giot, P. (2005), “An International Analysis of Earnings, Stock Prices, and
Bond Yields”, European Central Bank Working Paper Series No. 515.
Engle, R.F. (1993), “Statistical Models for Financial Volatility”, Financial Analyst
Journal, Vol. 49, No. 1, pp. 72-80.
Engle, R. F; Bollerslev T.; and Nelson D.B. (1994), "ARCH Models”, in R.F. Engle and
D. McFadden (eds.), Handbook of Econometrics, Vol. 4, pp. 2959-3038.
Fama, E. (1965), "The Behaviour of Stock Prices", Journal of Business, Vol. 38, No.1,
pp. 34-65.
Feldstein, Martin (1980), “Inflation and the Stock Market”, The American Economic
Review, Vol. 70, No. 5, pp. 839-847.
French, K.; Schwert, G.; and Stambaugh, R. (1987), “Expected Stock Returns and
Volatility,” Journal of Financial Economics, Vol.19, No. 1, pp. 3-29.
Flannery, Mark; and James, Christopher (1984), “The Effect of Interest Rate Changes
on the Common Stock Returns of Financial Institutions”, The Journal of
Finance, Vol. 39, No. 4, pp. 1141-1151. French, K.; Schwert, G.; and
Stambaugh, R. (1987). “Expected Stock Returns and Volatility,” Journal of
Financial Economics, Vol.19, No. 1 , pp. 3-29.
Goyal, R. (1995), "Volatility in Stock Market Returns", Reserve Bank of India
Occasional Papers, Vol. 16, No. 3, pp. 175-195.
Grewal, S. S. (1986), Making Money on the Stock Market, Vision Books, New Delhi.
Gupta, Arindam; and Kundu, Debashis (2006), “A Study of the Impact of Union Budgets
on Stock Prices in India”, The ICFAI Journal of Applied Finance, Vol. 12,
No. 10, pp. 65-76.
Hamilton James (1994), “Time Series Analysis”, Princeton University Press.
Henry, Blair (2002), “Is Disinflation Good for the Stock Market?”, The Journal of
Finance, Vol. 57, No. 4, pp. 1617-1648.
Jaffe, Jeffrey; and Mandelker, Gershon (1976), “The "Fisher Effect" for Risky Assets:
An Empirical Investigation”, The Journal of Finance, Vol. 31, No. 2, pp.
447-458.
Knif, Johan; Kolari, James; and Pynnonen, Seppo (2003), “ Inflation News and the
Stock Market: Macro efficiency or Overreaction” EFMA 2003 Helsinki
Meetings Working Paper Series.
Linter, J. (1965), “The Valuation of Risky Assets and the Selection of Risky Investments
in Stock Portfolios and Capital Budgets”, Review of Economics and
Statistics, Vol. 47, pp.13-37.
Ljung, G. M.; and Box, G. P. E (1978). “On a Measure of Lack for Fit in Time Series
Models,” Biometrika, Vol. 66, pp.66-72.
Mazumdar, Tanushree (2004), “FII Inflows to India: Their Effect on Stock Market
Liquidity and Volatility”, The ICFAI Journal of Applied Finance, pp. 5-20.
Malhotra, V.J. (1987), Determinants of Equity Prices in India, Doctoral Thesis,
Himachal University, Shimla.
Merton, R.C. (1973), “An Intertemporal Capital Asset Pricing Model”, Econometrica,
Vol. 41, pp. 867-887.
Merton, R.C. (1980) “On Estimating the Expected Return on the Market: An Exploratory
Investigation”, Journal of Financial Economics, Vol. 8, pp. 323-361.
Mossin, J. (1966), “Equilibrium in a Capital Asset Market”, Econometrica, Vol. 41, pp.
768-783.
Nelson, D.B. (1991), “Conditional Heteroskedasticity in Asset Returns: A New
Approach,” Econometrica, Vol. 59, pp. 347-370.
Panda, Chakradhara (2005), “An Empirical Analysis of the impact of FII’s investment on
Indian Stock Market”, Applied Finance, January, pp. 53-61.
Patra, Deepak (2012), “Do Budgets affect Stock Market?”, http://www.the
sundayindian.com/en/story/do-budgets-affect-stock-market/3/31670/.
Pattanaik, S.; and Chatterjee, B. (2000), "Stock Returns and Volatility in India: An
Empirical Puzzle", Reserve Bank of India Occasional Papers, Vol. 21,
No.1, pp. 376.
Piatroski, D. Joseph (2004), “The Influence of Analysts, Institutional Investors, and
Insiders on the Incorporation of Market, Industry, and Firm – Specific
Information into Stock Prices”, The Accounting Review, Vol. 79, No. 4, pp.
1119-1151.
Ray, Prantik; and Vani, Vina (2007) ,“What Moves Indian Stock Market: A Study on the
Linkage with Real Economy in the Post-Reform Era”,
www.igidr.ac.in/~money/mfc _06/prantik.pdf.
Ray, K.K. (2009), “Foreign Institutional Investment Flow and Indian Stock Market
Returns: A Relationship Study”, Vilakshan ,XIMB Journal of Management,
Vol. VI , Issue No.1 , pp. 39-52.
Rao, K.L.N; Bansal, Gunjan; and Gupta, Neha (2008), “Escalating Inflation”, Portfolio
Organizer, June, pp. 15-21.
Rao, Narayana; and Bhole, L. M. (1990), “Inflation and Equity Returns”, Economic and
Political Weekly, Vol. 25, No. 21, pp. M91-M96.
Roy, M.K.; and Karmakar, M. (1995), "Stock Market Volatility: Roots & Results",
Vikalpa, Vol. 20, No. 3, pp. 37- 48.
Scharler, Johann (2004), “Understanding the Stock Market’s Response to Monetary
Policy Shocks”, Working Paper 93, Oesterreichische National Bank.
Schwert, W.G. (1989), “Why Does Stock Market Volatility Change Over Time?”
Journal of Finance, Vol. 44, No. 5, pp. 1115-1151.
Sharma, Renuka; and Mehta, Kiran (2010), “Impact of Budget Announcement on Stock
Price Behaviour in India”, GNA Journal of Management and Technology,
Vol. V, No.1, pp. 53-59.
Sharpe, W.F. (1964), “Capital Asset Prices: A Theory of Market Equilibrium under
Conditions of Market Risk”, Journal of Finance, Vol. 19, pp. 425-442.
Shiller, Robert (1987), “The Volatility of Stock Market Prices”, Science, New Series,
Vol. 235, No. 4784, pp. 33-37.
Srivastava, Aman (2008), “Volatility of Indian Stock Market: An Empirical Evidence”,
Asia-Pacific Business Review, Vol. 4, No. 4, pp. 53-61.
Srivastava, Sandeep; and Jain, P.K. (2006), “Comparative Analysis of Volatility
Forecasting Models in Indian Stock Market: An Empirical Study”, Fortune
Journal of International Management, Vol. 3, No. 2, pp. 1-48.
Srinivasan, P.; and Ibrahim, P. (2010), “Forecasting Stock Market Volatility of Bse-30
Index Using Garch Models”, Asia Pacific Business Review, Vol. VI, No. 3,
pp. 47-60.
Srinivasan K.; Deo, Malabika; and Murugesan B. (2010), “Testing Alternative Models
for Forecasting Volatility in Stock Futures Market: A Linear Approach”,
Asia-Pacific Business Review, Vol. VI, No. 4, pp. 5-14.
Sujit, K.S. (2010), "Indian Stock Market, FIIs and Mutual Fund Flows: Trends and
Causation”, Pragyaa Journal of Management, Vol. 1, Issue 1, pp. 1-13.
Summers, Lawrence (1981), “Inflation, the Stock Market, and Owner-occupied
Housing”, American Economic Review, Vol. 71, No. 2, pp. 429-434.
Surumpudi, Neeraja (2006), “Foreign Institutional Investors –Back in Action”, Portfolio
Organizer, December, pp. 25-30.
Tessaromatis, Nicholas (1989), “Stock Market Sensitivity to Interest Rates and
Inflation”. http;//papers.ssrn.com/sol3papers.cfm?absrtact_id=392589.
Thomas, Susan; and Shah, Ajay (2002), “Stock Market Response to Union Budget”,
Economic and Political Weekly, February, pp. 455-458.
Tribedy, Gopal (1991), “Inflation and Economic Growth: An Empirical Investigation”,
The Indian Economic Journal, Vol. 39, No. 1, pp. 96-104.
Upadhyay, Saroj (2006), “FIIs in the Stock Market and the Question of Volatility”,
Portfolio Organizer, May, pp. 22-30.
Verma, Ashutosh; and Agarwal, Neeti (2005), “Impact of Budget on Stock Prices: An
Event Study”, PCTE Journal of Business Management, pp. 17-23.
Whitelaw, R. (2000), “Stock Market Risk and Return: An Empirical Equilibrium
Approach”, Review of Financial Studies, Vol. 13, pp. 521-547.
Zahir, M.A.; and Khanna, Y. (1982), “Determinants of Stock Prices in India”, The
Chartered Accountant, Vol. 30, No. 8, pp. 521-523.