Key fundamental factors and long-run price changes in an emerging market--a case study of Karachi Stock Exchange (KSE).

Irfan, Chaudhary Mohammad ; Nishat, Mohammed

This paper identifies the joint effect that multiple factors exert on share prices in Karachi Stock Exchange (KSE), using annual balance sheet data of listed firms during 1981 to 2000. Out of six theory-suggested fundamental variables (dividend yield, payout ratio, size, asset growth, leverage, and earning volatility), the highly significant joint factors observed are payout ratio, size, leverage, and dividend yield. Together, these four factors explain one-fourth variation in share prices at KSE. The explanatory power of fundamental factors is found to be different in the pre-reform period (1981-1990) and the post-reform period (1991-2000). In the pre-reform period, three factors, i.e., payout ratio, size, and dividend yield, explain about half of the variation in share prices, whereas during the post-reform period these factors explain about one-third of the variability in these prices. The correct sign of size effect is only observed in the pre-reform period.

1. INTRODUCTION

Share prices are the most important indicator readily available to the investors for their decision to invest or not in a particular share. Theories suggest that share price changes are associated with changes in fundamental variables which are relevant for share valuation like payout ratio, dividend yield, capital structure, earnings size of the firm and its growth, [Wilcox (1984); Rappoport (1986); Downs (1991)]. Linter (1956) linked dividend changes to earnings while Shapiro valuation model (1962) showed dividend streams discounted by the difference in discount rate and growth in dividend should be equal to share price. This predicts direct relation between pay out ratio and the price-earning multiple. Conversely it means that there is an inverse relation between pay out ratio and share price changes. Several event-based studies established direct relation between share price changes and either earnings or dividend changes [Ball and Brown (1968); Baskin (1989)]. Sharpe (1964) and Hamada (1972) suggested direct relation between share price changes and capital structure. Beaver, Kettler and Sholes (1970) showed that firms appear to pay less of their earnings if they have higher earning volatility. This suggests payout ratio as relevant factor for share price changes. Investigations of share price changes appear to yield evidence that changes in fundamental variable(s) should jointly bring about changes in share prices both in developed and emerging markets. However, the actual fundamental factors found to be relevant may vary from market to market. For example, changes in asset growth of firms are significant in the case of Japanese shares while earnings appear to be universally a relevant factor [Ariff, et al. (1994)]. However, it is widely agreed that a set of fundamental variables as suggested by individual theories is no doubt relevant as possible factors affecting share price changes in the short and the long-run [Ariff and Khan (2000)].

The link between fundamental factors and share price changes has been extensively investigated over short horizons but only few studies attempted to model it over lengthy periods of time. Studies over short windows commonly apply cross-sectional tests using event-based research methodology. The studies examining this relation cross-sectionally or inter-temporally are few, and these have one common feature i.e., the fundamental factors used in a specific study are either one or two although there is a long list of such factors. Furthermore, while price revisions at the time of announcements of price relevant disclosures are valid as announcement effects shown over short horizons, it is equally important to test the effect over a lengthier period of time using data over several years as measure of the variables [Ariff and Khan (2000)]. Black and Sholes (1973) support the idea, that the value of security will be higher, the higher is the volatility of security's return. The relation between dividend and earnings follows that greater the volatility of earnings of a firm, the less is the likelihood of dividend yield being changed by the firm's management. Hence earning volatility is directly related to share price volatility.

Another relevant factor in affecting the share prices is the capital structure of the firm. The level of debt financing by the firm has impact on the value of firm's assets. Hamada (1972) and Sharpe (1964) specify their theories regarding the capital structure. A high-risk firm (a firm with debt) must generate high return consistent with the investor's expected return. It follows that with higher debt firm should have greater rate of change in its share price. Hence capital structure (DA as debt to asset ratio) changes must be directly related to the share price volatility. Modigliani and Miller (1958) emphasised that in competitive capital markets the value of a firm is independent of its financial structure. But if markets are imperfect (transaction cost, taxes, informational asymmetry, agency cost etc.) then capital structure matters and influences the share prices.

Size of a firm does have effect on the valuation of the firm's assets. Smaller stocks have higher average returns. Introduction of size, as a multiplicative term to dividend, provides a significant improvement in the explanation of share prices [Karathanassis and Philiappas (1988)]. The size of the firm if captured through total capital employed, is expected to influence the share prices positively as large firms are better diversified than smaller ones and thus are less risky [Benishy (1961)]. Atiase (1985) showed that as the size of the firm increases, their share price volatility declines.

Ariff, et al. (1994) established the joint linear effect of six firm specific variables for the three markets using data relating to samples of firms over 16 or more years in Japan, Malaysia and Singapore. In general, the six variables are significantly related to share price volatility in the three markets, although some were not significant in particular markets. In the case of more analytically intensive Japanese market, changes in the fundamental factors accounted for two-fifth of the variation in share price volatility. The same was not the case in the less analytically intensive developing markets of Malaysia and Singapore. Obviously, larger portions of price variation appear not to be explained by the variation in the six firm-specific fundamental variables in the less developing markets. Perhaps, prices in the latter two markets, it may be suggested, are more responsive to macroeconomic factors, which were not included in the cross-sectional tests. Alternatively, investors in such markets are not pricing the shares on the basis of fundamental factors, perhaps preferring to price on speculative information. The US Investors are known to price the securities much more on the basis of analysis made widely available by the financial analyst community and the mass media. In another study Ariff (2000) on a sample of hundred homogenous industrial firms, four out of these six factors were found significant and explained two-third of share price volatility over a window of twenty years for US market.

Karachi stock exchange is an important emerging capital market of the region, among the developing countries. KSE is termed as high-risk high return market where investors seek high-risk premium [Nishat (1999)]. Few studies have attempted to analyse the long run behaviour of the market [Nishat (1991)] and no work has been done to explore the fundamental variables affecting the share prices. Factors affecting share prices have been identified for the short run only [Nishat (1995); Nishat and Saghir (1991)]. It is also important to study these factors in the Pakistani context after the introduction of reforms, which emphasised more towards openness to foreign investor and competition. Under reforms emphasis has been on information disclosure by companies, documentation, increasing role of brokerage houses and investment companies which provide more feedback to investor. The objective of this study is to investigate the joint effect fundamental factors may have on share prices in the long run. It also attempts to see the impact of these factors in two sub periods to assess the reform impact on the share price changes. This paper is a modest first attempt in Pakistan in this direction using the data of all the firms' (160) that are continuously listed for the last 20 years.

The paper is organised into five sections. The second section describes the econometric model and estimation methods. The data description and variable construction is discussed in section three followed by results and discussion in section four. Section five provides summary and conclusion.

2. ECONOMETRIC MODEL AND ESTIMATION METHODS

Keeping in view these six factors an approximation of the relation between share price volatility (PV) and the identified fundamental variables namely dividend yield (DY), payout ratio (POR), leverage (DA), asset growth (ASg), size of the firm (SZ) and earning volatility (EV) can be given in the following form:

PV=f([DY.sup.-]; [POR.sup.-]; [DA.sup.+]; [ASg.sup.-]; [SZ.sup.-]; [EV.sup.+])

The directions of the predicted relation between share price changes and the factors are indicated by the + or - signs placed as superscripts above the symbols used for the factors. These factors will be identified on an annual basis over a lengthy period of twenty years to estimate a valid relation using all the continuously twenty year listed companies at KSE from 1981 to 2000. The analysis utilised cross-sectional least squares regression. First test involved regressing the dependent variable PV against all the independent variables separately. This provides a crude test of the relationship between common stock volatility and the theory suggested fundamental variables individually. Thus providing the impact of each variable on stock price change if no other factor is considered.

In an attempt to find the collective impact of these six factors on price variability all the factors are regressed against the dependent variable PV. The test model can now be specified using observations per firm 'j', where j = 1, ... 160 during t period where t = 1981 to 2000.

[PV.sup.jt] = [[lambda].sub.ot] + [[lambda].sub.1][(DY).sub.jt] + [[lambda].sub.2][(POR).sub.jt] + [[lambda].sub.3][(DA).sub.jt] + [[lambda].4][(EV).sub.jt] + [[lambda].sub.5][(ASg).sub.jt] + [[lambda].sub.6][(SZ).sub.jt] + [[epsilon].sub.jt]

The expectation was that the DY, POR, ASg and SZ would be negatively related to the PV whilst EV and DA would be positively related to PV. The sign for the ASg and SZ and EV found to be contrary to the theory.

The firms selected are heterogeneous in character ranging from big firms as in energy sector to smaller ones as in textile sector. Because of the presence of heteroscedasticity in the data we are using Generalised Least Squares (GLS). The first set of estimation is from running the regression using all the variables. Adding more variables biases the researcher in favour of acceptance of the test model. Therefore, it is necessary to carry out a selection procedure to choose a parsimonious model. We use the Akaike selection statistical procedure to eliminate errors that could arise from multi collinear independent variables. The results after this selection, which is a step-wise regression procedure, are presented separately.

3. DATA AND VARIABLES CONSTRUCTION

Price change (PV) is represented by a price volatility measure, using extreme value method developed by Parkinson (1980) somewhat similar to the standard deviation, but superior to the traditional measure of standard deviation. Parkinson (1980) showed that the extreme value method developed by him is 2.5 to 5 times superior as a measure of variance in a variable. The year's high and low share prices of a firm are used to calculate the rate of changes in prices over the test period. This measure is appropriate to capture the changes in share prices on an annual basis, and it is expressed as a rate of change measure but it has greater amplitude than in case with the more common standard deviation. Tests showed that this variable has produced reliable results reported in other studies.

The six independent variables are measured on an annual basis, each firm providing a total of twenty annual observations. Dividend yield for the cross sectional regression test using the sample of firms will be the averages of the dividend yields of individual firms, the average being taken over the estimation years per firm.

Dividend Yield

This variable was calculated by summing all the annual dividends paid to common shareholders and then dividing this sum by the market value of the common stock. The average of all available years was utilised.

Earning Volatility

First, the average of the total earnings to total asset ratios for all years was obtained. Then, the average of the squared deviation from the overall average was calculated. A square root transformation was applied to the mean squared deviation to obtain estimates of standard deviations.

Payout Ratio

Total cumulative individual company earnings and dividends were calculated for all years. Payout is the ratio of total dividends to total earnings of the firm. The use of this procedure means that the problem of extreme value in individual years due to low or possibly negative net income is reduced. Furthermore the payout ratio was set to one in cases where total dividends exceeded total cumulative profits.

Size

The variable size was constructed in a form that reflects the order of magnitude in real terms. The variable was constructed by taking the average market of common stock for the period 1981 to 2000. The value (in millions) of real size was averaged over the period.

Leverage

The ratio of long term debt to total assets was calculated. The average of all years was used in the analysis.

Growth in Assets

The yearly growth rate was calculated by taking the ratio of the change in total assets. The average over all years was used in analysis.

The behaviour of the variables over the long run as defined in an earlier section is presented in Table 1 the descriptive statistics of the data in overall, pre-reform and post-reform periods respectively. All the firms that are continuously listed on the Karachi Stock Exchange from 1981 to 2000 has been taken for the purpose. The annual data of these firms is taken from the various issues of "balance sheet analysis" published by State Bank of Pakistan. Price data has been taken from the annual reports and annual publications of Karachi Stock Exchange.

4. RESULTS AND DISCUSSION

Table 2 reports the test results of regression between price change variable (PV) and each of the six independent variables. This generates the kind of results most commonly found in existing literature investigating one fundamental variable at a time. The results of this regression shows that two variables Earning volatility (EV) and Asset growth (ASg) are not significantly explaining the price volatility. The other four variables are significant but not substantial enough to explain a large portion of price variation. The explained variation ranges from as low as 1 percent in case of asset growth to as high as 16 percent for payout ratio. Earning volatility, which is significant factor in most of the markets, is insignificant in case of KSE. Individually the highest impact is of pay out ratio, which is 16 percent while lowest is of size if we consider the significant variables only. Highest impact is of POR followed by leverage, dividend yield and size. Contrary to theory, size has a sign that is opposite to that predicted by theory in this simple regression test. This is due to market imperfections as already reported in an earlier study on KSE [Nishat (1999)]. Sign of earning volatility is also wrong theoretically but the variable is statistically insignificant. Taken together these results are consistent with prior studies for developing markets.

The joint effect on share price variable from the six variables is tested running regression presented in Table 3. The explained variation in the price is not very substantial, about 28 percent. However this is consistent with the findings regarding the developing markets where firm related factors are not priced fully. Rather speculative forces and macro economic environment plays more dominant role [Ariff, et al. (1994)]. Noticeably the largest impact comes from payout ratio (coefficient of 0.26) followed by dividend yield, size and leverage closely. Thus four of the six factors are statistically relevant for the pricing of the shares in the Karachi stock exchange in this join test using a lengthy period data.

To control the multicollinearity and to select only significant variables, the procedure suggested by Mendenhall and Sincich (1989) is used. The results shown in Table 4 are obtained by applying the step-wise regression to remove the effect multiple factors may have on the model's specificity. These results indicate that the joint-effect of only four variables dominate this relationship. However there are a number of important differences. The variables selected are more reliable as the variable selection is carried out using step-wise regression method of entering one variable at a time, then adding variables only if the variable being added marks a substantial difference to the results. Overall, four variables namely payout ratio, size of the firm, dividend yield and leverage are statistically significant in that order of importance as determined by the size of the coefficient. Overall the model fit is significant with the F-ratio of 29.962. The explained variation by these four variables is 25.9 percent. Our results are consistent with the results observed in less developed markets of Singapore and Malaysia where goodness of fit obtained were about 25 percent [Ariff, et al. 1994). Individual factors were found to be significant, consistent with theoretical predictions, even in the institutionally less developed markets. However, compared with the results for developed markets the goodness of fit of the model is far inferior. It is very likely that the stable economic environment in developed markets coupled with their more analytically inclined investing institutions drive prices more than 60 percent on the basis of key fundamental variables. Such may not be the case of investors pricing shares in the more volatile economic

environments in the less analytically intensive investment, dominated by individuals trading on uninformed basis in the less developed capital markets like Pakistan.

The results for pre-reform and post-reform periods are presented in Tables 5 to 8. The significance of three fundamental factors is found to be more prominent, namely pay out ratio, size and dividend yield. In pre-reform period (1981-1990) as presented in Tables 5 and 6 the explaining power of these variables is higher as compared to overall study period. These three variables jointly explain more than two-fifth of the price variation in share prices in the market. Also, the sign of the size variable is different from the sign observed for the overall results. In pre-reform period the negative sign of size shows that larger firms were having lower volatility in their share prices while in post reform period the positive sign indicates that the larger firms are more volatile after the introduction of reforms. The results appear to vindicate the intuition given by common belief that fundamental factors drive share prices to change in the long run, though the power of model to explain is the price variation is not as robust as found in case of developed markets. This may be due to greater instability in the economic environment as well as both insufficient institutional progress in investment culture and the apparent lack of analytically intensive market practices in Pakistan.

5. SUMMARY AND CONCLUSIONS

This study attempts to explain the price changes due to the six theory-suggested fundamental variables (dividend yield, payout ratio, size of the firm, leverage, earning volatility and asset growth) in Karachi Stock Exchange during 1981 to 2000 using annual balance sheet data. This paper identifies the joint-effect multiple factors exert on share prices in the long run. The significant joint factors observed are payout ratio, size, leverage and dividend yields. Together these four factors explain one-fourth variation in share prices at KSE. The explanatory power of fundamental factors is found to be different in pre-reform (1981 to 1990) and post-reform (1991 to 2000) periods. In pre-reform period three factors pay out ratio, size and dividend yield explain more than two-fifth of the variation whereas during post reform period these factors explained about only one-third of the variability in prices. The correct sign of size effect is only observed in the pre-reform period, which shows that market has become more volatile in the post reform period, which made prices of larger firms more volatile. The explanatory power of the variables has reduced in the post reform era. The analysis indicates that variables other than fundamental variables may be more important and relevant to explain the share price variation in Pakistan and need further investigation.

Comments

* The study is a good attempt to measure the extent to which the changes in the six factors (dividend yield, payout ratio, size of the firm, leverage, earning volatility, and asset growth) explain the changes or volatility in the stock prices of the companies.

* The results show that Dividend Yield, Payout Ratio, and Size of the Firms have significant impact on price volatility. Results are not robust though and do not necessarily support the theory.

* While comparing the pre- and post-reform period, the paper finds that impact of fundamental factors is lesser in post-reform period. Post-reform period is taken as after 1990s, whereas it is pertinent to mention that major reforms in the capital market took place after 1999.

* Intuitively, we all know that KSE is a very narrow market dominated by few speculators, who do not really take a lead from financial results of companies and other information in the annual reports. Therefore, even before reading the study, one could have presumed that the study would show that most of the volatility in the stock prices could not be attributed to the fundamental factors. Supporting this perception, this is exactly what the study tries to prove on the whole and there are no counter-intuitive results.

* The authors have used regression analysis to measures the effect of these factors (independent variables) on the share price (dependent variable). The [R.sup.2] or coefficient of determination of the regression models is from 26 percent to 28 percent. This means that about 26 percent to 28 percent of the volatility in the share prices of the companies is explained by the six factors. That is, the six fundamental factors do not explain about 72 percent to 74 percent of the variability in the stock prices. * It would be important to point out certain peculiar characteristics of KSE.

Trading in only few stocks, highest level of volume as compared to capitalisation, thirty stocks dominate the whole market and actual settlement is still not more than 10 percent.

* It is also pertinent to note that market only started looking at fundamentals in "post-reform" period when foreign portfolio managers came to Pakistan.

* To use regression, authors needed a large number of data points for the same company. So they selected 160 companies, which were continuously listed at KSE from 1981 to 2000. Now, we do know that liquidity at KSE is concentrated to a high extent in scrips that were NOT listed during the twenty years. This means that most of the companies in the sample were highly illiquid. Price discovery is the most important function of any market. For a company's stock price to reflect its fundamentals, it should be a large capitalisation company with a significant free float and good fundamentals so that it is followed by a large number of analysts and investors. Moreover, the stock ought to be liquid--Hubco for instance. The sample companies do not fit this description. Therefore, to some people it would be surprising, that in such a sample the model still tends to explain as much as a quarter of the variability in the stock prices.

* News hits stock markets every day. Investors assess the impact of the news on the profitability of the companies and then take buy, sell, hold, and etc., decisions. The stock prices respond to these news items every day as directed by the investors. The information on the six fundamental ratios is coming from the annual reports, which were made available to the investor only once a year. But the volatility in the share price is of the whole year or at least not specific to the days, in which Annual Reports were made available to them. Therefore, there is a mismatch between relevant time period for the share price and the fundamental ratios. The model would be appropriate if the annual reports were the only information made available to the investors.

* Example: In May 2002, when there was a threat of Indo-Pak war, prices collapsed across the board because the whole economy was under threat. Later, when the war threat disappeared, prices rebound, war did not happen, so the profitability of the companies was not affected. It is important to take note of such extraordinary events that have phenomenal impact on price volatility. Therefore, it would be advisable to exclude the May period while compiling results of the study. This would keep the focus of study on relationship between volatility and annual report based fundamental factors.

* That fundamentals like dividend yield determine the stock price is only one way of looking at the stock prices. Another view (technical) is that share prices respond to the demand and supply. For instance, for the last month or so, the current bull-run at KSE is predominantly technical. More and more money and people are turning to KSE bidding up the prices of the stocks, even though no such news is coming every day that would influence the fundamentals of the companies. Study focused on fundamental factors in a highly speculative and narrow market like KSE. It set out to prove quantitatively what we already knew but intuitively.

* Although, there were few capital market reforms carded out in nineties but the actual reforms came in the period of 1999-2002. These reforms addressed the structural issues of governance, transparency, and efficiency. It would be interesting to see impact of fundamental after these comprehensive reforms.

* Psychology of local investors is also critical. There are three types Of investors, institutions, high net-worth speculative individuals and small investors. I am afraid only institutions are the one who would probably go for fundamentals. The remaining investors are either speculators or possess bank depositors' mentality. In vesting in stocks is a risk business and growth should be the objective. As some say, it is 90 percent psychology and 10 percent fundamentals.

* I think a study should be conducted to see to what extent share price volatility is explained by technical or non-Fundamental factors at KSE.

Haroon Sharif

Securities and Exchange Commission of Pakistan, Islamabad.

REFERENCES

Ariff, M., and W. Khan (2000) Key Fundamental Factors and Long Run Stock Price Changes. APFA 2000 Meeting, April. (Working Paper.)

Ariff, M., M. Shamsher, and M. N. Annuar (1994) Finding the Factors Associated with Stock Price Volatility: A Comparative Study of Developed and Developing Share Markets. In D. Ghosh and Khaksari (eds.) Managerial Finance in the Corporate Economy. London: Routledge.

Atiase, R. K. (1985) Pre-disclosure Information, Firm Capitalisation and Security Price Behaviour Around Earnings Announcement. Journal of Accounting Research 23:2, 215-235.

Ball, R., and P. Brown (1968) An Empirical Evaluation of Accounting Income Numbers. Journal of Accounting Research 6:2, 157-178.

Baskin, J. (1989) Dividend Policy and the Volatility of Common Stocks. Journal of Portfolio Management 15:3, 19-25.

Beaver, W., P. Kettler, and M. Scholes (1970) The Association between Market Determined and Accounting Determined Risk Measures. The Accounting Review 45:3, 325-349.

Benishay, H. (1961) Variability in Earnings-price Ratios of Corporate Equities. American Economic Review 51, 81-94.

Black, F., and M. Sholes (1973) The Pricing of Options and Corporate Liabilities. Journal of Political Economy 81: 4, 637-654.

Downs, T. W. (1991) An Alternate Approach to Fundamental Analysis: The Asset Side of the Equation. Journal of Portfolio Management 17: 2, 6-17.

Gordon-Shapiro Model is Often Referred to as Gordon, M. J. (1962) The Investment Financing and Valuation of the Corporation. Homewood, Illinois: Irwin.

Hamada, R. S. (1972) Portfolio Analysis, Market Equilibrium and Corporation Finance. The Journal of Finance 24:1, 13-31.

Karathanassis-Philippas (1988) Estimation of Bank Stock Price Parameters and the Variance Components Model. Applied Economics 20:3, 497-507.

Linter, J. (1956) Distributions of Incomes of Corporations among Dividends, Retained Earnings and Taxes. American Economic Review 46: 1, 97-113.

Mendenhall, W., and T. Sincich (1989) A Second Course in Business Statistics: Regression Analysis. Macmillan, London, New York.

Modigliani, F., and M. Miller (1958) The Cost of Capital, Corporation Finance and the Theory of Investment. American Economic Review 48: 261-297.

Nishat, M. (1999) The Impact of Institutional Development on Stock Prices in Pakistan. Doctoral Dissertation, Auckland Business School, University of Auckland.

Nishat, M. (1995) Determinants of Stock Prices in Pakistan. International Journal of Development Banking 13:2, 37-42.

Nishat, M., and A. Saghir (1991) The Stock Market and Pakistan Economy. Saving and Development, (2) -1991-XV.

Parkinson, Michael (1980) The Extreme Value Method for Estimating the Variance of the Rate of Return. Journal of Business 53:1, 61-65.

Rappoport (1986) The Affordable Dividend Approach to Equity Valuation. Financial Analysis Journal 42: 4, 52-58.

Sharpe, W. (1964) Capital Asset Prices: A theory of Market Equilibrium. The Journal of Finance 19:1, 425-442.

State Bank of Pakistan (Various Issues) Balance Sheet Analysis (1984, 1988, 1993, 1998).

Wilcox, J. W. (1984) The P/B--ROE Valuation Model. Financial Analysis Journal January-February, 58-66.

Chaudhary Mohammad Irfan is working towards the M. Phil. at the Applied Economics Research Centre, University of Karachi. Mohammed Nishat is Professor and Chairman, Department of Economics and Finance, Institute of Business Administration, Karachi.

Table 1 Descriptive Statistics

Variable Min Max Mean Std

(a) Overall Period
 PV 0.123 0.906 0.497 0.146
 DY 0.000 0.162 0.041 0.032
 EV 0.135 0.866 0.031 0.141
 POR -0.533 1.858 0.265 0.304
 SZ 0.718 8666.291 395.174 1107.773
 DA 0.000 3.478 0.152 0.319
 Asg -0.048 12.181 0.193 0.959
(b) Pre-Reform Period
 PV 0.121 0.927 0.453 0.153
 DY 0.000 0.217 0.054 0.043
 EV 0.000 0.401 0.018 0.057
 POR -0.326 3.716 0.326 0.043
 SZ 0.495 2067.124 115.641 252.791
 DA 0.000 1.333 0.132 0.172
 Asg -0.155 7.883 0.134 0.108
(c) Post-Reform Period
 PV 0.013 1.133 0.54 0.22
 DY 0.000 0.165 0.0302 0.0346
 EV 0.000 1.142 0.0422 0.1798
 POR -1.401 1.72 0.204 0.3295
 SZ 0.783 16692.99 674.71 2031.17
 DA 0.000 6.875 0.172 0.58
 Asg -1.74 23.297 0.245 1.8408

Where

PV : Price Volatility.
EV : Earning Volatility.
DY : Dividend Yield.
DA : Leverage.
POR : Payout Ratio.
ASg : Asset Growth.
SZ : Size of the Firm.

Table 2

Estimated Relation between Share Prices and Fundamental Variables:
Simple Regressions

Variables Coefficients t-value R-squared

DY -0.30 -3.94 0.09
EV -0.04 0.54 0.02
POR -0.40 -5.47 0.16
SZ -0.17 2.25 0.03
DA 0.32 4.29 0.10
Asg 0.10 1.30 0.01

[PV.sub.jt] = [[lambda].sub.ot] + [[lambda].sub.1]
[(DY).sub.jt] ... ... ... ... ... (2)

[PV.sub.jt] = [[lambda].sub.ot] + [[lambda].sub.1]
[(EV).sub.jt] ... ... ... ... ... (3)

[PV.sub.jt] = [[lambda].sub.ot] + [[lambda].sub.1]
[(POR).sub.jt] ... ... ... ... ... (4)

[PV.sub.jt] = [[lambda].sub.ot] + [[lambda].sub.1]
[(SZ).sub.jt] ... ... ... ... ... (5)

[PV.sub.jt] = [[lambda].sub.ot] + [[lambda].sub.1]
[(DA).sub.jt] ... ... ... ... ... (6)

[PV.sub.jt] = [[lambda].sub.ot] + [[lambda].sub.1] [(ASg).sub.jt]

Where

PV : Price Volatility.
EV : Earning Volatility.
DY : Dividend Yield.
DA : leverage.
POR : Payout Ratio.
ASg : Asset Growth.
SZ : Size of the Firm.

Table 3

Joint Effect of the Fundamental Variables on Share Price Variable

[PV.sub.jt] = [[lambda].sub.0t] + [[lambda].sub.1][(DY).sub.jt] +
[[lambda].sub.2][(POR).sub.jt] + [[lambda].sub.3][(DA).sub.jt] +
[[lambda].sub.4][(EV).sub.jt] + [[lambda].sub.5][(ASg).sub.jt] +
[[lambda].sub.6][(SZ).sub.jt] + [[epsilon].sub.jt]

 Coefficients t-values

DY -0.231 -2.985
EV -0.044 -0.598
POR -0.266 -3.350
SZ 0.219 2.971
DA 0.192 2.547
ASg 0.073 1.009

R-squared = 0.28.
Adjusted R-squared = 0.25.
F- ratio = 10.07 D.W = 2.02.

Where
PV : Price Volatility .
EV : Earning Volatility.
DY : Dividend Yield.
DA : Average.
POR : Payout Ratio.
Asg : Asset Growth.
SZ : Size of the Firm.

Table 4

Parsimonious Test Results of Price to Fundamental Relation:
Step-wise Regressions

 Variables Variables in Coefficient
Steps Entered Test Model (t-values) R-squared

1 POR -- -0.400 0.155
 (-5.471)
2 SIZE POR -0.410
 (-5.71)
 SZ 0.193
 (2.685) 0.187
3 DY POR -0.332
 (4.449)
 SZ 0.236
 (3.308)
 DY -0.232
 (3.046) 0.22
4 DA POR -0.259
 (-3.331)
 SZ 0.2477
 (3.526)
 DY -0.220
 (-2.947)
 DA 0.203
 (2.726) 0.25

F-ratio for step 4 is 29.96 significant at or better than 0.0001 level.

[PV.sub.jt] = [[lambda].sub.0t] + [[lambda].sub.1][(POR).sub.jt] ... (1)

[PV.sub.jt] = [[lambda].sub.0t] + [[lambda].sub.1][(POR).sub.jt] +
[[lambda].sub.2][(SZ).sub.jt] ... (2)

[PV.sub.jt] = [[lambda].sub.0t] + [[lambda].sub.1][(POR).sub.jt] +
[[lambda].sub.2][(SZ).sub.jt] + [[lambda].sub.3][(DY).sub.jt] ... (3)

[PV.sub.jt] = [[lambda].sub.0t] + [[lambda].sub.1][(POR).sub.jt] +
[[lambda].sub.2][(SZ).sub.jt] + [[lambda].sub.3][(DY).sub.jt] +
[[lambda].sub.4][(DA).sub.jt] ... (4)

Where

PV : Price Volatility.
DY : Dividend Yield.
POR : Payout Ratio.
SZ : Size of the Firm.
DA : Leverage.

Table 5

Joint Effect of the Fundamental Variables on Share Price
Variable in the Pre-Reform Era

[PV.sub.jt] = [[lambda].sub.0t] + [[lambda].sub.1][(DY).sub.jt] +
[[lambda].sub.2][(POR).sub.jt] + [[lambda].sub.3][(DA).sub.jt] +
[[lambda].sub.4][(EV).sub.jt] + [[lambda].sub.5][(ASg).sub.jt] +
[[lambda].sub.6][(SZ).sub.jt] + [[epsilon].sub.jt]

 Coefficients t-values

POR -0.371 -5.994
SZ -0.474 -6.794
DY -0.427 -6.76
Asg -0.001 -0.015
DA -0.009 -0.128
EV 0.075 1.132

R-squared = 0.471 F = 22.797
Adj. R-squared = 0.450 DW = 1.89

Where

PV : Price Volatility .
DY : Dividend Yield.
EV : Earning Volatility.
POR : Payout Rado.
SZ : Size of the arm.
Asg : Asset Growth.
DA : Leverage.

Table 6

Parsimonious Test Results of Price to Fundamental Relation
for the Pre-Reform Period: Step-wise Regressions

 Variables Variables in Coefficients Adj.
Steps Entered Test Model (t-values) R-squared

1 POR POR -0.359
 (-4.839) 0.124
2 SIZE POR -0.397
 (-5.838)
 SIZE 0.392
 (-5.773) 0.272
3 DY POR -371
 (6.304)
 SIZE 0.493
 (-8.173)
 DY -.442
 (-7.341) 0.456

F-slat is 23.418 for the third step at or better than 0.0001.

[PV.sub.jt] = [[lambda].sub.0t] + [[lambda].sub.1]
[(POR).sub.jt] ... (4)

[PV.sub.jt] = [[lambda].sub.0t] + [[lambda].sub.1] [(POR).sub.jt] +
[[lambda].sub.2] [(SZ).sub.jt] ... (5)

[PV.sub.jt] = [[lambda].sub.0t] + [[lambda].sub.1] [(POR).sub.jt] +
[[lambda].sub.2] [(SZ).sub.jt] + [[lambda].sub.3] [(DY).sub.jt] ... (5)

Where

PV : Price Volatility.
POR : Payout Ratio.
SZ : Size of the Firm.
DY : Dividend Yield.

Table 7

Joint Effect of the Fundamental Variables on Share Price Variable for
the Post-Reform Period

[PV.sub.jt] = [[lambda].sub.0t] + [[lambda].sub.1][(POR).sub.jt] +
[[lambda].sub.2][(SZ).sub.jt] + [[lambda].sub.3][(DY).sub.jt] +
[[lambda].sub.4][(AS).sub.jt] + [[lambda].sub.5][(DA).sub.jt] +
[[lambda].sub.6][(EV).sub.jt] + [[epsilon].sub.jt]

Variables Coefficients t- values

POR -0.274 -3.284
SZ 0.375 5.371
DY -0.252 -3.069
Asg 0.054 0.802
DA 0.083 1.222
EV -0.129 2.033

R-squared : 0.403 F-slat : 17.117
Adj. R-squared : 0.38 D.W : 1.94

Where

PV : Price Volatility.
POR : Payout Ratio.
SZ : Size of the Firm.
DY : Dividend Yield.
ASg : Asset Growth.
DA : Leverage.
EV : Earning Volatility.

Table 8

Parsimonious Test Results of Price to Fundamental Relation in
the Post-Reform Period: Step-wise Regressions

 Variables Variables in Coefficients Adj.
Steps Entered Test Model (t-values) R-squared

1 DY -- -0.475 0.221
 (-6.761)
2 SIZE DY -0.441
 (-6.687)
 SIZE 0.325
 (4.925) 0.321
3 POR DY -0.262
 (3.202)
 SIZE 0.392
 (5.896)
 POR -0.289
 (-3.503) 0.367

F-slat is 31.514 for the third step at or better than 0.0001.

[PV.sub.jt] = [[lambda].sub.0t] + [[lambda].sub.1]
[(DY).sub.jt] ... (1)

[PV.sub.jt] = [[lambda].sub.0t] + [[lambda].sub.1] [(DY).sub.jt] +
[[lambda].sub.2] [(SZ).sub.jt] ... (2)

[PV.sub.jt] = [[lambda].sub.0t] + [[lambda].sub.1] [(DY).sub.jt] +
[[lambda].sub.2] [(SZ).sub.jt] + [[lambda].sub.3]
[(POR).sub.jt] ... (3)

Where

PV : Price Volatility.
POR : Payout Ratio.
SZ : Size of the Firm.
DY : Dividend Yield.