LEVEL OF GROWTH AND ACCOUNTING PROFITABILITY IN CORPORATE VALUE CREATION STRATEGY

By Erni Ekawati ABSTRACT This research examines association between level of growth and accounting profitability drawn from corporate value creation strategy. Results demonstrate that although the accounting profitability measures generally rise with sales growth, an optimal point exists beyond which further growth contributes to value destruction and adversely affects profitability.

I. Introduction Investment industry demands that managers have to maximize the companys sales or earning growth over time. This presumption is based on the belief that growth can increase the accounting profitability, and is synonymous with shareholder value creation. Corporate strategies can then be assessed in term of their expected effect on accounting profitability and growth. As a consequence, traditional incentive schemes of compensation is often tied to the managers ability to beat budgeted increase in sales or earnings as measures of growth and in turn increasing accounting profitability ratios, such as return on equity, return on assets, and return on investment. This view contracts with perspectives that accounting profitability ratios alone cannot be used as an indicator of a profitable business. As Hax and Majluf (1984:214) point out that it is economic, and not accounting profitability that determines the capability for wealth creation on the part of the firm. It is perfectly possible that the company is in the black, and yet its market value is way below its book value, which means that from an economic perspective, its resources would be more profitable if deployed in an alternative investment of similar risk. Based on the financial management perspective, the ultimate goal of the company is to maximize shareholders wealth by maximizing the firm value. The indicator of firm value is stock price. The higher the stock price the higher the value of the firm. Value must be created through companys business over time in order to reach the maximum value of the firm above the book value. Value creation strategy provides a conceptual and operational framework for evaluating corporate strategies ( Albert and MacTaggart, 1979 and Fruhan, 1979 ). This strategy establishes firm value using two determinants of value, namely, growth and profitability. Interest in value creation strategy is increasing in practices. In fact the use of bonuses tied to market value of company stock, as opposed to the traditional incentive scheme using accounting profitability only, has made value creation a priority issues in many firms. In addition, academicians have considered value creation issues related to company sales and asset growth in order to maximize the value of the firm (Fruhan, 1984, Higgins and Kerin, 1983). Ramezani, Soenen, and Jung (2002) found that maximizing the growth of the sales or earnings is not identical to maximizing profitability and in turn maximizing the value of the firm. The ability of the

company to determine the optimal growth of sales or earnings could drive the company through maximizing the value of the firm. Value creation could be obtained along the way up to reaching the optimal growth rate. The further growth rate would not contribute to the value creation. This finding was consistent with Fuller and Jansen (2002) warning about the dangers of conforming to market pressures of growth. This research is concerned with three broad question that are of considerable theoretical and practical interests. First, what is the effect of sales growth on measures of corporate profitability? Second, how are the combined sales growth and profitability influenced the companys value creation strategy? Third, is the optimal growth rate to maximize profitability the same as that of to maximize the value of the firm? Understanding the nature of these effects will shed light on whether an optimal growth rate that maximizes profitability exists, and whether the optimal level of the growth rate is different from that of for maximizing firm value. If the level of growth is different within two conditions of profit and value maximization, managers should choose the level of growth that maximizes the value of the firm. Above the optimal level of growth the company will not generate value creation but value destruction. Despite the apparent popularity of value creation strategy, its empirical validity has yet to be demonstrated for Indonesian companies. Without comprehensive empirical validation, managers may justifiably by skeptical of the viability of value creation strategy, as a framework for choosing among strategic alternatives. Empirical evidence related to the relationships between growth, profitability, and value creation of US companies are available (Varaiya, Kerin, and Weeks, 1987; Shin and Stulz, 2000; Ramezani, Soenen, and Jung, 2002; Fuller and Jensen, 2002). All the results are consistent that the level of optimal growth rate exists to maximize the value of the firm. However, evidence from companies in emerging market is hardly found. It is possible that under inefficient market condition, the measure of value derived from the capital asset pricing model may not reflect the fair value of the company. Therefore, should the same phenomena be empirically found from companies in emerging market, the study would indirectly provide the test of indication of emerging market following efficient market behavior. Using Ramezani, Soenen, and Jung (2002) methodology, this study would provide an additional empirical evidence from public companies listed in Jakarta Stock Exchanges, one of emerging market in Asia, and demonstrate whether the similar phenomena applies. II. Theoretical Background and Hypotheses Development A. Growth and Profitability Product life cycle concept is used to explain the relationship between growth and profitability. Products move through four identifiable phases-introduction, growth, maturity, and decline. During the introduction, a firm focuses on product development and market development. The objective is to gain market acceptance. During the growth phase, the focus move to enlarge capacity and increase market share. During maturity phase, as competition becomes more intense, the emphasis shifts into reducing costs through improve capacity utilization ( economies of scale) and more efficient production process. During the decline phase, firms exit the industry as sales declines and profit opportunities diminish. In line with the product life cycle explanation, the sales growth will reach the optimal level on the third phase or maturity in which the growth of sales will maximize profit. Beyond the maturity phase the further growth will diminish profit. However, due to the traditional incentive schemes of compensation that is tied with the growth of sales earnings, managers tend push the growth beyond the optimal point. Managers should have slowed down the growth when reaching the maturity stage. They should have

started to plan some other alternative lines of business before it will be too late to quit from the existing business. Unfortunately, the growth tied incentive scheme inhibits managers to take preventive action to maintain the right level of profitability. Selling and Stickney (1998) found that profitability measured as return on asset (ROA) differed across time and through time as product pass through different stages in their life cycles. Ramezani, Soenen, and Jung (2002) tested the effect of sales growth companys profitability. In their study, the sample of the companies are sorted into four quartiles based on their growth of sales and earnings. The results indicated that the companies profitability increased as the growth of sales increased. However, on fourth quartile, declining on profitability occurred as sales continued to grow. This study shows that there is an optimal growth that should be reached before diminishing profitability occurred. The enforcement of further growth beyond the optimal level resulted in even more diminishing profitability. Accordingly, the following hypothesis is stated: H1: Up to some certain level, increase in sales growth will increase corporate profitability, however, beyond the optimal level increase in sales growth will have a reversed effect.

B. Growth and Value Creation Growth requires additional investment on operating assets and working capital. To ensure that the growth of sales results in value creation for the company, each additional investment project must have a positive net present value (NPV) of expected future net cash flows after initial investments. Project having a positive NPV would contribute to the stock price increase (Mc. Connel and Muscarella,1985). The increase of the stock price will increase stock return. The increase of stock return above the required rate of return will increase stockholders wealth beyond their expectation, termed as abnormal return. Thus, abnormal return can be used as a measured of companys value creation which is the difference between actual return and expected return. Further study by Mitra, Owers and Biswas (1991) showed the relationship between NPV of expected net cash flows and stock price. Tobins q method was used to determine whether the company was a value creator or over investor. The value creator company with the Tobins q ratio greater then I experienced an increase in the stock price as the company invested in the new project. On the contrary, over investor company indicated by Tobins q ratio (less than I) experienced a stock price decline as the company decided an additional investment, while divestiture could increase the stock price. These findings show that the relation between growth and value creation follows a similar direction as the relation between growth and profitability. Accordingly, the following hypothesis is stated: H2: Up to some certain level, increase in sales growth will increase corporate value creation, however, beyond the optimal level increase in sales growth will have a reversed effect.

C. Accounting Profitability and Value Creation

Literatures in financial management claimed that the goal of company to maximize accounting profitability is not synonymous with that of to maximize the value of the firm. This contention is based on some limitations exposed by the measures of accounting profitability such as return on equity (ROE), return on asset (ROA), and return on investment (ROI). The limitations are claimed due to the use of accounting profit which does not fully reflect cash flows, and the ignorance of cost of capital consideration required to create profit. However, based on the previous studied (Dodd and Johns, 1999; Garvey and Millbourn 2000; Chen and Dodd, 2002), profitability ratios had a positive relationshipwith stock prices. Should profitability ratios have a positive relationship with stock prices, the level of growth maximizing the profitability should be the same as that of maximizing the value of the firm. However, the weaknesses on the accounting measures of profitability ratios will direct into a conjecture that the level of growth that maximizes the value of the firm may not be the same as that of maximizing the companys profitability. This notion is based on the consideration of ignoring the cost of capital in the computation of accounting profitability ratios. Varaiya, Kerin, and Weeks (1987) used the term spread to identify the ROE minus cost of equity capital, in which positive spread would contribute to the values creation, while negative spread to value destruction. Thus, increase in ROE alone would not guarantee to the increase in firm value. They empirically demonstrated the combination of expected growth and expected positive spread that are associated with higher firm value. This finding suggested that value creation clearly resulted from pursuing growth strategies that were not only profitable but also creating a positive spread. It is possible that the sales or earning growth still contributes to the increase in profitability but not to the increase in companys value. This condition occurs because additional growth requires higher costs than the benefits created from it. Accordingly, the hypothesis is stated as follows: H3: The level of sales growth that maximizes the value of the firm is lower than that of maximizing the companys profitability.

D. Other Factors Influencing Profitability and Value Creation Previous studies show that there are many factors that influence the companys performance, whether the performance is measured by profitability ratios or value driven indicators such as economic value added and market value added. This study includes variables that have been cross-sectionally identified by the previous researches affecting the companies performance (Philips, 1999; Campbell and Shiller, 1998; Shin and Stultz, 2000; Opler, Pinkowitz, Stulz, and Williamson, 1999). The variables affecting companys profitability are economic condition, firm size, and systematic risk. These variables are included in the analysis as control variables so the effects of growth levels on profitability and value creation can be identified without interfered by other factors that have been identified by the previous researches affecting the variables being studied.

III.Research Method A. Data and Sample

The companies used as sample in this study are public companies listed in Jakarta Stock Exchange from the year of 1998 to the year of 2002. The periods being included in the sample are based on the availability of the capital market data from Accounting Development Center of Gadjah Mada University, and financial data from Indonesian Capital Market Directory. Following the standard practice, companies data from utilities and financial industry are excluded. The purposive sampling procedures are used to determine the companies being included in the sample. The criteria used are as follows: The company has to be listed until the year 2002.

The companies financial statements and the data required to compute all the operational variables must be available. There are 493 companies meeting the required criteria.

B. Data Analysis Procedures For each year data in the sample, the companies are placed into quartiles based on their sales growth. The first quartile with contains the companies with the lowest level of sales growth, the second quartile with the higher level and so forth. The sales growth on year t is measured by the average annual sales growth for the last three years (year t-3 to t). some companies could move from one growth quartile to another from year to year. To ensure the persistence of the result, persistent sample is formed. Persistent sample contains only companies that belong to the same growth quartile for within at least three year in a row. There are 224 companies meeting this criteria. Accounting profitability is measured by the following ratios: Return on equity (ROE) = Net Income/Total Equity Return on asset (ROANI) = Net Income/Total Assets Return on asset (ROAOP) = Operating Profit/Total Assets

ROAOP shows rate of return that come from the companys business operation only. ROANI measures rate of return from core business activities and other activities such as investment in other companies and contribution of financial leverage. These ratios are computed in the end of each year. Following Bacidore, Boquist, Milburn, and Thakor (1997), value creation is measured using the Jensens alpha derived from capital asset pricing model. (Rit Rft) = i + i(Rmt Rft)+eit Rit stock return of company i at period t, Rmt is market return at period t, and Rft is risk free rate at period t. i is an intercept used as a measure of abnormal return or value creation. One year estimation periods of stock daily return data are used to compute the abnormal return in the end of each year 1. When alpha is

positive, shareholders have received compensation above their risk-adjusted opportunity cost of capital. Conversely, when abnormal returns are negative, they have, been inadequately compensated for risk. Below is the list of variables influencing companys profitability. Variables Economic conditions Firm size Book-to-market Earning-price ratio Dividend payments Operational flexibility Proxy Dummy variables for each year observation. Total sales. Book value per share /(Price per share x number of shares). Earnings per share/Price per share Cash dividend per share Fixed assets/Total assets

Expected signs for regression of firm size, dividend payment, and operating flexibility against profitability are positive, while negative signs are expected for book-to-market and earning price ratio. Below is the list of variables influencing the abnormal return or value creation Variables Proxy Economic conditions Dummy variables for each year observation Profitability ROAIN or ROAOP Firm size Total sales Systematic risk Sensitifity of company return to the market return, estimated by one year period daily return of the companys stock return against market return. Expected signs for regression of firm size, profitability and systematic risk against abnormal return are positive. C. Model To investigate the relationship between level of sales growth and companys profitability, the following multiple regression model is employed. Model I : ROA=a+

q=2

bq Dq +

2002

cy Dy + dSIZE + eBM + fEP + gDIV + hFXA +

y=1999

where : ROA : a vector provitabolity measure in which element represent a company in a particular year (panel data). Two profitability measures are used, namely, ROAIN and ROAOP. a: the regression intercept. It measures the conditional mean of the ROA for the first quartile in year 1998. q: a quartile bq : coefficient differentiating the quartiles. It measures difference in ROA across quartiles after other factors are controlled for. The null hypothesis was that companies in the second through the fourth quartiles are no different from companies in the first quartile ; bq=0 for q = 2,3,4.

Dq : cy : Dy : SIZE : BM : DIV : FXA : :

1 for companies in the qth quartile of sales growth and zero otherwise (q=2,3,4) influence of each year on ROA. a dummy variable for the year, Dy= 1 when y= 1999, 2000, 2001, and 2002, and 0 otherwise. firm size as a control variable having impact on ROA book-to-market ratio as a control variable having impact on ROA dividend payments as a control variable having impact on ROA operating flexibility as a control variable having impact on ROA regression error

Regression coefficients of Dq are used to test H1. As moved to the higher quartiles the coefficients of Dq should increase then decrease after reaching an optimal level. To investigate the link between level of growth and value creation, the following multiple regression model is employed. Model II : a= k + b
q=2 4 2002
q

Dq +

Dy + IROA + mSIZE + nRISK +

y=1999

where, a: a vector with elements representing a companys estimated alpha in a particular year (panel data) k: estimate of the conditional mean of alpha; it measures difference in value creation across quartiles, after other factors have been controlled for. n: regression error; it is assumed to be independent identically distributed (i.i.d). regression coefficients of Dq are used to test H2. As moved to the higher quartiles the coefficients of Dq should increase then decrease after reaching an optimal level. H3 is tested by observing the difference of the optimal level obtained from model I and Model II. IV. Results and Discussions A. Descriptive Statistics Table 1 presents a comparison of performance measures, financial attributes, and asset pricing parameters for companies in the sample classified into four quartiles based on sales growth. Insert Table 1 about here Each company in the sample could move from one quartile of sales growth to another from year. Mean and median of sales growth rate rise across quartiles based on sales growth. The standard deviation sharply rises on the fourth quartile indicating high sales growth rates are accompanied by high volatility. Performance measures reported are ROE, ROANI, ROAOP, net profit margin (NPM), and operating profit margin (OPM). As expected, the performance measures using ROE, ROA NI, and NPM increases as the level of sales growth moves from the first quartile to the third one, however it decreases in the fourth quartile. ROAOP increases up to the second quartile and slowly decreases afterward. OPM

increases all the way from the first to the fourth quartiles. Since the sample are driven from the condition of economic crises and recovery periods, there are many companies having negative book value during the simple periods. ROE may not have a real meaning in measuring profitability. Thus, the profitability measures used in the data analysis are ROANI, ROAOP. As hypothesized, the relationship between profitability and sales growth shows highly non linear relationship, on average, the companies in the inner quartiles perform better. Variables of financial attributes provide broad pictures of firm size, operating flexibility, and future growth potential. High correlations are found between total assets and total sales. As shown in the table 1, the firm size increases up to third quartile and decreases in the fourth quartile. It resembles an inverted U-shaped across the sales growth quartiles. Operating flexibility as measured by fixed assets/total assets is, on average, similar across quartiles. Book-to-market and earning-price ratios have, on average, negative values due to many companies having negative book value and earning during the sample periods. This phenomena is attributed to the periods of Indonesian economic recession and recovery. The exclusion of the negative numbers would result in a very small sample size, therefore, earning-price and book-market ratios rather than price-earning and market-to-book ratios are used to maintain the meaning of these ratios. The measure of these ratios follow Fama and French (1995) study in which they included negative book equity value companies in forming a market portfolio. Table 1 shows that book-to-market and earning-price ratios reach the highest value on the third quartile indicating that the highest growth potential occurs in the third quartile. Further growth beyond the third quartile will no longer be optimal. Asset pricing parameters presented in table 1 show measures of value creation attributed by annualized CAPM alpha. As expected, the alpha value increases from the first quartile to the third one and decreases on the fourth quartile. Inverted U-shaped phenomena is also found on the relation between abnormal return and level of sales growth. The systematic risk measured by CAPM beta are similar across companies in each quartile.

B.

Results

To further investigate the relationship between level of growth and profitability, the formal test of regression of ModeI is employed. As shown in table 2 the regression output lends support to the analysis from the descriptive statistics. Insert Table 2 about here The regression of model I is estimated using ordinary least squares (OLS). Dependent variables used are ROANI and ROAOP. The majority of estimated coefficients are highly statistically significant. A number of diagnostic tests have been performed and the standard OLS assumptions be rejected.

As shown in Table 2, intercept coefficient estimates are conditional mean of either ROA NI or ROAOP for companies in the first growth quartile in year 1998 (used as a base value). The coefficients on the other quartile dummies represent the difference of average ROA of the corresponding quartile from the base value. Given this explanation, the coefficient of growth quartile dummies can be interpreted that ROANI incrases from Q1 to Q2, reaches the maximum on Q3, and decreases on Q4. The coefficients of Q2, Q3, and Q4 are 4.68%, 8.22%, and 6.68%, respectively. This result support H1 that profitability (ROANI) increases up to a certain level of growth is on Q3. The result cannot clearly be seen from the regression of ROAOP, because the coefficients of growth quartile dummies are risen as moving through the higher quartiles. The coefficients of Q2, Q3, and Q4 are 5.31%, 7.63%, and 7.95% yet, they still indicate diminishing marginal profitability. Thus, an optimal level of growth exists even though it cannot be shown by the result. The optimal level accurs some where beyond Q4 To test the persistence of the result due to the possibility of some companies moving from one quartile to another through the years, the regression of model I is also run using persistent sample. The results are robust for both measures of profitability. All controlled variables have estimated coefficient signs as expected, except for earning-price ratio and book-to-market ratio. Earning-price (EP) and market-to-book (BM) ratios have a positive relationship with profitability. Companies having lower EP tend to have lower profitability. It is suspected that low (high) EP and MB may not reflect the companies potential growth but they may indicate market over (under) valuation. Table 3 shows the results of the regression of abnormal return coefficient using ROANI and ROAOP as profitability measures. Insert Table 3 about here. Following the previous interpretations, the intercept coefficient estimate are conditional mean of abnormal return for companies in the first growth quartile in year 1998 (use as a base value). The coefficients on the other quartile dummies represent the difference of average abnormal return of the corresponding quartile from the base value. Using ROANI as a profitability measure, the abnormal return reaches the maximum value on Q3. The coefficients of Q2, Q3, and Q4 are 9.51%,23.60%, and 23.10%, respectively. Using ROAOP as a profitability measure, the coefficients of Q2, Q3, and Q3 are 6.83%, 21.00%, and 19.70%, respectively. The decreasing abnormal return associated with the fourth quartile of growth is again a significantfeatures of the results. Companies in the third quartile have the highest conditional abnormal return for both profitability measures. Moreever, the inverted U-shaped relationship between abnormal return and growth is similar to that of profitability and growth. The persistent sample also shows a similar pattern. This results supports H2 that value creation increases up to a certain level of growth, beyond that level the converse is true. As expected profitability ratios and measures of systematic risk have a positive significant relationship with abnormal return. However, firms size have no significant relationship with value creation also varies over times as reflected by the significant coefficients of year dummy variables. Thus, economic conditions do have affects on profitability and value creation.

Based on the results of the regression using ROANI as a profitability measure, it seems that the level of growth for maximizing firm value and profitability are the same, on Q3. However, further examinations on the profitability measured by ROAOP could shed light on the possibility of different level of growth that maximizes profitability and firm value. Of the regression of ROA OP. Profitability still increase through Q4, and will reach a maximum point some where beyond Q4, but the abnormal return reaches a maximum value on Q3 (see: Table 3). The same result is also found the persistent sample. This result demonstrates that maximizing accounting profitability does not necessarily enhance shareholders value. It is true that accounting profitability measures generally rise with the sales growth, yet an optimal level of growth exists, beyond which further growths destroys shareholders value. Therefore, this study also provides a support to the H3 that the level of growth that maximizes firm value is lower than that of maximizing profitability.

V. Conclusions and Limitations The empirical results of this study indicate that maximizing level of growth does not maximize accounting profitability and value creation. On the contrary the companies with moderate growth in sales have the highest profitability and value creation. This empirical findings also demonstrate that the level of growth that maximize the corporate profitability is higher that that of maximizing the value of the firm. Thus, managers should be cautious in planning the level of future growth. The optimal level of growth should maximize the firm value not maximizing the accounting profitability alone. These results suggest that corporate managers need to abandon the habit of blindly increasing company size, and investors need to carefully consider the drawbacks of diseconomies of scales resulted from pushing level of growth beyond the optimal point. The incentive schemes of compensation should not merely rely on growth and accounting profitability measures. Value creation issues related to company sales and asset growth in order to maximize the value of the firm should be considered. These suggest that measures of value creation such as economic value added, market value added should be developed and become important alternative measures in value creation strategy. The limitation of this study is the scheme of partitioning the data into quartiles and applying OLS to identify the non linearity of the relationship between growth and either profitability or value creation. An important question that deserves further investigation is what additional factors can be used to discriminate among the quartiles in an ex ante sense. Related to the OLS, another alternative such as nonlinear and non parametic procedures may be worth to examine in the future research.

REFERENCES

Albert, William and MacTaggart. 1979. The Divestitures Discusion: an Introduction. Merger and Acquisitions, Fall 1979: 18-30.

Armitage, Howard M. Ellen Wong, Alan Douglas. 2003. The Pursuit of Value. CMA Management, Vol. 77, Iss. 7 (Nopember):34 Bacidore, J.J. Boquist, T. Milburn, and A. Thakor. 1997. The Search for the Best Financial Performance Measure. Financial Analysts Journal, Vol. 53, no 3 (Mei/Juni): 11-21. Bannister, James W, Paul H Mihalek, Carl S Smith. 1997. Performance Plan Adoption and Corporate Performance. Managerial Finance, Vol. 23, Iss. 5: 28-39. Campbell, J. Dan R Shiller. 1998. Valuation ratios and the Long-Run Stock Market Outlook. Journal of Portfolio Management, Vol. 24, no.2 (Winter): 11-26. Chen, Shimin dan JamesL Dodd. 2002. Market Efficiency, CAPM, and Value-Relevance of Earning and EVA: A reply to the comment by professor Paulo. Journal of Managerial Issues, Vol. 14, Iss. 4 (Winter) : 507-512 Dillon, R. dan J. Owers. 1997. EVA as a Financial Metric Attributes, Utilization, and Relationship to NPV. Financial Practice and Education, Vol. 7 no. 1 (Spring/Summer): 32-40. Dodd, James L dan Jasen Johns. 1999. EVA reconsidered. Business and Econimic Review. Vol. 45, Iss. 3 (Apr-Jun): 13-18. Fama, Eugene and Kenneth R. French. 1995. The Cross-Section of Expected Stock Return. The Journal of Finance, Vol. 47, no. 2, June : 427-465. Farsio, Farzad, Joee Degel, Julia Degner. 2000. Economic Value Added (EVA) and Stock Returns. Financier, Vol. 7, Iss. 4 : 115-118. Fuller, J. Dan M. Jensen. 2002. Just Say No to Wall Street. Journal of Applied Corporate Finance. Fruhan, William E. Jr. 1979. Financial Strategy: Studies in the Creation, Transfer and Destruction of Shareholder Value. R.D. Irwin, Homewood, IL. Fruhan, William E. Jr. 1984. How Fast Should Your Company Grow? Harvard Business Review, Jan-Feb: 84-83. Garvey, Gerald T dan Todd T Milbourn. 2000. EVA versus Earnings : Does It Matter Which is More Highly Correlated with Stock Return? Journal of Accounting Research, Vol. 38 : 209-245. Hax, Arnaldo and Nicholas Majluf, 1984. Strategic Management : An Integrative Perspective. PrenticeHall, Englewood Cliffs, NJ, 1984 : 209-242. Higgins, Robert C. and Roger A. Kerin 1983. Managing the Growth-Financial Policy Nexus. Journal of Retailing, Fall : 19-48. Jensen, M. 1986. Agency Costs of free Cash flow, Corporate Finance, and Takeovers. American Economic Review, Vol. 76, no. 2 (May) : 323-329.

McConnel, John J. dan Chris J. Muscarella. 1985. Corporate Capital Expenditure Decision and the Market Value of the Firm. Journal of Financial Economics 14 (No. 3 September) : 399-422. Mitra, Devashis, James E. Owers dan Ashoke Biswas. 1991. A. Direct Test of the Free Cash Flow Hypothesis. Financial Management 20 (no. 1 Spring). Muphy, Kevin J. 1985. Corporate Performance and Managerial Remuneration : An Empirical Analysis. Journal of Accounting and Economics, (April) 7 : 11-42. Myers, S.C. 1993. Still Searching for Optimal Capital Structure. Journal of Applied Corporate Finance, Vol. 6 , no. 1 (Spring) : 4-14. Opler, T. Pinkowitz, R. Stulz, dan R. Williamson. 1999. The Determinants and Implications of Corporate Cash Holdings. Journal Of Financial Economics, Vol. 52, no. 1 (April) : 3-46. Paulo, Stanley. 2002. Operating Income, Residual Income, and EVA : which Metric is More Value Relevant A Comment. Journal of Managerial Issues, Vol. 14, Iss. 4 (Winter) : 500-506. Perez-Quiros, G., dan A. Timmerman. 2000. Firm Size and Cyclical Variations in Stock Returns. Journal of Finance, Vol. 55, no. 3 (Spring) : 1229-62. Philips, T. 1999. Why Do Valuation Ratios Forecast Long-Run Equity Return? Journal of Portfolio Management, Vol. 25, no. 3 (Spring) : 38-44. Ramezani C.A, L. Soenen, dan A. Jung. 2002. Growth, Corporate Profitability, and Value Creation. Financial Analysts Journal, Nov/Dec : 56-67. Selling, Thomas I. and Clyde P. Stickney. 1989. The Effects of Business Environment and strategy on a Firms Rate of Return on Assets. Financial Analysts Journal. Jan-Feb : 43-51. Shin, H. dan R. Stulz. 2000. Firm Value, Risk, and Growth Opportunities. Working Paper, Fisher college of Business, Ohio State University. Tilles, G. 1963. The Managers Job a Systems Approach. Harvard Business Review, Vol. 73, no. 1 (janFeb) : 74-82. Varaiya, Nikhil, Roger A. Kerin, and David Weeks. 1987. The Relationship between Growth, Profitability, and Firm Value. Strategic Management Journal, Vol. 8 : 487- 497.

VALUE RELEVANCE OF ACCOUNTING INFORMATION AND THE PRICING OF INDONESIAN INITIAL PUBLIC OFFERINGS
By Drs Tatang Ary Gumanti,M.Bus.Acc.,Phd Abstract The purpose of this study is to examine the relation between accounting measures of total firm risk and the value of firms making initial public offerings (IPOs). The existing explanations of the pricing of IPOs suggest that the value of IPOs is related to a number of fundamental accounting variables of the firms. A sample of 149 Indonesian IPO firms that went public during the period 1989-1997 is examined and is used to test the null hypothesis that the IPOs price is not related to firm specific accounting risk measures. The results provide evidence that the IPOs accounting information is value relevant to the market price of the IPO. Earnings Per Share (EPS), Book Value of Equity (BVE), and financial leverage are significantly related to the level of offering price. The study concludes that the IPOs accounting information is value relevant about the offer and market prices of the IPO, which is in support to that of Kim et al. (1995) and Klein (1996). Keywords: Initial public offerings, IPO pricing, accounting variables, risk 1. Introduction Analytical and empirical evidence of the association between accounting numbers and the value of IPOs are provided in numerous papers, for example, Downes and Heinkel (1982), Hughes (1986), Titman and Trueman (1986), Krinsky and Rotenberg (1989), Kim et al. (1994, 1995), Klein (1996), and Kim and Ritter (1999), amongst others. Kim et al. (1995) and Klein, in particular, show that information in the prospectus is value relevant concerning the IPO. The value of IPOs is also related to the signals taken by the entrepreneurs, such as the level of ownership retention or the quality of underwriters and auditors. Theoretical and empirical evidence has indicated that certain accounting measures can be used as proxies for total firm risk, that is, they could determine the riskiness of a corporation (Lev, 1974; Bowman, 1979; DeAngelo, 1990, among others). The literature also suggests that accounting information is relevant in determining the value and thus the riskiness of a corporation through the use of accounting analysis (Brealy & Myers, 1996; Benninga & Sarig, 1997; White et al. 2003, among others). Some scholars have also advocated the possibility of using accounting information in assessing the value of firm making an IPO (Beaver et al., 1970; Foster, 1986; Lev, 1989; Berstein & Wild, 1998; Noland & Pelvik, 1998). Moreover, Ryan (1997), based on his survey relating accounting numbers and company risk, notes the possibility of incorporating accounting information for measuring the risk of a firm making an IPO in

the absence of ex post risk measures prior to the offering. Loughran and Ritter (2002) suggest that public information is only partially incorporated into the pricing of an IPO. This could be interpreted that the pricing of IPO is not efficient leading to possibility of obtaining abnormal return. However, this assertion is in contrast to Benveniste and Spindt (1989). Lowry and Schwert (2004) also found that the pricing of IPO is efficient with respect to public information conveyed by the underwriter's treatment of public information. These conflicting findings need to be elaborated. A sample of 149 Indonesian IPO firms that went public during the period 1989-1997 is examined and is used to test the null hypothesis that the IPOs price is not related to firm specific accounting risk measures. The findings suggest that the IPOs accounting information is value relevant about the market price of the IPO, which is in support to that of Kim et al. (1995) and Klein (1996). This paper is organised as follows. Section two presents the review of existing literature. This is followed by the methodology used in the study. Section four presents empirical results. Final section concludes and provides direction for future studies. 2. Literature Review 2.1 Related Literature There have been a number of empirical studies suggesting the consistent and significant role of accounting numbers as a determinant of stock returns, known as fundamental analysis (see for example, Lev & Thiagarajan, 1993; Bernard, 1995; Abarbanell & Bushee, 1998, among others). Also, recent theoretical explanation has shown that accounting numbers indeed have value properties in explaining the value of companies, i.e., its stocks (Ohlson, 1995; Feltham & Ohlson, 1995). Moreover, an investor could earn abnormal returns by performing fundamental analysis (See for example, Chan et al., 1991; Lev & Thiagarajan, 1993; Setiono & Strong, 1998; Chung & Kim, 1994; Ballester & Livnat, 1997). Bauman (1996) provides a review of the recent interest in the role of fundamental analysis. The literature suggests that accounting numbers are potentially useful in the process of price determination of IPOs. For example, Foster (1986) asserts there are many contexts in which estimates need to be placed on the value of companies that are not traded on organized markets, for example (b) when determining the price at which a company could go public (p. 422). Berstein and Wild (1998) suggest Reliable estimates of value enable us to make buy/sell/hold decisions regarding securities, determine prices for public offerings of a companys securities (p. 641). Berstein and Wild also suggest the use of financial fundamental ratios in estimating equity values of companies whose stocks are not traded in active markets. Supports of the potential use of accounting information in the price determination of IPOs are provided in many empirical studies. For example, Beaver et al. (1970) conclude accounting risk measures can be applied to decision-settings where market determined risk measures are not available. Two such situations immediately come to mind: privately held firms going public for the first time... (p. 680). Lev (1989) challenges What is the role of financial variables in unusual circumstances when market variables are nonexistent or are of limited usefulness. These include the role of financial variables in the valuation of new public firms (initial public offerings) ... (p. 179, emphasise added). Finally, Nolland and Pavlik (1998) question How are market prices determined for private stock, and, in particular, what is the role of accounting numbers in the valuation estimate (p. 94). Gumanti (2003) provides a review on the importance of accounting information (variables) in explaining the variation of uncertainty surrounding an IPO.

Early studies on the valuation of IPOs generally depart from the signaling hypothesis proposed by Leland and Pyle (1977). Leland and Pyle argue that issuers of IPO use the level of ownership they retained as a signal of firm value. Issuers with higher quality future projects are likely to sell a lower portion of their ownership. Some studies provide strong support for this hypothesis. For example, Downes and Heinkel (1982) test Leland and Pyles model and conclude that firm value is an increasing function of the issuers ownership retention. Clarkson et al. (1991) document a significant association between ownership retention and the value of IPOs. How and Low (1993) find support for such an association in Australian IPOs. Other IPO valuation studies extend their examination beyond the ownership retention with inclusions of other explanatory variables, such us auditor and underwriter prestige and some accounting variables contained in the prospectus. These studies have generally supported the notion that the accounting variables are partly or jointly associated with the value of IPOs (Downes & Heinkel, 1982; Ritter, 1984; Krinsky & Rotenberg, 1989; Clarkson et al., 1991; Keasey & McGuinness, 1992; McGuinness, 1993; How & Low, 1993; Kim et al., 1994). Most of the studies find that the value of the IPOs is significantly related to the level of ownership retention, the level of debt, the quality of underwriters or investment bankers, the proposed use of the proceeds, firm size, excess returns, and accounting earnings. This evidence indicates that some characteristics of IPO firms are value relevant. 2.2 Previous Studies Two studies have been directed to examine the pricing of IPOs, in Korea (Kim et al., 1995) and in USA (Klein, 1996) 1. These papers are probably the only studies that explicitly provide direct tests of the pricing of IPOs using information available in the issue prospectus, with reliance more on accounting information. These two studies use the offer price and perform regressions using various variables obtained from the prospectus. Kim et al. and Klein conclude that the IPO firms financial variables significantly affect the market price of the issue. This finding supports the notion suggested in the literature that accounting information is used as input into the pricing of an IPO. Klein (1996) argues that the limited information available about the issuing firm in an IPO has made information contained in the issue prospectus a potential source of firm-specific information in the estimation of future growth and risk of the firm. Therefore, the prospectus, which is an audited document publicly available to outsiders, represents further assurance of the accountability of information since misleading information could lead to punishment or legal lawsuits. Departing from the Miller and Modigliani framework, Klein develops a testable model expressing equity value as a function of accounting earnings, book value of equity, expected earnings growth variables, and market or firmspecific risk factors. The IPO offer price is used to represent equity value. Unlike seasoned equities where the proxies for risk variables are easily identified, for most IPOs these proxies are unavailable, given the limited data and short operating history. One potential source available is the firm prospectus. Thus, risk proxies must be obtained from the prospectus. Klein (1996) uses the level of ownership retention as the proxy for the firm-specific variable. The identity of the firms auditor and underwriter is used as the proxy for firm-specific growth and risk variables. Three risk proxies are used by Klein, with a dummy variable being given to each of the proxies. The first risk proxy is whether a risk warning is presented in the cover of the prospectus, where a value of one is given for the presence of warning, and zero otherwise. The second risk proxy is the age of the firm, where a value of one is given if the firm presents the operating history for three or more full fiscal years in the prospectus, and zero otherwise. The final risk proxy is whether warrants are issued, where a value of one is given for

a firm offering stock only, and zero if the firm offers a combination of stocks and warrants. The firms pre-offering earnings per share, the pro forma book value of equity per share, and the net proceeds of the issue are used as proxies for accounting variables. Finally, the Nasdaq composite index is used as the proxy for the market factor and industry factor and is included to seek the effect of industry membership. The dependent variable is the issue price. A sample of 193 US IPOs is examined alongside 40 IPOs as the holdout sample. Klein (1996) finds support for the proposition that the prospectus provides value-relevant information about the IPO. More specifically, Klein finds that the firms offering price and aftermarket price are positively related to earnings per share, book value of equity per share, the level ownership retention, underwriter quality, and the issue proceeds. The display of the risk explicit reference on the issue prospectus is negatively related to the price. The model explains 62% of the variation of the issue price and 44% of the market price. The finding of a significant relationship between offer price and accounting variables is interesting given that existing IPO studies have largely ignored such possibilities, with attention placed on the cross-sectional variance of IPO initial returns. Kleins study also provides further support for the literature that accounting variables are value relevant about the IPO. To test the robustness of the model, Klein applies the model to 40 IPO firms and finds that the model is successful in predicting initial and aftermarket prices. Kim et al. (1995) test the external validity of Kleins model using Korean IPOs. A modification is advanced to capture institutional differences between Korean and US markets. They propose two models: (1) the benchmark model, and (2) the augmented model. The benchmark model consists of four variables, namely earnings per share, issue size, industry index, and the type of offer. Three signalling variables are added to the augmented model. The models are tested on a sample of 147 Korean IPOs that went public between June 1985 and March 1990. The tests are run on two groups of IPOs: those that went public before (labelled BEFORE) and after (labelled AFTER) the IPO price liberalisation which took place on 25 June 1988. The models explain about 40% of the variation of the pricing of Korean IPOs. The issue prices are significantly related to earnings per share, the industry index, and issue proceeds. On the other hand, none of the signalling variables is significantly related to the IPO price. Similar to Klein (1996), Kim et al. (1995) also test the predictive ability of the models using holdout samples. The models are tested on sixteen IPOs of the BEFORE group and thirty-six IPOs of the AFTER group. The results show that the models appear to be able to predict the market price of the holdout samples as indicated by a relatively high degree of correlation between the actual and predicted prices. The findings reported in Klein (1996) and Kim et al. (1995) provide further support to the contention that accounting information is impounded in the determination of IPO issue price and is value relevant about the IPO. Thus, the main objective of this additional analysis is to test the external validity of the assertion that accounting information is value relevant to the Indonesian IPO market. 3. Research Methodology 3.1 Sample Selection The sample firms used in this study must satisfy the following basic criteria: 1. The firms must have made an initial public offering during the period 1989-1997. 2. The firms must not be in the real estate, property, and construction industry (Code 61-69), the finance and insurance industry (Code 81-89), or an investment company (Code 98). 3. The prospectus of the IPO firms must be available.

4. Firms must have sufficient trading activities after the offering. The first criterion is applied for two reasons. First, the terminal year of 1989 is imposed because prior to that year the market was not highly regulated and also the market experienced low trading liquidity and high volatility. A number of market deregulations that were introduced in late 1988 have attracted firms to list on the stock exchange with the first issue occurring in July 1989. Limiting the sample of IPO firms to 1997 should minimise the effects of the economic crisis experienced by Indonesia, which started in the second half of 1997. The second criterion is employed to obtain a relatively homogenous sample with respect to the state of financial accounting reports. Firms in the real estate, property, and construction industry, the finance and insurance industry, and firms classified as investment companies have significant differences compared to other industries, which in turn may affect their valuation. Such differences include the structure of capital used in their operation. In addition to this, since the main issue analysed in this study is the role of accounting information in determining the magnitudes of price, differences in capital structure and also components of balance sheet and profit and loss statements may affect the calculation and definition of variables. Table 1 shows the number of firms remaining after each of the selection criteria was applied. As can be seen in Table 1, two hundred and sixty four firms made an IPO during the periods of 1989-1997 (Criterion 1). Of these 264 firms, 86 are those that come from the real estate, property, and construction industries (JSX Code 61-69: 26 firms), the finance and insurance industries (JSX Code 81-89: 54 firms), and investment companies (JSX Code 98: 6 firms). These IPO firms were excluded from the sample (Criterion 2). Four IPO firms were dropped, as their prospectuses were unavailable (Criterion 3). Finally, 25 firms were dropped due to lack of liquidity as shown by the relatively unchanged price in days after the offering. ------- INSERT TABLE 1 HERE------Table 2 shows the industry classifications and the year of IPO for the sample firms. The sample firms examined in this study represent 56.65% of the total number of IPOs made by firms during the period of analysis of 1989-1997. Most of the sample firms come from manufacturing industries (Code 3139, 41-49, and 51-59) which provided 109 firms or 73.15% of the total number of IPO firms examined. These industries involve basic industries and chemicals (48 firms), miscellaneous industries (35 firms), and consumer goods industries (26 firms). ------- INSERT TABLE 2 HERE------3.2 Measurement of Variables Similar to Kim et al. (1995), some modifications are needed to test Kleins (1996) model, as not all the variables examined by Klein are available in the current study. For example, it is uncommon for Indonesian IPO to provide specific warnings about risk on the front cover of the prospectus. Also, none of the IPOs provide an audited financial report of more than three fiscal years in their prospectuses. Most of the prospectuses in 1989 and 1990 consist only of the last two years of audited financial reports. Given this firm-specific difference, I use the natural logarithm of firm age, natural logarithm of gross proceeds, and the firm financial leverage (measured as the ratio of total debt to total assets plus the equity initial market value) as the measures for firm risks. The auditor variable is not examined as more than 90% of

the sample firms are audited by prestigious auditors (Refer to Footnote 11). The other explanatory variables are earnings per share, and price to book value of equity. The model used to test the pricing of IPOs is as follows (predicted signs are in parentheses): LPj = a0 + a1LEPSj + a2BVEj + a3LGPj + a4FLj + a5UWj + a6OWNj +a7LAGEj + 8INDj + uj

(+)

(+)

(+)

(-)

(+)

(+)

(+)

(?)

where P is either offer price or market closing price of week one, EPS is earnings per share, BVE is price to book value of equity, GP is gross proceeds, UW is underwriter quality, FL is financial leverage, OWN is the level of ownership retention, Age is the IPOs operation in years, and IND is industry membership. Definitions of variables used in the regression are presented in Table 3. ------- INSERT TABLE 3 HERE------4. Empirical Results 4.1 General Profile and Descriptive Statistics of the Sample Panel A of Table 4 provides the general profiles of the full 149 sample firms. The mean and median of sales for the most recent year (twelve months) are Rp. 106.66 billion and Rp. 59.01 billion, respectively. The sales range from Rp. 2.82 to Rp. 1,363.85 billion with a standard deviation of Rp. 165.09 billion. The mean of total assets of Rp. 141.44 billion is significantly higher than the mean of sales (p < .10). ------- INSERT TABLE 4 HERE------The mean offering price of the sample firms is Rp. 5,336.4. A deeper examination reveals that the offering price during the early market boom (i.e., 1989 and 1990) is significantly higher (p < .001) than for the subsequent periods (1991 to 1997). The mean and median offering prices are Rp. 8,850 and Rp. 8,550 for the years 1989-1990, compared to Rp. 3,554 and Rp. 3,200 for the years 1991-1997. The mean offering price for all IPOs made during 1989-1997 is Rp. 5,234.1 (Panel B of Table 4), which is insignificantly different from the mean offering price of the sample firms (t= 0.278, p < .390). The sample firms examined in this study offer on average of 23.01 million shares, with a minimum number of shares offered of 0.53 million and a maximum number of 430.77 million. The mean number of shares offered for the sample firms is lower than the mean of all IPO firms during the periods of examination of 38.05 million (Panel B of Table 4). A similar pattern is shown for the amount of proceeds from the issue. The mean gross proceeds for the sample firms is 75.64 million. The figure stands at 91.96 million for all IPOs. The mean and median ages of the firms in this study (i.e., the years in operation) are 16.41 years and 16.00 years, respectively, suggesting that the sample firms have been in existence for quite a long time. The years in operation range from 4 years to 79 years, with a standard deviation of 10.36 years. The descriptive statistics for the continuous variables examined in this study are presented in Table 5. For all of the accounting ratios, the values are based on the firms most recent financial data available in the prospectus. The sample firms have a mean (median) of financial leverage (measured as total debt divided by total assets plus the initial market value of equity) of 40.08% (39.60%). ------- INSERT TABLE 5 HERE-------

The means of offer price and the first week closing price are Rp 5,336.44 and Rp 5,828.21, respectively which is statistically significant. This indicates that the IPO experience an average increase price after the issue. 2 Earnings per share (EPS) of the sample firms ranges from Rp 41.00 to Rp 1,150.00 with a standard deviation of Rp 200.00. Book value of equity per share (BVE) stands from a minimum of Rp 434.78 to a maximum of Rp 14,117.55. The mean and median of the number of shares held by the initial issuers after the issue (i.e., the level of retained ownership) for the sample firms are 76.53% and 76.90%, respectively. This figure is comparable with IPOs in some Asian countries, such as 74.18% for Malaysian IPOs (Paudyal et al., 1999), 68.73% for Singaporean IPOs (Lee et al., 1996), and 69.71% for Korean IPOs (Kim et al., 1995). More than three-quarters of the sample firms examined in this study (78.5% or 117 firms) used prestigious underwriters when they went public. Just over two-third of the sample firms (64.4% or 96 firms) went public during the boom periods (1989, 1990, 1993, and 1994). 4.2 Findings and Discussion Table 6 provides the results of the multivariate analysis, whilst Table 7 provides matrix correlation of the variables examined in this study. For comparison purposes, two regressions for each of the models are provided. Because Whites test rejects homoscedasticity at the 5% level in all four regressions, the significance levels of the parameter estimates are adjusted for Whites (1981) heteroscedasticity consistent covariance matrix. Panel A presents evidence that the pre-IPO accounting information is related to the pricing of IPOs. The IPO issue price is positively and significantly related to earnings per share, price to book value of equity, and gross proceeds of the issue and is negatively related to the level of financial leverage (as risk measure). The signs for all accounting variables are in the expected direction. Thus, IPO firms offering higher issue prices, on average, tend to have higher earnings per share, price to book value of equity, gross proceeds, and lower risk. This finding is qualitatively similar to both Klein (1996) and Kim et al. (1995). The signals taken by the issuers appear to be impounded into the pricing of the IPO. The quality of underwriter is positively and insignificantly related to the price of the issue. This suggests that IPOs underwritten by prestigious underwriters offer relatively higher prices than IPOs underwritten by less prestigious underwriters, and, thus are more valued. However, the lack of a significant association makes such an assertion unwarranted. The evidence also suggests that the higher the fraction of ownership held by the initial issuers, the higher is the price of the IPO (p < .001). The positive and significant relationship between the level of ownership retention and the issue price found in the current study supports the Leland and Pyles (1977) signaling hypothesis. Industry membership does not appear to be an influential variable. The coefficients of IPO age are positive, but insignificant, in both regressions. The market model generates similar conclusions as the offer price model (Panel B of Table 6). The market price model is based on the first weeks closing price. The only difference is that the risk variable is no longer significant. Price differences may have affected the influencing factor of risk. The mean aftermarket price of Rp. 5,852.5 is significantly higher than the offer price of Rp. 5,336.4 (t=1.2866, p < .10), but the variance difference is insignificant (F= 0.8209, p < .1156). Nevertheless, the general conclusion is similar to the offer price model in that accounting information is value relevant in pricing in the new issues market. To test the predictive ability of the models, fitted values of offer prices and market price are estimated for non-sample of IPO firms from the initial sample firms that do not meet the selection criteria

and are those that went public in period 1992-1995. There are thirteen firms with available data that are selected arbitrarily to test the models. Table 8 provides the results of the forecast error both for the offering price and the market models. The mean forecast errors for the offering price and market model (not reported in the Table) are 9.14% and 6.94%, respectively. Both means are insignificantly different from zero. The median forecast errors for the offering price and the market model are 0.36% and 3.95%, respectively. The null hypotheses that the median forecast errors for both models are equal to zero can not be rejected at the 5% level (two-tailed) under the Wilcoxon signed-rank test. Table 9 provides the test statistics for the accuracy of the predicted prices. The correlations between actual and predicted offer prices, and between actual and predicted market prices, are 91.26% and 84.59%, respectively. The high correlation coefficients for the predicted and actual prices for both models provide further support for the accuracy of the models. The predictive ability of the offering price model appears to be stronger than for the market price model. The relatively weaker predictive ability of the market price model is also indicated by the sum of squared forecast errors which is 2.57 times greater than for the offering price. This finding confirms the assertion that the models are successful in predicting both the offering prices and market prices. Two implications germane to this finding are: (1) Information in the issue prospectus has implications for the future growth and risk of the firm, and (2) The estimated models are indifferent to time and industry membership. They could be applied to a variety of industries and over time. The findings reported in the current study support those of Klein (1996) and Kim et al. (1995). Thus, the proposition that accounting information is value relevant in the pricing of IPO is also verified in the Indonesian IPO market. The test of the pricing of IPOs yielded the conclusion that accounting variables available in the issue prospectus are value relevant to the IPO. Thus, the IPO price level is a function of its accounting variables. 5. Conclusions, Limitations, and Possible Direction for Future Studies The analysis on the pricing of IPOs reveals that the model used by Klein (1996) explains more than three-quarters of the variation of the offer price and the aftermarket prices. Thus, the accounting variables are value relevant to the pricing of IPOs. The predictive ability of the market model is weaker than for the offering price model. Overall, the findings provide further support that accounting information is impounded in the pricing of an IPO. It obviously provides confirmation of anecdotal evidence advocated in the literature that accounting numbers are used as input into the pricing of IPOs (Perez, 1984; Bloch, 1986; Sutton & Bendetto, 1988; Buck, 1990). Three limitations were identified in this study. The model used in this study is not developed through extensive derivation relating to accounting risk proxies and the extent of the IPOs pricing. This study focuses its analysis on the role of accounting information in determining the value of an IPO. It has been suggested that accounting information is not the only element in the total information system needed to assess the riskiness of a security. Accounting information does not have a monopoly in supplying information to the market (Dyckman & Morse, 1986). Thus, the selected accounting risk proxies examined in this study may not capture all of the effects of company risks. Specification error may also be present, and, thus, the exact contribution of each of the accounting risk proxies is not warranted. As noted by Anderson et al. (1995) that the success in testing the underpricing equilibrium depends heavily on the success in selecting the proxies, it may be argued that the selected accounting risk proxies examined in this study may not represent all the possible risks relating to the issuing firm. Thus, another

avenue for future study would be to test the model in other IPO settings, or the inclusion of other accounting risk measures. The second limitation of the study is related to the characteristics of the market. An emerging market may exhibit different motives for the investors valuation model and their investment decisions. That is, their justification on buy- sell decisions might not be driven by their extensive valuation methods, but instead, merely by illogical or intuitive reasoning. Yet, this notion might be incorrect, as most of the active market players in the Indonesian stock market are foreign investors who arguably use more knowledge-based analysis than intuition. Finally, as this study examines one of the emerging capital markets in the Southeast Asian region that has experienced a relatively high growth rate, future studies could also examine other capital markets in the region. This will enable us to make a judgment on the impact of the nature and the influence of institutional differences in explaining the underpricing phenomenon. REFERENCES Abarbanell, J.S. & Bushee, B.J. (1998), Abnormal returns to a fundamental analysis strategy, Accounting Review, 73 (1), 19-45. Anderson, S.C., Beard, T.R., & Born, J.A. (1995), Initial Public Offerings: Findings and Theories, Boston, Kluwer Academic Publishers. Ballester, M. & Livnat, J. (1997), The association between security prices and financial information in the Spanish stock market, Journal of International Financial Management and Accounting, 8 (2), 114136. Bauman, M.P. (1996), A review of fundamental analysis research in accounting, Journal of Accounting Literature, 15, 1-33. Beaver, W.H., Kettler, P. & Scholes, M. (1970), The association between market-determined and accounting determined risk measures, Accounting Review, 45(3), 654-682. Benninga, S.Z. & Sarig, O.H. (1997), Corporate Finance: A Valuation Approach, New York, McGrawHill. Benveniste, L., and P. Spendt. (1989). Hoe investment bankers determine the offer price and allocation of new shares, Journal of Financial Economics, 24, 343-362. Bernard, V.L. (1995), The Feltham-Ohlson framework: Implications for empiricists, Contemporary Accounting Research, 11 (2), 733-748. Berstein, L.A. & Wild, J.J. (1998), Financial Statement Analysis: Theory, Application, and Interpretation, 6th ed., Singapore, McGraw Hill. Bloch, E. (1986), Inside Investment Banking, Homewood Illionis, Dow Jones-Irwin Bowman, R.G. (1979), The theoretical relationship between systematic risk and financial (accounting) variables, Journal of Finance, 34 (3), 617-630. Brealy, R.A. & Myers, S.C. (1996), Principles of Corporate Finance, 5 ed., McGraw-Hill. Buck, G.C. (1990), Pricing initial public offerings. In: Khun, R. (ed.), Capital Raising and Financial Structure, Richard D Irwin. Chan, L.K.C., Hamao, Y. & Lakonishok, J. (1991), Fundamentals and stock returns in Japan, Journal of Finance, 46 (5), 1736-1764. Chung, H.Y. & Kim, J.B. (1994), Financial statement analysis and the profitability of risk-based

investment strategy: Evidence from the Korea stock exchange, Advance in International Accounting, 6, 117-132. Clarkson, P.M., Dontoh, A., Richardson, G. & Sefcik, S.E. (1991), Retained ownership and the valuation of initial public offerings: Canadian evidence, Contemporary Accounting Research, 7 (1), 115-131. DeAngelo, L.E. (1990), Equity valuation and corporate control, Accounting Review, 65, (1), 93-112. Downes, D.H. & Heinkel, R. (1982), Signalling and the valuation of unseasoned new issues, Journal of Finance, 37 (1), 1-10. Dyckman, T.R. & Morse, D. (1986), Efficient Capital Markets and Accounting, New Jersey, Prentice-Hall Inc. Feltham. G. & Ohlson, J.A. (1995), Valuation and clean surplus accounting for operating and financial activities, Contemporary Accounting Research, 11 (2), 689-731. Foster, G.R. (1986), Financial Statement Analysis, 2nd ed., New-Jersey, Prentice Hall. Gumanti, T.A. (2003). Can Accounting Information Act as A Proxy for Ex Unte Uncertainty in Initial Public Offerings. Gadjah Mada International Journal of Business. 5 (2): 249-269. Hair Jr, J .F., Anderson, R. E., Tatham, R. L., & Black, W. C. (1995), Multivariate Data Analysis with Readings, 4th ed., New Jersey, Prentice Hall. How, J.C. & Low, J.G. (1993), Fractional ownership and underpricing: Signal of IPO firm value?, Pacific-Basin Financial Journal, 1, 47-65. Hughes, J.S. (1986), Signaling by direct disclosures under asymmetric information, Journal of Accounting and Economics, 5, 119-142. Hughes, J.S. (1989), Discussion of the valuation of initial public offerings, Contemporary Accounting Research, 6 (2), 519-525. Keasey, K. & McGuinness, P. (1992), An empirical investigation of the role of signalling in the valuation of unseasoned equity issues. Accounting and Business Research, 22 (86), 133-142. Kim, J., Krinsky, I. & Lee, J. (1994), The valuation of initial public offerings and accounting disclosures in prospectuses: New evidence from Korea, The International Journal of Accounting, 29, 46-61. Kim, J., Krinsky, I. & Lee, J. (1995), The role of financial variables in the pricing of Korean initial public offerings, Pacific-Basin Finance Journal, 3, 449-464. Kim, M. & Ritter, J.R. (1999), Valuing IPOs, Journal of Financial Economics, 53, 409-437. Klein, A. (1996, Fall), Can investors use the prospectus to price initial public offerings, Journal of Financial Statement Analysis, 2, 23-39. Krinsky, I. & Rotenberg, W. (1989), The valuation of initial public offerings. Contemporary Accounting Research, 5 (2), 501-515. Lee, P.J., Taylor, S.L. & Walter, T.S. (1996), Expected and realised returns for Singaporean initial public offerings: Initial and long run analysis, Pacific-Basin Finance Journal, 4, 153-180. Leland, H.E. & Pyle, D.H. (1977), Information asymmetric, financial structure, and financial intermediations. Journal of Finance, 32 (2), 371-387. Lev, B. & Thiagarajan, S.R. (1993), Fundamental information analysis, Journal of Accounting Research, 190-215. Lev, B. (1974), On the association between operating leverage and risk, Journal of Financial and Quantitative Analysis, 9 (3), 627-641. Lev, B. (1989), On the usefulness of earnings and earnings research: Lessons and directions from two decades of empirical research, Journal of Accounting Research, 153-192. Loughran, T., and Ritter, J.R. (2002), Why dont issuers get upset about leaving money on the table in

IPOs?.Review of Financial Studies, 15: 413-443. Lowry, M., and Schwert, G.B. (2004). Is the IPO pricing process efficient, Journal of Financial Economics, 71 (1): 3-26. McGuinnes, P. (1993), The market valuation of initial public offerings in Hong Kong. Applied Financial Economics, 3, 267-281. Noland, T.R. & Pavlik, R.M. (1998), The Underpricing of initial public offerings: Review, critique, and integration, in C. Lee (Ed.), Advances in Quantitative Analysis of Finance and Accounting, Vol. 6, pp. 73-102. Ohlson, J.A. (1995), Earnings, book values, and dividends in equity valuation, Contemporary Accounting Research, 11 (2), 661-687. Paudyal, K., Saadouni, B. & Briston, R.J. (1998), Privatisation initial public offerings in Malaysia: Initial premium and long run performance, Pacific-Basin Finance Journal, 6: 427-451. Perez, R.C. (1984), Inside Investment Banking, New York, Prager. Ritter, J.R. (1984), Signalling and the valuation of unseasoned new issues: A comment. Journal of Finance, 39 (4), 1231-1237. Ryan, S.G. (1997), A survey of research relating accounting numbers to systematic equity risk, with implication for risk disclosure policy and future research, Accounting Horizons, 11 (2), 82-95. Setiono, B. & Strong, N. (1998), Predicting stock returns using financial statement information, Journal of Business, Finance and Accounting, 25 (5/6), 631-657. Titman, S. & Trueman, B. (1986), Information quality and the valuation of new issues, Journal of Accounting and Economics, 159-172. White, G.I., Sondhi, A.C., & Fried, D. (2003), The Analysis and Use of Financial Statements, 3rd ed., New York, John Wiley & Sons, Inc. White, H. (1980), A heterokedasticity