Chapter 5: Research Findings, Analysis and Discussions, Conclusion and Recommendations

Transcription

1 Chapter 5: Research Findings, Analysis and Discussions, Conclusion and Recommendations 5.1. Research Findings 5.2. Discussion of Research Findings Hypotheses 1: Firm related IPO characteristics Hypotheses 2: Issue related IPO characteristics Hypotheses 3: Market related IPO characteristics Hypotheses 4: Findings for the Predictive Model Hypotheses 5: Impact of Recent Regulatory changes on IPO underpricing Hypotheses 6: Investors Attitude towards IPO underpricing 5.3. Conclusion and implication of the research 5.4. Recommendation for future research

2 The study explored and investigated the determinants of IPO (Initial Public Offering) underpricing in India, with the empirical part focused on several theories presented by researchers in the past. Through an extensive literature review, the researcher has categorized the varied IPO characteristics studied so far and some more that were enumerated during the study into firm related, issue related and market related variables. These IPO characteristics were then detailed to study their impact on IPO underpricing. There are two types of investors: uninformed investors who were assumed to neither hold any private information nor any analysis and forecasting powers and informed investors who could draw matured decisions based on secret (private) information available and their predictions of the future market conditions. In the current study, different models were studied to find out the prediction powers of each of the comprehensive IPO characteristics on the decisions of informed investors and the impact of the publicly available information before subscription close date on the decisions of uninformed investors. Based on detailed study on the past trends on initial underpricing, the researcher has tried to suggest some investment strategies to the investors. Further, the research study elaborated the impact of the recent regulatory changes in the Indian primary markets. In the current chapter, the research findings have been presented in the first part in the light of the hypotheses derived from the literature. The research findings were then discussed and analyzed in detail in part two of this chapter. In the third part, conclusions and implications have been detailed while the recommendations for future research have been presented in the last section Research Findings: To present the research findings, the null hypotheses have been restated as follows: 1. The firm related variables like age of firm, firm s price earning, industry price earning, number of employees, sector to which issue belongs, book value to market value, and the financial information like cash and bank balance, earnings before interest and tax, debt levels do not predict IPO underpricing. 2. The issue related variables like offer price, market capitalization, share overhang, marketing expenses, listing lag, subscription levels, IPO grading, underwriter s

3 reputation, loan repayment as objective of firm and offer price on upper side of price band do not predict IPO underpricing. 3. The market related variables like hot and cold market, pre subscription market returns, market returns and pre list market returns do not predicts IPO underpricing. 4. The firm related, issue related and market related variables together do not predict IPO underpricing in India. 5. The recent regulatory changes do not have any impact on IPO underpricing. 6. The retail investors are not able to judge the degree of IPO underpricing with the available information. The research findings in response to the above hypotheses have been stated as follows: 1. The firm related variable Book to market ratio (BVMV) possesses a significant negative relationship with IPO underpricing. Firm related IPO characteristics BVMV predicts 11.3% of IPO underpricing (refer table 4.30). All other variables studied namely age of firm (AGE), firm s price earning (FPE), industry price earning (IPE), number of employees (EMP), sector to which issue belongs (DVSEC), book value to market value (BVMV), and the financial information like cash and bank balance (CASH), earnings before interest and tax (EBIT), debt to total assets (DEBT) are not found to have any significant impact on IPO underpricing in Indian primary market. The direction of relationship is similar to that predicted according to literature (refer table 5.1). 2. The issue related variables like underwriters reputation (DVUW), marketing expenses (MEXP), listing on upper side of offer price (DVOP), listing lag (LL), proportion of equity sold (PROPSOL), subscription rate in non institutional category (HNI), qualified institutional category (QIB), retail category (RETAIL) and market capitalization (LNMCAP) are found to be significant determinants of IPO underpricing. All these issue related IPO characteristics predict 43.1% (refer R 2 in Table 4.37) of IPO underpricing in Indian primary market. The remaining IPO characteristics namely offer price (LNOP), share overhang (SHANG), overall oversubscription (OSUBS), IPO grading (DVRATE), loan repayment as an objective of issue (DVLOAN) fail to possess any significant relationship with IPO underpricing. The direction of relationship for significant variables are similar for all the variables except LNMCAP (refer table 5.1).

4 3. The market related IPO characteristics first day return on stock (FRET), pre list market return for 9 days (PLIST 9) and pre subscription market return for 30 days (PSUBS30) significantly predict 29.4% (refer R 2 in table 4.41) of IPO underpricing. Other market related IPO characteristics namely market return (MR) and hot and cold market conditions (DVHOT) are not significantly related with IPO underpricing in Indian primary markets. The direction of relationship for all the significant variables is similar to the predicted relationship (refer table 5.1). 4. The firm related, issue related and market related IPO characteristics have been studied in a comprehensive manner to actually determine the prediction powers of all the variable together. It is important to study the IPO characteristics in a comprehensive manner as the categorized variables taken together could create the problem of autocorrelation and multi collinearity and thus, the predicting powers could be adversely affected. Refer table 4.42, model 4 has been studied to include only the IPO characteristics that were significant in independent tests and the entire variables were found to be significantly related to IPO underpricing except for PLIST9 and PSUBS30. The combined prediction powers are 59.4% (see R2, Model 4, Table 4.42) with exactly similar direction of relationship as predicted in independent tests. The comprehensive IPO characteristics have also been studied to test the reliability of the results in Model 5 (refer table 4.42). The researcher found that there is a very limited increase in the prediction powers of the model (refer R 2, Model 5, Table 4.42). Also, the variables significant above were only found to be significant, with little additional prediction powers displayed by number of employees (LNEMP). This tested the reliability of the results as both Model 4 and Model 5 were corroborating the results of Model 1, Model 2 and Model 3. Thus, the final determinants of IPO underpricing in Indian primary markets assuming the availability of entire information are Book to market ratio (BVMV), underwriters reputation (DVUW), marketing expenses (MEXP), listing on upper side of offer price (DVOP), listing lag (LL), proportion of equity sold (PROPSOL), subscription rate in non institutional category (HNI), qualified institutional category (QIB), retail category (RETAIL) and market capitalization (LNMCAP) and first day return on stock (FRET).

5 The actual direction of relation of significant variables to underpricing is similar to the predicted relationship except for market capitalization (LNMCAP). 5. The two recent regulatory changes have been studied. One was the reduction in the listing days from 22 working days to 12 working days and second was the necessitated IPO grading by the issuer firm. The trends in changes in listing lag and the underpricing presented week though significant positive relationship. This proved that with the decrease in the listing lag, the underpricing also decreases. Thus, the impact of this regulatory change was in same direction as was hypothesized by regulators and the move helped in reducing the uncertainties in terms of reduced market volatility. The second regulatory change - IPO grading (up to grade 4) depicted strong negative relationship with underpricing. However, issuers with grading of 5 explicated positive relationship with underpricing. Thus, the results proved that the better grading eliminates uncertainty and thus yield less underpricing. However, best or superior firms with IPO grading 5 underprice their issue to leave a good taste in the mouth of investors. 6. To investigate the investors attitude towards underpricing, a crucial factor of oversubscription has been considered. The investors invest not only the amount of stocks allocated to them; rather the complete amount to the extent the issue has been oversubscribed. Thus, the actual realized returns are less than the underpricing. The realized returns have been defined as the allocation weighted initial returns (AWIR). Table 4.43 depicts that the AWIR are negative in the hot markets of 2006 and 2007 which indicate that the hot markets have been crowded with both good (underpriced) and bad (overpriced) issues and the average realized return for the investors is negative or negligible in hot markets. The researcher further studies the determinants of AWIR and investigates only information that is publicly available and known to an uninformed investor. The results found that the investors must study the fundamentals like firms earnings (EPS, NP, EBIT) and size of the firm in terms of number of employees (LNEMP) and the flow of the market (HOT).

6 Table 5.1: Table representing the actual direction of relationship vis-à-vis the predicted Abbreviations Variables Definition Predicted Actual Firm Related Natural logarithm of the number of days between subscription open date and LNAGE Age of Firm incorporation date +/- BVMV Book Value to market value Ratio of Post issue total net asset value per share divided by first day close price - - FPE Firm's Price Earning Firm's offer price divided by earning per share + IPE Industry Price Earning Average market value of competitors set divided by average earnings per share + LNEMP Number of Employees Natural logarithm of the total number of Employees +/- CASH Cash to Total Assets Ratio of Cash and bank balance to Total Assets +/- EBIT EBIT to Total Assets Ratio of EBIT to Total Assets +/- DEBT DEBT to Total Assets Ratio of long term debt to Total Assets + LNTA Size of the firm Natural logarithm of Total Assets + DVSEC Sector Dummy variable from 1 to 12 according to sector to which issue belongs +/- Issue Related OP Offer Price Natural logarithm of the offer price of issuing firm -/+ LNMCAP Market Capitalization Natural logarithm of the post issue number of shares multiplied by list day - +

7 close price SHANG LNMEXP PROPSOL LL OSUBS QIB HNI RETAIL DVOP Share Overhang Marketing Expense Proportion of equity sold Listing Lag Overall Subscription levels Times subscribed by Qualified Institutional Bidders Times subscribed by Non Institutional Investors Times subscribed by retail investors Upper side of price band Proportion of equity retained by promoters to post issue equity -/+ Proportion of Marketing Expenses to offer size - - Proportion of equity issue in current issue + + Difference between the number of days between subscription close date and list date + + Number of shares demanded divided by the shares offered + Number of shares demanded by QIBs divided by the shares offered to QIBs + + Number of shares demanded divided by the shares offered + + Number of shares demanded by Retail investors divided by the shares offered + + Dummy variable 1 if offer price is greater than or equal to average of price band + +

8 Dummy variable 0 for no grading DVRATE IPO grading assigned and from 1 to 5 as poor to good + Dummy variable 1 if one of the Underwriters' underwriters is amongst top underwriters DVUW Reputation list of Bloomberg - - Loan Dummy variable of 1 is assigned if loan repayment as repayment is not amongst the objectives DVLOAN objective of issue. +/- Market Related Dummy variable 1 if IPO is the month above the median value of the 3 months Hot market centered moving average of quarterly HOT issue number of shares, else 0 + Pre subscription Market Nifty return from -1 to -30 days prior to PSUBS30 Return subscription open date + + Pre list Market Nifty return from -1 to -9 days prior to PLIST9 Return listing date + + First day Return on List day return on stock from open to FRET Stock close + + Market Return on market index from list day MR Return close to subscription close Discussion of Research Findings: This section analyses and discusses the research findings of each of the hypothesis and the results thereof. For this purpose, each of the IPO characteristics categorized as firm related, issue related and market related IPO characteristics have been discussed for its resultant impact on IPO

9 underpricing. The section ends with the discussion of the detailed underpricing predictor model derived from the current study for Indian primary markets. The degree of underpricing has reduced over the years in Indian primary markets. The initial return as reported by Shah (1995) has been 105.6% for the issues between 1991 and The underpricing of 22.62% for 55 new issues on NSE has been recorded from March 2004 to October 2006 (Pande and Vaidyanathan, 2009). The current study shows 15.90% average underpricing for 250 issues from January 2004 September The year by year underpricing in table 4.1 clearly shows the tremendous decrease in underpricing from 2004 to The decrease in underpricing is good for the issuer firms as underpricing is an indirect cost to the issuer firms. The following paragraphs extend the answers for such a decrease in IPO underpricing in Indian primary market context. The discussion revolves around two contradictory theories of underpricing. One is ex ante uncertainty that suggests that big, stable and profitable firms do not underprice their issues as little risk is involved in subscribing their issues (Baron, 1982) and second is signaling theory that suggests that big, stable and profitable firms underprice their issues to signal their firms quality and retain investor s trust for times of their subsequent offerings (Leland and Pyle, 1977) Hypothesis 1: Firm Related determinants of IPO Underpricing: 1. Age of the firm and IPO underpricing: According to the ex ante uncertainty theory, age of the firm is negatively related to underpricing (Carter et al., 1998; Ritter, 1984; Ritter, 1991; Megginson and Weiss, 1991; Chambers and Dimson, 2009; Falconieri, 2009; Zouari et al., 2011). Age of the firm has been assumed as an indicator of stability of the firm and stable and aged firms need not to underprice their issues. Contrary to the results from empirical studies, following facts have emerged from the research findings: Though insignificant, there has been weak non-monotonic positive relationship as depicted by the deciles in table 4.27 and consecutive chart The deciles represent that with the increase in the age of the firm, the underpricing graph is depicting the non monotonic upward trend.

10 The observation also illustrates that the average underpricing has always been greater than 10% when the age of the firm exceeds 7 years. 7 years is a reasonable time period to investigate the financial performance of the firms and provide a satisfactory platform to make sound investment decisions. The results support the argument that the older firms tend to underprice more to leave a good taste in the investor s mouth so that the investors remain motivated to invest at times of further public offerings (FPOs) as well. If the investors do not earn good returns at times of IPO, they would be reluctant to apply for FPOs. 2. Book Value to Market Value (BVMV) and IPO underpricing: Chambers and Dimson (2009) investigated book value to market value ratio (BVMV) and found negative relation between BVMV and IPO underpricing. The investors perceive little risk for high BVMV firms and thus little underpricing. Thus, the empirical results of Chambers and Dimson support the ex ante uncertainty theory rather than signaling theory. Following facts have been enumerated from the current study: As hypothesized, there is a significant negative relationship between BVMV and UP. Greater the BVMV, lower is the underpricing. Higher BVMV eliminates risk for the investors as their investment was rightly justified by the high book value of the firm. The scenario is clearly depicted in the chart 4.24 and table 4.27 and table Table 4.27 signifies underpricing of around 8% or more if BVMV is less than Thus, the firms with less than 0.50 BVMV underprice their issues to reward their investors. 3. Firm s Price Earning (FPE) and IPO underpricing: The price earning of the firm has been considered as the base for pricing of a new issue. The firms price earnings are basically compared with the industry s price earnings and if FPE is less than IPE, such firms were assumed to do better in future. Also, higher PE firms were assumed to be risky, thus greater underpricing. Following have been the research findings:

11 The deciles chart 4.24 and the deciles table 4.27 fails to depict any constant relationship between the movement of the underpricing and the firms PE ratio. The results of current study support the findings of Engelen and Essen (2010). Contrary to above findings, table 4.29 and table 4.30 depicts insignificant negative relation between the FPE and IPO underpricing. 4. Industry Price Earnings (IPE) and IPO Underpricing: The current study investigated the impact of Industry price earnings (IPE) on IPO underpricing. The IPE is disclosed by the issuer firms in their prospectuses in the section Basis for issue price and the company s price earnings could be compared to the industry price earnings to comment on the future prospects of the company. Following have been the results of the investigation: There is no significant relationship between the industry price earnings ratio and the company s underpricing (Refer Table 4.29 and Table 4.30). 5. Cash to Total Assets (CASH) and IPO underpricing: CASH is an important indicator of firm s liquidity position as both excess cash and deficient cash positions are not good for firms. CASH position depends upon the life cycle of business and the expansion opportunities available to business. For example, if the business is expanding or in early stages of its life cycle, the cash positions tend to be negative as well. Thus, a detailed analysis is required to comment upon the businesses cash position and its impact on underpricing. The results of the current investigation have been described as follows: The deciles chart 4.24 and the deciles table 4.27 represents that there is weak positive relationship between cash to total assets ratio (CASH) and IPO underpricing. Table 4.30 clarifies that there is no significant relationship between CASH and underpricing. Thus, other factors need to be considered to determine IPO underpricing. 6. Earnings before Interest and Tax to Total Assets (EBIT) and IPO underpricing: Literature on the impact of earnings on IPO underpricing suggests contradictory views. The ex ante uncertainty followers suggest that there is a negative relation between earnings and

12 underpricing (Welch, 1989; Allen and Faulhaber, 1989). However, Welch and Ritter (2002) supported signaling theory that explains that good firms underprice their shares to signal their firm s quality which differentiates them from the lower quality firms. The results from our investigation have been explained as follows: According to deciles table 4.27 and corresponding chart 4.24, IPO underpricing fails to depict any relationship with earnings. The detailed deciles table fails to comment specifically about the extent of underpricing vis-à-vis EBIT. Table 4.29 and table 4.30 also corroborate the results discussed above and present no significant relationship between the two. The IPO characteristic EBIT has been concluded as significantly negatively related with underpricing when determining the allocation weighted initial returns (refer table 4.45). 7. Debt to Total Assets (DEBT) and IPO underpricing: Ex ante uncertainty theory suggests that there is a positive relation between debt levels and IPO underpricing as huge debt is an indicator of uncertainty. But debt levels have also been dependent on the type of industry, business and market conditions. Some investors also perceive that raising equity to repay the debt re-structures the capital of the firm and the firms thus perform better in future. The inferences from current study have been as follows: Contrary to ex ante uncertainty theory, table 4.27 and corresponding chart 4.24 depicts weak negative trend in underpricing with the increase in debt to total assets (DEBT). With no debt, the underpricing goes up to 38.82%. Table 4.29 and table 4.30 fails to depict any significant relationship between DEBT and underpricing. 8. Total Assets (TA) and IPO underpricing: The assets of the firm have been considered as the proxy for the size of the firm and greater the size, lower uncertainty and thus little underpricing according to ex ante uncertainty theory (Megginson and Weiss, 1991, Ibbotson et al., 1994; Carter et al., 1998). The findings from the current study have been explained as follows:

13 The deciles table 4.27 and regression table 4.30 fails to depict any particular trend between the assets of the issuing firm and its underpricing. 9. Number of Employees (EMP) and IPO underpricing: The number of employees in the firm has been an indicator of the size of the firm. Larger number of employees signify huge firm which is presumed to be less prone to risk and thus little underpricing according to ex ante uncertainty theory and reverse for the signaling theory. The inferences from current study have been explained as follows: The results fail to indicate any relationship between the number of employees and IPO underpricing. The study of the number of employees across different categories in organization study might yield significant results. Thus, the researcher recommends this as the scope of further study. 10. Sector wise classification and IPO underpricing: In Indian scenario, all the major sectors yield underpricing of more than 8% (refer table 4.28). However, the sector classification does not signify any impact on IPO underpricing in India (refer table 4.30) Hypothesis 2: Issue Related determinants of IPO Underpricing: 1. Offer Price (OP) and IPO underpricing: Several researchers have given different opinions regarding the relationship between the offer price and the initial returns. The school of thought that superior firms must underprice their issues so as to leave good taste in the mouth of the investors to motivate them to invest in their further public offerings as well recommends that there is a positive relationship between the two (Daily et al., 2003) and the second school of thought that the lower offer price suggests small firm with more uncertainties must underprice their issues to compensate the investors for the risk (Jain and Kini, 1999; Ibboston et al., 1988). The current study bestow following results:

14 Though insignificant, the offer price has been positively related to the IPO underpricing in Indian primary markets. Indian primary markets suggest that higher price is an indicator of firm s quality and thus they must underprice their issues so as to signify their firms quality and motivate them to invest in their subsequent offerings as well. 2. Market Capitalization (MCAP) and IPO underpricing: Contrary to the findings of many researchers as of negative relationship between the market capitalization and the IPO underpricing, the results of current study support the school of thought that the bigger firms underprice their issues to signify the firm s quality. Thus, following have been subsequent findings: Market capitalization has been positively related to IPO underpricing for Indian primary markets (Table 4.31 and corresponding chart 4.27 and table 4.37). In consensus to the results of offer price, market capitalization also supported the same school of thought that big firms tend to underprice their issues which is true for Indian primary markets as well. 3. Share Overhang (SHANG) and IPO underpricing: Literature suggests contrary views regarding the relationship between share overhang and IPO underpricing. One school of thought suggests that there is a positive relation between share overhang and IPO underpricing by arguing that the firms with greater share overhang face little uncertainty (Ritter and Welch, 2002; Welch, 1989). The second school of thought suggests that there is a negative relationship between the two as there can be possibility of cash flows used by porters for their self interest (Bozzolan and Ipino 2007; Michaely and Shaw, 1994; Allen and Faulhaber, 1989). Following have been the inferences from the current investigation: Though insignificant, there has been a negative relationship between share overhang and IPO underpricing. The reason for such a support could be that the investors feel more confident when a majority proportion is held by promoters as they retain stocks of the firms

15 with good future prospects only which reduce uncertainty and thus little underpricing. 4. Marketing Expenses (MEXP) and IPO underpricing: Marketing expenses help disseminate the information and reduce information asymmetry. Many researchers found marketing expenses as a significant factor in determining IPO underpricing. Following have been the inferences from the current study: The results of the current study depict significant negative relation between the two as hypothesized (refer table 4.31 and corresponding chart 4.27 and table 4.37) and the results have been corroborated by earlier studies (Alok and Vaidyanathan, 2009; Frieder and Subrahmanyam, 2005 and Habib and Ljungqvist, 2001). The results support the school of thought that reduced uncertainty by disseminating information reduces risk which is not required to be compensated with huge underpricing. 5. Proportion of Equity Sold (PROSOL) and IPO underpricing: Proportion of equity sold (PROPSOL) has not been studied in the existing literature. However, this variable has been investigated in the current study as this is a mandatory disclosure in the red herring prospectuses and the IPO advertisement in India. PROPSOL is an indicator of the paid up capital the company would raise after the issue and the extent of public holdings in the company. Following have been the inferences from the current study: There has been a significant positive relationship between the proportion of equity sold and the IPO underpricing (refer table 4.37). The assumption regarding the PROPSOL is that greater the proportion of equity sold, greater is the underpricing as the promoters have not retained the management powers with themselves and the transaction costs would have been limited for further offerings.

16 6. Listing Lag (LL) and IPO underpricing: A unique contribution of the current paper has been the study of recent regulatory changes in Indian primary markets. One such change has been the reduction in the number of days in listing a new issue. Securities and Exchange Board of India has reduced the listing time for the IPOs to 12 working days from existing 22 days w.e.f. 1 May, 2010 (DNA, 2010). The reduction in listing lag has been presumed to eliminate the impact of market movements on the listing price of the issue. Table 4.18 displays the listing lag from 2004 to 2011 and the listing lag decreases from around 21 days in 2004 to 14 days in The underpricing also decreases with the decrease in listing lag. According to deciles table 4.31 and corresponding chart 4.27, though the relationship between listing lag and IPO underpricing is weak and non monotonic, however it indicates that with decrease in listing lag, the underpricing decreases. The results were corroborated by the multiple correlations table 4.36 and the multiple regression table4.37 which shows significant positive relationship between the underpricing and listing lag. The results suggest that with the decrease in 1 day in listing, the underpricing decreases by 1.3%. Panda and Vaidyanathan (2009) depicts significant positive relation between two variables in Indian context and our results have been in consensus with the earlier results. Thus, the study have presented following subsequent facts: The decrease in listing days would be followed by the decrease in underpricing. The regulation change of reduction in listing lag in Indian primary markets has been successful in reducing the underpricing of new issues. 7. Over Subscription and IPO underpricing: The current study considered the oversubscription levels in varied investors categories namely overall oversubscription (OSUBS) and its categories Qualified Institutional Bidders (QIBs), Non institutional Bidders (HNI) and Retail bidders (RETAIL). The study examined their independent impact on IPO underpricing. The results of all the categories have been similar and displayed monotonic positive relationship with IPO underpricing. Following subsequent facts have been disclosed:

17 There has been a strong positive relationship between the overall and categorical subscription levels and the IPO underpricing (table 4.36 and table 4.37). On an average, the subscription levels in Retail and OSUBS category should be above 3 and in HNI category it should be above 2 and the QIB category the subscription levels should be around 6 to yield returns of 6% or more in most cases. The investors in QIB category are mainly large institutions who have been expected to possess hidden or crucial business information as well about the new issues. Thus, the huge oversubscription in this category definitely increases the probability of getting huge initial returns. 8. Offer Price on upper side of Price band (DVOP) and IPO underpricing: Literature argues that the issue has been in great demand if the stock is allocated on upper side of the price band. The relationship thus was assumed to be positive on the basis that only reputable issues would be demanded more and such reputed firms tend to underprice their issues to signify their firm s quality. The following have been the research findings: There has been a significant positive relationship between the DVOP and IPO underpricing (refer table 4.37). Table 4.33 corroborated the detailed results by depicting negative underpricing for lower demand firms and vice versa, maybe they have been small firms that tend to overprice their issues and raise public fund. The results support the school of thought that reputed and better prospects firms have been underpriced so as to signify their firms quality and retain investors faith for the subsequent public offerings. The results also supports that the Indian IPOs have generally been in great demand. Refer table 4.33, 208 issues out of 250 issues have been allocated at upper side of the price band and enjoyed an average underpricing of 20.71%.

18 9. Issuing Firm Grading (DVRATE) and IPO Underpricing: Securities and Exchange Board of India (SEBI) necessitated the assignment of grading by issuing firm w.e.f. May 1, 2007 (Pathak, 2008). The IPO grading is an indicator of the firm s fundamentals, though the grading does not comment anything on the pricing and the market value of the issuing firm. The current study has investigated the impact of assigned grading on IPO underpricing and following have been the inferences: Table 4.32 supports the school of thought that the firms with poor grading tend to underprice more to compensate the investors for the risk undertaken by them. The underpricing tends to decline as the grading increased from 0 to 4. Table 4.32 also supports the Welch s school of thought that very superior firms with grading of 5 also underpriced their issues tremendously to convey a signal of their firm s quality to investors and retain their faith for subsequent offerings. Table 4.37 however clarifies that there is no significant relationship between IPO underpricing and assigned grading. 10. Loan repayment as one of the issue objective (DVLOAN) and IPO underpricing: Loan repayment as one of the objectives of the issue has not been studied as prospective reason for IPO underpricing. The loan repayment as an issue objective could be perceived in two ways. One is in negative sense that the issuing firm has not been able to manage its solvency well till date and the additional money raised would not be used for productive purposes. Second is in positive sense that after repayment of loan, the issuing firm would re-structure its capital and would work well in future. Thus, based on these assumptions, DVLOAN has been investigated for its probable affect of underpricing and following inferences have been drawn: There is no significant relationship between loan repayment as an objective of issue and IPO underpricing (refer table 4.37). Table 4.34 presents that the firms that have loan repayment as an issue objective underprice less than the firms that do not have loan repayment as an objective. Thus, the results support the school of thought that firms that employ the raised

19 funds in productive purposes have been superior in quality and tend to underprice to signify their firm s quality. 11. Underwriter s reputation (DVUW) and IPO underpricing: The IPO characteristic Underwriter s reputation has been widely studied in international context. Due to lack of data for Indian underwriters, the researcher has used the list compiled by Bloomberg in 2011 regarding the underwriters reputation. Thus, dummy variable of 1 has been assigned if any one of the underwriters amongst issuing firm underwriters has been in the top list of underwriters presented by Bloomberg in 2011, else 0. Following have been the results from the current investigation: In corroboration with earlier studies (Johnson and Miller, 1988; Booth and Chua, 1996; Kim and Ritter, 1999; Wang, 2005; Chang et al., 2008; Lamberto and Rath, 2010), the results found significant negative relationship between Underwriters s reputation and IPO underpricing (refer table 4.37) Hypothesis 3: Market Related determinants of IPO Underpricing: 1. First day return on Stock (FRET) and IPO underpricing: In consensus with the previous studies (Gupta, 2011), the results support the hypothesis that the Indian primary markets are efficient enough and impact the initial returns of an investor. Table 4.38 and corresponding chart 4.34 presents that the investors earn huge returns if the first day there is a positive trend between the list price and the list day close price. The results support the following findings: There has been a significant positive relation between the first day IPO return and the IPO underpricing. The Indian markets are efficient and adjust the price of the new issue by the end of the day and investors can strategize their investments during the first day as the movements could be predicted.

20 The investors who fail to buy at the time of new issue could be benefitted by purchasing on the list day as well. In fact, they could investigate the list day price movement to manipulate their investment decisions. 2. Pre-subscription market return for 30 days (PSUBS30) and IPO underpricing: A unique contribution of this research has been to study if the new issue price could fully incorporate the market movements. The results have been similar to the earlier results (Loughran and Ritter, 2002). There has been a positive relationship between the Pre-subscription market return and the underpricing (Table 4.41). This means that the new issues do not incorporate the entire market movement and the investors thus could earn huge initial returns by subscribing to issues preceded with positive market movements. 3. Pre-list market return for 9 days (PLIST9) and IPO underpricing: The pre-list market returns have not been known at the time of subscribing the new issue in the primary market, but it is easy to forecast the near future market movements. The market returns immediately preceding the listing of an issue has a huge impact on the underpricing. Following have been the subsequent findings: There is a significant positive relationship between pre list market returns and the IPO underpricing (Table 4.41). 4. Hot and Cold Market and IPO underpricing: There is a dissonance of the results of the current study with the earlier studies (Ibboston and Jaffe, 1975; Ritter, 1984; Loughram and Ritter, 2002; Gupta, 2011) that the huge number of IPOs leads to huge underpricing. Table 4.39 illustrates that 50 new issues have been listed in cold market and 200 new issues have been listed in the hot market, but the average underpricing has been almost the same, being 15.42% for cold markets and 16.02% for hot markets. The results have been in consensus to the findings on Indian primary markets of Majumdar (2003). Thus, following have been the findings:

21 There is no relation between the market conditions in terms of number of issues and underpricing in Indian primary markets. The reason appended could be that only firms with sound fundamentals dare to list their issues in cold markets. This was indicated by a little number of firms that got listed in cold markets. In Hot market conditions, the market was flooded with both good and bad issues and thus, the returns were averaged Hypothesis 4: Findings for the Predictive Model: A predictive model has been developed using the significant IPO characteristics of each of the firm related, issue related and market related characteristics (refer chart 4.33). The model developed has further been tested for its accuracy by applying a regression model considering all the IPO characteristics in holistic manner. The purpose is to remove any incidents of auto correlation and collinearity across the variables in three categories. The results have been almost similar and the holistic model 5 determines 62.6% (adjusted R 2 is 57.4%) of IPO underpricing in India. The results have been far better than earlier studies of Pande and Vaidyanathan (2009) that determines only 18.54% of IPO underpricing in India. The contribution for enhanced determination of underpricing goes to the extensive variables considered in the study and the time period taken for the study. The comprehensive model derived has been described as follows: Underpricing = EMP BVMV DVUW MEXP DVOP LL PROPSOL HNI RETAIL LNMCAP FRET. Where, EMP BVMV DVUW MEXP = Number of Employees of issuing firm = Book Value to Market Value = Dummy variable of 1 if underwritten by reputed underwriters = Marketing Expenses

22 DVOP LL PROPSOL HNI RETAIL LNMCAP FRET = Dummy variable of 1 if issue allocated on upper side of price band, else 0. = Listing Lag between subscription close day and list day = Proportion of equity sold = Subscription in Non Institutional Category = Subscription in Retail Category = Natural Logarithm of post issue market capitalization = First day return on stock. The model is statistically significant and all variables discussed above determine 57.4% of IPO underpricing. The residual determinants of underpricing could be attributed to human behavior and irrational decisions. The participation of human behavior in making financial decisions renders finance more as an art than a science. Thus, the predictability of financial models depends up on the human decisions as well as on uncontrollable external environment Hypothesis 5: Impact of Recent Regulatory changes on IPO underpricing Indian primary markets witnessed three major regulatory changes in the recent past. The recent amendments by SEBI include the increment of the investment limit for retail investors from Rs.1lakhs to Rs.2lakhs w.e.f. May1, 2010; reduction in the listing time for the new issues to 12 working days w.e.f. Nov. 2010; and the necessary IPO grading w.e.f. May 1, Thus, it is imperative to conclude the impact of these regulatory changes on IPO underpricing. In the current study, the researcher has studied the impact of two regulatory changes due to limitation of time and data necessitated IPO grading and reduction in listing time. The study on IPO grading has not depicted any significant relationship with IPO underpricing. However, the close observation suggested that the underpricing decreases with the increase in grading up to grade 4, which is apparent from the fact that improved grading bestows more confidence amongst investors and reduces risk. Thus, the investors gain little underpricing. The detailed observation also highlighted that the issuers with grading 5 allowed huge underpricing

23 to investors. The reason that became apparent from the scenario is that superior firms underprice their issues to signal their firm s quality and earn investors faith for the subsequent offerings (refer table 4.32 and chart 4.28). The study of reduced listing days depicted significant positive relationship with IPO underpricing, meaning the reduction in listing lag reduced IPO underpricing (refer table 4.37) Hypothesis 6: Investors attitude towards IPO underpricing: The actual investment for the investors in the new issue is not equal to the stocks allocated, rather the amount for which the stocks have been subscribed. The Indian primary markets have been characterized by huge oversubscription (refer table 4.19). The securities and exchange board of India recently amended the rule specific to the maximum amount that could be subscribed in the retail investor s category. On March 29, 2005, SEBI redefined the retail individual investor as one who applies or bids for securities of or for a value not exceeding Rs 1 lakhs. This limit has been recently enhanced to Rs. 2 lakhs w.e.f. May 1, 2010 (DNA, 2010). The above discussion motivated the researcher to study the actual realized returns and its determinants. The proxy for realized returns for investors after considering the oversubscription level has been defined as allocation weighted initial returns. Table 4.43 indicates that the average allocation weighted initial return has been 1.2% and table 4.18 indicates that the average listing lag in India has been around 20 days. Thus, the returns could be re-invested to compound the earnings. Thus, the investors in Indian primary markets were not at economical loss. The researcher extended the study to find the determinants of allocation weighted initial returns using only the information that has been publicly available to investors before subscribing the issue. The researcher found that study of fundamentals (earnings- EPS, NP and EBIT), size (number of employees LNEMP) and market conditions (HOT) indicates the prospective allocation weighted returns to the uninformed investors. 5.3.Conclusion and Research Implications: The current research has elaborated the literature on underpricing scenario across the world and the varied theories supplemented in the literature as a probable cause for underpricing across the

24 world. However, the results for determinants of underpricing have not been in consensus across different markets and the time periods. Thus, based on backdrop of existing literature on IPO underpricing and its reasons, the current study derived two schools of thoughts that have been founded to be providing contradictory opinion. These two schools of thoughts derived have been elaborated as follows: 1. One school of thought recommends that larger, more experienced and more stable firms with employment of reputable underwriters offer low risk new issues and thus, should experience little underpricing. The argument appended has been that such firms are easier to value, offer little risk to investors, disseminate appropriate information and thus, investors do not expect price protection against valuation errors. This school of thought basically coincides with the Baron s (1982) model of ex ante uncertainty which has been further explained and elaborated by several studies by Rock (1986), Beatty and Ritter (1986), Welch (1992), Amihud et al. (2003) amongst others. 2. The second school of thought suggests that larger, more experienced and more stable firms intentionally underprice their issues to leave a good taste in the mouths of investors and retain their trust for their subsequent offerings. This school of thought has its origin in Signaling theory originally developed by Leland and Pyle (1977) and further elaborated by Downes and Heinkel (1982), Ritter (1984b) and Welch (1989). Further, with the backdrop of the existing literature and experts opinion in the field, the researcher has classified the IPO characteristics as firm related, issue related and market related. Firm related IPO characteristics comment on the firm s fundamentals and financials which is outwardly disclosed in new issue prospectus and incorporate variables like age of the firm (AGE), book value to market value (BVMV), firm s price earnings (FPE), industry price earnings (IPE), number of employees (LNEMP), cash to total assets (CASH), earnings before interest and tax to total assets (EBIT), debt to total assets (DEBT), total assets (LNTA), and sector to which issue belongs (DVSEC). The prefix LN has been assigned where natural logarithm of the variable is taken and DV has been assigned where dummy variable is taken (refer table 3.1 for detailed definitions). The variables discussed above have been the indicators of firm s size and financial performance.

25 Issue related IPO characteristics elucidate factors related to pricing, underwriting, subscription and others that have been specific to that particular issue. Issue related IPO characteristics include offer price (LNOP), market capitalization (LNMCAP), share overhang (SHANG), marketing expenses (MEXP), proportion of equity sold (PROPSOL), listing lag (LL), overall oversubscription and in varied categories namely retail, non-institutional and qualified institutional bidders (OSUBS, RETAIL, HNI, QIB), allocation on upper side of price band (DVOP), grading assigned to IPO (DVRATE), underwriter s reputation (DVUW) and loan repayment as an objective of issue (DVLOAN). Market related IPO characteristics incorporate the prevailing market conditions and include hot and cold market conditions (HOT), pre-subscription market returns for 30 days (PSUBS30), prelist market returns for 9 days (PLIST9), market return (MR) and first day return on the stock (FRET). All the IPO characteristics mentioned above have been tested for its impact on IPO underpricing using the descriptive study of each of these variables vis-à-vis underpricing, deciles of each of these variables and the corresponding movement in underpricing, and further multiple correlations and multiple regressions have been used to derive a predictive model. All the IPO characteristics have been discussed in the light of two schools of thoughts discussed above as well to comment that which school of thought is more pertinent for the Indian primary markets. Albeit the detailed findings as discussed above, following have been the imperative unique recommendations derived from the study: The findings from the current study have dissonance with the earlier studies in terms of the degree of underpricing. The degree of underpricing has been decreasing over years which act as a good indicator for the issuers as underpricing is an indirect cost to issuers. Though the degree of underpricing has been decreasing over years, but then the investors are not at economic loss by investing in primary market in India. The average underpricing has been15.90% and the average allocation weighted initial return has been 1.2% which is for the average listing lag of 20 days. Thus, the returns

26 could be re-invested to earn more than the normal rate of interest of around 8% for Indian markets. Several recent amendments by Securities and Exchange Board of India (SEBI) have been studied for its impact on IPO underpricing. SEBI necessitated grading of IPOs by the rating agencies w.e.f. May 1, 2007 (Pathak, 2008). The current research studied the variable and found no significant impact of grading on IPO underpricing. However, the detailed analysis through some light on the trend of underpricing based on assigned grading. Table 4.32 depicts that the underpricing decreases with the increase in IPO grading from 1 to 4 which support the ex ante uncertainty theory. However, it has also been visible from the table that issues with excellent grading of 5 also underprice their issues to maintain their reputation and retain investor s trust for their subsequent offerings. Another recent change by Securities and Exchange Board of India (SEBI) has been the reduction in the listing time for the IPOs to 12 working days from existing 22 days w.e.f. 1 May, 2010 to reduce the valuation errors due to secondary market volatility (DNA, 2010). Thus, the current research incorporated the IPO characteristic Listing Lag (LL) in the study to investigate its impact of IPO underpricing. The results in table 4.37 report significant positive relationship between the listing lag and IPO underpricing. This implies that regulatory decision to reduce listing days has been well received by the market in reducing the risk due to market volatilities during long listing times. An important observation was the underpricing during hot and cold market conditions (refer table 4.39). The investors in hot market were no better than the investors in cold market. The probable explanation is that the hot markets have been flooded with both good and bad issues whereas the cold markets experience only good issues. Thus, poor or overpriced issues during hot market compensate for the returns earned in good or underpriced issues. The results from cross-sectional analysis depicts significant negative relationship with book value to market value (BVMV), underwriter s reputation (DVUW), marketing expenses (MEXP), and number of employees (LNEMP) and significant positive relation with allocation on upper side of price band(dvop), listing lag (LL),