Long Run Performance of IPO-Empirical Evidence from Indian Capital Market

 

Dr. Amit Hedau

 

ABSTRACT:

IPO is the buzz word in the capital market across the world. Though the offer price is calculated using historical financial information, offer price do not sustain for long period after IPO. In most of the countries including India, it is observed majority of IPO are trading at a price lower than the offer price. Using the CAAR, BHAR and Wealth Relative Index, this paper analysed the long run performance (up to 18 months post listing period) of the IPOs offered in the Indian capital market during January 2007- December 2015. Findings of the paper are in consistent with the international evidence. This study can be useful to the corporations who are planning IPOs in the near future to understand dynamics of Indian capital market.

 

 

INTRODUCTION

In March 1602, the "VereenigdeOost-IndischeCompagnie" (VOC), also known as Dutch East India Company was formed. The VOC was the first modern company to issue public shares, and it is this issuance, at the beginning of the 17th century, that is considered as the first modern Initial Public Offer (IPO). The company had an original paid-up share capital of 6,424,588 guilders¹. The ability to raise this large sum is attributable to the decision taken by the owner’s to open up an access to share ownership to a wide public. Everyone living in the United Provinces had an opportunity to participate in the Company. Each share was worth 3000 guilders. All the shares were tradable, and the shareholders received receipts for the purchase. A share certificate documenting payment and ownership, such as we know today, was not issued but ownership was entered in the company's share register. In the United States, the first the public offer was made by the Bank of North America in around 1783.

 

In India, the Bombay Stock Exchange (BSE), the oldest stock exchange in India and Asia established in 1875. D.S. Prabhudas& Company (now known as DSP Merrill Lynch, a joint venture partner with Merrill Lynch) is the first company to register with the BSE. National Stock Exchange (NSE), jointly owned by financial institutions, insurance companies, banks and other financial intermediaries and third largest exchange in the world, came into existence in 1993 and started trading in mid 1994. A wide body of research spanning thousands of offerings over decades and across several markets around the world shows that new issue public offerings tend to do well on the first day of their listing or in the initial period after listing (the under-pricing effect), but underperform the broader market over the long run. The present paper aims to analyse the long run performance, up to 18 months during post listing periodwith the help of Cumulative Average Abnormal Returns (CAAR), Buy and Hold Average Returns (BHAR) and Wealth Relative Index. The present study may be considered as one of the comprehensive research work in terms of period and sample size of the study.

 

LITERATURE RIVIEW:

Behavioral economists have demonstrated that individuals often violate Bayes' Rule and rational choice theories when making decisions under uncertainty in experimental settings (Kahneman and Tversky, 1974). In a similar vein, financial economists have also discovered long-run pricing anomalies that have been attributed to investor sentiment. Behavioral theories posit that investors give too much weight to recent results and trends. Eventually, overoptimistic investors are disappointed and subsequent returns decline.

 

To cite several important illustrations, Alvarez, Susana and Gonzalez (2001) analysed all the new issues in Spanish market during 1987-1997, with a sample of 56 firms to provide evidence on initial under pricing and long run underperformance of new issues. They used Buy and Hold Returns (BHR), Calendar time portfolios and the Fama and French three factor Model for their analysis. The results of study showed that the existence of long run underperformance depended upon the methodology used. There existed long run underperformance when BHR were used and not when mean calendar time returns were employed.

 

Gompers and Lerner (2003) analysed the performance of nearly 3661 new issues in the United States from 1935 to 1975 for up to 5 years. The results displayed the evidence of underperformance when event time buy and hold abnormal returns were used. The study indicated that the initial returns of recent new issues contained information on the market’s valuation of future public issues.

 

Ritter and Welch (2002) examined three main aspects i.e. why firms go public, why they reward first day investors with considerable under pricing and how public issues perform in the long run. The paper explained the “Life cycle theories‟ and “Market timing theories‟. This study presented both theoretical and empirical evidence for short run and long run underperformance of public offerings of the new issues and showed that the under pricing was sensitive to methodology and to the time period chosen.

 

Moshirianet al (2010) examined the post-issue stock price performance of initial public offerings (IPOs) from Asian markets from 1991 to 2004. They provided a comparative assessment on the short- and long-term stock performance of Asian IPOs with comprehensive international evidence. They used several different methods to examine the strength of IPO performance. They revealed that while there is initial under pricing in Asian IPOs, the existence of long-run underperformance for the Asian IPOs depends resoundingly on the methodology used for assessment.

¹A guilder is currency of Netherlands having currency symbol as ANG and currency code as NAƒ or NAf or ƒ. In the long run, the evidence is that of underperformance, i.e., negative returns accrue to the investors holding these IPOs. Numerous studies in various countries (see Table 1) have confirmed underperformance after one year (Aggarwal and Rivoli, 1990), three years (Ritter, 1991 and Loughranet al., 1994), five years (Loughran and Ritter, 1995).  Ibbotson (1975) studied the negative relation between initial returns and long run share price performance for a sample of US public offerings issued during the period 1960-69. He reported that there was a general positive performance in the first year, negative performance in the next three years and a general positive performance in the fifth year. The distribution of returns was highly skewed, indicating that these investments were individually very risky.

 

Dawson (1987) examined the one year market adjusted returns for initial public offerings in Hong Kong, Singapore, and Malaysia during 1978-1984, and found those in Hong Kong were down 9.3per cent, and those in Singapore were down 2.7per cent. However, neither decline was statistically significant. In contrast, there was a positive, statistical significant over performance in Malaysia of 18.2 per cent.

 

Aggarwal and Rivoli (1990) examined the long-run performance of their sample of 1598 public offerings offered from 1977 to 1987. They found that investors investing in the public offerings at the end of the first day of trading and holding them to the 250th trading day (roughly one calendar year) earned a negative cumulative average abnormal return of 213.75 per cent. Ritter (1991) reported that newly listed companies substantially underperform a set of seasoned firms matched by size and industry in stock returns for the first three years of their listing. His sample consists of 1,526 initial public offerings in the U. S. market between 1975 and 1984. He interpreted his finding as a result of investor’s temporary over optimism of future growth and firms time the market to go public. The latter argument is supported by the fact that new issues tend to be time clustering.

 

Levis, M (1993) investigated the long run performance of a sample of 712 UK public offerings issued during 1980-1988. He reported underperformance of –11.4 per cent after three years. He reported long run returns based on three alternatives benchmarks: the Financial Times Actuaries All share Index, the Hoare Govett Small Companies Index and All Share Equally Weighted Index. Loughran and Ritter (1995) investigated the post-issue stock returns of 4,753 initial public offerings and 3,702 seasoned equity offerings from 1970 to 1990. They reported that the 3 and 5 year buy and hold returns in the post-issue period were 26.9 per cent and 50.7 per cent less than non-issuing firm with the closest size to them respectively. The underperformance was found significant both statistically and economically. Their results are robust to some alternative benchmarks, such as equal-weighted and value-weighted returns on market indices.

 

 

Table 1: Various Studies on Long Run Under Performance

Source-Authors Compilation

 

Rajan and Servaes (1997) examined initial public offerings from 1975-1987. They found a five-year raw return of 24 per cent. This represented a 47 per cent underperformance when compared against NYSE/AMEX index, a 17 per cent underperformance against the smallest decile from the NYSE/AMEX, and a 41 per cent underperformance against firms matched by size and industry. Jog (1997) documented long run performance of 308 public offerings listed in Toronto Stock Exchange during the period 1971 to 1995. The study reported that Canadian new issues underperformed benchmark (TSE 300) in first three aftermarket years and improved their performance from the fourth aftermarket year.

 

The present paper aims to analyse the long run performance, up to 18 months during post listing period with the help of Cumulative Average Abnormal Returns (CAAR), Buy and Hold Average Returns (BHAR) and Wealth Relative Index. The present study may be considered as one of the comprehensive research work in terms of period and sample size of the study.

 

RESEARCH METHODOLOGY:

Sources of Data:

The present paper seeks to analyze the long run performance (up to 18 months from IPO) of the IPO’s offered in the Indian capital market during 2007 to 2015 (January to December). The information about the IPO offered during period of research is taken from official website of National Stock Exchange (NSE).

 

Table 2 – Sample construction

	No. of New ssues
All listed new issues during the period of study ( January 2007 – December 2015)	314
Follow on Public Offers (FPOs)	(67)
Information not available in the data base	(21)
Valid Cases Selected for the Study (71.97%)	226

 

Methodology for Long Run Performance Analysis:

The present study also analyzed the long-run performance of (226) new issues floated in Indian Capital Market during the research period (January 2007-December 2015). The data for post-issue performance is obtained from Prowess database maintained by CMIE. To analyze the post listing performance of new issues, the standard event study methodology is applied. The abnormal returns of the sample are computed using the Cumulative Average Abnormal Returns (CAAR) and Buy-and-Hold Abnormal Returns (BHAR). There continues to be disagreement regarding the measurement of long-run abnormal return performance of the new issues. Barber and Lyon (1997) Lyon et al. (1999), Kothari and Warner (1997) and Fama (1998) analyze the difference of these two methods, and find that there is no consensus on the preferred one. Lyon et al. (1997) document that BHARs should be used if the research question is whether or not investors earn abnormal stock returns by holding stocks over a particular time horizon. While the CAAR approach should be employed to answer: Do sample firms persistently earn abnormal monthly returns?

 

Though CAARs implicitly assume frequent portfolio rebalancing, Fama (1998) justifies its use since it would produce fewer spurious rejections of market efficiency than would the use of BHARs. There also exists a greater knowledge of the distribution properties and the statistical tests for CAARs. Since in India, the majority of investors are individual, and they trade much more frequently than those in other markets, to guarantee the robustness of the results, present study increased the frequency of measurement i.e instead of yearly; monthly abnormal return are applied using both the measures. Research in the US has frequently used a matching firm approach to measure long-run abnormal returns. However, many new issues floated in India during research period are first in the industry; no matching firm’s data is available. In case if matching firms is available, the number of matching firms is very small. The small number of companies’ available means there would be a bias caused by the repeated use of matching companies.

 

Motivated by the existing international practice, this research used both BHAR and CAAR to evaluate long term performance for a period of 18 months from the date of listing. The BHAR is the difference between the holding-period returns of stock ‘i’ and the market return. The first measure used is the eighteen months buy-and-hold market-adjusted returns (BHAR), defined as:

BHAR_iT=-

Where, r_{it =} monthly return of the stock

 

                r_{mt =} monthly return of the of the benchmark index

 

After calculating BHAR for each month, mean  is computed by using following formula

 

A simple t-test is employed to test the null hypothesis of zero mean market-adjusted buy-and-hold return

 

Where,σ BHAR is the standard deviation of the buy-and-hold market-adjusted returns, and n is the sample size.

The second measure used is the eighteen months cumulative average market-adjusted abnormal returns (CAAR). The CAAR is calculated in the following manner.

 

First, the raw returns of the stock i in the event month t is measured as follows:

                (P_c-P_i)

Rit= --------------

                    P_i

Where R_it is the monthly raw return for the company i in the event month t where the starting price for each company is its last price for the month of listing, excluding the initial return, P_c is the closing price on the following month of listing, P_i is closing price of the previous month (P_c – 1). Thus, monthly returns for next 18 months are calculated. Second, the benchmark index returns is calculated in the same way as the raw returns over the same period as follows:

 

                    (Pc-bench- Pi-bench)

R bench =-----------------------------

                                Pi-bench               

 

Where, R bencht is the monthly benchmark return from the index, Pi-bench and Pc-bench is the monthly returns of the benchmark index in the previous and following the event month respectively. As an additional robustness check, another measure, Wealth Relative Index using the procedure employed by Ritter and Levis is calculated. The magnitude of this measure is an indication of the performance of new issues vis-a-vis the market. A wealth relative greater than unity implies that new issues outperformed the market in that period, while a wealth relative below 1, indicates under-performance.

 

WRit for a sample of n stocks from offer date, to date “t” is calculated using the formula:

 

Where R_it is the return for individual stock i at time t, R_mt is the market index return for the corresponding period. The total size of new issues for the discussion is represented by N.

 

Analysis of Long Run Performance:

This section of the paper discusses about the findings of Long Run performance of IPOs ina step wise manner.

 

Buy and Hold Average Return (BHAR):

From the observed sample size, it is found that 143 (63%) new issues with negative BHAR. These negative returns can be interpreted as post listing under performance of new issues compared to benchmark market index. A positive buy and hold return indicate over performance in relation to the market index. This suggests that there is significant under performance of the new issues during period of research. The mean buy and hold average returns for the first 18 months after listing are shown in table 3 as follows:

 

Table 3 – Average BHAR during Post Listing Period of 18 Months

Source: Author’s Computation

 

It can be observed from the above table that mean buy and hold average return for 14 months, out of 18, is negative showing consistent under performance. For the first 7 months it is persistently showing negative returns. These findings are in contrast to the international evidence Aggarwal Lie and Rhee (2008); Alvarez and Gonzalez (2005); and Ritter and Welch (2002), where the underperformance even continued up to 36 months. Indian new issues floated during research period showed significant percentage of under pricing. It means if investor had bought shares from the secondary market, his investment amount would be higher for same quantity of shares. And average buy and hold returns for the same period would be much higher in negative terms than indicated in the above table. Therefore it can be interpreted that it is only those investors who acquire shares through primary market are able to earn excess returns compared to the investors from the secondary market.

The result obtained from this study showed a strong relationship between under pricing and long run performance of the new issues. More is the under pricing lesser is the buy and hold returns. However if new issues are fairly priced or overpriced, investors would have lost the more money. Investors investing in new issues through primary market are better-off compared to the investors investing on the listing day.

Following table shows the results of one sample t test. The BHAR is statistically significant at 5% level of significance.

 

Table 4 One-Sample Test of BHAR at test value = 0

Source: Author’s Computation

 

Cumulative Average Abnormal Returns (CAAR):

The overall results show that the returns given by the new issues floated in India during period of research have been much lower. Among the 18 months, only 10^th months shows negative CAAR less than 50%. Highest (27.28%) and lowest (15%) negative CAAR is observed in the 1^th and 11^th month respectively. The table 4 reports the Cumulative Average Abnormal Returns (CAAR) for the 18 months during post listing period.

 

Table – 5 CAAR of New Issues during Post Listing Period of 18 Months

Month	No. of New Issues with		Total	% of Negative Returns	Highest Positive Return	Lowest Negative Returns
	Negative Returns	Positive Returns
1	153	73	226	67.84	30.31	-27.28
2	142	84	226	62.87	31.51	-20.24
3	125	101	226	55.56	31.65	-19.24
4	138	88	226	61.11	25.66	-20.37
5	128	98	226	56.73	35.71	-21.04
6	128	98	226	56.73	29.29	-22.06
7	140	86	226	61.99	38.48	-26.76
8	119	107	226	52.92	29.47	-18.41
9	140	86	226	62.28	36.64	-20.8
10	111	115	226	49.12	34.62	-19.98
11	150	76	226	66.67	36.77	-15
12	120	106	226	53.37	38.76	-18.85
13	153	73	226	67.94	34.98	-19.47
14	128	98	226	56.93	35.13	-19.64
15	140	86	226	62.24	42.78	-20.2
16	129	97	226	57.4	33.68	-21.84
17	139	87	226	61.69	40.62	-17.63
18	153	73	226	68.07	32.13	-17.98

Source: Author’s Computation

 

The results are in line with the previous literature all over the world that proves the underperformance of public offerings. This is an indicator of informational inefficiency of capital markets in India. A simple t-test is employed to test the null hypothesis of zero mean cumulative average adjusted returns (CAAR).

 

Where,σ CAR is the standard deviation of cumulative abnormal returns for the sample of n firms and n is the number of new issues.

 

 

 

Table 6 One-Sample Statistics of CAAR

Source: Author’s Computation

 

Table 7 One-Sample Test of CAAR at test value = 0

	t	df	Sig. (2-tailed)	Mean Difference	95% Confidence Interval of the Difference
					Lower	Upper
CAAR	-6.957	226	.000	-11.868	-15.22	-8.51

Source: Author’s Computation

 

Results of one sample t test for monthly CAAR are produced as follows: 

Table 8 One-Sample Test of Monthly CAAR at Test Value = 0

Source: Author’s Computation

 

The table gives the results of t test applied to test significance of CAAR. In the 18 months event window, the CAAR found to be statistically significant in the 1^st, 2^nd 4^th, 6^th, 7^th, 10^th, 11^th, 12^th, 13^th and 18^th month.

 

Wealth Relative Index (WR):

In the year 2007, 2008 and 2014all the new issues offer during the research period are under performed.. For the remaining years, majority of the new issues gave wealth relative of less than unity. The following table shows the year wise wealth relative index of the new issues in the respective years.

 

Table 9 Year Wise Wealth Relative

Year	Total no. of New Issues	Wealth Relative
		< 1	> 1
2007	67	67	0
2008	26	26	0
2009	15	8	9
2010	51	32	19
2011	29	18	11
2012	9	7	2
2013	8	6	2
2014	6	6	0
2015	15	10	5

Source: Author’s Computation

The findings suggest that greater parts of the new issues are continuously underperformed up to 18 months of their listing. Alternatively it can be said that the portfolio of the new issues is constantly losing their values with respect to market benchmark index up to 18 months of post listing period. It can also be interpreted in such way that initial day traders who have purchased the new issues at the offer price from the primary market and have hold it for 18 months cannot expect positive market adjusted return. Hence it can be concluded that investors should sell their IPO allotment on the day of listing to earn the short term abnormal returns. Otherwise they have to wait for a period of more than 18 months to earn positive abnormal returns. The findings of this study are in consistent with international evidence Ritter (1991); Levis (1993); Almeida and Duque (2000) and Jaskiewiczet al (2005). To wrap up, it can be said that there may be IPO related factors which may contribute towards discouraging long run performance of IPO, operational performance during post listing period should be given more weighted to evaluate the same. A corporate entity is always surrounded by issues outside its control, including stock market, economy and fluctuations within the industry. The management can focus on factors within their control, meeting or beating investor’s expectations and delivering on promises remains as crucial as ever.

 

CONCLUSION:

Like rest of the world, the poor performance of IPOs over long run is observed in Indian Capital Market too. The anomaly of long run under performance is witnessed in Indian as well as leading stock markets across the world.  Out of the studied sample, The Buy and Hold Return (BHAR) of 142 new issues (63%) of sample size, is negative up to 14 months confirming persistent under performance of new issues.

 

In the eighteen months event window, more than half of the new issues are showing negative Cumulative Average Adjusted Return (CAAR) which validates the consistent under performance during period of research. For sample size of 226, the wealth Relative (WR) index of 180new issues is less than unity indicating under performance of new issues in long run. Even in year wise analysis, majority of new issues are showing WR less than unity.

 

Since most of issues are underpriced, investors gain substantial returns on listing day. However, these high prices are not sustainable for long term. In case of decline market, IPOs decline faster than the market index but failed to match the pace of market improvement when the market starts recovering. Hence the IPOs underperform in the long run. Though, it cannot be generalized, as some IPOs performed well after the listing.

 

REFERENCES:

1.        Alvarez, Susana and González, Victor, Long-Run Performance of IPOs in the Spanish Capital Market (undated).EFMA 2001 Lugano Meetings. Available at SSRN:  http:// ssrn.com/abstract =274086 or http://dx.doi.org /10.2139/ssrn.274086

2.        Aggarwal, R., Rivoli, P. (1990), “Fads in the Initial Public Offering Market”, Financial Management (Winter), pp 45–57.

3.        Dawson, S. M. (1987), “Secondary Stock Market Performance of Initial Public Offers, Hong Kong, Singapore and Malaysia: 1978–1984”, Journal of Business Finance & Accounting, 14: pp 65–76.doi: 10.1111/j.1468-5957.1987.tb00529

4.        Gompers, Paul A. and Lerner, Josh (2003), “The Really Long-Run Performance of Initial Public Offerings: The Pre-Nasdaq Evidence”. Journal of Finance, Vol. 58, pp. 1355-1392

5.        Kahneman D. Tversky A Science, (1974),” Judgment Under Uncertainty: Heuristics and Biases” New Series, Vol. 185, No. 4157. pp 1124-1131

6.        Loughran, T., Ritter, J., Rydqvist, K. (1994). “Initial Public Offerings: International Insights,” Pacific-Basin Finance Journal, 2, pp 165–199.

7.        Loughran, T., Ritter, J. (1995), “The New Issues Puzzle,” Journal of Finance, 50, pp 23–51.

8.        Moshirian. F et al. (2010), “Model specification and IPO performance: New Insights from Asia.” Research in International Business and Finance, (24), pp 62-74.

9.        Ritter, J. (1991), “The Long-Run Performance of Initial Public Offerings”, The Journal of Finance, 46, pp 3-27.

10.     Rajan, R., Servaes, H. (1997), “Analyst Following of Initial Public Offerings,” Journal of Finance 52, pp 507-529

 

 

 

 

 

 

Received on 04.02.2018          Modified on 24.02.2018

Accepted on 01.03.2018           ©A&V Publications All right reserved