ASAC 2007 Ottawa, Canada Sebouh Aintablian School of Business Lebanese American University Suzanne Mouradian (student) Institute of Financial Economics American University of Beirut INITIAL PUBLIC OFFERINGS IN CANADA: A TEST OF THE UNDERPRICING THEORIES AND AFTERMARKET PERFORMANCE 35 We study a sample of 199 Canadian IPOs for the period 1993-2001 and find that high underpricing in certain industries may explain the long-term underperformance. Also, issues offered during hot-issue periods have higher under-pricing and initial return, but lower long run returns. This result is consistent with the window of opportunity hypothesis. I. INTRODUCTION The under-pricing and the long-run performance of initial public offerings (IPOs) have been widely documented. In general, the literature confirms that IPO are under-priced thus causing abnormal positive initial returns (returns above those of the benchmarks) and that the long-run IPO returns fall behind the returns of the benchmark. Evidence also suggests that investors are over-optimistic about companies issuing equity, and this is true despite the form of issuance: IPO, private or seasoned placement. The reputation of the underwriter is also found to affect the under-pricing and the long run return of the IPO. In this paper, we examine the under-pricing and the aftermarket performance of Canadian Initial Public Offerings (IPOs) for the period January 1, 1993 to December 31, 2001. The study aims at finding a relationship between IPOs and their long run performance. The paper is organized as follows: Section II provides a literature review on under-pricing and long-run performance of IPOs in U.S and Canada. Section III discusses going public activities, in Canada. The types of equity issues, decisions to go public and its steps are explained. This section also provides a brief overview of the Toronto Stock Exchange and regulatory agencies. Section IV describes the data and the methodology used in this paper. Section V presents the empirical results and their analysis. We conclude the paper with a summary of the main findings. II- LITERATURE REVIEW A substantial body of work examines initial public offering of stocks, focusing on the process of going public, its benefits and costs, the initial and the long run returns of IPOs, and the variables that affect the returns of these IPOs such as the industry, ex ante uncertainty, proceeds, cold/hot issues, underwriter reputation, etc. In this section, we survey selected literature on U.S. and Canadian studies on 35 140
IPOs. A. The U.S. Evidence Ritter (1984) 36 explains why hot issue markets occur. He investigates the behavior of initial public offerings (1,028 issues) that occurred over the period 1977-1982. During the hot-issue market of 1980 (15 month period), He finds that the average initial return is 48.4%, whereas during the rest of the 1977-1982 period the initial return is 16.3%. Following Rock s (1982) model of underpricing of IPOs, he examines whether the behavior of the IPOs could be explained by the hot issue market hypothesis of 1980, with a positive equilibrium relation between risk and expected initial return and an increase in the riskiness of the average IPOs. Ritter (1984) finds almost no evidence of the occurrence of a hot issue market. The evidence doesn t support the equilibrium phenomenon hypothesis, but the monopoly power hypothesis; where the underwriters exploit startup natural resource firms during the boom of oil and gas in 1980. In another study, Ritter (1991) examines the long-run performance of 1,526 IPOs that went public in the 1975-1984 period. He finds that 3 years after going public the issuing firms significantly underperformed a set of matched firms. The evidence is consistent with the two explanations for under-pricing: (i) over-optimism or fads, when periodically investors become overoptimistic about young growth companies and (ii) the windows of opportunity, when companies take advantage of the over-optimism- theories. Shiller (1990) relates under-pricing to asymmetric information and not just compensation. He explains that underpricing results in high initial returns giving an impression that the underwriter is providing good advice; this new theory is called the Impresario hypothesis. According to Shiller, a hot market occurs when an underwriter notices that the public is prone to follow a fad in IPOs. The underwriter will underprice the issues and permit high initial returns for a certain period of time to create publicity for itself. Underwriter reputation also affects both the initial and the long run return of IPOs. Carter, Dark and Singh (1998) find that the underperformance of IPOs over a 3 year holding period is more severe for IPOs with less prestigious underwriters. They also find that IPOs with more reputable underwriters have less short-run under-pricing. B. The Canadian Evidence Jog and Srivastava (1997-98) study the long-run performance of Canadian IPOs. They find that two years after the issue Canadian IPOs lagged 23% behind the TSE, and four years after issue they lagged 40% behind the TSE. This is bad news for both investors and start-ups. Long-term investors will be discouraged and thus it will be harder for start-up companies to raise capital. Ursel (2000) studies a sample of 30 IPOs on the Toronto Stock Exchange for the period 1997-99. She finds that average under-pricing was 7.06%, an increase compared to previous years, but still less than that of the U.S. IPOs. As for the long-run returns, Ursel finds that on average, the Canadian IPO returns for this period falls behind the TSE by 3% only. Kooli and Suret (2001) test for long-run performance of Canadian IPOs for the 1991-1998 period. The authors find that the average initial returns on the first day is 20.57%, and that the investor who buys immediately after listing and hold the shares for 5 years will loose 24.66% on Equally Weighted basis relative to a similar investment in the control firms. 36 R. Jay. Ritter. The Hot Issue Market of 1980. The Journal of Business 57(2)(April 1984): 215-240. 141
A. Basic Procedures of New Issues in Canada III. Going Public Activities in Canada After taking the decision to go public, a firm issuing securities must follow a series of steps. In general, the basic procedure for new issues in Ontario (and the rest of Canada) is as follows: Step 1: Management should acquire the approval of the board of directors to issue any securities. If the number of authorized shares of common stocks needs to be increased, then the shareholders should vote for it. Step 2: Copies of a preliminary prospectus should be prepared and distributed to the Ontario Securities Commission (OSC) and prospective investors. This preliminary prospectus (also known as red herring, due to the bold red letters that are typed on its cover, indicating that the OSC has not approved nor disapproved of the securities) contains financial information which will also be incorporated in the final prospectus, but it doesn t contain the offer price of the new security. Within two weeks time, the OSC studies this prospectus and informs the issuing company of any required changes. Step 3: The revisions are made, and the final prospectus is sent to the OSC. When approved, a price is set and the selling efforts begin. Note that delivery of securities or confirmation of sale should accompany any final prospectus. Step 4: During and after the waiting period, underwriters make Tombstone advertisements for the new issue. B. The Prompt Offering Prospectus (POP) System The OSC introduced the POP system in 1983 to reduce repetitive filing requirements for companies. The five provincial securities commissions (Alberta, British Columbia, Manitoba, New Brunswick, and Nova Scotia) have compatible legislation. According to the POP system, certain securities issuers will be granted fast access to the securities market, since they will not be required to follow the basic steps of preparing preliminary and final prospectus before the distribution of the securities. Only large companies that have been following the requirements of the OSC for a long time (36 months) can benefit from the POP system. C. Multi-Jurisdiction Disclosure System (MJDS) The MJDS was introduced in the early 1990s, by the SEC in the United States and the securities regulators in Canada. The MJDS allows large companies in both U.S. and Canada to issue securities in the two countries following disclosure requirements that are satisfactory for the SEC and the Canadian securities regulators. This opened up new opportunities for issuing companies. D. The Toronto Stock Exchange (TSX) 142
The Toronto Stock Exchange Group is composed of the Toronto Stock Exchange 37 (TSX), which is Canada s largest exchange and the Toronto Venture Exchange (TSXV). The Toronto Stock Exchange Group ranks the third in North America and eighth in the world in terms of market capitalization. It is considered worldwide as one of North America s premier stock exchanges. There are many benefits of listing a company on the Toronto Stock Exchange. A TSX listing brings visibility, index eligibility, institutional investment, lower costs, a well-regulated and fully automated marketplace among other benefits. E. Regulatory Agencies All top Canadian companies are listed in the TSX although it has the strictest listing requirements in Canada. Currently, there is no federal regulatory agency to regulate the securities industry; thus, Canada relies on provincial securities commission for regulation; the Alberta Securities Commission, British Columbia Securities Commission, Manitoba Securities Commission, New Brunswick Securities Commission, and Nova Scotia Securities Commission. The fact that only five provinces have commissions is because of the absence of exchanges in some of the other provinces. The goal of the regulatory commissions is to promote efficiency of the information flow about the securities in the market and to enhance their functioning. In 1994 and by the Securities Amendment Act, the Ontario Securities Commission (OSC) was granted rulemaking powers and authorities similar to those that Securities Exchange Commission (SEC) has in the US. The OSC regulates the TSX; therefore, all the companies listed on the TSX fall under the jurisdiction of the OSC. Alberta Securities Commission and British Columbia Securities commission jointly oversee the regulation of the TSXV. Efforts have been put to address the issue of creating a national securities agency to regulate the whole securities industry. In this direction, the Minister of Finance of Canada established the Wise Persons Committee. The Minister assigned this committee the task of developing an independent assessment regarding the structure of the securities regulatory body that would best serve Canada. The Committee reported its assessment in December 2003; however, to this date, nothing has materialized. A. Data IV. Data and Methodology The data on Canadian IPO announcements (date, amount, issue price, proceeds ) was extracted from The Financial Post of New Issues. The daily prices of the stocks were extracted from CFMRC (Canadian Financial Markets Research Center) database. The original sample of IPO announcements covered the period January 1, 1993- March 31, 2006. The IPO announcements of last 2 years were removed from the sample, to match with the data on the daily closing prices available from CFMRC (up to December 31, 2004). For the 3-year long run returns, the sub-sample covered the Canadian IPO announcements from January 1, 1993 to December 31, 2001 38. 37 Toronto Stock Exchange; available from http://www.tsx.com Internet; accessed 19 April 2006. 38 The long run returns for stocks that were delisted before their 3 rd anniversary, was calculated by using the last none zero closing price before being delisted instead of the 3 rd year closing price. 143
Next, the sample was filtered further by the following constraint variables: Market= Canada, Security type general=equity, Security type specific=common, IPO=Yes, Status= completed, Issue Price 1 Canadian Dollar, Convertible= NO. The initial sample of IPOs was 2069. After filtering, we reached to a clean sample of 199 IPOs. The S&P/TSX Composite Index which includes 208 of Canada s largest public companies was chosen as the benchmark. The IPOs are categorized according to industry into the following eight categories: T&M (software and programming industries, computer and peripheral communications, equipment motion pictures and other media related industries), Mining, Financial (includes the Consumer Financial Services and insurance industries), O&G (i.e. Oil and Gas), Real Estate, Life Science (mainly the healthcare sector which includes biotechnology and drugs industries), Product (all types of manufacturing consumer and industrial and retail) and other (includes mainly transportation and forestry) 39. Table 1 presents the sample of Canadian IPOs categorized according to industry and type of underpricing. Table 1 Sample of Canadian IPOs Categorized According to Industry and Type of Underpricing Industry Number of IPOs T&M (Technology and Media) 64 O&G (Oil and Gas) 20 Products 54 Financial 17 Mining 10 Life Science 18 Real Estate 6 other 10 Total 199 Underwriting Type Firm Commitment 159 Best Efforts 33 Bought Deals 6 Unknown Type 1 Total 199 B. Methodology The main estimation model of the study is similar to that of Kooli and Suret (2001) and Ritter (1991): Long Run Return i = α o + α 1 Mining i + α 2 Life Science i + α 3 Tech. i + α 4 Und. i + α 5 Hot/Cold i + ε I (1) 39 The industry categorization in this paper is similar to that of the Survey of Public Offerings in Canada, conducted by PriceWaterhouseCoopers; available from www.pwc.com. 144
The dependent variable (Long Run Return i ), is taken to be the 3 year return on each stock 40. It is calculated in two ways: (i) (3 rd year closing price 1 st day closing price)/ (1 st day closing price) (referred as Longrunreturn1) (ii) Log (3 rd year closing price) - Log (1 st day closing price) (referred as Longrunreturn2) Under-pricing is an independent variable of this model, Und. which is also calculated in two ways similar to the long run returns. A. Descriptive Statistics V. Empirical Results Table 2 provides a summary of the descriptive statistics on issue price, gross proceeds, and under-pricing of 199 Canadian IPOs. The mean offer price of the sample IPOs is Can$8.26 per share, with a standard deviation of 4.32 %. The skewness (1.05) is positive, indicating that the distribution is skewed to the left and has a long right tail. The kurtosis (6) indicates that the distribution is leptokurtic relative to the normal distribution. The Jarque-Bera (111.32) is very large and the probability is very small (0.00%) implying that the series is not normally distributed. Table 2 Descriptive Statistics: Issue Price, Gross Proceeds and Underpricing of 199 Canadian IPOs (1993-2001) Issue Price Gross Proceeds Underpricing (1) 41 Underpricing (2) 42 Mean $ 8.258920 $ 39388268 5.2842 % 1.9095 % Median $ 8.000000 $ 24999997 0.5982 % 0.2590 % Maximum $ 29.00000 $ 80.0000 % 25.5273 % 486000000 Minimum $ 1.000000 $ 1300000-45.0000 % -26.0356 % Std. Dev. 4.317809 58.2944 % 15.0356 % 5.6243 % % Skewness 1.049427 4.900931 1.950744 0.913652 Kurtosis 6.003362 33.10431 9.875876 8.954532 Jarque-Bera 111.3187 8311.121 515.6191 320.0628 Probability 0.000000 0.000000 0.000000 0.000000 The average gross proceeds of the sample IPOs is $39,388,268, with a standard deviation of 58.29 %. The skewness (4.9) and kurtosis (33) also indicate that the series is not normally distributed. 40 24 out of the 199 IPOs got delisted before their 3 year anniversaries. For these stocks the last closing price before getting delisted was used instead of the 3 rd year closing price. 41 Measured by 1 st day closing price minus issue price. 42 Measured by the log of the 1 st day closing price minus the log of the issue price 145
Table 2 shows that the mean under-pricing is 5.28%, with a standard deviation of 15.04%. The average under-pricing in other Canadian studies ranges from a minimum of 3.6% to a maximum of 20.57% (see Table 3). The mean of under-pricing, as measured by the log returns is 1.91%, with a standard deviation of 5.62%. Table 3 Historical Under-pricing in Canadian IPOs Study Period Studied Average underpricing (per cent) Krinsky and Rotenberg (1989) 1971-83 11.6 Jog and Riding (1989) 1971-83 9.96 Jog and Srivastava (1994) 1971-92 7.87 Jog (1997) 1971-94 8.43 Clarkson and Merkley (1994) 1984-87 6.44 Kryzanowski and Rakita (1996) 1984-93 4.18 Ursel and Ljucovic (1998) 1987-94 3.64 Kooli and Suret (2001) 1991-98 20.57 Ursel (2000) 1997-99 7.06 Current Paper (2007) 1993-2001 5.28 Table 4 summarizes the descriptive statistics for long run returns. The average long run return is 2.24%, with a standard deviation of 102.85%. The average long run return measured as the log of 3 rd year closing price minus log of 1 st day closing price, is -19.81% with a standard deviation of 52.38%. We notice that the long run returns of IPOs in our sample are not normally distributed. Table 4 Long Run Returns of Canadian IPOs (1993-2001) Long Run Return 1 43 Long Run Return 2 44 Mean 2.2424% -19.8127% Median -26.137 % -11.3987% Maximum 525% 108.6360% Minimum -99.7315% -257.1126% Std. Dev. 1.028491 0.523756 Skewness 2.284120-0.979902 Kurtosis 9.786432 4.906850 Jarque-Bera 552.1270 61.68446 Probability 0.000000 0.000000 Table 5 summarizes the initial and long run returns of our sample of 199 IPOs according to 43 As measured by the conventional way of 3 rd year closing price minus the 1 st day closing price divided by the 1 st day closing price. 44 As measured by Log of 3 rd year closing price minus Log of 1 st day closing price. 146
different industries. Oil & Gas has the highest initial return (i.e. under-pricing) followed by Technology and Media (T and M).On the other hand, the products industry has the highest long run return. Table 5 Average Initial and Long Run Returns of Canadian IPOs (1993-2001) According to Industry T&M Financial O&G Mining Life Science Real Estate Products Sample Size N=64 N=17 N=20 N=10 N=18 N=6 N=54 IR1 1.3344% 0.7762% 2.4396% -0.0020% 0.0265% 0.0002% 0.6821% IR2 0.4770% 3.2510% 0.8874% -0.0117% -0.0080% -0.0003% 0.2729% LR1-0.3715% 1.4286% 2.0607% -1.8537% -1.1091% 0.2546% 4.9074% LR2-10.0442% 1.2458% -0.5522% -2.3770% -3.2953% 0.0096% 2.4950% B. Regression Analysis In the first stage of testing, we regress the long run return with independent variables such as industry dummies, a dummy for under-pricing, and a dummy for hot/cold issue. Table 6 presents the results of the regression. The coefficient of the under-pricing is negative and insignificant. The dummy for the mining IPO and the dummy for the technology IPO are negative and significant at the 1% level. The life science dummy is also negative and significant at the 5% level. The coefficient of the under-pricing is negative but insignificant, implying an inverse relationship between under-pricing and the long run performance. The dummy for the hot and cot issue period is significant at the 10% level. The result indicates that when we have a hot issue period this causes a decrease in the long run return. 147 Table 6 Summary of the Results for the 3 year Long Run Return Dependent Independent Variables Variable Constant Mining Life Science T&M UND Hot/Cold R2 adj % N LR-Return (3 years) -0.0424-0.4296-0.3110-0.2426-0.5820-0.1727 5.77 199 (-0.7881) (-2.5312)** (-2.3748)** (-2.9895)** (-0.8888) (-1.4317) - -0.4724-0.3522-0.2795-0.7553-0.1958 5.95 199 - (-2.9410)*** (-2.9377)** (-4.2192)*** (-1.2260) (-1.6755)* *** significant at 1% ** significant at 5% * significant at 10% To further explore the relation between underpricing and other relevant factors, we run another regression by having the first day underpricing as the dependant variable. The variation in under-pricing is explained by three independent variables: the issue price, a dummy for Financial Industry and a dummy for Oil&Gas Industry. The following equation is estimated: UND2=C(1)+C(2)*ISSUEPRICE+C(3)*FDUM+C(4)*ODUM (2) The constant is positive and significant at the 5% significance level. The coefficient for the issueprice is negative and significant at the 10% level. The coefficient of the dummy for Financial
Services industry is negative and significant at the 5% level. Also, the coefficient of the dummy for the Oil&Gas industry is positive and significant at the 5% level. Table 7 Regression Results for Under-pricing as the Dependant Variable Dependent Independent Variables Variable Constant Issue Price Financial Oil&Gas R2 adj % N Underpricing 0.0209-0.0015-0.0333 0.0752 19.22 199 (2.5481)** (-1.7266)* (2.5108)** (6.1546)*** *** significant at 1% ** significant at 5% * significant at 10% VI. Summary and Conclusion We study a sample of 199 Canadian IPOs for the period 1993-2001 and find that high underpricing in certain industries may explain the long-term underperformance. Also, issues offered during hot-issue periods have higher under-pricing and initial return, but lower long run returns. This confirms the window of opportunity hypothesis. 148 References Canadian IPO announcements, The Financial Post of New Issues. Canadian Securities Administrators; available from http://www.csa-acvm.ca; Internet; accessed 23 April 2006. Canadian Stock Exchange; available from http://en.wikipedia.org; Internet; accessed 26 April 2006. Carter, R., Dark, F. and A. Singh, Underwriter Reputation, Initial returns and the Long-Run Underperformance of IPO Stocks, The Journal of Finance, 53(1), (February 1998), 285-311. Daily Closing Stock Prices, CFMRC.
Draho, J. The IPO Decision, Edward Elgar. Fehmi, D. Initial Public Offerings with Reference to Emerging Arab Markets. A Project to the Economics Department, American University of Beirut, March 2006. Hertzel, M., Lemmon, M., Linck, J. and L. Rees, Long-Run Performance following Private Placements of Equity, The Journal of Finance, 57(6), (December 2002), 2595-2617. Kooli, M. and J. Suret, The Aftermarket Performance of Initial Public Offerings in Canada, Scientific Series, (September 2001), 1-34. Rahman, H. and K. Yung. Insurance IPOs- A Test of the Underpricing Theories. Journal of Insurance Issues 22(1)(1999): 61-77. Ritter, J., The Long Run Performance of Initial Public Offerings, The Journal of Finance, 46(1), (March 1991), 3-27. Ritter, R. Jay., The Hot Issue Market of 1980, The Journal of Business, 57(2), (April 1984), 215-240. Ross, S., Westerfield, R. and Jaffe, J. Corporate Finance (Canadian Edition), Sixth Edition, New York: McGraw-Hill Irwin. Sadek, N.G., "'Issuing securities to the public'. Initial Public Offerings and Seasonal Equity Offerings with Reference to Lebanon and the Arab Markets", A Project to the Money and Banking Department, American University of Beirut, February 2004. Shaw, D. The Performance of Primary Common Stock Offerings: A Canadian Comparison, The Journal of Finance, 26(5), (December 1971), 1101-1113. Shiller, R. Speculative Prices and Popular Models Journal of Economic Perspectives 4 (Spring 1990): 55-65 Survey of Public Offerings in Canada, PriceWaterhouseCoopers; available from http://www.pwc.com; Internet; accessed 25 April 2006. Toronto Stock Exchange; available from http://www.tsx.com; Internet; accessed 19 April 2006. Ursel, N., Priced to Sell the Evolution of Underpricing in Canadian Initial Public Offerings, Canadian Business Economics, (February 2000), 15-20. Vijay, J. and A. Srivastava, The Mixed Results of Canadian IPOs, Canadian Investment Review, (Winter 1997/98), 23-27. 149