Investment bank reputation and relaxed listing requirements: Evidence from infrastructure firm IPOs in Hong Kong

Pacific-Basin Finance Journal 9 Ž2001. 101–117 www.elsevier.comrlocatereconbase

Investment bank reputation and relaxed listing requirements: Evidence from infrastructure firm IPOs in Hong Kong Kathryn L. Dewenter a,) , Laura Casares Field b a

Department of Finance, UniÕersity of Washington, Box 353200, Seattle, WA 98125, USA b Penn State UniÕersity, UniÕersity Park, PA, USA

Abstract In early 1996, the Stock Exchange of Hong Kong allowed firms focusing on infrastructure projects to issue initial public offerings ŽIPOs. under a relaxed set of listing requirements, allowing these firms to go public with a shorter history or lower profitability levels. We provide evidence that these firms are no more speculative than firms listing under the regular requirements. To the contrary, we find that firms listed under the relaxed requirements are taken public by reputable investment banks and that these firms have characteristics that otherwise mitigate their lack of earnings history. These patterns are consistent with investment banks avoiding highly speculative issues to protect their reputations. q 2001 Elsevier Science B.V. All rights reserved. JEL classification: G15; G24 Keywords: Initial public offerings ŽIPOs.; Hong Kong; Infrastructure finance; Investment bank; Relaxed listing requirements

1. Motivation The need for funds to develop infrastructure in emerging markets is huge and growing. For example, the World Bank estimates that Latin America and the Caribbean currently need to invest 4.5% of GDP each year on power, transport, )

Corresponding author. Tel.: q1-206-685-7893; fax: q1-206-685-9392. E-mail address: [email protected] ŽK.L. Dewenter..

0927-538Xr01r$ - see front matter q 2001 Elsevier Science B.V. All rights reserved. PII: S 0 9 2 7 - 5 3 8 X Ž 0 1 . 0 0 0 0 6 - 3

102

K.L. Dewenter, L.C. Field r Pacific-Basin Finance Journal 9 (2001) 101–117

telecoms, sewage, and water supply upgrades.1 These funding requirements are far beyond what multilateral agencies and federal governments can provide. Therefore, emerging market countries have had to turn to private sources to help finance these projects. To date, private equity funds have been the primary vehicle for infrastructure investments. Market participants, though, argue that public funds, via stock market listings, are needed to both expand the pool of available money and to provide an exit strategy for the initial private investors. Several countries, recognizing the need for public funds and the short lives of most infrastructure firms, have developed relaxed listing requirements for infrastructure firm initial public offerings ŽIPOs.. Hong Kong, Malaysia, and Singapore, for example, all announced plans to establish relaxed listing requirements for infrastructure firms in late 1995.2 By relaxing the requirements on trading record period and profit levels, Hong Kong’s relaxed listing requirements, called sub-rule 8.05Ž2., allow infrastructure project companies to list on the Stock Exchange of Hong Kong ŽSEHK. at an earlier stage of their lives. Before the stock exchange implemented the relaxed rules, all firms had to meet the basic, or regular, listing criteria. Among these rules, the SEHK requires that, AA new applicant must normally have a trading record of not less than 3 years under substantially the same management. The profit attributable to shareholders must, for the most recent year, be not less than HK$20 million and, for the 2 preceding years, be in aggregate not less than HK$30 million.B 3 Under the relaxed rules for infrastructure project companies, firms are allowed to issue an IPO even if they have fewer than 3 years of audited financial statements andror lower profit levels than the formal thresholds. Since implementation of the special rules for infrastructure project companies, infrastructure firms can choose whether to issue stock on the exchange early in their lives by listing under the relaxed requirements, or, alternatively, to wait until they are more seasoned and list under the regular requirements. One might ask how investors react to stock issues by firms under the relaxed rules. Are the relaxed listing firms considered more risky or speculative than comparable regular listing firms? Or, do the investment banks, concerned about their reputations, only sponsor those firms that somehow compensate for the lack of a comparable earnings history? This paper provides evidence on Hong Kong’s experience to date with relaxed listing requirements for infrastructure firm IPOs. We study infrastructure listings 1

Economist, APrivacy please: Infrastructure in Latin AmericaB ŽJuly 27, 1996. 54–55. In the late 1990s, listing requirements in several exchanges were also lowered for internet firms Žfor example, Mothers in Japan.. 3 ATrading recordB refers to years of operation. This quote is from the SEHK website,www.sehk.com.hk. Throughout the 1990s, the Hong Kong dollar was pegged at HK$7.8 per $US1. 2

K.L. Dewenter, L.C. Field r Pacific-Basin Finance Journal 9 (2001) 101–117

103

made on the Hong Kong exchange during 1996 and the first half of 1997 to examine whether there are any noticeable differences in those firms issuing IPOs under the regular listing requirements vs. those issuing IPOs under the relaxed listing requirements.4 We look at the characteristics of the offerings to see if there is any evidence that the IPOs issued under the relaxed requirements are more speculative than the regular requirement listings. We examine how the relaxed offerings were received by investors compared to infrastructure offerings issued under the regular guidelines, and at the post-IPO stock-price reaction to accounting performance announcements. Since our post-IPO period includes the Asian financial crisis, we are also able to compare how the two groups fared during extraordinarily poor economic times. Finally, we look to see whether the relaxed listing firms have characteristics that otherwise mitigate their lack of earnings history. We begin with the population of Ži. all infrastructure firms issuing IPOs on the Hong Kong Stock Exchange under sub-rule 8.05Ž2. from January 1996 through 1997, and Žii. all infrastructure firms issuing under the regular listing requirements over the same period. During this time, there were four relaxed and 13 regular listing infrastructure firms. Three of the relaxed listing firms focused exclusively on road projects in China, one focused on port development. The regular listing firms focused on railroads Ž1., power stations Ž1., telecoms or satellites Ž3., roads in China Ž4., or were diversified across infrastructure projects Ž4.. Since we knew we were going to have a small sample, no matter what criteria we used, we decided to construct the cleanest possible set of matching firms. As a result, we restrict our analysis to the seven firms that focused exclusively on road projects in China. Three of the relaxed listing firms fit this criterion. One firm, Road King ŽIPO issue date: July 4, 1996., was forced to use the relaxed requirements because it had earned negative net profits within the previous 3 years, and two firms, GZI Transport ŽJanuary 30, 1997. and Jiangsu Expressway ŽJune 27, 1997., had been in business less than 3 years. The four regular listing firms are: Anhui Expressway ŽNovember 13, 1996., Shenzhen Expressway ŽMarch 12, 1997., Zhejiang Expressway ŽMay 15, 1997., and Sichuan Expressway ŽOctober 7, 1997.. A more complete description of all seven firms is provided in Appendix A. While this sample is small, it does provide two sets of firms with very similar profiles. It allows us an early comparison of the listing and subsequent stock price performance of the relaxed vs. regular requirement firms. We find no evidence of any striking differences consistent with the relaxed listing firms being relatively more risky or speculative than the regular listing firms. Instead, we find evidence that the relaxed listing firms were, in some respects, actually less speculative than

4

Given the recent market turmoil in Asia, no firms issued IPOs under sub-rule 8.05 from June 30, 1997 through June 30, 1999.

104

K.L. Dewenter, L.C. Field r Pacific-Basin Finance Journal 9 (2001) 101–117

the regular listing firms. Indeed, well-known, reputable investment banks underwrote the relaxed as well as the regular listing firms. Moreover, prior to the IPO issue, the relaxed listing firms were larger or more profitable than the regular listing firms despite the fact that they did not meet the regular listing requirements. All of this evidence is consistent with investment bankers avoiding highly speculative issues to protect their reputations. Below, Section 2 discusses the empirical evidence. In the conclusion, Section 3, we discuss whether or not the findings of this small sample can be generalized.

2. Empirical evidence This section compares the relaxed and regular listing firms along several different dimensions. Unless otherwise noted, the data on the issues come from the prospectuses or press reports, while the stock price data come from Datastream International. 2.1. IPO structure Table 1 provides descriptive data on the IPOs for the three firms using the relaxed rules as well as the four firms listing under the regular rules. The bottom row of the table provides the difference in the averages for the relaxed vs. regular firms. The results of a non-parametric Wilcoxon rank test for differences in the averages are reported in the table. The difference is never statistically significant using a t-test with pooled variances Žresults not reported in the table..5 Column 1 provides the first trading date. These IPOs occurred between July 1996 and October 1997. Column 2 provides the offer price. The two groups seem to have substantially different offer prices with an average of HK$4.91 for firms under the relaxed rules and HK$1.98 for firms under the regular rules. Indeed, the offer price ranges from HK$3.11 to HK$8.40 for firms under the relaxed rules, while it is only HK$1.55– 2.38 for the regular listing firms. Over 1996–1997, the mean offer price across all SEHK IPOs was HK$3.06, with a range of HK$1.00–19.80.6 A Wilcoxon rank test rejects that the two sets of prices for these infrastructure firm IPOs come from 5

Note both the t-test and the rank sum test assume independence. According to Angel Ž1997., the median US stock price is about US$40.00, while a typical Hong Kong share sells for about US$2.00. Angel argues that differences in average stock prices across countries are at least partly due to different rules on tick sizes. While the US uses a single absolute tick size that applies to most stocks, AHong Kong has the most extreme version of a step function, with 10 different tick sizes in its rule book.B Žp. 658.. 6

First trading date

Offer price ŽHK$.

Funds raised ŽHK$ million.

Fraction sold Ž%.

International tranche Ž%.

Subscription rate Žtimes.

1

2

3

4

5

6

Relaxed rules firms Road King GZI Transport Jiangsu Expresswayy Average

7r4r1996 1r30r1997 6r27r1997

8.40 3.23 3.11 4.91

1180.5 928.6 3800.4 1969.8

26.0 25.0 25.0 25.3

85.0 75.0 61.0 73.6

2.1 528.4 2.4 177.63

Regular rules firms Anhui Expressway Shenzhen Expressway Zhejiang Expressway Sichuan Expressway Average

11r13r1996 3r12r1997 5r15r1997 10r7r1997

1.77 2.20 2.38 1.55 1.98

872.6 1644.5 3412.5 1387.7 1829.3

35.0 33.9 30.0 35.0 33.5

85.0 90.0 90.0 65.0 82.5

1.0 36.6 118.0 1.7 39.33

2.94 a

140.5

y 8.2 a

Difference (relaxedIregular)

y 8.9

138.3

The sample consists of all infrastructure finance firms focused exclusively on road projects in China which issued an initial public offering ŽIPO. of equity on the Hong Kong Stock Exchange during the period of July 1996 through October 1997. Column 1 provides the offering dates. Column 2 provides the offering prices, in Hong Kong dollars. Column 3 provides the Hong Kong dollars raised in the offering, in millions. Column 4 provides the fraction sold in the offering, measured as the ratio of shares sold to total shares outstanding after the IPO. Column 5 provides the fraction of the offering sold abroad. Column 6 provides the subscription rate, which is measured as the number of shares requested from underwriters divided by the number of shares offered. During 1996–1997, the US DollarrHK Dollar exchange rate was approximately 1:7.8. a Indicates significant difference in a non-parametric Wilcoxon rank test at the 5% level.

K.L. Dewenter, L.C. Field r Pacific-Basin Finance Journal 9 (2001) 101–117

Table 1 Background on infrastructure firm IPO issues

105

106

K.L. Dewenter, L.C. Field r Pacific-Basin Finance Journal 9 (2001) 101–117

the same sample at the 5% level. Thus, we find that the relaxed firms have relatively higher offer prices than the regular firms. In the US, Seguin and Smoller Ž1997. show that more speculative IPO offers tend to have lower offer prices. Despite the apparent difference in offer price, the averages of the funds raised Žcolumn 3. across the two groups are substantially similar: HK$1970 million for firms listing under the relaxed rules, compared to HK$1829 million for firms listing under the regular rules. Looking at the firms individually, there appear to be no major differences in the funds raised across the two groups. Firms listing under the relaxed rules sold an average of 25% of the firm in the IPO, compared with 33% for firms listing under the regular requirements Žcolumn 4.. In fact, all three firms listing under the relaxed requirements offered between 25% and 26%, while all four firms listing under the regular rules issued at least 30%, with two offering 35% of the firm to the public. A Wilcoxon rank test rejects that these two sets of numbers come from the same sample at the 5% level, providing evidence that the relaxed firms offered a smaller portion of their firms than did the regular listing firms. In the signaling model of Leland and Pyle Ž1977., the fraction of equity retained by an entrepreneur upon taking his firm public signals firm value. A lower retained portion signals lower quality. Leland and Pyle Ž1977. would thus interpret the pattern in column 5 as evidence of relatively higher firm value for the relaxed listing firms.7 The fraction of the offer sold outside of Hong Kong, in column 5, seems to be fairly similar across the two groups, with an average 74% for firms listing under the relaxed rules compared to 83% for firms listing under the regular rules. The range for firms under the relaxed rules is 61–85%, while for firms under the regular rules the range is 65–90%. In other words, all of these IPOs have a large international tranche. The final column in Table 1 provides the subscription rate. GZI Transport Žlisted under the relaxed rules. was severely oversubscribed with a subscription rate of 528 times, as were Shenzhen Expressway and Zhejiang Expressway Žboth listed under the regular rules. at 37 times and 118 times, respectively. ŽOver 1996–1997, the average subscription rate across all SEHK IPOs was 102, with a range of 0.66–1276.. The average subscription rate for firms listing under the

7

In a generalization of Leland and Pyle Ž1977., Grinblatt and Hwang Ž1989. provide a model in which both the mean and variance of future cash flows are unknown to investors. In contrast to Leland and Pyle, in the Grinblatt and Hwang model the issuer’s fractional holding alone is insufficient to signal firm value. Grinblatt and Hwang argue that two signals are required: Ži. the fraction of shares retained by the entrepreneur, and Žii. the degree of underpricing. Thus, our finding of a significantly larger proportion retained for relaxed rules firms is consistent with these firms being of higher quality, as in the Grinblatt and Hwang model. However, as shown below in column 1 of Table 2, we do not find a significant difference in underpricing for firms listing under the relaxed vs. regular rules.

K.L. Dewenter, L.C. Field r Pacific-Basin Finance Journal 9 (2001) 101–117

107

relaxed rules is 178 times, far more than the subscription rate of 39 times for firms listing under the regular rules. The difference, however, is driven exclusively by GZI Transport’s huge subscription rate and is not significant. In sum, Table 1 does not provide much evidence of differences in the offerings of the relaxed listings vs. the regular listings. The offer characteristics are quite similar across the two sets of firms. The only evidence we do have of differences, that the relaxed firms had higher offer prices and offered a smaller portion of their firms, is inconsistent with these firms being more speculative than firms issuing under the regular requirements. 2.2. Short-run stock price moÕements Table 2 provides data on short-run stock price movements after the IPO. All numbers are expressed as percentages. Column 1 provides the 1-day initial return after the IPO issue. The initial return is defined as the percentage change from the fixed IPO offer price to the closing price on the first day of trading. Column 1 in Table 2 shows that on average, firms issuing under the relaxed requirements earned a 10.39% initial return, but only one firm, GZI Transport, earned a positive initial return Ž51%.. The other two firms listing under the relaxed rules earned negative initial returns, with Road King earning y8.33% and Jiangsu Expressway earning y11.57%. On average, firms listing under the regular rules earned a very small initial return of only 0.37%, but again there was considerable variation across firms. Two firms earned positive initial returns, Shenzhen Expressway with 25.90% and Zhejiang Expressway with 5.04%, while two earned negative initial returns, Anhui Expressway with y10.73% and Sichuan Expressway with y18.71%. The difference between the two sets of returns is not significant with either a t-test or a Wilcoxon rank test. Overall, it appears we cannot infer that the initial returns differ across the two groups Columns 2 and 3 of Table 2 provide the standard deviation of stock returns for the first 20 days of trading after the firm’s IPO, and for the first common 6-month period for all seven firms ŽOctober 1997–April 1998.. Ritter Ž1984. uses the standard deviation of stock returns in the immediate post-offer period to proxy for ex-ante uncertainty. He finds that this variable is positively related to initial returns. If there is more ex-ante uncertainty about the relaxed listings, one might expect these listings to have a larger standard deviation of returns. In column 2, the mean standard deviation for the first 20 days is 2.90% for the relaxed listing firms and 5.40% for the regular listing firms.8 The relaxed firms

8

For comparison, the range of standard deviations for the middle one-third of the IPOs in Ritter Ž1984. is 3.3–5.7%.

108

Standard deviation of returns Ž% .

Asia financial crisis event days: firm’s 1-day stock return Ž% .

Initial return First 20 days October 1997 – April 1998 8r15r1997 9r4r1997 10r20r1997 10r21r1997 10r22r1997 10r23r1997 10r24r1997 1 Relaxed rules firms Road King GZI Transport Jiangsu Expressway Average Regular rules firms Anhui Expressway Shenzhen Expressway Zhejiang Expressway Sichuan Expressway Average Difference (relaxed I regular )

2

3

4

5

6

7

8

9

y8.33 51.08 y11.57 10.39

2.11 2.76 3.81 2.90

y10.73 25.90 5.04 y18.71 0.37

2.66 3.97 3.78 11.17 5.40

10.02

y2.50

4.35 5.17 5.02 4.85

y1.86 y1.48 y6.34 y3.23

y8.89 y1.55 y4.79 y5.08

y0.74 y2.56 y8.09 y3.80

y4.98 y5.90 y4.95 y5.28

y3.40 y7.14 y14.76 y8.44

y20.60 y24.39 y5.60 y16.86

y2.73 10.92 4.68 4.29

5.53 6.07 6.43 6.06 6.02

y5.72 y3.84 y6.19 NA y5.25

y6.09 y7.60 y3.16 NA y5.61

y3.80 y6.64 y5.36 y10.22 y6.50

y1.60 y7.85 y5.03 y11.38 y6.47

y2.38 y18.02 y11.92 y3.99 y9.08

y13.99 y12.08 y13.53 y18.57 y14.54

4.78 7.97 13.04 16.43 10.55

2.02

0.53

2.70

1.19

0.64

y2.32

y6.26

y1.18 a

10

The sample consists of all infrastructure finance firms focused exclusively on road projects in China which issued an initial public offering ŽIPO . of equity on the Hong Kong Stock Exchange during the period of July 1996 through October 1997. Column 1 provides the initial return, which is measured from the offering price to the first day of trading. Column 2 provides the standard deviation of returns measured over the first 20 days of trading, while column 3 provides the standard deviation of returns for the first common 6-month period for the entire sample. Columns 4–10 provide the 1-day stock returns Ž=100 . for the infrastructure firms over various days during the Asia Financial Crisis. 8r15r1997 Žthe date over which returns are measured for Column 4 . and 10r20r1997 – 10r24r1997 ŽColumns 6 – 10 . were days of intense speculation against the Hong Kong dollar and stock market. 9r4r1997 Žthe date over which returns were measured in Column 5 . was the date that the Malaysian government announced the cancellation of several high profile infrastructure projects. All numbers are in percentages. a Indicates significant difference in a non-parametric Wilcoxon rank test at the 5% level.

K.L. Dewenter, L.C. Field r Pacific-Basin Finance Journal 9 (2001) 101–117

Table 2 Short-term stock price movements

K.L. Dewenter, L.C. Field r Pacific-Basin Finance Journal 9 (2001) 101–117

109

appear to have lower variability, although a t-test and a Wilcoxon rank test do not reject equality of these two sets of numbers. The highest variability in column 2 is recorded for Sichuan Expressway Ž11.17%., a firm listing under the regular requirements. The 20-day period covered in this calculation includes a week of great market turmoil on the whole SEHK. ŽSee the discussion of the Asia financial crisis below.. This relatively high number might bias the inferences. As a result, we also calculate the standard deviation over a longer and common period for all seven firms. The mean standard deviation for the relaxed firms over the 6-month period of October 1997 through April 1998 is 4.85% and for the regular firms is 6.02%. A Wilcoxon rank test rejects that the two sets of numbers come from the same distribution at the 5% level. This finding, that the variability of returns is relatively lower for the relaxed listing firms, suggests that the relaxed listing firms are no more speculative than firms listing under the regular requirements. A third dimension of short-run stock returns that we examine is how these IPO offers behaved during the Asian financial crisis. All seven of the offers occurred shortly before or during the Asian financial crisis that began when Thailand devalued its currency, the Baht, in early July 1997. Even though the primary victims were in Thailand, Indonesia, and Korea, the crisis had relevance for the seven Hong Kong infrastructure firms we are looking at. First, a common factor contributing to the problems was bloated federal budget deficits. Beginning with Malaysia in September 1997, several of the countries announced drastic budget cutbacks, including the cancellation of many high profile infrastructure projects. This news had negative implications for all infrastructure firms in the region. Second, the Hong Kong dollar and the Stock Exchange of Hong Kong came under intense speculative pressure as the resolve of China to defend Hong Kong’s currency was tested. If the relaxed listing firms were considered more risky than the regular listing firms, then we might expect relatively larger stock price reactions for these firms on the key Asia crisis event dates. In Table 2, columns 4–10, we provide the 1-day stock return for the seven infrastructure firms for the date when Malaysia announced the cancellation of several infrastructure projects ŽSeptember 4, 1997., and for two periods of intense speculation against the Hong Kong dollar ŽAugust 15, 1997 and October 20–24, 1997.. These dates do appear to be significant events for these firms. For example, the mean drop in firm value on October 23, 1997 was 16.86% for the relaxed firms and 14.54% for the regular firms. However, there is no striking difference in returns across the relaxed vs. regular listing firms. A t-test and a Wilcoxon rank test fail to reject equality across the two groups for all of these event dates. A fourth and final dimension of short-run stock price movements that we examine is reaction to earnings announcements. Pownall and Waymire Ž1989, p. 103. argue that managers are more likely to release earnings forecasts when substitute sources of information are not available. This argument suggests that the relaxed listing firms might release forecasts more frequently and that these

110

K.L. Dewenter, L.C. Field r Pacific-Basin Finance Journal 9 (2001) 101–117

announcements may be more informative due to the lack of alternative, or historical, information. We conjecture that firms issuing under the relaxed requirements may try to compensate for the lack of knowledge or history by making more frequent, voluntary interim earnings announcements. In addition, the stock price reaction for any given announcement would be larger for firms issuing under the relaxed rules Ži.e., each announcement would contain more information.. From IPO issue to mid-1999, six of the seven firms had biannual earnings announcements reported in the South China Morning Post corresponding to the publishing of mid-year and year-end results.9 The exception, Sichuan Expressway, a regular listing firm, has only issued annual announcements. Across the relaxed listing firms there were three press reports of interim, voluntary earnings announcements. Across the regular listing firms, there were four reports of interim, voluntary earnings announcements. Thus, there is little evidence that one set of firms provided more frequent earnings announcements than the other. Across all seven firms, we were able to calculate stock price reactions for 16 annual earnings announcements that also provided the percentage change in earnings. These calculations are 3-day market adjusted buy and hold returns around the date the announcement appeared in the South China Morning Post. We use Datastream’s Total Market index for Hong Kong to calculate market returns. The mean return for the six announcements made by the relaxed listing firms is 1.86%, with a mean change in reported earnings of 14.56 times. This includes an observation for Jiangsu Expressway which reported an earnings jump of 83 times in 1997. Without this observation, the mean change in reported earnings for the relaxed firms is 88.03%. The mean return for the 10 announcements made by the regular listing firms is 0.122% with a mean change in reported earnings of 253.0%. To test whether the stock price reactions were larger for the relaxed listing firms, controlling for the change in earnings, we ran the following OLS regression: Return i s b 0 q b 1 Earnings i q b 2 Relaxed i q ´ i where Return i s the 3-day market adjusted buy and hold return for firm i, Earnings i equals the reported percentage change in earnings for firm i, and Relaxed i is a binary variable set equal to 1 for the six announcements made by the relaxed listing firms and zero otherwise. For the full sample of 16 observations, or for a sample of 15 that excludes the Jiangsu observation, the beta coefficient estimates for Earnings or for Relaxed are never significantly different from zero. Thus, there is no evidence that stock price reactions to earnings announcements by the relaxed listing firms are larger, a finding that would have been consistent with

9

One of these firms, Jiangsu Expressway has no announcements after April 1998 Ži.e., it is missing a mid-year and an annual announcement..

K.L. Dewenter, L.C. Field r Pacific-Basin Finance Journal 9 (2001) 101–117

111

relatively more information in those announcements. The lack of earnings history for the relaxed rules IPOs does not appear to affect the informativeness of their later earnings announcements. 2.3. Long-run returns Although no cohesive theory on long-run performance of IPOs exists, there is some evidence that IPOs issued by more speculative firms perform especially poorly in the long run. Ritter Ž1991., Loughran and Ritter Ž1995. and Brav and Gompers Ž1998. find that smaller issues perform especially poorly in the long run. Michaely and Shaw Ž1994. find that IPOs managed by high prestige investment bankers have less negative returns over a 2-year period, while Carter et al. Ž1998. find similar results over a 3-year period. Ritter Ž1991. and Loughran and Ritter Ž1995. argue that firms going public take advantage of Awindows of opportunityB by going public when investors are Aoverly optimistic.B Firms issuing under the relaxed requirements may do so because they are anxious to go to the market immediately, rather than wait until they are seasoned enough to list under the regular requirements. Thus, firms listing under the relaxed requirements may be those most likely to be taking advantage of Awindows of opportunity,B rather than just waiting a year or two until they meet the requirements for normal listing. In this case, one might expect firms listing under the relaxed requirements to experience poorer performance in the longer term. Fig. 1 provides the stock performance over a longer time period, from October 7, 1997 through June 30, 2000. October 7, 1997 is the first date that all seven firms are listed. As can be seen from Fig. 1, there is no discernible difference between the three relaxed listing IPOs and an equally weighted index of the four IPOs under the normal requirements. Had one invested HK$100 in each on October 7, 1997, one would have HK$68.27 worth of stock in Road King, HK$57.38 in Jiangsu, and HK$39.58 in GZI. Alternatively, an equally weighted index composed of the regular listing IPOs would have been worth HK$56.81 on June 30, 2000. Overall, the longer-run returns do not seem to be different between the regular listing IPOs and IPOs listing under the relaxed requirements. 2.4. Compensating characteristics The data above provide no evidence supporting the contention that the relaxed listing firms were more speculative than the firms that listed under the regular requirements. The two sets of firms are surprisingly similar along almost all dimensions. Moreover, where there appear to be differences, they indicate lower uncertainty or variability for the relaxed listing firms. These results suggest that the relaxed listing firms may have been taken public by reputable investment banks who only sponsored relaxed listing firms that

112

K.L. Dewenter, L.C. Field r Pacific-Basin Finance Journal 9 (2001) 101–117

Fig. 1. Long-run stock price performance. The sample consists of all infrastructure finance firms focused exclusively on road projects in China which issued an initial public offering ŽIPO. of equity on the Hong Kong Stock Exchange during the period of July 1996–October 1997. The figure provides a graph of the stock price performance for the three relaxed-listing firms, Road King, GZI and Jiangsu, as well as an equally weighted index of the returns of the firms listing under the regular rules over the period from October 7, 1997 Žthe first day all of these infrastructure firms are all traded. through June 10, 2000.

somehow compensated for the lack of a comparable earnings history. Chemmanur and Fulghieri Ž1994. argue that in assessing the credibility of investment banks, investors use the past performance of previous issues. Beatty et al. Ž1998. provide evidence that SEC investigations of underwriters impose indirect penalties on both the underwriter and its past IPO clients. Kroszner and Rajan Ž1994. find that prior to the passage of the Glass Steagall Act which formally separated commercial and investment banks in the US, banks appeared to have reacted to the public’s awareness of potential conflicts of interest by underwriting relatively high-quality securities. Gompers and Lerner Ž1999. and Hamao et al. Ž2000. investigate the issuance of equity by firms where the underwriter has a prior equity stake through a venture capital subsidiary and find no evidence of conflicts of interest. All of these papers suggest that reputation concerns will affect bank behavior.

K.L. Dewenter, L.C. Field r Pacific-Basin Finance Journal 9 (2001) 101–117

113

Column 1 of Table 3 provides the Listing Sponsor for all seven offers. This is the investment bank that formally sponsored the firm’s listing on the SEHK. Usually, the listing sponsor is the same as the lead underwriter.10 Where appropriate, the parent bank of the listing sponsor is noted in the footnote at the bottom of the table. Most of the investment banks, or their parents, are readily recognizable names. Securities Data’s investment bank rankings for these firms, based on 1996 global equity offerings, range from 3 for Merrill Lynch to 47 for HSBC.11 ŽFor our calculations, we use the lower number Žcloser to one. ranking for either the investment bank or its parent.. The range of rankings Žnot shown in the table. across relaxed listing firms is 9 to 47, with a mean of 24.6. The range across regular listing firms is 3 to 27, with a mean of 17.0. A t-test and a Wilcoxon rank test do not reject that these rankings are from the same sample. So, the quality of investment banks appears to be similar across the two sets of IPOs. Columns 2–4 in Table 3 provide evidence on whether the relaxed listing firms had characteristics that might compensate for the lack of a comparable earnings history. Note, though, that t-tests and Wilcoxon rank tests never reject equality of the relaxed and regular sets along these three dimensions at the 5% level. Column 2 indicates that, at the time of the IPO, the firms listing under the relaxed rules had 5.3 operational projects on average, compared with only 2.5 operational projects for firms listing under the regular rules. In fact, two relaxed listing firms, Road King and GZI Transport, had nine and six operational projects, respectively. Jiangsu Expressway had only one operational project. By contrast, the largest number of operational projects for firms listing under the regular rules was four Žfor both Shenzhen Expressway and Zhejiang Expressway.. The other two firms listing under normal rules had one project each. Examining the 1997 turnover Žtotal revenue., firms listing under the relaxed rules had average turnover of HK$391.2 million compared to HK$294.6 million for the regular listing firms. Examining turnover on a firm-by-firm basis, it appears that the difference is driven entirely by relaxed listed firm Jiangsu Expressway. Indeed, Jiangsu Expressway had a 1997 turnover of HK$725 million–almost twice that of the largest turnover for any firm listing under the regular rules ŽZhejiang Expressway with HK$408.9 million.. So, even though Jiangsu had only one operational project, it was almost two times larger than any other road firm.

10 For all firms except Shenzhen Expressway, this firm was identified in the press as the Listing Sponsor. For Shenzhen, Merrill Lynch was referred to as the Global Coordinator with no explicit reference to either a listing sponsor or a lead underwriter. For four of the other six firms, the listing sponsor was also identified as the lead underwriter. For Jiangsu Expressway and Sichuan Expressway, no firm was ever mentioned in the press as the lead underwriter for these two IPOs. 11 Securities Data’s League Tables rank underwriter quality by gross proceeds of equity underwritten globally during 1996, with full allocation given to the lead underwriter.

114

Listing sponsor a

Operational projects Žas of IPO.

1997 Turnover ŽHK$ million.

Net profit ŽHK$ million.

1

2

3

4

Relaxed rules firms Road King GZI Transport Jiangsu Expressway Average

SBC Warburg Peregrine HSBC Inv. Bk.

9.0 6.0 1.0 5.3

241.4 b 207.1 725.0 391.2

389.0 186.1 311.4 295.5

Regular rules firms Anhui Expressway Shenzhen Expressway Zhejiang Expressway Sichuan Expressway Average

Crosby Cap. & CEF Cap. Merrill Lynch BZW Asia JP Morgan Sec.

1.0 4.0 4.0 1.0 2.5

209.9 242.7 408.9 317.0 294.6

129.9 196.7 275.5 147.7 187.5

2.8

96.6

108.0

Difference (relaxedIregular)

The sample consists of all infrastructure finance firms focused exclusively on road projects in China which issued an initial public offering ŽIPO. of equity on the Hong Kong Stock Exchange during the period of July 1996 through October 1997. The Listing Sponsor is the investment bank that formally sponsored the firm’s listing on the SEHK. Usually, the listing sponsor is the same as the lead underwriter. Where appropriate, the parent bank is noted in the footnote at the bottom of the table. 1997 turnover is total revenue. a SBC is associated with Swiss Bank, Crosby is associated with Societe Generale, CEF is associated with Canadian Imperial Bank, and BZW is associated with Barclay’s Bank. b Road King had income from joint ventures of HK$193.8 m that was not included in turnover.

K.L. Dewenter, L.C. Field r Pacific-Basin Finance Journal 9 (2001) 101–117

Table 3 Additional comparisons of infrastructure IPO firms

K.L. Dewenter, L.C. Field r Pacific-Basin Finance Journal 9 (2001) 101–117

115

Looking at net profit in column 4, the average net profit for firms listing under the relaxed rules was HK$295.5 million compared with HK$187.5 million for firms listing under the regular rules. Two of the firms listing under the relaxed rules, Road King Žwhich listed under the relaxed rules because of negative net profit in 1994. and Jiangsu Expressway, earned net income larger than any of the firms listing under the normal rules, with 1997 net profit of HK$389.0 and HK$311.4 million, respectively. The third firm listing under the relaxed rules, GZI Transport, earned HK$186.1 in net profit for 1997, which was larger than the 1997 net income for two of the four firms listing under the regular rules. In sum, Table 3 provides evidence that the relaxed listing firms had more projects, or were larger or more profitable than the regular listing firms. These patterns are consistent with the investment bankers choosing firms which otherwise compensate for the lack of a comparable earnings history.

3. Conclusion In this paper, we provide evidence on the Stock Exchange of Hong Kong’s experience with relaxed listing requirements for infrastructure firm IPOs Žsub-rule 8.05Ž2... By comparing the characteristics and subsequent performance of the relaxed listing IPOs to IPOs listing under the regular requirements, we show that there is no discernible difference in performance across the two groups. Although the firms listed under the relaxed listing requirements must do so due to a lack of operating history or profitability, it appears that the firms taken public under these requirements to date are otherwise no more speculative than issues under the regular requirements. We do find evidence, however, that firms going public under the relaxed requirements sell a smaller fraction to the public at a higher offer price and have lower variability in stock prices. These findings suggest relatively less risk for the relaxed listing firms. There is also evidence that, prior to the IPO issue, these firms had a larger number of operational projects, more turnover, or higher net income than the regular listing firms. These patterns are consistent with investment banks choosing not to take highly speculative issues public in order to protect their reputations. Clearly, this sample is small and the results cannot necessarily be generalized to other markets or industries. Indeed, there are several examples of exchanges with relatively relaxed listing rules that have failed, at least in part, due to the failure of some highly speculative issues. ŽFor example, see Aggarwal and Angel Ž1999. for a discussion of the American Stock Exchange’s Emerging Company Marketplace.. The SEHK’s experience with relaxed listing rules for infrastructure firms, however, provides evidence that is consistent with reputation concerns prompting investment banks to carefully screen which firms to sponsor under the relaxed listing rules.

116

K.L. Dewenter, L.C. Field r Pacific-Basin Finance Journal 9 (2001) 101–117

Acknowledgements We thank Phil Berger, Harold Mulherin, Cathy Schrand, and an anonymous referee for helpful comments. Gerald Tsui at the Stock Exchange of Hong Kong provided data on Hong Kong’s IPO issues.

Appendix A. Description of the sample firms Road King was a spinoff of Wai Kee Development, a Hong Kong based firm involved in quarrying, civil engineering, and construction. All of Wai Kee’s People’s Republic of China ŽPRC. road projects were consolidated into Road King in October, 1994. At that time, AIG-AIF, a subsidiary of American International Group, acquired a 40% interest in Road King. In mid-1996, Road King had eight toll roads in Guangdong, the PRC province next to Hong Kong, and two projects in Jiangsu. Road King was the first firm to issue an IPO under sub-rule 8.05Ž2.. It listed under sub-rule 8.05Ž2. because it did not have 3 years of positive earnings. Earnings were negative in the 12 months ending 1994, but positive for 1995 and 1996. The prospectus does state that the firm was listing under the relaxed listing rules. A search of the press for 6 months prior to the IPO Žin the Asia Pacific Library of LexisrNexis. reveals 10 articles on the Road King IPO, with only three indicating that Road King was listing under the new listing rules. GZI was the first Ared chipB infrastructure company to be listed in Hong Kong. Red chip companies are firms with PRC ownership or management. GZI is a spinoff of Guangzhou Investment, the investment vehicle of the Guangzhou municipal government. At listing, GZI had six toll road projects in Guangzhou. It also had right of first refusal for all future road projects in the region. Press reports indicate that management was considering diversification into other infrastructure areas, including ports and freight forwarding. GZI listed under sub-rule 8.05Ž2. because it did not have 3 full years of audited statements. The prospectus reports positive earnings for 10 1r2 months of 1994, a full year of 1995, and 8 months of 1996. The prospectus clearly states that the firm was listing under sub-rule 8.05Ž2.. A search of the Asia Pacific library in LexisrNexis, however, finds no mention of the listing status in the 6 months prior to the IPO. Jiangsu Expressway’s IPO was an H-share issue. H-shares are shares of companies already incorporated in the PRC that are to be listed and traded on the Hong Kong Stock Exchange. Jiangsu Expressway, which runs the Jiangsu section of the Shanghai–Nanjing expressway, was the only listed toll road operator in the Jiangsu Province. Its primary owner is Jiangsu Communications Investment Co., the investment arm of Jiangsu Communications Bureau. The proceeds from the IPO were to go towards the purchase of other roads in the province.

K.L. Dewenter, L.C. Field r Pacific-Basin Finance Journal 9 (2001) 101–117

117

We were not able to get a prospectus for the Jiangsu Expressway IPO. Press reports suggest that the firm listed under sub-rule 8.05Ž2. because it did not have 3 years of audited statements Žrather than negative earnings.. In the 6 months prior to the IPO, the Asia Pacific library in LexisrNexis has 10 articles on the Jiangsu IPO, with only two mentioning that is listing under the relaxed rules. Regular Listing Firms: All four of the firms issuing stock under the normal rules were H-share firms. The provincial governments control all these firms. They fully or partially own from one to four expressways in their respective provinces.

References Aggarwal, R., Angel, J., 1999. The rise and fall of the Amex emerging company marketplace. Journal of Financial Economics 52, 257–289. Angel, J.J., 1997. Tick size, share prices, and stock splits. Journal of Finance 52, 655–681. Beatty, R.P., Bunsis, H., Hand, J.R.M., 1998. The indirect economic penalties in SEC investigations of underwriters. Journal of Financial Economics 50, 151–186. Brav, A., Gompers, P.A., 1998. Myth or reality? The long-run underperformance of initial public offerings: Evidence from venture and nonventure capital-backed companies. Journal of Finance 52, 1791–1821. Carter, R.B., Dark, F.H., Singh, A.K., 1998. Underwriter reputation, initial returns, and the long-run performance of IPO stocks. Journal of Finance 53, 285–311. Chemmanur, T., Fulghieri, P., 1994. Investment bank reputation, information production, and financial intermediation. Journal of Finance 49, 57–79. Gompers, P., Lerner, J., 1999. Reputation and conflict of interest in the issuance of public securities: Evidence from venture capital. Journal of Law and Economics 42 Ž1., 1–28. Grinblatt, M., Hwang, C.Y., 1989. Signalling and the pricing of new issues. Journal of Finance 44, 393–420. Hamao, Y., Packer, F., Ritter, J., 2000. Institutional affiliation and the role of venture capital: Evidence from initial public offerings in Japan. Pacific-Basin Finance Journal 8, 529–558. Kroszner, R.S., Rajan, R.G., 1994. Is the Glass–Steagall Act justified? A study of the U.S. experience with universal banking before 1933. American Economic Review, 810–832. Leland, H., Pyle, D., 1977. Informational asymmetries, financial structure, and financial intermediation. Journal of Finance 32, 371–387. Loughran, T., Ritter, J.R., 1995. The new issues puzzle. Journal of Finance 50, 23–51. Michaely, R., Shaw, W.H., 1994. The pricing of initial public offerings: Tests of adverse-selection and signaling theories. Review of Financial Studies 7 Ž2.. Pownall, G., Waymire, G., 1989. Voluntary disclosure choice and earnings information transfer. Journal of Accounting Research 27, 85–105, Supplement. Ritter, J.R., 1984. The ‘hot issue’ market of 1980. Journal of Business 57, 215–240. Ritter, J.R., 1991. The long-run performance of initial public offerings. The Journal of Finance 46, 3–27. Seguin, P.J., Smoller, M.M., 1997. Share price and mortality: An empirical evaluation of newly listed Nasdaq stocks. Journal of Financial Economics 45, 333–363.