Role-of-asymmetric-information-and-moral-hazard-on-IPO-underpricing-and-lockup_2014_Journal-of-International-Financial-Markets-Institutions-and-Money.docx

《Role-of-asymmetric-information-and-moral-hazard-on-IPO-underpricing-and-lockup_2014_Journal-of-International-Financial-Markets-Institutions-and-Money.docx》由会员分享，可在线阅读，更多相关《Role-of-asymmetric-information-and-moral-hazard-on-IPO-underpricing-and-lockup_2014_Journal-of-International-Financial-Markets-Institutions-and-Money.docx（25页珍藏版）》请在淘文阁 - 分享文档赚钱的网站上搜索。

1、Int. Fin. Markets, Inst. and Money 30 (2014) 81105 Role of asymmetric information and moral hazard on IPO underpricing and lockup Haz Hoque University of York, The York Management School, Freboys Lane, Yorkshire YO10 5GD, United Kingdom a r t i c l e i n f o a b s t r a c t Article history: Received

2、 25 January 2013 Accepted 4 February 2014 Available online 12 February 2014 JEL classication: G24 G32 Keywords: Information asymmetry Moral hazard Lockups Underpricing UK market This paper analyses the role of asymmetric information and moral hazard on IPO underpricing and lockups. I document that h

3、igh information asymmetry is related to underpricing while the lockup length and lockup expiration return is related to moral hazard. Accordingly, lockup length and underpricing work as substitute signals. These results relate to the UKs unique institutional sett- ings: long and diverse lockups, and

4、 a close relationship between the issuing company and the corporate broker. Moreover, director ownership affects lockup length and lockup expiration return in a non-linear fashion. 2014 Elsevier B.V. All rights reserved. 1. Introduction Initial public offerings (IPOs) are subject to asymmetric infor

5、mation and moral hazard problems. Prior work has shown that asymmetric information is mitigated by IPO underpricing (Rock, 1986; Benveniste and Spindt, 1989) and installing lockup contracts on the insiders (Brau et al., 2005). Brav and Gompers (2003) test three explanations for lockups, namely, mora

6、l hazard, information asymmetry I would like to thank Chen Su, Philip Linsley and the seminar participants at Swansea University for their helpful suggestions. Any remaining errors are my responsibility. Tel.: +44 7515890949. E-mail addresses: haz.hoqueyork.ac.uk, haz al http:/dx.doi.org/10.1016/j.

7、intn.2014.02.001 1042-4431/ 2014 Elsevier B.V. All rights reserved. Contents lists available at ScienceDirect Journal of International Financial Markets, Institutions & Money journal homepage: 82 H. Hoque / Int. Fin. Markets, Inst. and Money 30 (2014) 81105 and rent seeking, but nd support only for

8、 the moral hazard hypothesis. However, Brau et al. (2005) provide support for the moral hazard and information asymmetry hypotheses. Recently, Yung and Zender (2010) bridge the gap between Brav and Gompers (2003) and Brau et al.s (2005) studies on lockups by showing that asymmetric information is a

9、dominant consideration for one set of rms, while moral hazard prevails for others. The objective of this paper is to analyse the roles of asymmetric information and moral hazard on IPO underpricing and lockups using unique hand-collected data on IPO lockups from the UK market. Though all IPOs suffer

10、 from moral hazard and information asymmetry problems, in some rms, moral hazard may be dominant, whereas information asymmetry may be dominant in others. While lockups are mechanisms to mitigate information asymmetry or moral hazard, underpricing only alle- viates information asymmetry.1 By compari

11、ng high information asymmetry rms (small rms, IPOs underwritten by low-ranked underwriters and Alternative Investment Market AIM rms) with high potential for moral hazard rms (high director ownership rms2), I am able to determine if lockups are more closely associated with information asymmetry or m

12、oral hazard. At the same time, by com- paring high information asymmetry rms and high potential for moral hazard rms (henceforth, high moral hazard rms), I am able to reconrm that underpricing is related to information asymmetry, not moral hazard.3 This sample polarisation is at the heart of my rese

13、arch design, as it leads to different predictions for the high asymmetric information subsample compared to the high moral hazard subsample. For instance, if information asymmetry drives underpricing4 and lockup length in the high asymmetric information subsample, changes in moral hazard in those rm

14、s should not affect the underpricing and lockup length. On the other hand, if moral hazard drives lockup length in the high moral hazard subsample, changes in information asymmetry should not affect the lockup length. Examining the high asymmetric information and high moral hazard subsamples should

15、enlighten us as to the role of asymmetric information and moral hazard on IPO underpricing and lockups. I need an identication strategy to separate out the high asymmetric information subsample and the high moral hazard subsample.5 I use following proxies for information asymmetry. I use size as a p

16、roxy for information asymmetry by following Beatty and Ritter (1986). Small IPOs are subject to more information asymmetry, while large IPOs are subject to less information asymmetry. The second proxy is underwriter reputation, as in Carter and Manaster (1990), where reputable underwriters mitigate

17、information asymmetry. The third proxy used is the IPO market. AIM is a market for small growth companies, where information asymmetry is high. The ban on insider sells for a certain period of time is frequently alleged to be a necessary commitment device to induce the public to buy shares at the of

18、fering which reduces moral hazard. Because insiders have better information about the rms future prospects, they may try to take advantage of that information at, or soon after, the initial public offering. If insiders knew that the price immediately after the offering was excessively high, they mig

19、ht wish to cash out at a high price. The lockups are likely to act as a commitment device to alleviate the 1 One of the important assumptions of my analysis is that post-IPO markets are fundamentally accurate, and this is reected in underpricing. If post-IPO prices are not fundamentally correct, eit

20、her because they are driven by investor behavioural biases or because there is a systematic mispricing of the IPO rms, then underpricing measures may be severely biased. 2 Thanks to an anonymous referee for suggesting that director ownership is a good proxy for moral hazard, and sample bifurcation b

21、ased on moral hazard could be done using director ownership in the IPO. 3 To the best of my knowledge, no study considers underpricing as a result of the moral hazard problem. The survey evidence is in line with this (see, for example, Jenkinson and Ljungqvist, 2001; Ibbotson et al., 1994). Since un

22、derpricing is resolved immediately after the issue, it is difcult to imagine how underpricing can mitigate moral hazard. However, Brenan and Franks (1997) argue that underpricing is a mechanism through which managers can protect their private benets by allocating shares strategically when raising mo

23、ney through IPOs. 4 One should note that asymmetric information is not the only determinant of IPO underpricing. There are other explana- tions for IPO underpricing: prospect theory (Loughran and Ritter, 2002), corruption (Loughran and Ritter, 2004), informational cascades (Welch, 1992), lawsuit avo

24、idance (Hughes and Thakor, 1992), signalling (Allen and Faulhaber, 1988) and IPO as a marketing event (Chemmanur, 1993). For the analysis of the current paper to be valid, it needs asymmetric information to be one of the determinants of IPO underpricing. 5 This does not mean that in the asymmetric i

25、nformation subsample there is no moral hazard problem, or that in the moral hazard subsample there is no information asymmetry. Rather, it means that in the asymmetric information sample, information asymmetry is the dominant problem, and in the moral hazard sample, moral hazard is dominant. H. Hoqu

26、e / Int. Fin. Markets, Inst. and Money 30 (2014) 81105 83 moral hazard problem. Firms which have higher potential for moral hazard problem (i.e., high director ownership rms) would have to accept a longer lockup in order to convince investors to buy stock in the offering (Brav and Gompers, 2003). In

27、 sum, small IPOs, IPOs underwritten by non-prestigious underwriters and IPOs on the AIM serve as the high asymmetric information subsample, and IPOs where director ownership is high serve as the high potential for moral hazard subsample. Existing research provides little support in favour of the sig

28、nalling role of underpricing6 in the US but supports the signalling role of underpricing in the UK (see, for example, Garnkel, 1993; Jegadeesh et al., 1993; Michaely and Shaw, 1994 for the US; and Levis, 1993 for the UK). Hence, if IPO underpricing is a signalling mechanism, the rms that signal thro

29、ugh underpricing may impose a less stringent lockup. In that case, we expect a negative relationship to exist between IPO underpricing and lockup length. If moral hazard drives the lockup length, and as there is no evidence that moral hazard drives IPO underpricing, there should not be any relations

30、hip between underpricing and lockup length in the rms with high moral hazard. This paper tests these conjectures. Although a number of studies on IPO underpricing in the UK exist, the research on lockups in the UK is very limited. The lockups research mostly focuses on the US (Field and Hanka, 2001;

31、 Brav and Gompers, 2003; Brau et al., 2005; Yung and Zender, 2010). In the UK, Espenlaub et al. (2001) analyse the diversity of the lockups. Furthermore, Hoque (2011) analyses the choice and role of different types of lockups in the UK. However, to the best of my knowledge, no study has examined the

32、 reasons for lockups and IPO underpricing conjointly in the UK. Lockups and issue prices are set at the beginning of the IPO, and therefore those signalling devices might be complements or substitutes. Since the prevalence of both lockups and underpricing has not been considered together in the cont

33、ext of the UK, this study lls the gap in the literature by analysing data from the UK market, which is different from the US market. The US lockup length is relatively short (median 180 days), and little information is produced and disclosed after the IPO. In addition, for the US IPOs, there is a qu

34、iet period7 when analysts afliated with underwriters are not allowed to issue research reports and recommendations. The lockup length in the UK is much longer (median 598 days), where several earnings announcements are likely to be made during the lockup period. Moreover, unlike the US, there is no

35、quiet period in the UK. As more information is produced and disclosed in the market during the lockup period, asymmetric information may be less of a consideration in the UK market. Hence, I test whether moral hazard is the dominant consideration for the UK lockups. Ritter (2003) summarises the diff

36、erences between the IPO markets of the US and Europe in terms of price setting, the IPO allocation mechanism, allocations to individual and institutional investors, analyst coverage, etc. There are also differences between the European and US IPO markets in terms of the fees charged by underwriters

37、(Abrahamson et al., 2011). The difference is even more striking when we compare the UK and the US. The length of lockup in the US is concentrated at 180 days (see, for example, Field and Hanka, 2001), whereas the length varies widely in the UK. Furthermore, the relationship between underwriters and

38、issuing rms is closer in the UK than in the US, as UK IPO rms need to have a corporate broker to maintain stock exchange listings. Other corporate nance activities, like stake building in target companies, further equity issues and director trading also need to be done through corporate brokers. Und

39、erwriters have more power in the UK to set the issue price, as argued by Chambers and Dimson (2009). The UK market serves as a good out of sample test for Brav and Gompers (2003) and Yung and Zender (2010), where the lockup contracts are heterogeneous and where the IPO market setting is distinct. Th

40、e different types of lockups and institutional settings in the UK present an interesting empirical issue, as to whether lockups and IPO underpricing emanate from information asymmetry and/or moral hazard. Our results show that information asymmetry is a determinant of IPO underpricing for any of the

41、 IPOs. For the high asymmetric information subsample, the proxy for information asymmetry is a determinant of underpricing, while this is not the case for the high moral hazard subsample. I nd 6 Underpricing is not the only signal rms can use at the time of IPO. The alternative signalling mechanisms

42、 include reputable underwriters (Booth and Smith, 1986), auditors (Titman and Trueman, 1986) and venture capitalists (Megginson and Weiss, 1991). 7 The quiet period ends 40 calendar days after IPO. Before 2002, it was 25 calendar days. 84 H. Hoque / Int. Fin. Markets, Inst. and Money 30 (2014) 81105

43、 no evidence that information asymmetry is related to lockup length for any of the IPOs. Even in the high asymmetric information subsample, the proxy for information asymmetry is not related to the lockup length. The evidence shows that moral hazard drives the lockup length in the UK market. These r

44、esults are consistent with the notion that compared to the US, lockups in the UK are longer and much more information is produced during the lockup period. Hence, in the UK, lockups do not emanate from the information asymmetry problem; rather, they mitigate the moral hazard problem in IPOs. When I

45、analyse lockup expiration returns, the results show that the measure for information asymmetry is related to the drop in share price for the high asymmetric information subsample, and director ownership drives the lockup expiration returns in the high moral hazard subsample. Partial correlation anal

46、ysis shows that lockups and underpricing are substitute signals. This is consistent with previous research which has shown that underpricing works as a signalling device in the UK. I show that lockups and underpricing are substitute signals, as underwriters set the lockup agreement and the issue pri

47、ce at the time of IPO. This paper contributes to the literature in three major ways. First, I show that information asym- metry is one of the determinants of IPO underpricing and the result is driven by the high asymmetric information subsample, while director ownership is one of the determinants of

48、 lockup length and lockup expirations returns and it is driven by the high moral hazard subsample. These results are in contrast to Yung and Zenders (2010) nding that in some IPOs asymmetric information drives lockups and in others moral hazard drives lockups. My results highlight the distinctive- n

49、ess of the UK market, as the lockup lengths are longer, thus mitigating information asymmetry to a certain extent. One of the important aspect about the UK data is that the lockups are not clustered at one or two time periods (180 and 360 days), but are highly diverse. This variation allows me to tease out the difference for the UK and US rms. This is a major reason to do the analysis on the UK rms. Second, lockups and underpricing are substitute signa