Underpricing at the time of listing is a topic that for some time has attracted the interest of both academics and practitioners. It is a widespread and generalized phenomenon, extensively examined in literature, with a number of theories and models put forward to explain it.
This paper focuses on underpricing in IPOs (Initial Public Offerings) by Italian banks, with a view to establishing the relevance of bank specificity.
The analysis is pertinent for two reasons. First, it represents an in-depth study of the results reported in IPOs in a specific and important sector (banking). Second, this analysis is an "indirect" test of Baron's theory (1982), according to which underpricing stems from information asymmetry between the issuer and the intermediary participating in the placement. Although similar tests have previously been undertaken on this topic, none has thus far referred to European markets; hence, the importance of testing the theory in markets that, due to the role that banks assume, are more exposed to information asymmetry as Baron reported.
In general terms, we test if an explanation of underpricing may lie in the nature of the financial system, which can be intermediary-oriented (as in Europe) or market-oriented (as in the US).
In order to test our hypothesis, the mean and median underpricing of banks that were "self-placed" on the Italian stock market between 1985-2007 was calculated. This value was then compared with the mean and median underpricing registered by other firms listed in the same period. This comparison was undertaken because if underpricing occurs as a result of information asymmetry in favour of the intermediaries participating in the placement--when intermediaries themselves participate in the placement of their own securities--a significant reduction in underpricing in the listings should ensue.
A similar study conducted in 1989 in the United States (Muscarella and Vetsuypens, 1989), found no significant differences between banks and other firms in terms of underpricing. The US financial system, however, differs extensively from the European system, especially in Italy, where the capital market is relatively undeveloped with numerous small and medium enterprises. In Italy, and generally in the whole of continental Europe, banks are the prime source of funding of firms (Shleifer and Vishny, 1997; La Porta et al., 2000; Baran, 2008). As such, we could expect that the greater importance of banks in intermediary-oriented contexts would allow them to play a very different and much more important role in setting the placement price, as well as benefitting from greater information asymmetry (Bessler and Kurth, 2007). The purpose of the present study is to verify this hypothesis and, indirectly, to test Baron's model in a market oriented towards financial intermediaries.
Our findings reveal significantly lower underpricing for banks than other companies. This result appears to be consistent with Baron's hypothesis and suggests that this model may contribute to explaining underpricing in Italy and, potentially, in other intermediary-oriented markets.
The paper is organised as follows. The first section reviews the most relevant literature on underpricing, while the second section presents the Italian stock market sample surveyed. Section three presents the data and methodology. The results, demonstrating that banks in Italy, unlike other firms, have succeeded in minimising the indirect cost of underpricing in listings, are reported in section four. The concluding section discusses the validity of Baron's model for the Italian market.
The theme of underpricing has been widely discussed in literature (Ritter and Welch, 2002). Much attention has been paid to Ibbotson's noted definition (1975), stating that underpricing is the difference between the price of the first trading day and the price at which the security is offered on the market, which represents "money left on the table'"
Ibbotson's study measures underpricing by taking the closing price at the end of the first week of listing as the reference for a comparison with the offering price. Subsequent analyses, however, have shown that this underpricing is already manifest at the end of the first day of listing (Miller and Really, 1987), while other studies (Barry and Jennings, 1993) have demonstrated that in 90% of cases, underpricing is already present in the difference between the offering price and the opening price on the first day of listing.
Underpricing occurs, albeit to different extents, in numerous countries. The ample literature on the topic suggests that the phenomenon can also be particularly frequent in given periods defined as hot issue markets (Ritter, 1984), and while numerous reasons have been put forward, none appear to be entirely satisfactory. The main explanations proposed (Ljungqvist, 2006) can be summarized into the following four main groups: risk aversion, asymmetric information, signalling and certification requirements with regard to securities and price support activities.
With regard to risk aversion, two main observations can be made. First, it is in the issuer's interests, as well as that of the underwriting syndicate, to ensure that the offer does not remain unsold. To prevent this, the issuer may deliberately decide to offer the securities at an advantageous price and thus favour the underwriting syndicate. This is not however confirmed by market evidence, where IPOs are typically oversubscribed. It is therefore unlikely that underpricing is attributable to the fear of being unable to place securities.
Secondly, risk aversion may also be present post-IPO. Tinic (1988) found that underpricing could be a way of limiting the risk of investors accusing the issuing company and the underwriting syndicate of not having fully and appropriately disclosed all the information available and correctly setting the offering price. Underpricing would thus appear to be a form of insurance cover in the event of legal proceedings. Although Tinic supports his hypothesis with empirical evidence, it is not confirmed by other studies, for instance Drake and Vetsuypens' (1993).
Models based on information asymmetry focus on two aspects: the relationship between issuers and financial intermediaries and the relationship between investors. The first type of model is usually based on that developed by Baron (1982), which sees underpricing by firms on the market on the first day of listing as a consequence of information asymmetry in favour of the banks involved in the placement. These banks offer the issuer's securities at the time of the IPO at a lower price than their real effective value in order to encourage purchasing and to avoid maintaining a large number of shares in their portfolio from the same issuer. These issuers do not have the opportunity to verify either the intermediary's efforts or the appropriateness of the price proposed, and would therefore be victims of information asymmetry, while the financial intermediaries involved in the placement could use this to their advantage. This theory was tested with reference to the US market by Muscarella and Vetsuypens (1989) and, more recently, by Dempere (2008). Muscarella and Vetsuypens start from the assumption that underpricing represents a cost for the issuing firms and that this is the result of the deliberate underestimation of the price of the security by the placement intermediaries. If underpricing is performed by banks, it is logical to assume that the banks themselves, when placing their own securities, will not wish to bear the cost, or, will at least attempt to reduce this to a minimum. Muscarella and Vetsuypens verified the mean discount practised during the IPO in a sample of self-underwritten bank offerings, and compared the results with those of firms in other sectors. The difference proved to be statistically insignificant, and did not provide empirical confirmation of Baron's theory. The results of Dempere (2008) support those of previous studies on self-underwritten IPOs.
With regard to the second subgroup of explanations based on information asymmetry, a further and significant theory was formulated by Rock (1986). According to this hypothesis, there are two types of investors on the market: the well informed and the uninformed. The informed investor, thanks to his superior knowledge, only participates in the IPOs of "good" firms, while the uninformed investor, unable to distinguish between them, participates in all of them. The uninformed investor thus obtains a sizeable portion of the securities of "bad" firms and only a residual portion of the IPOs of "good" firms. Underpricing, in Rock's hypothesis, is the instrument that enables uninformed investors to remain on the market, allowing them to obtain adequate returns on newly listed securities. In this regard, Beatty and Ritter (1986) drew up...