投资银行-机构投资者关系、“捧场”与IPO中的利益问题

投资银行是直接融资过程中最重要的中介机构,作为融资产品的“卖方”代表,其核心竞争力源于它所构建的投资者关系网络,但现有文献对于投行-投资者关系在证券发行中的作用还缺乏系统的实证研究。本文利用机构投资者在中国IPO新股发行中的完整询价记录,基于报价参与和报价水平两个维度提出了一种新的动态关系强度测算方法,并由此细致刻画了投行-机构关系对股票发行、定价的作用机制。结果发现:(1)投行可以驱动关系机构主动认购其承销的IPO新股,并引导关系机构给出与投行估值水平保持一致的高位报价,进而有效缓解了IPO拍卖制下的“投资者参与不确定性问题”。(2)关系机构的捧场报价显著提高了股票发行价格、增加了投行的承销收入;但是当新股交易价格在长期内逐渐向其内在价值水平收敛时,受到关系机构捧场支持越多的IPO新股,其长期回报率相对越低。(3)投行与机构的关系在本质上是互惠的,一旦投行被赋予新股分配权力,它明显倾向于将新股抑价发行带来的巨大利益分配给在前期捧场的关系机构,二者的关系越强,机构报价被认定为“有效报价”的概率越高,而一旦被认定为有效报价,关系机构的新股需求更是将被优先满足。

The Relationship between Investment Banks and Institutional Investors,Boosts, and Benefit Transfer in IPOs

Investment bank is one kind of the most important intermediaries in direct financing. As the representative of security issuer, an investment bank's competitiveness partly originates from its networks of relationship investors, which have been rarely discussed in the literature. Based on the complete records of institutional investors' participation in Chinese IPOs‘ book-building process, we have designed a new dynamic measure of the relationship between an investment bank and institutional investors. Furthermore, we have analyzed in detail how this relationship affects IPO pricing. The main results are as follows. (1) An investment bank can drive related investors to take part in the IPO process and submit high bids that are consistent with the underwriter analyst’s valuation. This implies that the investment bank can alleviate the uncertainty of investor participation in IPO auctions by its networking ability. (2) The boost from related investors can significantly increase IPOs' offering price and the under-writing fee paid to the investment bank. However, when the trading price converges to its intrinsic value gradually in the long run, the appearance of more related investors in the IPO process will predict lower long-term return to investors. (3) The relationship between the investment bank and institutional investors is reciprocal in essence. Once the investment bank is granted allocation discretion, it will be inclined to allocate money left on the table to its related investors. The stronger the relationship, the larger probability that bids will be considered affective. Given that their bids are judged affective, related investors' demand for new issues will be satisfied preferentially.
This paper makes three contributions to the literature. First, we present a new mechanism for interest distribution in the IPO process. The traditional literature usually assumes that the principal-agent problem exists between the IPO issuer and the investment bank (IB) responsible for the underwriting. The IB is inclined to underprice the IPO relative to its intrinsic value. Then, due to its allocation discretion, it will allocate the underpriced new issues to its connected investors, who will transfer part of the revenues to the underwriter, such as in the form of brokerage fees. This quid pro quo thus conflicts with the issuer's interests. However, if underwriting fees gained by the IB are positively related to the IPO offering price, the interest of the IB should be bundled with the IPO issuer and there will be no serious principal-agent problem. We find that the IB has a significant incentive to overprice the IPO according to the high bids of its connected institutional investors, so the issuer can obtain more IPO proceeds. In addition, the IB can favor connected investors over time. For example, once underwriting IBs are granted allocation discretion, they will allocate underpriced new shares to their connected investors who have supported the bidding of prior IPOs.
Second, this paper provides a new method to measure the relationship between investment banks and institutional investors, which considers both IPO issuers' participation and the relative bidding prices of investors. Based on brokerage fee data, the traditional method can only identify one special investor (fund)-underwriter relation. In contrast, our new method can measure underwriters' connection with all kinds of investors. Furthermore, compared to the method based on only participation frequency, our method also considers whether investors' bidding prices are consistent with underwriters' expectations, which means that connections can be more accurately measured.
Third, our findings extend the understanding of the efficiency of IPO auction methods. The classical literature usually concludes that auction methods lack efficiency in pricing IPOs because the underwriters have no allocation discretion, which is a key instrument to encourage investors to produce information on IPOs in the book-building process. One problem in IPO auction is investors' participation uncertainty. However, our empirical research finds that underwriters can solve this problem through their connections with institutional investors. It suggests that networks between IBs and investors can replace underwriters' allocation discretion to some extent. Furthermore, this paper shows that an underwriter will be inclined to allocate underpriced new issues to its connected investors once it gains allocation power. Thus, from the perspectives of fairness and efficiency, IPO auction methods will perform better than the traditional literature predicts if the function of the underwriter-investor relationship is considered seriously.