Routledge Revivals: The Efficiency of New Issue Markets (1992)

Kyran McStay

• Publisher: Routledge
• Year: 2018
• Topic: Business
• eBook ISBN:
  9781351273022
• Subtopic: Economic Theory
• Index: Business

First published in 1992, The Efficiency of New Issue Markets provides a theoretical discussion of the adverse selection model of the new issue market. It addresses the hypothesis that the method of distribution of new issues has an important bearing on the efficiency of these markets. In doing this, the book tests the efficiency of the Offer for Sale new issue market, which demonstrates the validity of the adverse selection model and contradicts the monopsony power hypothesis. This examines the relative efficiency of the new issue markets and in turn demonstrates the importance of distribution in determining relative efficiency. The book provides a comprehensive overview of under-pricing and through this assesses the efficiency of new issue markets.

Chapter 1

The Efficiency of New Issue Markets: Introduction

Introduction

The most striking anomaly concerning the operation of new issue markets is the apparent under pricing of unseasoned common stock offerings. The term under pricing refers to the observation that the initial performance of unseasoned new issues is, on average, significantly positive. Initial performance is usually measured by the change of the closing share price on the first day or first month of trading relative to the initial offer price. The observation of positive average initial returns on a randomly chosen group of new issues suggests that they are offered at an expected discount relative to the price which is established in after market trading. The average initial performance of SEC-Registered unseasoned common stock issues is generally in the range of 15 to 20 percent with the standard deviation of initial returns in the 30 to 40 percent range. The existence of positive initial returns has been taken as prima facia evidence of the inefficiency of new issue markets.

The aim of this thesis is to shed further light on the subject of the under pricing and thereby assess the efficiency of new issue markets. There have been many non-rigorous explanations advanced for under pricing which amount to no more than the statement that “new issues are under priced to assure the success of the issue”. Ibbotson (1975) cites some of these—for example, an explanation “prevalent in Wall Street” is that “under pricing new issues ‘leaves a sweet taste in investors’ mouths’ so that future underwritings from the same issuer could be sold at attractive prices” (Ibbotson 1975, p 264). To date there have been only two internally consistent and rigorous explanations offered to account for under pricing. The first is based on a long held belief of the exploitation of monopsony power by the investment banking industry: investment banks individually, or in collusion, exploit the inexperience of issuers and under price unseasoned offerings. The second line of research, which is of more recent origin, is based on the specification of an information asymmetry whereby some investors have better information—less uncertainty—concerning the valuation of the issue than others. This uneven distribution of information gives rise to an adverse selection bias against unsophisticated or lesser informed investors. This research is due to the pioneering work of Rock (1982, first published in 1986).

The intuition behind the adverse selection model of the new issue market is the following: there is a group of informed investors who have superior abilities at recognising issues which will decrease in price in initial trading. These investors, who are likely to be professionals with access to equity research, apply only for issues which they assess to be undervalued at the offer price. As a result unsophisticated investors face a higher probability of receiving an allotment in issues which are overvalued. New issues are, therefore, under priced to compensate the unsophisticated investor for the bias in the probability of allotment between good and bad issues.

This chapter examines the alternative explanations of under pricing (section I) and reviews existing empirical evidence relating to the pricing of unseasoned common stock offerings (section II). Section III outlines the research contained in this thesis.

I. Alternative Explanations of Under Pricing

The monopsony power and adverse selection hypotheses are the only theoretically consistent explanations of under pricing which have been offered to date. This thesis takes the adverse selection hypothesis as the maintained hypothesis and the monopsony power hypothesis as the alternative hypothesis.

Monopsony Power and Inefficiency

The following argument is typical of the monopsony power explanation of under pricing: investment banks individually, or in collusion, exploit inexperienced issuers by under pricing. The gains are passed onto investors in return for side payments: the investment bank rations shares in over subscribed issues to its regular clients and receives a side payment in the form of higher charges for the other services which it provides. Adherents to this view point to the widespread use of side payments in the investment banking/securities industry: for example, many services, such as research, are provided on a ‘soft dollar’ basis, where costs are recouped from security dealing charges. Ibbotson (1975) discusses various monopsony power explanations. Baron (1982) provides a rigorous formulation of the monopsony power hypothesis.

Baron’s model is essentially a model of the demand for investment bank advising and distribution services based on information asymmetries between the issuer and the bank. The bank is better informed than the issuer about capital market conditions and the issuer cannot observe the distribution effort of the bank. The advising function has value since the issuer can delegate the pricing decision to the bank so that it uses its superior information about capital market conditions. Distribution effort has value to the extent that the bank can increase the demand for the issue. The term delegation contract is used for the contract under which the issuer engages the bank to price and distribute the issue.

The problem faced by the issuer is to choose the optimum delegation contract given rational expectations about the bank’s incentive to mis-represent its information about capital markets and its incentive to shirk in its distribution effort. The optimum delegation contract involves the investment bank choosing an offer price below the first best offer price—defined as the price which the issuer would set if it knew the investment bank’s information had and if it could observe distribution effort. Investment banks will, however, under price new issues in order to economise on distribution effort thereby increasing the expected return from new issues activity. From the issuer’s perspective, this equilibrium is superior to that where the it does not employ the services of the bank to price the issue since it gains some advantage from the bank’s information and distribution effort—intuitively, the issuer is better off being exploited by the investment bank when compared with the case where it does not employ the investment bank.

Baron demonstrates that the value to the issuer of delegating the offer price decision is an increasing function of the issuer’s uncertainty about the market demand for the issue and that the offer price chosen by the bank is a decreasing function of this uncertainty: the greater the bank’s information advantage the greater the extent of under pricing. As Baron’s model hypothesises, the bank’s information advantage about the demand for the issue and capital market conditions is positively related to the uncertainty of the issuer so that the extent of under pricing is also related to uncertainty.

There are several conceptual difficulties with the monopsony power explanation of under pricing. The first lies in the findings of many studies which point to the competitive nature of the investment banking industry (see for example Hayes (1971, 1979) and Hayes et al (1983)). Also, one of the longest anti-trust trials in history failed to find any evidence of collusion between 17 of the largest investment banks in the US (see Medina 1953). Judge Medina concludes in his opinion:

If they had in fact acted in combination or as a unit to divide the business among themselves, and to form a monopoly vis-avis the other firms in the industry, as alleged, the pattern of such combination, no matter how cleverly disguised or concealed, must surely have emerged, after such a long and continuous scrutiny as has gone on in this case for almost three years. But it did not. (Medina 1953 pp. 415–6)

To ensure survival in such a competitive market the investment bank must balance the desires of issuers and investors—provided that exploitation of issuers’ inexperience is detectable. Under pricing resulting from exploitation is unlikely to survive in a competitive investment banking industry since information on the extent to which new issues are under priced becomes public information. An investment bank could establish a reputation of not exploiting the issuer by excessive under pricing and thus increase its market share. Beatty and Ritter find “that investment banks pricing off the line [of average under pricing] in one sub-period do in fact lose market share in the subsequent period, although the relation is a noisy one.” (1986 p. 227)

Another difficulty encountered by the monopsony power hypothesis is found in the fact that the pattern of under pricing appears to have been unchanged since the regime of fixed security dealing commissions ended in the US in May 1975. One of the more convincing explanations of why the investment bank should pass on the gains from under pricing to investors was that they received side payments in the form of higher than competitive security dealing charges. There is no evidence to suggest that the extent of under pricing has decreased despite the large falls in dealing expenses which have occurred since fixed commissions were abolished in the US in May 1975 (see Ritter 1984).

A further observation which casts doubt on the monopsony explanations is that under pricing exists even where the initial public offering is made by competitive tender (see Dimson (1979) and Jacquillat and McDonald (1978)). This illustrates that under pricing is a feature of the competitive operation of new issue markets. Investment banks generally set the offer price at a level where there is some excess demand given the orders submitted—for example, for the Offer for Sale by tender for Morgan Grenfell (summer 1986) the strike price was set at a level were the issue was about four times over subscribed. Despite this fact the issue realised negative initial returns.

To date there has been only one attempt to explain the role of under pricing in a competitive market for investment banking services. This explanation is contained in the adverse selection model developed in Rock (1982 and 1986).

The Adverse Selection Model

The adverse selection model of the new issue market assumes that the issuer and the investment bank are uncertain about the share price which will be established in after market trading. The problem created for the organisation of the new issue market is that some investors have superior information about the value of the firm. These investors apply only for issues which they believe are undervalued at the public offer price. The investors who do not have superior information apply for all issues or at random with respect to the realised initial return for the issue.

Uninformed investors, therefore, face a bias in the probability of allotment for “good” and “bad” issues: issues which experience positive initial returns will be more heavily subscribed than those with negative initial returns because of the selective participation of informed investors. The uninformed will face a higher probability of allotment for bad issues—reflecting the lower demand—than for the good issues. In this case if all issues were offered at their expected value—that is, they are not under priced—the expected return to the uninformed investor would be zero.

The central hypothesis of the adverse selection model of the new issue market is that issues are under priced on average to compensate the uninformed for the bias in the probability of allotment between good and bad issues: the expected gain from under pricing compensates the uninformed for the expected loss associated with adverse selection. Dimson (1979) recognised this point in the introduction to his thesis but it was Rock (1982) who provided the first formal modelling of the theory.

In Rock’s formulation of the market for initial public equity offerings the issuer and investment bank are uncertain about the true value of a share, v, but produce an unbiased estimate which is a random drawing from the distribution f(v). The issuer must set an initial offer price, p, and solicit orders at that price. The offer price may not be changed in the light of realised demand. Rock’s model is based on the following main assumptions:

1. Informed investors have perfect information about v and apply only for issues which are undervalued (ie. v > p); and

2. Allotment is by lottery—orders are drawn at random and filled in their entirety until there are either no more orders (the issue is under subscribed) or no more shares (the issue is over subscribed). All investors, informed or uninformed, face the same probability of allotment assuming that they apply. There is no preferential rationing to any group of investors;

Under these assumptions, the uninformed face a higher probability of allotment in the bad state than in the good state when the informed participate in the market. Consequently, without under pricing, the average returns to the uninformed are negative. Rock demonstrates that there exists an expected discount which compensates for this bias. For the case where f(v) is uniformly distributed the model yields the conclusion that the issue price is set at the level for which uninformed demand equals issue size—the optimum offer price is the full subscription price.

The main testable implication of the adverse selection model which has been implemented to date is the prediction of a positive relationship between the uncertainty of the inves...

Table of contents

• Cover
• Half Title
• Title Page
• Copyright Page
• Original Copyright Page
• Dedication
• Table of Contents
• Chapter 1 The Efficiency of New Issue Markets: Introduction
• Chapter 2 The Adverse Selection Model of the New Issue Market
• Chapter 3 Distribution and the Efficiency of New Issue Markets
• Chapter 4 Testing the Efficiency of New Issue Markets
• Chapter 5 Distribution and the Efficiency of New Issue Markets in the US and the UK
• Chapter 6 Summary and Conclusions
• Appendix 1 The Nature of the Information Asymmetry in New Issue Markets
• Appendix 2 Mathematical Appendix
• Appendix 3 Hot Issue Markets
• Bibliography
• Index