Statistical methods for assessing the fairness of the allocation of shares in initial public offerings

Abstract

In the late 1990s the stocks of technology companies, especially those involved with the Internet, were highly profitable. Customers of investment firms requested many more shares of new issues or initial public offerings (IPOs) than were available. Some customers allegedly made questionable arrangements with brokers for favourable treatment. This article describes statistical measures for assessing the fairness of an investment firm's or broker's allocations of IPO shares to their customers. One approach compares the success rates of different groups of customers, e.g. those alleged to be favoured with those of other groups. The second method incorporates the prior business each customer gave the firm into the comparison as an allocation criterion. This metric is an objective quantification of the fact that it is economically sensible for a firm to allocate more IPO shares to its ‘best’ customers.