Other Anomalies

The Size Effect | Announcement Based Effects | Corporate Transactions
Insider Transactions | The S&P Game

Some studies have shown that small firms (capitalization or assets) tend to outperform. The small stock affect was first documented by Rolf W. Banz ¹. He divided the stocks on the NYSE into quintiles based on market capitalization. The returns from 1926 to 1980 for the smallest quintile outperformed the other quintiles and other indexes. Others have argued that its not size that matters, its attention and number of analysts that follow the stock.

This anomaly is subject to intense debate over whether an opportunity to generate excess returns actually exists. Ibbotson data (See Historical Data) and other studies show that small capitalization stocks outperform large stocks in the United States as well as in foreign markets. However, others argue that its not reasonable to assume that investors can realize those returns. Professor Jeremy J. Siegel argues that the period from the end of 1974 through the end of 1983 accounts for the whole outperformance of small caps and according to John C. Bogle, since December 1978, small caps and large caps have earned exactly the same returns (Source: Wall Street Journal 2/10/97).

James O'Shaughnessy argued in What Works on Wall Street that the returns in small stocks are attributable almost entirely to micro-cap stocks with market capitalizations under $25 million. Small stocks typically have large spreads and commissions and cannot be bought by institutional managers without significantly moving the share price. Therefore, he argues that even so called "small company" funds have difficulty taking advantage of small capitalization stocks. Mark Hulbert of the Hulbert Digest argues in "The small-cap myth" (Forbes - 3/10/97) that after accounting for commissions when buying small stocks, there is no advantage. In the same issue of Forbes, David Dreman (in "When statistics lie") points out that the Banz study deals only with stocks from the NYSE which are larger than small stocks from other exchanges. Dreman also states that much of the data is based on stocks that traded thinly or not at all, the point being that you couldn't really buy them in large quantities if at all at their quoted price. On the impact of trading costs and liquidity on the analysis of small-cap performance, Marc R. Reinganum ² recently commented that "Several academic papers have been written on this topic, and depending upon whose you read, some have negated the fact that a small-cap effect exists. Others support the notion that, even taking the transaction costs into account, small caps carry some premium. The answer depends on how the studies are structured." See also Interrelationships.

Price changes tend to persist after initial announcements. Stocks with positive surprises tend to drift upward, those with negative surprises tend to drift downward. Some refer to the likelihood of positive earnings surprises to be followed by several more earnings surprises as the "cockroach" theory because when you find one, there are likely to be more in hiding. Robert Haugen in his book The New Finance: The Case Against Efficient Markets argued that the evidence implies investors initially underestimate firms showing strong performance and then overreact. Haugen concluded that "The market overreacts--with a lag" and that " we apparently have a market that is slow to overreact." You can check today's earnings surprises from Yahoo and Zacks. See also Technical Anomalies and Interrelationships.

There is evidence, and try as they might, the accountants and financial people can't make it go away, that when you get an earnings surprise, somehow or other the market doesn't seem to absorb it all right away.
William F. Sharpe in Investment Gurus by Peter J. Tanous

Could this anomaly be shrinking? The 4/27/98 issue of the Wall Street Journal included an article by Greg Ip titled "Big News on Your Stock? Hold On to Your Hat." The article discussed a study by Robert Butman of TQA Investors LLC that analyzed the reactions of thousands of stocks to negative and positive earnings surprises from 1995 to 1998. The results were compared with an earlier study from 1983 to 1989. The price reactions that took three to four weeks in the 80's have occurred over only two days in the more recent period. It's possible that the use of faxes and the internet have contributed to faster dissemination of information and resulted in stocks adjusting much faster to news.

Numerous studies have concluded that Initial Public Offerings (IPOs) in aggregate underperform the market and there is also evidence that secondary offerings also underperform (See IPOs). Several recent studies have also documented arguably related market inefficiencies. Bala Dharan and David Ikenberry ³ found that firms listing their stock on the NYSE and AMEX for the first time subsequently underperform (Here is the abstract of their study). Tim Loughran and Anand M. Vijh recently found that acquiring firms that complete stock mergers underperform, while firms that complete cash tender offers outperform. See "Do Long-Term Shareholders Benefit From Corporate Acquisitions?" in adobe acrobat format from the Journal of Finance. The study implies that acquirers who use their stock to effect transactions may believe the stock is overvalued.

Stock repurchases, on the other hand, can be viewed as the opposite of stock issues, and studies have shown that firms announcing stock repurchases outperform in the following years ⁴. See The Market and Stock Repurchases (10/3/94) in Pensions & Investments. See also Mark Hulbert's "Putting their money where their mouths are" in Forbes (4/21/97). This evidence seems to confirm the theory that managers tend to have inside information regarding the value of their company's stock and their decisions whether to issue or buy back their stock may signal over or undervaluation. The implication of these studies seems to be that investors may do better buying stocks of firms that are repurchasing their own stock rather than from firms that are selling or issuing more of their own stock.

There have been many studies that have documented a relationship between transactions by executives and directors in their firm's stock and the stock's performance. Insider buying by more than one insider is considered by many to be a signal that the insiders believe the stock is significantly undervalued and their belief that the stock will outperform accordingly in the future. However, many researchers question whether the gains are significant and whether they will occur in the future. Mark Hulbert reported in Insider trading (Forbes 11/3/97) that none of the newsletters he follows that focus on insider behavior have done well. The article discusses a recent study by Josef Lakonishok and Inmoo Lee. (See Bloomberg's Insider Trading and Insider focus.)

"The S&P Game" involves buying stocks that will be added to the S&P 500 index (after the announcement but before the stock is added several days later). The fact that stocks rise immediately after being added to S&P 500 was originally documented by Andrei Shleifer as well as Lawrence Harris and Eitan Gurel in 1986. ⁵ A recent article about the topic is titled "An Anatomy of the S&P Game: The Effects of Changing the Rules" by Messod D. Beneish and Robert E. Whaley in the December 1996 issue of the Journal of Finance (Here is the abstract). Opportunities may also exist with other indexes.

This page is not a stand-alone page and should not be read or used without first viewing the main Anomalies page which includes important information and warnings about interpreting historical stock market anomalies.

Anomalies | Calendar | Fundamental | Technical

Please send suggestions and comments to Investor Home

Last update 4/31/98. Copyright © 1998 Investor Home. All rights reserved. Disclaimer