Deletions from the S&P 500 regularly outperform new additions

William Hester, CFA
March 2005
All rights reserved and actively enforced.

It's always a red-letter day for a company being added to the S&P 500 index. With more than $1 trillion tracking the index, inclusion promises more recognition, deeper trading liquidity, and often a pop in the share price. But for every company added, one has to go. And though a majority of spots open up because an index member is acquired, there are a growing number of companies that continue to trade after being removed from the benchmark. They're often taken out because their market values have shrunk, having long fallen out of favor with investors. Or they no longer represent the fastest growing areas of the economy.

Given such dour sentiment for stocks removed from the S&P 500, the group has shown surprisingly strong returns, consistently outperforming the shares of companies that have been added to the index. Since the beginning of 1998, the median annualized return of all stocks deleted from the index and held from their removal date through March 15 of this year was 15.4 percent. The median annualized return of all stocks that were added to the index was 2.9 percent. Similarly, the average annualized return for deleted companies was 11.4 percent, while those added to the index earned just 0.4 percent.

This performance data differs from the "index effect," a trading strategy that has grown in popularity along with passive investing. Stocks added to the index often experience a rise in the price of their shares from the announcement date through the effective date of the change. Like most profitable ideas that become widely known, the index effect may be diminishing. Recent research from Standard & Poor's found that companies added to the index earned just 3.6 percent cumulative excess returns from the announcement date to their effective date over the last two years, compared with nearly 9 percent in the late 90's. The authors of the report figure that many index fund managers are spreading their purchases of the newly added stocks over time, dampening the effect.

The getting-kicked-out-of-the-index effect is a different phenomenon, and a relatively new one. That's because prior to 1996 S&P only added firms to the S&P 500 when a spot opened up, usually after a company was acquired. But now the S&P Index Committee removes stocks from the benchmark for not being "representative." This usually means that a company no longer represents its industry, or its industry doesn't represent the makeup of the economy. In 2000, more than half the total number of stocks removed from the index were Old Economy type stocks, many of which were replaced with technology and communication companies. More on how that turned out later.

S&P emphasizes that any changes to their indexes do not "reflect an opinion on the investment merits of the company." Instead the decisions - according to the methodology posted on its website - are based on market capitalization, trading liquidity, financial strength, and sector considerations. But if the stocks don't reflect S&P's investment opinion, they do reflect a pattern seen in piles of investment research over the past few decades: Neglected and underappreciated companies often have more life left in them most investors expect.

These types of stocks have at least two advantages. First, well, they're neglected and underappreciated, which means investors have low expectations for them. Second, they often have depressed prices, which makes for a potentially favorable starting point. For example, Owen Illinois dropped 84 percent in the 12 months leading up to the announcement of its December 11, 2000 removal from the S&P 500. From the announcement date to the removal date, it dropped another 23 percent to $3 a share. Following improved company and industry financials, along with the hope of a federal asbestos fund, investors have bid the stock up to $24 a share.

For stocks being removed from the S&P 500, selling pressure between the announcement data and the removal date is common, say Messod Beneish and Robert Whaley, professors at Indiana University and Duke University, respectively. In research published in the Fall 2002 Journal of Portfolio Management they looked at all of the stocks taken out of the index for not being representative. Immediately following the announcement, the abnormal return (the difference from the S&P 500's return) for the group was -6.2 percent. From that point up through the removal date, the average abnormal return trailed by another -8.3 percent.

While low expectations are an advantage for neglected stocks, companies being added to the index can be burdened by high hopes. Investors usually price growth expectations into shares long before a company actually becomes a bigger part of its industry, or its industry becomes a bigger part of the economy. This shows in the average performance of newly added companies before the announcement date. Replacement companies had average abnormal returns of 289 percent over the prior two-year period, the professors found. During the prior 12 months, those stocks had abnormal returns of 91 percent.

Annualized Returns

The prior performance of both the companies being added and those being removed from the index probably helps explain some of our results. In the table below is the median and average annualized return of each year's additions and deletions, along with number of companies that continued to trade after being removed from the index.

The median annualized returns of the removed stocks outperformed the shares of companies that were added to the index in every year going back to 1998. The median returns were positive in each year, even during the three-year bear market. The average annualized returns of the deleted companies were nearly as impressive. In five of the seven years, deleted stocks outperformed those added to the index. The companies deleted in 2004 trailed those that were added, but the majority of their companies were included late in the year so they have registered only a few months of performance.

The other year where the average annualized return of those removed from the index trailed those that were added was 1999. The average return of the deleted companies in 1999 was -10.1 percent, which trailed the -1.6 percent return of the added stocks. This is not surprising. Value strategies underperformed that year as investors chased growth stocks higher and higher. In 1999 the Dogs of the Dow, the strategy of buying the highest yielding Dow stocks each year, had its worst year in a decade, according to a Barron's article at the time. Also, only five deleted companies continued to trade that year. One of them was Laidlaw Inc., the parent of Greyhound Lines, which filed for bankruptcy 18 months after it was removed from the S&P 500.

Laidlaw's bankruptcy highlights the risk of deep value strategies, where investors seek out companies that are currently under a dark cloud. There were other companies that filed for Chapter 11 after being removed from the index, including Polaroid and Bethlehem Steel. The returns presented above include the effect of losing 100 percent of an investment in these companies. However, stocks whose bankruptcies were known at the time of their removal, including Enron and Worldcom, were not included in the calculations.

Not surprisingly, some of the biggest individual gainers were stocks that seemed like they were heading toward bankruptcy, but were able to avoid it. For example, shares in AMR Corp, the owner of American Airlines, had a 12-month gain of 820 percent following its removal from the index.

The additions to the S&P 500 at the top of the market's bubble performed particularly poorly, even considering multi-year annualized returns. In 2000, about half of the stocks that were removed were Old Economy companies. They were replaced mainly with shares of technology and communication companies. The median annualized return of those companies added to the index in 2000 has been -11.2 percent, while those deleted rose by 8.8 percent annually. The timing of when some companies were added and then deleted from the index is noteworthy. Sapient, a consultancy that helps companies sell goods on the Internet, was added to the S&P 500 on May 5, 2000, just as tech stocks began to crater. Sapient lost 96 percent of its value during its two-year residence in the big cap index. Since it was taken out of the S&P 500 for lack of representation, the shares have gained more than 60 percent a year.

12-Month Returns

Investors didn't need to wait long to earn abnormal returns from stocks deleted from the index. The median 12-month return for stocks removed from the index was 14.46 compared to the median return for the added stocks of -0.24 percent. The median return of deleted stocks outperformed newly added stocks in each of the last five years. The average return of deleted companies was 31.4 percent, while those added returned just 2 percent. (Much of the one-year gains emerged in the first six months.)

This table also shows the poor performance of the companies added to the index in 2000 and 2001. The stocks added to the index in 2000 had a 12-month median return of -22.6 versus a gain of 16.6 percent for the companies deleted from the index. The stocks added in 2001 fell -10.6 percent, while the deleted group gained 17 percent. The 2003 median return of those companies deleted from the index (85.25 percent), while impressive, came mostly from the result of two stocks: AMR Corp and McDermott International. McDermott, the builder of offshore oil-drilling platforms, rose 160 percent in the year after it was removed from the index.

The B-Team or No Team

When S&P removes a company from the index it has two options. One is to put the stock into another one of S&P's benchmarks, most often the S&P Mid Cap Index or the S&P Small Cap Index. Or it can simply remove the company from the S&P 500 without offering it a consolation prize. Since 1998, roughly half of the companies removed from the S&P 500 have ended up in one of S&P's other indexes.

You probably guessed it - the more neglected and underappreciated a company was, the better it has done, at least considering this limited sample. The 30 companies taken out of the index and placed into either the S&P Mid Cap or Small Cap index had a median return of 13.7 percent, and an average return of 5.2 percent. The 27 stocks removed from the S&P 500 index and not invited into another index had a median return of 18.3 percent and an average return of 14.4 percent. (In 2002, seven companies were removed from the S&P 500 for not being based in the U.S. These were not included in the results.)

The Dow Jones Industrial Average

With fewer members, the Dow Jones Industrial Average (DJIA) has less turnover than the S&P 500. Lately, the editors of the Wall Street Journal, keepers of the 109-year old index, have made changes every few years, swapping three or four companies at a time. The DJIA was last changed a year ago when AIG, Pfizer, and Verizon took the places of International Paper, Eastman Kodak and AT&T. There were also changes in 1999, when Microsoft and Intel took the places of Sears and Goodyear.

The results below show the performance of the two portfolios of all additions and deletions since 1997, from their effective dates through March 15 of this year. The median annualized return of stocks deleted from the index was 3.7 percent while the median return of those stocks added to the index fell an annualized 3.9 percent. The average annualized return of those stocks deleted from the index was 2.8 percent while those added to the index fell 2.6 percent.