Slim pickings for Russell 2000 arbitrageurs

PHILADELPHIA — Arbitrageurs looking to profit from the annual reconstitution of the small-cap Russell 2000 Index next month will have a more difficult time of it this year, according to some industry experts.
That’s good news for the average Russell 2000 investor, whose gains are reduced by the arbitrage. In fact, conditions this year are such that “there’s no such thing as arbitrage around reconstruction anymore,” said Lori Richards, a senior product manager for Russell Investment Group of Tacoma, Wash.
After all, Russell has made several structural changes to the index, she said.
This is the first year Russell will apply “percentile banding” to reconstitution. This will reduce changes by permitting existing companies that have not grown or decreased significantly in size to remain members of their existing index, Ms. Richards said.
Declining market volatility also has worked to help keep turnover down in the index, she said.
Another important change initiated in 2004 has continued to reduce arbitrage opportunities. Russell began adding initial public offerings quarterly, in addition to reconstitution, Ms. Richards said.
Before, IPOs were added just at reconstruction, making it easier for arbitrageurs to jump in front of mutual fund and exchange traded fund purchases of IPOs at reconstruction.
“I think it is true the impact and opportunity for arbitrage is smaller,” said Dina Ting, a portfolio manager with the $11.45 billion iShares Russell 2000 Index Fund, an exchange traded fund offered by Barclays Global Investors of San Francisco.
Although arbitrage opportunities may be fewer, there still are such prospects, according to Gary Gastineau, managing director of ETF Consultants LLC in Summit, N.J.

Russell has “not remotely come close to solving the problem,” he said. And as long as opportunities exist, the Russell 2000 will underperform other small-cap benchmarks, Mr. Gastineau said.
The index is the most popular small-cap index in the United States with $43 billion in benchmarked assets.
“An index like this should not be the template for a small-cap fund being sold to retail investors,” Mr. Gastineau said. “It’s going to consistently underperform its sector.”
Because of inherent flaws in the index, long-term shareholders in mutual funds and ETFs that track it will continue to underperform the small-cap market, Mr. Gastineau said.
With the exception of IPOs, Russell makes changes to its indexes on the last Friday in June. But preliminary lists of companies being added or deleted are made public about a month in advance.
This year, reconstitution changes will be effective after the close on June 22, because the last Friday in June falls before a long holiday weekend, and liquidity is expected to be low.
It’s a much different strategy than that followed by Standard & Poor’s in New York, which adjusts its indexes throughout the year on a subjective, as-needed basis and telegraphs those changes only a few days in advance.

Performance cost
Russell’s approach gives arbitrageurs a chance to plan out their strategies well in advance of the changes that are made to the index, said Vijay Singal, a finance professor at Pamplin College of Business at Virginia Polytechnic Institute and State University in Blacksburg.
It’s a particular problem for the Russell 2000, because it generally has higher turnover than the broad-market Russell 3000 or large-cap Russell 1000 indexes.
Based on data collected between 1990 and 2002, investors in funds that follow the Russell 2000 lose about 1.3% a year in performance due to arbitrageurs — and sometimes as much as 1.84%, according to a 2006 study of which Mr. Singal was co-author.
Consistent with this finding, the study found that the Russell 2000 underperformed other small-cap indexes by more than 3 percentage points a year in the 1995-2002 period, even though comparable indexes did not entail more risk.
Russell has suggested that the study “cherry picks” its data and is not a true representation of the performance of the index.
Russell’s own data indicate that arbitrage is not the problem it once was.
The effect on the Russell 2000 Index return averaged 0.65% for the period between 2000 and 2006, while after controlling for cap size, sector and momentum, the average effect was only 0.08%, according to a report the company issued in April.
Russell “also found evidence that the reconstitution effect is diminishing over time, possibly due to the decreased number of index changes in recent low-volatility years, the changes in index methodology introduced by Russell and the increase in active strategies in recent years,” the company said in the report.
By “active strategies,” Russell is referring to the use of strategies that try to generate alpha by shorting stocks, Ms. Richards said.
“It makes it really hard to figure out what’s going to happen” at reconstruction, she said.
It does appear that the number of changes to the index is diminishing.
In 2000, there were 1,120 additions and deletions, according to Russell. There were 554 last year.
It’s a trend that’s expected to continue.
Merrill Lynch & Co. Inc. of New York constructed a list of companies which, if the re-balancing had taken place April 18, would have been added to, deleted from or migrated between the projected Russell 1000 and 2000 indexes. Based on this preliminary analysis, Merrill is projecting 414 Russell 1000 and 2000 additions, deletions and migrations to occur at this year’s re-balancing — 27% less than the 564 total changes in 2006.