Imagine ringing in the new year in New York City’s Times Square with a million of your closest friends. Shortly after the ball drops, everyone is ready to leave—at the same time. Demand for a ride in that area soars, and so does the price for the suddenly coveted service. This dynamic surge is a lot like the index reconstitution effect, an important yet less-visible source of costs associated with index investing.
Index funds seek to mirror the performance of an index. To do that, they need to mirror the changes in the index’s holdings at the time of its reconstitution, when stocks are added to or dropped from the index. This lack of flexibility can lead to increased trading volume and price pressure around that reconstitution, the so-called index reconstitution effect.
Two of the most widely tracked indices in the US are the S&P 500 Index and the Russell 2000 Index. At the end of 2020, nearly $5.4 trillion in assets tracked the S&P 500. This means for every $1 invested in S&P 500 companies, 17 cents were indexed.1 An additional $191 billion in assets tracked the Russell 2000 as of 2019, meaning about 10 cents indexed for every $1 invested in Russell 2000 companies.2 With all of these assets tracking the indices, collective movements to mimic index changes may materially impact a security’s trading volume and price.
To examine the impact of index reconstitution on traded volumes and prices, we identify stocks that were added to or deleted from these two indices at an index reconstitution event over the past 10 years, from 2012 through 2021.3 Our sample includes 2,998 additions and deletions in total, with 50 additions and deletions coming from the S&P 500.
Exhibit 1 presents the value-weighted average trading volume for rebalanced stocks as a multiple of the stocks’ volume 40 trading days prior to reconstitution. For both the S&P 500 and Russell 2000, additions and deletions experience a large increase in trading volume around reconstitution, as much as 20 times higher for the S&P 500 and over 30 times higher for the Russell 2000, relative to volume before the reconstitution period.
Since market prices are forward-looking and index rebalances are announced before reconstitution dates and often anticipated before announcement dates, the price impact due to index reconstitution often begins to accumulate before the actual reconstitution date. Indeed, we find that stocks that are added to an index tend to go up in price prior to rebalance, while deletions tend to go down.
Exhibit 2 presents the value-weighted average cumulative excess return to added or deleted stocks relative to the index. For both indices, we observe an increase in cumulative excess returns from the announcement date to the reconstitution date, followed by a reversal in excess return in the days following, on average. For example, for S&P 500 additions and deletions, the average cumulative excess return climbed to almost nine percentage points over the five days prior to reconstitution and dropped by more than two percentage points the following day.
Index investing has grown considerably in recent decades, with US equity index funds representing 52% of the US equity fund market at the end of 2021.4 While index funds typically offer lower expense ratios than actively managed funds, our study shows that there are potential hidden costs associated with rigidly following an index. Specifically, index fund investors end up buying index additions at high prices and selling index deletions at low prices. These results highlight the importance of flexibility and efficiency in portfolio design, portfolio management, and trading. A systematic daily investment process that rebalances thoughtfully and flexibly can help investors avoid the costs associated with demanding immediacy, while still providing many of the benefits of index investing.