Outperforming stocks often rally in the final week of a quarter, thanks to window dressing
I wrote about window dressing last quarter, but the topic merits a closer look now. The third quarter has often been the best quarter to take advantage of window dressing, which is when fund managers buy the hottest stocks near the end of the quarter and sell off the stocks that have been performing poorly. They do this because the fund holdings are disclosed at the end of the quarter, so anyone delving into the holdings of a fund might be fooled into thinking the manager had the foresight to invest in the recent outperformers. Additionally, they won’t find that the fund manager made a bad bet on recently poor stocks.
In the article below, I provide a mea culpa of sorts from last quarter, and find potential short-term trades based on the theory of window dressing.
Evidence for Window Dressing
First, let's revisit what was previously shown as evidence for window dressing. Using S&P 500 Index (SPX) stocks, I grouped them based on how well they performed in the six months leading up to the final week of the quarter. According to the window dressing theory, the best performers should have a strong week, as the manager buys up those stocks. Similarly, the worst performers should do poorly in that last week of the quarter.
The table below supports this theory, as the best performers average a gain and beat the S&P 500 55% of the time. The worst performers average a 0.18% loss in the final week, with barely half beating the S&P 500.
Past Performance Doesn't Predict Future Results
Using this data, how would you have done last quarter, buying outperformers and selling underperformers? The table below shows the results. Oops. It is the complete opposite of what our theory predicted.
The worst-performing stocks averaged a gain of over 2% in the final week, with 84% of those stocks beating the S&P 500. The best performers, which we expected fund managers to buy, averaged a loss of 1.8%, with barely a third beating the broad-market index. This should serve as a reminder that past performance is not always a predictor of future performance.
Third-Quarter Window Dressing
Last quarter, I noted how window dressing wasn’t apparent in the first and fourth quarters. However, evidence of it did exist in the second and third quarters. After last quarter’s devastating results, only the third quarter shows any sign of the dubious practice.
While focusing on S&P 500 stocks since 2012, the first table below shows the average return for stocks in the final week of the quarter, based on how well they performed in the preceding six months. The first thing to notice is that the final week of the third quarter looks to be a rough one for stocks. Each of the three brackets average a negative return.
The worst performers leading into the final week perform the worst in that final week, too. They average a loss of 1.34%, with only 44% beating the S&P 500. The data doesn’t show strong buying for the best performers, either. Those stocks average a loss bigger than the loss of the "other stocks."
Individual Stocks
The tables below are stocks that fund managers might be buying or selling before the end of the month, if they’re looking to manipulate their stocks holdings. The first table shows the S&P 500 stocks that have performed the best over the past six months. If the window dressing theory holds, these stocks could see some increased buying over the next week and a half. The second table shows the worst performers over the past six months. These stocks could see increased selling pressure.
