Blog - Investing Notes

September 14, 2007 - Update: End-of-Quarter Effect

Does the stock market offer a predictable pattern of returns around the ends of calendar quarters? Do funds deploy cash to bid stocks up before quarter ends to boost portfolio values at the end of reporting periods (with subsequent reversal)? Or, do fund liquidity needs tend to drive stock prices lower before ends of quarters? Is the end-of-quarter effect the same as the turn-of-the-month effect? To address these questions, we examine daily stock market returns from 10 trading days before to 10 trading days after the ends of calendar quarters. We also compare daily returns with those for turns of calendar months. Using daily closes for the S&P 500 index since 1/3/50, we find that...

The following chart shows the average daily returns for the S&P 500 index, with one standard deviation error bars, from 10 trading days before to 10 trading days after ends of all calendar quarters since 1950 (230 quarters). The average daily return of 0.03% for all 14,516 trading days in the sample serves as a benchmark. The chart suggests possible systematic weakness a few days before ends of quarters and possible systematic strength the first few days after ends of quarters.

As usual, differences between average daily returns and the average daily return for the entire sample are small compared to standard deviations of returns, making it very risky to trade potential anomalies.

Are results different for different calendar quarters?

The next chart shows the average daily returns for the S&P 500 index (without error bars) from 10 trading days before to 10 trading days after ends of each of the four calendar quarters since 1950 (57-58 quarters for each). Note that the scale differs from that above.

The Q4 (end-of-year) graph is least like the others, showing earlier gains. The Q2 and Q3 graphs are most alike.

Does a recent subsample confirm the patterns for all ends of quarters and for the ends of each of the four quarters?

The next chart compares the average daily returns for the S&P 500 index from 10 trading days before to 10 trading days after ends of all calendar quarters since 1950 with those since 1990 (70 quarters). While there are similarities, the daily returns for the recent subsample are more volatile, casting doubt on the reliability/persistence of the pattern for the entire sample.

Recent subsamples for the four calendar quarters separately (not shown), each with only 17-18 observations, are likewise more volatile than their longer-term counterparts, emphasizing the risk of depending on such calendar effects for trading.

Is the overall end-of-quarter pattern simply a turn-of-the-month effect?

The final chart compares the average daily returns for the S&P 500 index from five trading days before to five trading days after ends of quarters to those for ends of months, with (693 months) and without (462 months) ends of quarters. Evidence suggests that ends of quarters delay the turn-of-the-month effect by one or two days and sharpen it.

In summary, evidence suggests some weakness before ends of quarters and excess returns in the few days after ends of calendar quarters, but the effect is of low reliability and therefore risky to trade. The fourth quarter pattern is the most unique.

For related research, see Blog Synthesis: Calendar Effects and the Trading Calendar. See especially our blog entry of 7/20/06 on the turn-of-the-month effect.