IN INVESTING, TIMING COULD BE EVERYTHING

You’ve heard of the “October effect,” when the stock market tends to go down. And of course there is the “January effect,” when it tends to go up. Not to mention the axiom: “Buy on Friday and sell on Monday” (or is it the other way around?).

Now Frank Russell Co., a pension-fund consulting firm in Tacoma, Wash., is advancing the “TOM effect,” short for the turn-of-the-month, when Russell says stock prices tend to go up because that is when investors receive cash to put into the market.

The idea isn’t completely original. Earlier studies by analysts Norman Fosback, Yale Hirsch and Robert Ariel have come up with similar results. But the new study, by Russell analyst Chris R. Hensel and William T. Ziemba of the University of British Columbia, is one of the most extensive, covering the Standard & Poor’s 500-stock index and its predecessors from 1928 to 1993.

Defining the time

Hensel and Ziemba found that the turn-of-the-month period, defined as the last trading day of the month through the first four trading days of the next month, generated an average daily return six times that of the daily average for the rest of the days of the month.

“It appears that this happens because of the flow of funds,” says Hensel. “There’s a number of things that happen at the end of the month–people get paid, dividends are paid, principal payments,” as well as corporate contributions to pension or retirement funds.

The Russell study found average daily returns for the turn-of-the-month period were 0.1236 percent, as opposed to an average daily return of 0.0186 percent during the 65-year test period. The probability of that happening, say the authors, is below one in a thousand, making the phenomenon statistically significant.

The study follows daily closing prices on the Standard & Poor’s 500-stock index since 1957, when the value-weighted index grew to 500 stocks, and on its predecessor, the S&P 90-stock composite, since 1928.

Hensel says psychology plays a part, too. “People tend to aggregate things according to the calendar,” he says. “People observe what has happened to a stock during a month and then make decisions whether to invest or not.”

Birinyi Associates, a Greenwich, Conn., firm that monitors stock movements, finds similar results for the period 1915 to 1994. The average daily price change in the Dow Jones industrial average surges just before month’s end, when it jumps from minus 0.02 percent on day 28 to plus 0.04 on day 29, 0.12 percent on day 30, peaking at 0.19 percent on day two of the next month, then falling to minus 0.01 percent on day seven. The biggest loser, on average, is day 19.

One potential weakness in both the Russell and Birinyi data is that they represent price change only, and don’t count the dividends that an investor in such indexes would receive. One Birinyi analyst said that counting the dividends daily, which only became possible within the past few years, wouldn’t cancel out the turn-of-the-month effect.

The Russell study found evidence of other “effects” known to Wall Street, such as the January effect, which analysts have long attributed to the halt in year-end selling for tax reasons by individual investors as well as inflows of new money from corporate pension and retirement funds.

The October effect–when stocks go down–also turned up in the Russell study, with the October average daily return during the TOM period the lowest of any month. Thomas M. Keresey, chairman of Palm Beach Investment Advisers in Palm Beach, Fla., says one explanation for the October effect is that money managers tend to sell stocks during that month to raise cash before the end of their fiscal year, typically Oct. 31, in order to make once-a-year dividend payouts.

But Keresey says he ignores these kinds of effects. “It doesn’t make any difference to me. When I buy a stock, I buy it because I think it’s going to go to a certain price from a long-term perspective,” he says.

Several traders and money managers say they have heard of the TOM effect, but note that it is difficult to take advantage of it. “You can’t predict what’s going to happen to a particular stock,” says Anthony Conroy, vice president of global investment management at Bankers Trust.

Transaction costs

The Russell study’s authors take it a bit further and suggest that traders attempt to move into stocks or stock futures just before the turn of the month, and back into cash afterward. However, as the study acknowleges, their suggestion doesn’t include the impact of transaction costs, which would likely be considerable.

Hensel even tried doing something similar–shifting from stocks to cash and back–with the S&P index funds and money market funds at the Vanguard mutual fund group. But only until the Valley Forge, Pa., large mutual fund family caught on. “I was overtrading according to their rules, so they basically asked me to stop,” he says.

However, he says that the average investor can still time a monthly mutual fund contribution check, putting it in the mail later in the month, say around the 20th, so it will go into the fund just before stock-price returns historically turn up.

More Top Picks Inboard M1