A new art from the calendar: how to time your stock purchases

Imagine an investor eyeing a dusty photo calendar tacked over his basement workbench. ``Jumpin' Jehoshaphat! If 'tain't Friday and tomorrow's New Year's Day. Better git my broker on the horn and tell 'im to buy.''

Stock purchases based on calendar dates? Silly, right?

Well, in its quest to discern profitable patterns, Wall Street has been known to generate some fanciful folklore. But from the hallowed halls of academia comes new evidence that this method could have reaped fat payoffs in the past.

Wall Street, it appears, celebrates the day before a holiday. From 1963 through 1982, on average, more than one-third of the annual gains by the stock market accrued in just eight trading days.

Those returns were posted the trading day before New Year's, Presidents' Day (George Washington's Birthday), Good Friday, Memorial Day, Independence Day, Labor Day, Thanksgiving, and Christmas.

This ``pre-holiday effect'' is chronicled in a study by Robert A. Ariel, a doctoral candidate at the Sloan School of Management, Massachusetts Institute of Technology (MIT).

By far the greatest advances occurred before New Year's. But more stocks took part in the Independence Day ``rally,'' though returns were slightly lower. The day before Labor Day offered the next best returns for most stocks.

Mr. Ariel's study, as yet unpublished, backs up findings of Norman G. Fosback, president of the Institute for Econometric Research, Fort Lauderdale, Fla. Ten years ago, Mr. Fosback published a similar study of Standard & Poor's 500 stocks from 1927 to 1975.

``Alas, the rest of the world is now waking up,'' writes Fosback in a recent Market Logic newsletter. The pre-holiday trend, however, ``has not been as prevalent in the last 10 years as it has in the last 50,'' Fosback notes.

But don't discard that calendar. You have yet to pick which month, which day of the week, and what time to trade.

January stands as the best month for stock gains over the years. The widely known ``January effect'' actually begins on the last day of December and continues into first few weeks of January. Small companies tend to chalk up the biggest gains -- on average, 38 percent of their annual return, according to a study by Richard Roll, published in the winter, 1983, Journal of Portfolio Management.

January also sets a pattern for other months: The best returns are made in the first half of the month. From 1963 through 1981, ``all of the market's cumulative advance occurred around the first half of the month, the second half contributing nothing to the cumulative increase,'' says a 1984 MIT Working Paper published by Ariel.

Which day to trade? If you tend to arrive at work on Monday with a bearish outlook, take heart; so do most investors. On average, stock returns at Monday's open are significantly lower. Perhaps in anticipation of weekend merrymaking, stocks tend to close higher on Fridays. So if you can hold off selling until Friday, you might make a tad more money. This according to studies by Kenneth R. French, in Vol. 8 of the Journal of Financial Economics 1980 and by Michael R. Gibbons and Patrick Hess, Journal of Business, Vol. 54, 1981.

And you know that adage about the early bird getting the worm? It seems to apply to stocks, as well. Within a trading day, stocks on average make the highest gains in the first 30 minutes of trading (except on Mondays, when bears rule). But there is some compensation for the pillow-bound. During the last five minutes of trading, returns are also higher relative to the rest of the day. This is according to a study by Thomas H. McInish at the University of Texas, Arlington, and Keith Ord and Robert A. Wood at Pennsylvania State University. Results appear in the July Journal of Finance.

The flood of research on this subject has been spurred by a practical use for it (stock index options that allow speculators to bet on the direction of the market) and by some avid interest in punching holes in the ``efficient or random-market'' theory. The theory says that since stock prices immediately reflect all that investors know or expect to happen to a company, any system for choosing stocks is no better than throwing darts at stock listings.

If stocks move in January or on Friday or whatever, why do they?

The academics are still scratching their heads. ``We're just at the point of discovering these anomalies, we haven't yet explained them,'' says Mr. McInish.

Before holidays and on Fridays, ``It may be that traders don't want to go into closed periods with risky short positions. So there may be a high level of short covering before a closed period,'' speculates Fosback. And early-in-the-month gains may be related to receipt of monthly paychecks. But no one knows.

And ``If we could rerun world history 20 times, we might find out if this is just a statistical aberration,'' says Terry Marsh, a finance professor at MIT and researcher of these patterns.

Last week there were no fireworks in the Dow Jones industrial average before the Independence Day close (advances led declines overall, though). But the Dow rebounded Friday to close at 1,334.45, down 1.01 in four trading sessions. Chart: Interest Rates. *Yields; Source: Bank of Boston.

Percent Prime rate 9.50 Discount rate 7.50 Federal funds 8.06 3-mo. Treasury bills 7.00 6-mo. Treasury bills 7.08 7-yr. Treasury notes 10.05* 30-yr. Treasury bonds 10.27 *6-28-85 figure