Managers of growth-oriented mutual funds are declaring that the tide has finally shifted away from value to growth.

I wouldn't bet on it.

I admit that recent experience certainly supports their argument. Over the three months through October 31, the average growth stock has gained 9.73 percent, in contrast to 7.29 percent for the average value stock (as judged by the CBOE Russell 3000 Growth (TICKER:RHN) and CBOE Russell 3000 Value (TICKER:RHK) indexes).

This advantage in favor of growth is in marked contrast to the period dating all the way back to the beginning of the bear market in March 2000. Between then and this past July 31 (the beginning of this most recent three-month period of growth beating value), the Russell 3000 Growth index lost more than 42 percent, in contrast to a 56 percent gain for the Russell 3000 Value index -- an extraordinary spread of nearly 100 percentage points in favor of value.

But, there are at least two reasons to be skeptical of the argument that happy days are here again for growth-oriented investment strategies. The first: This isn't the first time that managers of growth funds have been declaring their imminent resurrection, and for most of the last decade, they've been wrong. Why should we believe them now? After all, their job is to sell their approach through thick and thin.

The second, more fundamental, reason to be skeptical: I can find no statistical basis for the belief that growth’s relative strength should continue.

To investigate the quarter-by-quarter patterns in growth’s relative strength, I turned to the database maintained by Dartmouth finance professor Kenneth R. French and University of Chicago finance professor Eugene F. Fama. This database is the standard academic database for the relative performance of growth and value and extends back to 1926.

For each three-month period since then, I calculated the relative performance of growth stocks versus value stocks, as defined by Professors Fama and French. For each of these periods, I then calculated how growth performed relative to value over the subsequent three months.

I could detect no statistically significant relationship between growth’s relative strength in one three-month period and in the subsequent one. This means that we can conclude nothing about how growth will perform relative to value over the next three months on the basis of growth’s relative strength over the last three months.

I reached the same conclusion when focusing on large-cap stock category, on which most mutual funds focus.

These findings should have been evident even without my going to the trouble of running several econometric tests. A cursory review of the data is all that is needed to reveal that, even though growth has greatly lagged value this decade, a number of shorter periods along the way outperformed value in terms of growth. And, yet, none of them meant that the tide had turned.

Consider the first quarter of 2003, for example. Over that three-month period, according to the Fama-French data, the large-cap growth sector outperformed the large-cap value sector by 8.7 percent -- by an even greater margin, in other words, than growth beat value over the last three months.

But, this did not presage a renaissance for the large-cap growth sector, despite the widespread arguments to the contrary, at that time, by managers of large-cap growth funds. Over the second quarter of 2003, the large-cap growth sector lagged the large-cap value sector by 4.9 percent.

Note carefully that the econometric tests I ran do not mean that growth will not outperform value over the next three months. Instead, the proper conclusion to draw is this: Whether or not growth outperforms value between now and the beginning of next year has nothing to do with how it did over the last three months.

The tests I ran prompted me to compile a list of the stocks, currently recommended by the newsletters, that had outperformed a buy-and-hold over the last 15 years. I chose 15 years because this is long enough to be fairly evenly balanced between years, where growth outperformed value and where the reverse was the case.

Eight stocks are currently recommended by at least four of the newsletters in this select group of market beaters. For each of the eight, I determined how Russell, Inc., the keepers of the Russell 3000 Value and Russell 3000 Growth indexes, categorized it along the growth versus value dimension. (Note that Russell doesn't require a stock to be either all value or all growth, but anywhere along the spectrum between these two polarities.)

The eight stocks are:

On average, these eight stocks are spaced about half-way between the extremes of growth and value.

The best performers, in other words, are not betting all or nothing on either one or the other of these two styles.