O'Shaughessy's book "Predicting the Markets of Tomorrow : A Contrarian Investment Strategy for the Next Twenty Years" is "on fire sale pricing" at Amazon -- the specific offer I pointed to is at $3.99 (with the advantage of fast free shipping if you're on Amazon Prime), others are as low as $2.53 (plus shipping, though). Anyway, for anywhere around that price, it's definitely worth getting!
The author's a guru in top-down, long-term strategies based on picking large-ish numbers of stocks based on capitalization, style (growth vs value), sector, &c; he claims that the last 20 years (1986-2006) have been historically "incredibly good" for large-cap growth stocks, and that markets revert to the mean so value stocks and small caps are much better choices for the next 20 years (he's got centuries of data to back up his contentions, too).
Of course, investors tend to think that the last 20 years "define" what's "normal, expected" behavior -- like the proverbial generals, always preparing to win the last war, most investors pick strategies that might have served them well had they been in place for the last 20 years... and are going to be woefully disappointed by applying them from here on. Whatever's least loved and appreciated now (because it underperformed recently) is going to be underpriced (exactly because unloved) and thus outperform going forwards -- that's the root of the "contrarian" in the book's title.
But but but... don't we all know that growth is the key determinant of a stock's value (through the computation of the latter as present value of the stream of all future earnings)? Nope -- that's a widespread fallacy. Growth may create value, but it may just as well (perhaps more often) destroy value -- because the pursuit of growth does not generally come for free, but rather costs capital, sacrifice on margins, and the like.
The best book I know to show this in analytical, accounting-popularization detail is Bruce Greenwald's recent but truly immortal classic "Value Investing: From Graham to Buffett and Beyond" -- really an incredibly good book (and far from costly!-) which I strongly recommend to every reader for many reasons, but what's really unique about it is exactly its cold-headed, precise analysis of growth and how much, exactly, is it worth to me as an investor.
You really have to read the book for all details and persuasive, worked-out examples, but, let me try to summarize. Growth can be indifferent, positive, or destructive of value -- it all depends on whether a company his growing "within its franchise" (with competitive barriers to competitors), in which case growth can indeed be good for the stockholder (but the amount of growth of this kind that is at all possible is delimited by the franchise's boundaries!); on a level playing field (no competitive barriers), in which case growth can at best be indifferent (negative whenever an acquirer overpays for an acquisition, negative whenever there's the slightest execution defect, ...); or against competitive barriers (against a strong and awake entrenched competitor), in which case growth destroys value for the stockholder (since the cost of capital to pay for and maintain that growth is higher than the growth's returns).
For the management of a company, growth is always great, as it enhances their power, builds up their little empires, increases their compensation (esp. the value of their generously awarded stock options): for the owners of a company, though (and that's the way you should think of yourself if you want to be a real investor, not "a trader"!-), one needs to be much more critical and selective about, which kind of growth you're buying! You can frame this as typical "agency problem": what's best for the agents (management) isn't necessarily so for the principals (stockholders).
Growth has its place -- both organically, i.e. "within the franchise", and by acquisition, should it ever happen that the acquirer doesn't wildly overpay for the acquiree (has that ever happened in the history of the world?!-) -- but nowhere as much as currently popular valuation mythology posits. (The "discounted stream of future payments" valuation model, no matter how neat and indisputable it may appear on its face, has this misconception to mark among its many other problems... but I guess that's a topic for another, future post!-).