That's an interesting paper that I haven't run across before, but the numbers cited in that paragraph are not annualized returns - they're the range
in returns between the best and worst-performing periods.
For stocks, as noted, the spread between the best decade (17%) and worst decade (-4%) is 21%. One would expect the spreads for bonds and bills to be lower, and they are, at 18% and 17%.
But what the authors are pointing out is that over 20-year and 30-year periods, a portfolio of stocks is less risky
than a portfolio of bonds or bills, because the multi-decade return range for stocks actually falls below the range for bonds and bills.
Thus, over a 30-year time horizon, the best and worst period returns are only 8% apart for stocks, compared to 9% for portfolios of bonds and bills. Stocks are now less risky than bonds and bills. This section of the paper is an endorsement of stock portfolios, because their riskiness drops much more than debt securities as one's time horizon widens.
Nowhere do the authors give the actual average returns of stocks and debt instruments over annual, 10, 20, or 30 year periods. I could find it easily with the statistical software we use at work, but I won't have access to Pertrac for some time.... I'll try to look it up, though.
But I would agree that actively managed funds are risky propositions. Every year, The Wall Street Journal picks a stock portfolio by throwing darts at stock pages; their random portfolios outperform 39% of actively managed funds. There's a more thorough explanation here
They attribute this conclusion to the "random walk" theory, which means that the market is inherently unpredictable over short periods, so active management is no more likely to generate good returns than picking any given portfolio (and considering trading and management fees, will likely underperform a "passive" portfolio).
The implication of this finding isn't to avoid the stock market. It's to invest for the long term in a well-diversified portfolio or in index funds that track an entire market. Mutual funds, day trading, and other active management techniques may well produce lower returns and cost you more than buying a Spyder and sitting on it. This doesn't mean that there aren't good money managers who can consistently outperform the market; I work for a company that, since 1984, has beaten the S&P's return with half its level of volatility. You just have to look hard for the consistent winners and keep close track of what they're doing.
[Edited 2004-01-13 04:50:04]
Keynes is dead and we are living in his long run.