Maybe it's time for a new stock index -- one that will give a better picture than the current indexes do of how the average investor, or the average portfolio, is doing in the market at any point in time.
That was the idea behind the first indexes. They were a shorthand way for investors to know roughly how their portfolios were doing.
We have plenty of indexes of the U.S. stock market now - from the Dow Jones industrial average to the Standard & Poor's 500 stock index, to the Russell 3000 and its component parts, to the Wilshire 5000.
But none of these is completely satisfactory, in part because most are capitalization weighted (some total capitalization and others float-adjusted). When a particular sector becomes overheated, it can dominate the index.
The classic example is the most recent - the impact of the tech stock sector on the S&P 500. The technology sector accounts for just 7% of the gross domestic product, but 36% of the S&P 500. Partly as a result of that weighting, in 1999 the S&P showed a total return of 21%, but the median money manager in the PIPER universe was up only 18.2%. This is because most of the gain in the S&P 500 was accounted for by just a few of the stocks in the index. Only about half of the S&P 500 stocks were up or flat in 1999. The others were down.
Many pension executives still use the S&P 500, or the Russell 1000 or 3000, when measuring the performance of their funds' total domestic equity portfolio. Many have been disappointed with their equity performance, compared with the S&P.
Ken Safian, president of Safian Investment Research Inc., White Plains, N.Y., became frustrated enough with the distortions of the cap-weighted indexes that at the beginning of 2000 he published his own benchmark -- the Safian Market Benchmark.
Mr. Safian, a veteran analyst who, with the late Ken Smilen, first split stocks into growth and cyclical classifications, argued a new index was needed because the S&P does not accurately represent the stock market. It also has had numerous component changes over the years, making it practically, if not useless, for historical analytical purposes.
His new index uses an adjusted GDP weighting. Mr. Safian and his researchers went through the current-dollar GDP line by line and classified each item (except such things as government spending) to a market sector.
After determining each market sector's GDP weight, the team adjusted each sector's weight for rate of growth, so the weights for fast-growing sectors, such as technology, are higher than their GDP weights. But to prevent any sector's getting further overweighted by the market price action during the year, each sector's weight is rebalanced annually.
But why weight the market sectors in the index by the sectors' shares of the GDP? Because, said Mr. Safian, numerous academic studies have shown the economy has significant influence on stock market returns.
Mr. Safian and his colleagues calculated the returns of the new index, without dividends, from 1995 on and compared them with the returns of the S&P 500 and the Wilshire 5000 indexes. They found the returns of the three indexes were very close until 1998, when a select group of large-cap growth stocks and the technology sector caused the S&P to outpace both the Safian index and the Wilshire 5000. In 1999, when small-cap technology stocks surged, the Wilshire 5000 outperformed the Safian index and the S&P 500.
However, in 1997, 1998 and 1999, the returns of the Safian index were closest to the returns achieved by the median money manager - just what Mr. Safian was trying to achieve.