Q. I've got a question for you and other indexing enthusiasts: When is an index grossly overvalued, how do you measure it, and would you sell or simply not buy more?   I don't think any index strategy will work if one continuously ignores market valuation.

---P.R., by e-mail from Dallas


A. Index strategies don't work because of market timing. They work because not making investment decisions works better than making investment decisions. No one on Wall Street wants to hear this because they make their living by making investment decisions and charging us for it.

The end result of Wall Street investment decisions is great for them (commissions, management fees, free trips, expense paid bachelor parties, etc.). But the people who put up the money--- you and me--- rarely benefit.

There are several reasons for this. First, it costs a lot of money to have someone make investment decisions. Wall Street has been charging all the traffic will bear for decades. Second, it costs money to execute those decisions--- in commissions and market maker spreads.

How this continues, year after year, is beyond me. If a group of surgeons learned (however reluctantly) that a procedure harmed their patients about 70 percent of the time, they would be duty bound to stop performing the procedure. Wall Street just keeps doing its "procedure," regardless of efficacy, because it makes their Mercedes payments.

When it comes to market timing, more decision makers claim success than achieve it. Many sell far too early. More sell far too late. There is no evidence that decision makers add any value by way of market timing decisions--- indeed the marching orders of most investment managers, institutional and retail, is to remain invested at all times.

As a practical matter, both managed and indexed portfolios have some protection against overvalued markets. That protection is called asset allocation.   If your portfolio is allocated 50 percent to equities and 50 percent to fixed income and the stock market returns 30 percent while the bond market returns only 5 percent, your portfolio won't be 50/50 after a year. It will be 55.3 percent equities, 44.7 percent fixed income. To rebalance the portfolio some of the high performing asset will be sold and some of the lower performing asset will be purchased.

Between 1995 and 1999 the U.S. Stock market soared every year. It beat the return on bonds by at least 20 percentage points each year. As a consequence, a regularly rebalanced index portfolio would have sold stocks in each of those years. In the bear market years 2000 through 2002 the same portfolio would have been rebalanced by selling bonds and buying stocks. It's not a perfect method, but it works better than market guessing.


  Q. My company was recently bought by a larger corporation, causing significant changes in my benefits. Of particular concern is my 401k retirement plan. Our new plan is based on Institutional funds. The other plan was based on mutual funds. In addition, we have only six choices and there is virtually no information on exactly how the funds are managed or by whom.

What is the difference between institutional funds and mutual funds? Are they managed differently? None of the new funds are rated any higher than 4 stars by Morningstar and it appears that the institutional funds have lower rates of return than comparable mutual funds. Is this normal?

---S.M., by e-mail


A. I think you may be looking at this through loyalty-tinted glasses. The biggest single liability of 401(k) plans is that they developed around a retail mutual fund marketing structure rather than the institutional structure that served big corporate pension plans. As a consequence, 401(k) plans tend to have much higher costs than traditional pension plans. One result: 401(k) plan performance tends to trail institutional fund performance by about 2 percentage points a year.

As 401(k) plans have grown into gigantic pools of assets, plan sponsors (like the corporation for which you work) have worked to reduce costs by moving from the expensive retail funds to less expensive institutional funds. The differences can be substantial. A retail investor in PIMCO Real Return fund, a single fund that sells through a variety of marketing channels, will buy "A" shares with a 3 percent front-end load and an annual expense ratio of 0.90 percent on a minimum initial purchase of $5,000. The institutional shares of the same fund have no commission and an annual expense ratio of only 0.45 percent--- half as much.

Alas, to get the institutional shares, you'll need to write a check for at least $5 million.

Odds are you will soon get plenty of information about who manages the money and its performance. My bet is that you'll be pleased with the lower expenses and the slight improvement in performance.