Economist Gary Shilling offers a radical notion.

Change in the financial services industry won't come from existing firms. Instead, just as Herb Kelleher and Southwest Airlines broke the business model in air travel, an outsider will invent the Southwest Airlines of financial services. Citing the steel and airlines industries in the February issue of his newsletter, Mr. Shilling points out how individuals and industries resist change.

"Before United filed for bankruptcy, its pilots only agreed to concessions with the understanding that their pay would jump 22 percent between 2004 and 2007 and fully restore the reductions," he writes. "American pilots want a 39 percent raise because they aren't the industry's highest paid."

Given the moribund condition of the industry, United and American pilots aren't likely to see their wishes come true.

Mr. Shilling's idea may be striking but it isn't new. In a study of innovation published over 30 years ago, John Jewkes showed that fundamental innovations seldom came from the industry most affected.   That's why a carbon paper company didn't invent Xerography, IBM didn't invent the mini or microcomputer, and book publishers didn't invent the World Wide Web.

With that in mind, let's take a look at financial services.

Three years into the worst market since the Great Depression employment in financial services is down. Industry employment peaked at 783,000 in 2000 and fell to 708,000 by the end of last year. The broad figures understate what's happening at the major brokers.   Employment at industry leader Merrill Lynch, for instance, is down nearly 30 percent.

  "… it may be very difficult for many existing institutions to cut their costs sufficiently to bring their charges in line with likely investment returns," Mr. Shilling notes. In fact, charges are rising. Fidelity Magellan, once the largest fund in the industry, has raised its charges from 60 basis points in 1999 to 88 basis points in 2002 --- in spite of trailing the S&P 500 Index over the period. While larger than most increases in the industry, it's an indication of how the industry is responding to adversity: they will add insult to the injury rather than innovate.

What we are heading for is a major collision between the fee structure of the industry and the tolerance of individual investors. Skeptics should consider how much of the investors' return the financial services industry takes:

•           In the variable annuity industry, money market sub-accounts can't earn enough to cover their average annual expenses of 1.60 percent. Average expenses of 2.07 percent for longer term fixed income sub-accounts now press current yields, taking 80 percent of the 2.59 percent yield on a 5 year Treasury and 57 percent of the 3.64 percent yield on a 10 year Treasury.

•           In broker distributed fixed income funds, the average expense ratio for "B" and "C" shares--- the ones that don't charge an up-front commission--- is 1.70 percent. This is 30 percent of the yield on high quality 10-year corporate bonds and 44 percent of the yield on current coupon GNMA mortgages. You can lower the annual expenses to 1.11 percent if you buy "A" shares but you'll have to pay a front-end load that averages 3.70 percent. That's more than a full year of interest income for most fixed income funds.

•           In broker sponsored wrap accounts, which have trended down in cost to just under 2 percent a year, investors are looking at the same problem--- the cost of managing fixed income assets is absorbing 30 to 100 percent of interest income.

•           Move to equities and the problem is, if possible, worse. Of the 4,393 funds with 5-year track records that invest in domestic equities, 2,443 have provided negative returns for the period. The average return has been an annual loss of 0.53 percent. During the same period, fund companies collected average expenses of 1.47 percent. The financial services industry gets the fee mine; the investor gets the risk shaft.

•           If yields continue to decline and equity returns run at sub-historic rates, as a growing number of academic studies suggest, an industry modeled on a cost structure of 2 percent annual fees could be taking 50 percent of fixed income yields and 25 percent of equity returns, assuming equity returns of 8 percent.

That's not likely to be acceptable to the investing public.

Bottom line?

We need Herb Kelleher to come out of retirement.