In the Stock Indexes We Trust?

So, the news buzz last week was that the New York Stock Exchange (search) is about to enter the modern world by going public and introducing electronic trading to accompany its traditional floor trading.

The major reason is competition from other major electronic markets, like Nasdaq (search) and Instinet.

But another subtler reason is trust. The recent charges against 15 floor traders have caused people to question whether they can trust the NYSE to execute their trades fairly. Having trust in the fairness of the Exchange is analogous to having trust in our stock market barometers. Do they tell us how the markets are really doing?

We believe they do everyday, but is it possible that the fluctuating U.S. dollar is "gaming" the markets, which would make them less-than-accurate barometers?

In the financial world, bias can creep in — even into the composition of the benchmark indexes — just like Carl Sandburg's (search) "fog that comes on little cat feet."

Take, for example, a chart that ran on the business page of a major daily newspaper on Jan. 2. It showed that most of the companies listed in the S&P 500 were actually doing much better than the index itself, which was still down 18 percent5 from its 2000 peak at the time. Because the index is weighted by market capitalization, and because a few large telecommunications and technology stocks remain down — such as Cisco, Intel, and Lucent — they drag down the whole index.

So the chart showed the fanciful path the S&P 500 would take without this weighting – that is, if all the 500 company stocks (including Starbucks and eBay) were equally weighted. In this scenario, the index would have gained 36 percent since 2004, to 54 points higher than the actual price. This interpretation suits those who want to view the market through a bullish lens, but is it unbiased? No. It is biased according to stock price, while the S&P 500 is "biased" according to capitalization.

If financial market indexes can be biased or even manipulated in such a not-so-subtle way, what about distortions that are subtle, which most of us cannot spot so easily? Consider currency fluctuations, for example. They can affect the apparent value of a market index. For instance, a weekly chart showing the S&P 500 denominated in U.S. dollars looks much different from the same index denominated in euros. Click on the thumbnail photo above to view the chart.

In both charts, the big hump looks about the same. But look at what happens since 2002. In the dollar-denominated chart, the S&P 500 Index increases about 50 percent. But in the euro-denominated chart, the S&P 500 goes from a low near 700 in March 2003 to about 900 in early 2005, an increase of less than 30 percent. What makes the difference? The dollar’s massive decline.

Viewed through this lens, more than 60 percent of the S&P’s rally over the past two-plus years can be explained by the dollar’s decline in value vs. the euro’s rise. And with the U.S. dollar having dropped spectacularly against most currencies over the past two years, it's clear that the rally in the S&P 500, in non-dollar terms, is nowhere near what it seems to those of us in the USA.

Stable Currency Benchmark

The question becomes then, how can we correct such distortions? To answer that question, take a quick quiz:

Which of these statements does not belong?

A.) To calculate the amount of rainfall, it's best to use an unchanging measure like a rain gauge.

B.) To measure the length of a car trip, it's best to use an unchanging measure like miles or kilometers.

C.) To quantify the S&P 500 Index, it's best to use a changing measure like the U.S. dollar or the euro or the Japanese yen.

It's easy to see that (c) doesn't belong with the others. Since no immutable measure exists to judge stock indexes (no doubt, if one did, we would already be using it), the next best strategy is to create a valuation benchmark that is at least more stable and less changeable than a single currency.

Does this kind of measure of value exist yet? Yes, it does. Technical analyst Robert Prechter of Elliott Wave International has created a new benchmark, which he calls the Stable Currency Benchmark (TM) (SCB), that includes equal-value portions for four currencies: the Swiss franc, the Singapore dollar, the New Zealand dollar and the U.S. dollar.

The point of the SCB is to create a measure that reflects stable global purchasing power. When one currency is weak, the others will usually be strong. So fluctuations tend to cancel one another out, leaving a consistent and truly global benchmark of value. The SCB speaks to those who recognize that the dollar is an imperfect unit of account by keeping it as one of four currencies, though not as the primary currency.

The SCB creates a new aggregate currency that anything in the world can be measured against. For instance, you can use it to factor out fluctuations and see truly comparative values since 1988 in the S&P, the Nikkei and the FTSE indexes against the SCB. Click here to view the chart.

Restoring Trust

Why does using a stable benchmark matter? It matters because all stock markets are global markets now, attracting a good deal of foreign investment. When we invest in other markets, the picture we see in local currency terms may be much different from true value. We can't be sure if our gains or losses are real or fiction.

Most people don't realize how the recent collapse in the U.S. dollar's value throws off our ability to accurately measure the changes in U.S. and global financial markets. But once the fog created by our fluctuating-dollar measure dissipates, we can see more clearly whether our stock market indexes are worthy of our trust.

Susan C. Walker writes for Elliott Wave International, a market forecasting and technical analysis company. She has been an associate editor with Inc.magazine, a newspaper writer and editor, an investor relations executive and a speechwriter for the Federal Reserve Bank of Atlanta. She received her B.A. in Classics from Stanford University. For more information about the Stable Currency Benchmark, please visit