The Washington PostDemocracy Dies in Darkness

The Dow Jones industrial average is ridiculous

Let's make that "S&P 2000." (Richard Drew/Associated Press)

The big (symbolic) news out of Wall Street today is this: Three of the 30 members of the Dow Jones industrial average are changing. Alcoa, Hewlett-Packard and Bank of America are out. Nike, Visa and Goldman Sachs are in.

It is tempting to use this occasion as an opportunity to examine the symbolism of what this says about Corporate America, and the global economy. Aluminum (Alcoa) is out! Athletic wear is in! Down with plain old conventional banking! Up with fancy high-finance investment banks.

But in reality, it really only shows the utter uselessness of the Dow Jones industrial average for measuring anything. It is an accident of history that the Dow is the most widely cited measure of how the overall stock market is doing, and a bad accident.

Let us count the ways: The Dow includes only 30 stocks. They are chosen by the fine people at S&P Dow Jones indices, intended to represent that breadth of the American stock market. They are weighted not based on the size or importance of the company, but by its per-share price. So IBM, with its $184 per-share price, counts more than four times as much as Coca-Cola, at $39 a share, even though the two have about the same stock market capitalization.

A century ago, that made sense. People had to calculate the value of the index by hand. In that context, limiting the list to 30 stocks, and using their per-share price to calculate the index, made a lot of sense. The index became entrenched as the most widely referenced measure of how the stock market was doing, in no small part because it was part of the same corporate empire as the nation's leading business newspaper. The Wall Street Journal had every incentive to emphasize the Dow as a measure of how the market was doing, and to duly celebrate various milestones in its pages (Dow 1,000 arrived in 1972; 10,000 in 1999). The rest of the business media followed.

The announcement of the changes shows the absurdity of the index. "The index changes were prompted by the low stock price of the three companies slated for removal and the Index Committee’s desire to diversify the sector and industry group representation of the Index," the company said.

Of course, the per-share price of a stock has absolutely nothing to do with its size, importance or representativeness. Bank of America is being booted, it would seem, for its sub-$15 per-share price, in favor of Goldman Sachs with a $164 share price. But Bank of America is a way bigger company! Its total market capitalization is $155.6 billion, to $74.5 billion for Goldman. It has 257,000 employees, to 32,000 for Goldman. It is engaged in banking and lending activity in basically every community in America, as opposed to Goldman's specialty investment banking business. But when you find yourself in the archaic trap of weighing companies based on their per-share price, that's the kind of absurdity you end up with.

The case for swapping Alcoa for Nike is stronger — the latter company has a $59 billion market cap to Alcoa's $8.5 billion — but this shows the opposite problem. What took so long? Alcoa has been a relatively small company for years, no longer what one would consider to be among the 30 most important companies or stocks. But the Dow mix is changed only rarely (the last time three members were changed at once was 2004), so it has lingered long past its time as a symbol of American industrial might.

A century ago, it made perfect sense to have an easy-to-calculate index that captured the overall trends in the U.S. stock market, even if it was a little arbitrary and weird. But we are living in an age where magical devices can instantly perform calculations on hundreds or thousands of stock prices at once. And we have the tools to select and weight stocks in an index based on their size — typically, as measured by their total stock market value v rather than the arbitrary per-share price.

In other words, we have the technical capacity to create and monitor the Standard & Poor's 500 index. Or the Wilshire 5000. Or really anything other than the Dow Jones industrial average. So that's what we should all do.