As is the case for sausage and legislation, if you like the Dow Jones Industrial Average you shouldn’t look too closely at how it’s made.
If you fall in this queasy category, you should stop reading this column now.
For the rest of you, I start with shining a spotlight on the big change to the DJIA’s composition that will take effect this coming Monday. The picture isn’t pretty.
That change, as no doubt you’ve heard by now, replaces Exxon Mobil (XOM) - Get Exxon Mobil Corporation Report with Salesforce.com (CRM) - Get salesforce.com, inc. Report, Raytheon Technologies (RTX) - Get Raytheon Technologies Corp Report with Honeywell International (HON) - Get Honeywell International Inc. (HON) Report, and Pfizer (PFE) - Get Pfizer Inc. Report with Amgen (AMGN) - Get Amgen Inc. Report. The primary reason for the change, according to the press release announcing it from Dow Jones S&P Indices, was “Apple Inc.’s decision to split its stock 4:1, which will reduce the [Dow Industrials] index's weight in the… Information Technology sector. The announced changes help offset that reduction.”
You have every right to wonder why Apple’s (AAPL) - Get Apple Inc. (AAPL) Report stock split will reduce its weight in the overall index. The reason is that the DJIA is what’s known as a price-weighted index, according to which a stock’s weight is a function of its price.
I have never seen a convincing rationale for why such a weighting scheme makes sense. As far as I can tell, it came about for the simple reason that, when the Dow was created in the late 1800s, it was easier to add up the prices of all the component stocks. But there is no investment rationale I know of for why a more expensive stock should be given more weight than a cheap one.
One way of appreciating how silly it is to price-weight the DJIA: Ask yourself which of the Dow 30 stocks had the highest price at the beginning of this year, and thus the greatest weight. Then also ask yourself which had the lowest price. In my experience, hardly anyone has a clue. And, yet, by focusing on the DJIA, we all implicitly are drawing conclusions based on the answers.
The answers are Boeing (BA) - Get Boeing Company Report and Pfizer, with beginning-of-year prices of $326 and $39, respectively. This meant that Boeing’s performance was given many times more weight in the Dow’s overall performance than Pfizer’s. And that’s a big reason why the DJIA has struggled this year. Boeing has lost 44.9% year to date.
If price weighting is silly, which alternate scheme should be used? That’s a complicated question, and a full theoretical discussion of it is beyond the scope of this column. But the most widely-used weighting scheme is cap-weighting, according to which a stock’s weight is a function of its market capitalization. If the Dow were to have used that scheme this year, then Pfizer this year would have had 18% more weight in the index than Boeing.
The accompanying chart shows what the Dow’s performance this year (through Aug. 24) would have been had it been a market-cap weighted index. Notice that, in contrast to a dividend-adjusted return of 0.8% for the price-weighted actual version, the cap-weighted Dow would have gained 10.7%. That’s in large part because the Dow stock with the biggest market cap -- Apple -- has turned in a huge gain so far this year: 72.6% through Aug. 24.
Another popular weighting scheme is to give each component stock equal weight. As you can also see from the chart, on an equal-weighted basis the Dow 30 this year lost 3.2%.
For context, I also included in the chart the year-to-date return of the entire stock market, as defined by the combined market caps of all publicly-traded stocks. Defined this way, the market’s year-to-date gain is 7.2%.
The conclusion I draw from these widely varying returns: The DJIA’s weighting scheme potentially has a larger influence on its performance than the stocks it has in it. The same 30 stocks this year have either gained 0.8% or 10.7%, or lost 3.2%, depending on the scheme chosen.
Keep this in mind the next time you turn to the DJIA as a benchmark of the “market.” Chances are good that it doesn’t represent what up until now you had thought it does.