MW DeepSeek panic shows the S&P 500 isn't the safe investment people think

By Brett Arends

Monday's plunge for the S&P 500 and Nasdaq is a warning to check your portfolio for this major mistake

Do you want the good news first about Monday's stock-market turmoil, or the bad news?

The good news is that this could easily be a one-day storm in a teacup and that in a couple of months, or even weeks, you'll have forgotten all about it.

Look at the 2% fall in the S&P 500 SPX, and the 3% fall in the Nasdaq COMP, on any long-term chart of the stock market - even one of just five years - and you can barely see them. By any measure that should matter to a long-term investor, these numbers are trivial.

The bad news, though, is that this turmoil should be a major wake-up call for investors.

The danger with 401(k)s, IRAs and other retirement accounts around the country isn't from Chinese AI competition like DeepSeek, or even from popularity of the so-called Magnificent Seven MAGS of Big Tech growth stocks. Or even from the new administration in Washington and, say, the risks of a trade war.

The danger comes from portfolio diversification. Or, more accurately, the lack of it.

The U.S. stock market has reached a ludicrous situation where just seven stocks account for one-third of the entire S&P 500 index. Most people investing in an index with the number "500" in it probably assume that they are diversifying their investments across 500 stocks.

You are - but only just. That's because the S&P 500 doesn't allocate your money equally across all the companies in the index, but allocates it based on their stock-market valuation. That means that the companies that sport the highest theoretical valuations, based on their booming stock prices, get a lot more of your money than the others. This concentration always happens in these indexes, but right now it's off the charts.

Nvidia $(NVDA)$ and Apple $(AAPL)$ alone account for 7% of the index each. The entire Magnificent Seven - including Microsoft $(MSFT)$, Meta $(META)$, Amazon $(AMZN)$, Alphabet $(GOOG)$ and Tesla $(TSLA)$ - add up to about 33% of the entire index. You thought you were betting your retirement on the fortunes of 500 companies, but one-third of your money is on just seven.

Oh, and they are all in similar fields, or at least are closely correlated. Congratulations. As Dirty Harry would say: Do you feel lucky?

As of Friday, anyone with money in an S&P 500 index fund had more of their money (6.8%) in Nvidia stock alone than they did in the bottom 200 companies in the index.

And they had more money in the Magnificent Seven stocks than they did in the bottom 400 companies in the index put together.

This is insane.

At the moment, anyone buying the index is betting more than 400 times as much money on one stock, Nvidia, as they are on such obscure, fly-by-night, minor companies as ... er ... Walgreens Boots Alliance $(WBA)$, Campbell's $(CPB)$, Hasbro $(HAS)$, MGM Resorts International $(MGM)$ or Caesars Entertainment $(CZR)$. You've got more than 500 times as many dollars bet on Nvidia as you do on Brown-Forman $(BF.B)$, maker of a niche brand of bourbon called Jack Daniel's that nobody has ever heard of.

I have no specific view on any of these stocks. Please don't bother writing in to explain why Nvidia is a better investment than Jack Daniel's. For all either of us knows, you may be right. But is it 560 times better? And if so, doesn't that actually mean the market isn't even remotely efficient? How can the market be setting prices correctly if one stock is 560 times a better bet than another?

The strategy of index funds would be like finding a roulette wheel where the odds favored the gambler, rather than the house, but then placing hundreds of chips on some numbers and only one or two on others. It makes no sense.

About 20 years ago, the late Lynn Stout, a professor of law at Cornell University, raised serious questions about the theory of "market efficiency" anyway.

There is an obvious alternative to investing in an index weighted toward the most expensive stocks: One that weights all the members equally. While supporters of "equal-weight" indexing sometimes call it a form of "smart beta," or even "value investing," maybe the real magic is simply its neutrality. You get the maximum diversification across the range of stocks. By this process, your Nvidia bet is 1/500th of your entire bet on the S&P 500. So, for that matter, is every other bet.

Does it work? You bet.

Over the past 25 years, MSCI's equal-weight U.S. index has outperformed its regular, size-weighted index by an average of 1.2 percentage points a year. Based on the gross returns of the two indexes, someone who invested in the equal-weight index at the end of 1999 would be one-third richer today than the person who invested in the regular size-weighted index.

This is true even though the equal-weighted index has done worse than the regular index for the past 10 years. (This is based on total shareholder returns, including reinvested dividends.)

MSCI also has price data, excluding dividends, for the equal-weighted index going back to the mid-1970s. A direct comparison with the regular size-weighted index shows that the equal-weighted index has won handily over half a century, by an average of just over 1 percentage point a year. Equal weight did better in two-thirds of all five-year periods.

Just over a decade ago, Rob Arnott and colleagues at fund advisers Research Affiliates stretched the analysis back even further. From 1964 to 2012, they found, equal-weight indexing beat traditional, value or "capitalization" indexing by an average of 1.8 percentage points a year.

Behavioral finance is easy - in theory. We all know about "the recency effect," which means the built-in bias in the human brain to give too much weight to recent events. Yet we are all prone to it just the same.

The regular, size-weighted S&P 500 index has beaten the equal-weighted version XX:SP500EW for the past 10 years. Therefore, goes the argument, it must be intrinsically "better" and will keep winning.

Sure, why not?

Meanwhile, anyone who figures it makes more sense to bet equal amounts on lots of company stocks has options. The Invesco S&P 500 Equal Weight ETF RSP bets equally across the entire large-company index, and charges 0.2% in fees a year. There's also Blackrock's iShares MSCI USA Equal Weighted ETF EUSA, which owns nearly 600 stocks equally and charges 0.09%.

Or you can keep wagering most of your money on a few big companies in similar fields, and hope that the next 10 years look much like the last 10.

-Brett Arends

This content was created by MarketWatch, which is operated by Dow Jones & Co. MarketWatch is published independently from Dow Jones Newswires and The Wall Street Journal.

(END) Dow Jones Newswires

January 28, 2025 11:35 ET (16:35 GMT)

Copyright (c) 2025 Dow Jones & Company, Inc.