The most valuable skill an investor can acquire is the ability to distinguish theory from law. Examples of theory include forecasts, the relationship between gold and inflation, or anything ever written by an economist. Examples of law include principles of diversification, statistical arbitrage, and mostly, the fact that anything good that ever gets posted online is done on those nights you’ve promised your girlfriend a nice dinner out and your full attention.
Michael Batnick, Director of Research at Ritholtz Wealth Management, posted a fantastic chart online last night, and my phone lit up like a Disney fireworks show.
Phil: “Sorry, we’ve got to get home earlier than expected”
V: “Ok, sure, it must be something important”
Phil: “I have to tweet and blog about company weights within the S&P 500”
Here is the chart that Mike posted:
The chart shows the relative weights of the top five companies in the S&P 500, alongside the bottom 282 on the left hand side. Some have quibbled that the chart doesn’t show the middle 213 stocks, which is a fair point, but that was fully disclosed and anyway not related to what Mike was showing. The key point here is that investors in a market cap weighted S&P 500 index fund have as much exposure to just five companies as they do to the smaller 282 companies combined.
The graphic struck a chord and went viral, as it should have. It had a specific relevance to me, since we launched the Reverse Cap Weighted U.S. Large Cap Index (index ticker: Reverse), which allocates to the exact same stocks as any traditional S&P 500 fund, but the exact opposite concept. In a reverse strategy, those five goliaths are the smallest weighted companies.
There are three key reasons why we launched the reverse strategy. The first is that by weighting smallest-to-largest, Reverse captures the size premium, or “small minus big” within large cap. By keeping the size bet in the large cap space, we believe that the quality, liquidity and valuation issues that have called the size premium into question are largely solved.
The second reason is that when we rebalance reverse, we take profits from the winners and reallocate that money to the stocks with the most room to run. It is a systematic buy-low-sell-high process that contributes outperformance in all but extreme-momentum cycles.
And the third, the one that I want to focus on here, is that a reverse index can be used as a tool alongside a market cap weighted S&P 500 index fund to flatten the distribution of the stocks. It is for this reason that Mike’s graphic was sent to me by a dozen clients, friends, and coworkers. We have been trying to highlight the concentration risk in market cap weighted funds for a long time, something Mike just did in a powerful way.
I recently posted a gif image here on Twitter showing how adding a reverse strategy along side an allocation to the S&P can drastically reduce portfolio concentration. As gif images aren’t supported everywhere, Ill use a couple of the source graphs below. Along the vertical axis are the company weights, and horizontal are the companies, largest to smallest as left-to-right. SPX (S&P 500 Index) & Reverse index relative weights are labeled on top. HHI stands for the Herfindahl-Hirschman index, measure of index concentration calculated by squaring the weights of each constituent, and then summing the resulting numbers. The lower the HHI the better the diversification.
If you look at the left hand side of the top slide (100% SPX) you can see those same outliers – Apple, Amazon, Alphabet, Microsoft and Facebook. When you get down to a 50:50 split between the two indices, those weights are just about cut in half – still overly represented, but in a way that drops the HHI all the way from 89.1 to 33.9.
HHI can be further maximized by investing in an equal weight portfolio – mathematically the most diverse allocation to an index – but the 50:50 SPX and Reverse index strategy has a slightly different return profile. The higher allocations to the largest companies reduce mapping error while the higher allocation to the smaller tail provides increased SMB exposure for your buck. Ultimately, any recommendation of an optimized portfolio would have to be client-specific.
Here is an illustrative look at combo weights versus cap weighting and equal weighting:
It is important to note the non-linear relationship between the weighting strategies. For SPX, only 114 companies are larger than average weight of 0.2%. For Reverse, 279 companies are larger in weight than average level of 0.2%.
As noted by many online and in Mike’s subsequent blog post, the concentration of the top holdings in SPX that we are seeing now is not a particularly unusual phenomenon; it has been this way, more or less, for some time. And just as Mark McGwire’s steroid use was a violation of the public trust, yet Andy Pettitte’s steroid use was evidence of his competitive spirit; the outrage meter on SPX concentration can vary wildly.
If you take a look at the historical annual HHI index for the S&P 500, the current score of 89.1 is not all that bad:
On a rolling basis, we come to a fairly similar conclusion:
What is fairly unusual is the similar risk profile for these top companies. We have rarely seen such sector concentration in the top holdings before (the last S&P sector to exceed the current weight of Info Tech…was Info Tech just before the .com bubble burst in 2000), and these companies in particular share some specific risk attributes. Alphabet just got hit with a $5 Billion fine from the EU. Microsoft has faced anti-trust lawsuits over the years. Facebook has had numerous regulatory violations and also has anti-trust risks. Amazon is a stated target of the current administration. All of these companies are depending on expansion into emerging markets to continue their growth, and all face similar obstacles for achieving that- risks that are forward-looking, not backwards-looking.
Also worth noting is that while the “top 5 weight %” is a constant, the companies that comprise that top five changes frequently. Jason Zweig pointed out in the Wall Street Journal “Of course, history also suggests that every firm that was expected to dominate indefinitely — from RCA in the 1920s to IBM in the 1980s to Nokia in the 1990s — has ended up slipping. And the idea of virtually limitless growth flies in the face of much of human experience. Trees can’t grow to the sky because they would be bent and crushed under their own weight first.”
At the end of the day, concentration itself may or may not be a bad thing for investors, just as being overly exposed to a handful of stocks could turn out wonderfully for investors, or it can go the other way. My crystal ball is no better than anyone else’s. I am just glad to see these issues rising to the forefront of the conversation, and that investors are becoming aware of the risks they are taking on when they invest in market cap weighted funds.
As Aaron Brown noted recently, “the downside is that market-cap weightings leave you overexposed to overvalued companies, and underexposed to undervalued ones… Inverse-cap gives even better factor exposures and has outperformed significantly since 1997, but who knows what the next 20 years will bring?”
Posted by: Phil Bak 07/19/2018
This blog or its content do not constitute an offer to sell or the solicitation of an offer to buy any security or investment product and should not be construed as such.