Table of Contents:
Warning: This is a long post with some finance theory, math, and intermediate investing concepts. I tried to make it palatable with a bunch of 90s music references when things get tough. Also, this is for educational purposes only. This is not investing advice. Consult your investment advisor for that.
Executive Summary:
You can beat index funds with index funds. By using fundamentally weighted index funds, we expect to improve return over time vs standard index funds. While these funds are slightly more expensive, academic research shows that they make up for the added cost with higher returns and similar levels of volatility. This holds across geographies and time periods.
When most people think of index funds, they think of mutual funds and ETFs that follow market capitalization-weighted indices like the S&P 500. I propose that market capitalization weighting is a suboptimal way of weighting a portfolio of stocks. The more overvalued a stock is, the more oversized its position in your portfolio.
By using alternatively weighted ETFs, our clients get similar volatility to standard index funds, but with likely long-term outperformance. It’s similar to repeatedly playing a 50/50 coin toss game with a coin that comes up heads 55% of the time.
Why Cedar Peak uses index funds, but not “index funds”
One of the characteristics that set CPWA portfolios apart from most other advisors’ is that we don’t use index funds. At least, not the same ones other advisors do.
Don’t get me wrong. Index funds are ok. They are, by definition, average. They are cheap. They are diversified. Buying index funds is like buying IBM in the 70s or casting Tom Hanks in a movie. No one will blame you if it doesn’t work out.
Index funds are fine.
In the words of Roy Kent, “Don’t you dare settle for fine.” [1]
You can do better. Index funds have a dirty little secret. When it comes to weighting the stocks in the portfolio, the most popular index funds use a terrible method - Market Cap-Weighting.
By using funds that improve upon the standard methods of stock weighting, we expect our funds to do slightly better than average while seeing similar levels of volatility.
What is portfolio weighting?
The weighting in a portfolio is what percent of the total portfolio the stock or ETF in question makes up. If I look at the Top 10 holdings of the S&P 500 ETF (SPY), it looks like this as of 11/17/2022.
Symbol | Company Name | Weight |
AAPL | Apple Inc. | 5.82% |
MSFT | Microsoft Corp. | 5.27% |
AMZN | Amazon.com Inc | 2.33% |
BRK.B | Berkshire Hathaway B | 1.75% |
GOOGL | Alphabet Inc A | 1.62% |
JNJ | Johnson & Johnson | 1.46% |
UNH | UnitedHealth Group Inc | 1.46% |
GOOG | Alphabet Inc C (Google) | 1.45% |
XOM | Exxon Mobil Corp | 1.36% |
JPM | JPMorgan Chase & Co | 1.24% |
So, if you own $100 of the S&P 500, you own $5.82 of Apple stock, $5.27 of Microsoft stock, $2.33 of Amazon stock, and so on all the way down to your $0.01 of Vornado Realty Trust. 5.82% is Apple’s weight in the S&P 500.
Why weightings are important
Weightings determine a fund's performance. If we change how much of each stock we own, we change the portfolio’s performance. Let’s say, just for funsies, that we swap the weightings of Apple and Johnson and Johnson. That would be the equivalent of selling $4.36 of Apple and using it to buy J&J (5.82% - 1.46%). If J&J’s stock outperforms Apple’s, our changed portfolio would outperform the S&P 500.
If we were to wake up tomorrow to find that Apple (5.82% of the index) was up +10% and the other stocks in the S&P 500 were unchanged, the S&P 500 would be up approximately 0.58% (10% x 5.82%). If we were to wake up tomorrow to find that UnitedHealth Group (1.46% of the index) was up 10% and the rest of the stocks were unchanged, the index would only be up 0.14%
When building a portfolio, your best ideas should have the highest weighting. This way, if you are right, you get the biggest benefit from your rightness.
How index funds determine weightings
We'll use the S&P 500 as our example because it is so widely known. Since an index fund replicates the performance of the index, an S&P 500 fund would replicate the weightings of the index. The S&P 500 is market cap weighted. This means a stock’s portfolio weight is based on the company’s share of the total market capitalization of all the stocks in the S&P 500. Yay! More definitions!
Market Capitalization = Price per share x # of shares outstanding. In other words, the market value of all of the stock for a given company.
Using this as the basis of weighting means the more valuable the company is, the more heavily weighted it is in the index. Apple is currently the most valuable company in the world, so it has the highest weighting in the index. Seems reasonable, right?
There’s a downside to that. As a stock’s market value rises, it becomes a more important part of an index - regardless of why the stock went up. If a stock rockets higher because a subreddit decides to make it a meme, the index funds will own more of it regardless of whether the increase is justified. In June 2021, Gamestop and AMC made up 2.1% of the Russell 2000 Value index (IWM), which had 1,495 stocks in it.
Market Cap Weighting seems logical. Why is it a bad thing?
It's not bad per se. There are just better ways to do it if you want to improve your investment performance. Who doesn't want that?
The five best arguments for weighting a portfolio by market cap are:
It’s easy
It adjusts holdings for company scale - i.e. Walmart is bigger than Target on most measures, so Walmart is a bigger holding in the S&P 500
It’s based on widely available information making it easy to track and replicate
It builds in some diversification since larger, more stable companies have a higher weight in the index
It's easy
Thinking that this is the ideal way to weight a portfolio embeds a critical assumption – All stocks are properly priced at all times. No stocks are ever underpriced or overpriced. It would be easy to go down a rabbit hole arguing this, but we can rewind to 2021, look at the meme stock mania and know that's not true.
One assumption and the goals of stock portfolio building
To examine market-cap weighting’s main weakness, we need to make one assumption.
Key Assumption: Every stock’s price is roughly, but not exactly, what it should be. Purely for discussion's sake, we'll say, +/- 20% of fair value. If a stock's fair value is $100, it is trading between $80-$120. We just don’t know if it’s too cheap or too expensive.
When we assume that stocks aren’t perfectly priced, market-cap weighting's weaknesses shine through.
When building a portfolio of stocks that are inaccurately priced, we should have two goals.
We want to own as much as we can of stocks that are too cheap because they will go up in price when people realize the stock is too cheap.
Conversely, we want to own less of the overpriced stocks for similar reasons.
The more skeptical among you may be asking, “Kevin, if you know which stocks are underpriced and which stocks are overpriced, why are you buying index funds? Shouldn't you be sipping a lager, reading a book on the beach of Kevin Island, which is what you would rename Oahu after you buy it all because you made so much money buying cheap stocks and avoiding expensive stocks?”
To that I say “Great question. I don’t know which are overpriced and which are underpriced.” It doesn't matter. If your goal is to beat "the market," you need to avoid market-cap weighting's big flaw and use a different weighting method.
What is this big weakness? Let's build a fake market-cap weighted index and find out.
The problem with market cap-weighted indices: A fake index example
To avoid compliance issues with using real companies, I'll build a fake index of fake companies.
Using my high school CD collection (most of which I still have) and a random number generator, I created a 10-stock portfolio with weightings that we can use for discussion purposes. Again, these are all fake weightings as these are all fake companies.
Fake Company Name | Index Weighting |
Weezer Respiratory Therapy | 10.6% |
Neutral Milk Hospitality Group | 11.0% |
Depeche Motors | 8.8% |
Pumpkin Reconstruction Partners | 15.0% |
Halen Van Lines | 2.7% |
Chumbawambank | 17.1% |
No Debt Financial Services | 6.7% |
N'Sink Plumbing Repair | 19.2% |
Jamiroquiet Soundproofers | 5.2% |
Nine Inch Nail Salon | 3.8% |
If there is a fund that follows this index, $19.20 of every $100 in it would be invested in N’Sink Plumbing Repair[2]. That’s almost 4x as much as would go into Jamiroquiet Soundproofers, which is virtual insanity.
We need to remember our key assumption and portfolio construction goals. Every stock is priced approximately right but is between 20% overpriced and 20% underpriced. We want to own more of the stocks that are underpriced and less of the stocks that are overpriced.
How do we know if a stock is underpriced or overpriced? Again, we don’t. Again, it doesn’t matter.
I used my random number generator to assign overvalued/undervalued amounts. Then, I calculated what the weights should be if all the stocks traded at their fair value. That's the Fair Value weight. Another way of phrasing that is the fair value weight is what the weights would be if none of the stocks were too cheap or too expensive.
Company name | Weight | Over/(Under) Valued | Fair Value Weight | Over / (Under) Weight |
|---|---|---|---|---|
Weezer Respiratory Therapy | 10.6% | 20.0% | 8.2% | 2.4% |
Neutral Milk Hospitality Group | 11.0% | (7.0%) | 11.5% | (0.5%) |
Depeche Motors | 8.8% | 2.0% | 8.4% | 0.4% |
Pumpkin Reconstruction Partners | 15.0% | (16.0%) | 16.9% | (1.9%) |
Halen Van Lines | 2.7% | (10.0%) | 2.9% | (0.2%) |
Chumbawambank | 17.1% | (20.0%) | 19.9% | (2.9%) |
No Debt Financial Services | 6.7% | (12.0%) | 7.3% | (0.6%) |
N'Sink Plumbing Repair | 19.2% | 10.0% | 16.8% | 2.4% |
Jamiroquiet Soundproofers | 5.2% | 19.0% | 4.1% | 1.1% |
Nine Inch Nail Salon | 3.8% | (5.0%) | 3.9% | (0.1%) |
Let’s dig into the details. We'll look at the most overvalued and most undervalued stocks in the fake index – Weezer Respiratory Therapy and Chumbawambank - bolded above.
Weezer is overvalued. Say it ain't so! We also own more of it than we would in an ideal world (10.6% vs 8.2%). Conversely, Chumbawambank is the most undervalued stock in our index. At 20% below fair value, it got knocked down. But it will get up again. You’re never going to keep it down. That’s why we’re sad we own less of it in the index than we would in a perfect world (17.1% vs 19.9%).
If we plot Over/Undervalued against the Over/Underweight of each stock, we see a strong correlation.
We can draw a handful of conclusions from this example, which applies to all market-cap weighted indices:
Every stock that is overvalued is one where we own more than we should
The more overvalued a stock is, the more outsized the position is
We own less of every undervalued stock than we should
The more undervalued a stock is, the more underrepresented it is in our fake index fund
Reviewing those two tenets of portfolio construction
It’s better to own underpriced stocks, not overpriced stocks
The more underpriced something is, the more of it you should own
The market cap-weighted portfolio violates our two main tenets of portfolio construction. Yet most advisors use market cap-weighted index funds as either the core or the exclusive funds in their portfolios.
Better ideas for index funds - Lots of them, in fact
In a paper titled “The Surprising Alpha From Malkiel’s Monkey and Upside-Down Strategies,” Research Affiliates tested a bunch of different weighting methods over long periods of time and many geographies.
Almost any other idea did better than market-cap weighting. Let’s start with a very simple one. Equal Weight. 500 stocks in the S&P 500? 1/500th of the portfolio goes into each stock.
I have some quibbles with the equal-weight methodology. My main beef is that it doesn’t adjust for the scale of the company. For example, Mcdonalds' is far bigger than Domino’s Pizza by almost any measure - # of stores, sales, profits, etc. Should they be equal size positions in your portfolio?
Quibbles aside, how does the equal-weight S&P 500 do compared to market-cap weighted S&P 500? Very Well.
If we put $10,000 each into the S&P 500 and the S&P 500 Equal Weight 30 years ago, we end up with $68,000 more in the equal weight portfolio than in the regular S&P 500 portfolio. (10.9% vs. 9.6% annualized returns). That’s pretty darn well.
Research Affiliates tested a bunch of different weighting methods. Almost every alternative weighting method outperformed the standard cap-weighted benchmark. Not only that, but the inverse of those methods outperformed the cap-weighted benchmark, too.
Weighted by Volatility? That beat Market-cap weighting
Weighted by Book Value? That beat Market-cap weighting
Weighted by the inverse of Book Value? That beat Market-cap weighting
My key takeaway from this paper is that the key to beating market-cap-weighted index funds like the S&P 500 is to break the link between portfolio weighting and market capitalization. If you use an index fund that weights its positions using a different method, you should outperform the standard index fund over a long period of time.
Fundamental Weighting - Our preferred weighting method
Where many advisors would use an S&P 500 ETF (Ticker: SPY), we use a fundamentally weighted US Large Cap ETF. It holds almost all of the same stocks as the S&P 500 ETF. It just holds different amounts. It breaks the link between weighting and market capitalization.
Fundamental indices weight positions based on fundamental metrics – revenue, net income, and cash flow.
It's easier to understand if we boil it down to a two-stock portfolio. As of the close on 12/23/22, the market capitalizations of Netflix and AT&T are similar. If it were only those two stocks, the market cap-weighted portfolio would be evenly split.
| Market Cap ($B) | Weighting |
Netflix | $131.3B | 50.2% |
AT&T | $130.5B | 49.8% |
However, if we look at Net Income (e.g. Profits over the last 12 months), AT&T was responsible for 80% of the profits of those two companies. One could make an argument that AT&T should have a bigger weighting in the portfolio because it generates 4x more profit than Netflix.
| Net Profit ($B) | Weighting |
Netflix | $5.04B | 20.1% |
AT&T | $20.04 | 79.9% |
The quantitative brains behind the Fundamental Indices create a blend of revenue, profits, and cash flow. They weight based on that. The more of those three things that companies generate, the greater the weight in the index. This addresses the scale issue with equal weight index construction while breaking the link to market capitalization. It also does it in a logical way.
Research Affiliates analyzed the performance of a Fundamental-Weighted portfolio in the US from 1964-2012, finding that the Fundamental Index would have outperformed Market Cap weighting by 1.94% (11.60% vs 9.66%) with almost identical volatility.
In exchange for that outperformance, you pay a slightly higher expense ratio. The biggest S&P 500 ETF (SPY) charges 0.09%/year in management fees while Schwab’s US Fundamental Large Company ETF (FNDX), which is based on RAFI's Fundamental index, charges 0.25%/year. To me, the additional fee is well worth the additional expected return.
When we put the portfolios of SPY and FNDX side-by-side, we can see that they own the same stocks, just in different amounts. Here are the Top 10 holdings in SPY and their weights in both portfolios as of mid-January 2023
When Research Affiliates applied the fundamental weighting method outside the US, it saw similar outperformance compared to a market-cap weighted benchmark. More expected return out of the same stocks with similar volatility. That’s why we use these products as our core holdings for the international equity portion of our portfolios, too.
These funds are not guaranteed to beat the market. They work by tilting your portfolio toward parts of the market that tend to outperform – value and small-caps - while maintaining a lot of overlap with the standard index. It’s akin to proposing a 50/50 bet with a weighted coin that you know will come up heads 55% of the time. You’ll still lose sometimes. It won’t feel like something is ‘off,’ but over time you should come out ahead.
Note: Research Affiliates, the same people who wrote the Malkiel's Monkeys paper maintain the RAFI indices that Schwab’s Fundamental ETFs are based on. They make money when people own their ETFs. In fact, RAFI stands for Research Affiliates Fundamental Index.
Conclusion: Wrap it up, nerd!
I admit it. Index funds isn’t the sexiest topic. I'm amazed you made it this far if we aren't related and you aren't reading this out of obligation.
Good investing is done over the long run. Using an alternative weighting method won’t make you rich overnight, but it is an excellent way to tilt the odds in your favor. A small increase in return compounds into huge differences in money over long periods. Think back to the S&P 500 vs. S&P 500 Equal Weight example. A 1.0% difference in returns amounted to making 22.4x your money instead of 15.6x over 30 years.
Alternatively weighted index products should give our clients higher returns with similar amounts of volatility. If thinking differently helps clients reach their financial goals, we’re going to take that opportunity for them. Every time.
Disclosure: This commentary is intended only for educational use only. It should not be taken as financial advice. Please consult an investment professional if you want advice. As of the original publication date of 3/16/23, principals and clients of Cedar Peak Wealth Advisors had positions in the following tickers mentioned in this article: FNDX, BRK.B, GOOG, JPM.
Kevin Shuller is the Founder/Chief Investment Officer of Cedar Peak Wealth Advisors. Sometimes, he reads academic papers like the one in this blog post just for funsies.
[1] The only Roy Kent quote I could use without dropping an F-bomb.
[2] Buy! Buy! Buy!