[This column of mine originally published on MD Magazine and comes out of a conversation I had on Sermo about fundamental index funds. It seems that any fund that can possibly call itself an index fund these days is doing so, in part to capitalize on the good reputation some index funds have acquired due to low costs and excellent long-term performance. However, there are significant differences even between cap-weighted index funds and “fundamental index funds” are something else entirely.]
It has been more than 40 years since the institution of “Bogle’s Folly,” the original Vanguard 500 index fund for retail investors. The results speak for themselves as index funds have repeatedly and continuously outperformed the vast majority of the actively managed mutual funds in their asset class over the long term. The obvious truth behind Bogle’s Cost Matters Hypothesis, along with the significantly better overall performance, has convinced both professional and personal investors of the merits of this passive investing approach. However, since 2004 there have been a growing number of “fundamental index funds” seeking to capitalize on the popularity of more traditional index funds.
Fundamental Versus Cap-Weighted Index Funds
Capitalization-Weighted Index Fund
To understand the idea behind a “fundamental index fund” you have to look under the hood to understand how a traditional index fund is designed. Traditional stock indices, and the funds that track them are composed of stocks in a “capitalization-weighted” manner. That means that if the value of Apple makes up 2% of the value of the stock market as a whole, then 2% of the value of the stock index, along with 2% of the money in the index fund following that index, will come from the performance of Apple. The main benefits of this approach are that it is very easy to run this type of fund, it is very inexpensive to do so, and you can essentially guarantee that investors will achieve the same return as the market they are investing in. If the stock market goes up 10%, so does the value of the investment.
Fundamental Index
A proponent of a fundamental index, however, would argue that there is a better way than market capitalization to decide how much money to invest in each stock in a fund. For example, one method has been to invest the same amount of money into each company, so that an investor has the same amount of money invested in Apple, currently the largest (worth $653 billion) US company [Update 2019: Apple's market cap is now just under $1 Trillion], as in Telephone and Data Systems, Inc., the 1,000th largest (worth $3 billion) company in the US and in Information Services Group, Inc., worth just $140 million. Although it might not seem smart to invest the same amount of money into two companies, one of which is 5,000 times larger than the other, the results of doing that seem promising, at least when viewed through the rearview mirror of back-tested data.
Although market capitalization could be considered a “fundamental,” for the most part these strategies involve deciding how much to invest in each stock based on book value, dividends, sales, or earnings. This avoids some of the flaws of an “equal-weighted” approach while still avoiding the “bubble problem” of a capitalization-weighted index. That is to say, the more overvalued a stock becomes, the more of it a capitalization-weighted index automatically owns. Proponents of a fundamental index approach claim long-term benefits (using back-tested data) of as much as 2% per year over a capitalization-weighted index fund.
The Data
While the back-tested data tells a compelling story, I find it far more interesting to see how fundamental index funds have done going forward. There isn’t a lot of long-term data, unfortunately, as none of these funds have been around for more than 10 years. One of the more well-known fundamental index funds with a fairly long track record, is the PowerShares FTSE RAFI US 1000 ETF (PRF), which follows an index which contains the 1,000 largest stocks in the US weighted on multiple fundamental factors including book value, cash flow, sales, and dividends.
PRF can be reasonably compared to a cap-weighted index fund such as Vanguard’s Large Cap Index Fund (VLCAX) which tracks the 750 largest stocks in the US based on market capitalization. Over the last five years, VLCAX has outperformed PRF by 14.23% to 13.60% per year. Over the last 10 years, PRF outperformed by about 0.5% per year. Rather disappointing if you were expecting to outperform by 2% a year, although 10 years is not necessarily the long term. [Update 2019: Over the last five years, VLCAX has outperformed PRF by 11.12% to 8.80% and over the last 10 years 13.05% to 11.86%.]
Just a Small/Value Tilt?
Active Management in Disguise?
In many ways, fundamental indexing is little different from an actively managed mutual fund. Many active managers pick stocks based on fundamentals. In fact, many “quantitative” fund companies have been using computer models to pick stocks for decades, just like a fundamental index fund does. The long-term data on active management, whether a person picks the stocks or a computer does, is quite clear. Very few mutual funds outperform their respective capitalization-weighted index over the long term, especially when adjusted for the Fama/French factors of small and value, and it is impossible to pick them in advance. Critics say that even if you call it “fundamental indexing” or its newer term, “smart beta,” it is really just active management in disguise.
Costs Matter
Finally, studies have shown that the best predictor of future mutual fund returns is cost, not past performance. Fundamental indexing simply costs more than capitalization-weighted indexing. The funds are not as large, and thus do not benefit as much from economies of scale. Rather than rebalancing automatically as a cap-weighted index does, a fundamental index must be rebalanced from time to time, incurring additional transaction (and tax) costs.
In addition, there is no mutually-owned mutual fund company (such as Vanguard) offering a fundamental index fund, adding the cost of corporate profit to the mix. Thus, it is no surprise that the expense ratio of even a popular and relatively low-cost fundamental index fund such as PRF is 8 times higher than a total stock market index fund at Vanguard. You cannot predict future performance, but is very easy to predict future costs.
In conclusion, while there are far worse ways to actively manage a mutual fund, fundamental index funds do not appear to be the revolution its proponents would have you believe they are. You can get the same performance at lower cost simply by tilting a portfolio toward small and value using a cap-weighted total stock market index fund and a cap-weighted small value index fund.
Do you invest in fundamental index funds? Why or why not? Do you think it's worth paying extra to avoid some of the issues with cap-weighted index funds? Comment below!
Great article. I did read about these fundamentally weighted funds and was happy to see you write an article about it.
We are waiting for your next book on amazon.
Merry Christmas!
I hope to get some more books out in the future, but life gets busy, you know? Between the newborn, everything else going on at WCI, and the fact that I can’t drop shifts for another 6 months are all conspiring against my bookwriting!
Is this the same thing as “smart beta”? The new fancy term being pushed by wall st money managers like BlackRock? Personally I prefer to have these additional options in my 40X plans since I can hold individual stocks in them.
I mean’t “since I can’t hold individual stocks in them.”
Yes, basically the same thing.
Maybe you can do an article on value of tilting / Fama and French benefits. This is a great article, how people disguise small value tilt and charge extra money for it.
I guess I never have written an article specifically on the concept of a small value tilt, a la Fama/French. Probably because that information is so readily available elsewhere and in no way physician-specific. I’ll try to keep it in mind for future posts though.
I thought this was a good and accurate read! I have wondered about using the RAFI index as a small value tilt for small cap value. There are plenty of domestic small cap value funds, but I haven’t been able to find a true international small cap value fund. Do you think using the Schwab RAFI int. small cap ETF (FNDC) as an international small cap value fund is a good idea? Currently I’m just using the vanguard international small cap ETF. Thanks!
I think that’s a reasonable choice. I like how you look at it though. It’s a way to get a small/value tilt you can’t otherwise easily get. If that’s worth the extra ER to you, then go for it. I use what you use, the Vanguard international small cap ETF, as I haven’t tried to small/value tilt my international stocks (only my domestic ones.)
I agree that fundamental indexes and smart beta products are just an expensive way of getting a small and value tilt to your portfolio. I assume the returns you mention for fundamental funds are pre tax. If you have these funds in a taxable account, the higher turnover in these funds will create a tax grab further lowering returns.
On a similar note, this article from ETF.com looks at Dimensional Funds (which are similar to fundamental and smart beta funds, but with some additional portfolio management features) and cap weighted Vanguard funds that have similar small and value loadings. They concluded that in the past Dimensional Funds did have an edge as their funds held smaller stocks and more “valuey” stocks, but now that Vanguard has funds accessing these factors with higher loadings, (eg. micro cap funds) it is possible to put together a cap weighted portfolio that has the same small cap and value loadings as Dimensional Funds. Dimensional still has features such as a higher degree of securities lending and patient trading etc, which add value, but the funds are more expensive so the returns end up much the same when compared to cheaper cap weighted funds.
http://www.etf.com/publications/journalofindexes/joi-articles/24234-cloning-dfa.html?nopaging=1
Despite the well known disadvantages of cap weighted funds, they’ll always be cheaper than fundamental and smart beta funds, particularly after tax, and as we know cost is the best predictor of fund performance. I think the later are largely smart marketing, designed to collect more fees from investors. I remember Larry Swedroe commenting on Bogleheads once that anything other than cap weighting is active management.
There is certainly a lot of marketing going on. I’m curious about your reference to a Vanguard microcap fund. Did I miss something new?
Oops – the microcap fund in the article was actually an ishares product, IWC. 85% VBR plus 15% IWC gives you the same result as DFA’s small cap value fund.
My impression, although I don’t have data to back it up, is that there is a much greater awareness of the small cap value premium now than there was 10 years ago. Once again, my impression is that there’s much more money trying to obtain that small cap value premium.
Small cap value is 3% of the total stock market. It won’t take much for increased flows to effect small cap value prices.
Now if the small cap value premia are risk premia, they should continue, despite the increased flows. But the existence of a small cap premium, that one can make money on after expenses, is debatable. One can make money from the value premium, and the value premium increases as market cap decreases. But I’m not convinced that the value premium is a risk premium. Please show me the data that value stocks are more risky.
If the value premium is not a risk premium, then the SCV premium may decrease, and that’s ignoring the headwind of increased costs associated with it.
I still think that trying to obtain the value premium is a good idea. Bubbles have always existed and always will. The value premium takes advantage of that. So your return will likely be greater with value stocks, although my guess is that it will decrease. But more importantly, value investing is risk management. A value investor would not have 41% of all stock market exposure in stocks with an average PE10 of greater than 90. But a devout indexer would. I’m describing the Japanese stock market at the end of 1989.
One way to get the SCV premia is a DIY SCV fund. As mentioned, the value premium decreases as market cap decreases. A firm, such as DFA with $378 billion in assets, has difficulty trading smaller stocks that an individual investor wouldn’t have.
A disadvantage to such an approach is tax inefficiency. In an ETF wrapper, cap gains disappear. I’ve seen an ETF with 50% turnover have no cap gains distributions. However, I wouldn’t be surprised if the governments take away the tax advantaged status of ETFs. As ETFs grow in popularity, others may want the playing field levelled.
Bernstein seems to agree with you about value stocks in this piece written over a decade ago: http://www.efficientfrontier.com/ef/902/vgr.htm
About William Bernstein’s piece, he correctly points out that growth beat value 1929-1932 and 1989-1990; the opposite holds true for 1973-1974 and 2000-2002. I haven’t seen data for 2007-2009.
But take a look at the differences between growth and value. When growth beats value, the difference is smaller than the opposite.
Also, there is a consensus that value companies are more levered than growth companies. In 1929-1932, there was deflation. A company that has a fixed interest loan to pay, but has declining prices for its products, has a problem. But with a fiat currency, deflation is much less of an issue. It can happen, as shown by Japan more recently. But even in Japan, the deflation hasn’t been impressive.
With fiat currencies, one should be much more concerned about inflation than deflation. With inflation, a heavily indebted company will do well. In fact, William Bernstein recommends value stocks as an inflation hedge.
So with value stocks you get bubble protection and inflation protection, albeit imperfect. But I just don’t see where the increased risk justifying the value premium comes from.
Good Article! Smart Beta, Fundamental Indexing, Enhanced Indexing are all the same concept repackaged by clever marketing departments. When Rob Arnott’s RAFI indexes came out there was a long standing battle between Research Affiliates and WisdomTree over the patent rights to the terminology “fundamental indexing.”
There are an endless number of possibilities, many of which have academic grounds and others that were created because, like many financial products, they’re easy to sell. Identifying which ones are viable can be difficult. Value can be defined with a variety of different metrics (low price-to-earnings, price-to-book, and price-to-sales ratios, and high dividend yields.) Using dividends isn’t necessarily a bad thing; it’s just less than optimal because not all companies pay dividends.
Many of these products are targeting well know academic factors such as size, value or momentum. Let’s say that you believe in this research and want to build a factor tilt into your portfolio. If it makes sense to target a single factor, then it may make sense to target multiple factors. Some like momentum and value can work well together. However, using two separate products in an attempt to reap the reward may not work so well because of the different ways that the underlying index is defined. Therefore, to achieve the desired exposure, any argument for smart beta should be in the same context of how one fund interacts with the rest of the portfolio.
There’s also what Bernstein refers to as Rekenthaler’s rule: “if the bozos know about it, it doesn’t work any more.” Sometimes I wonder how far to apply that to back-tested-data based strategies. I feel that Bernstein is an astute market historian and is able to differentiate trivial trends from fundamental trends, but sometimes I wonder how kindly historians will judge his analyses. Maybe trends that seem trivial will turn out to be fundamental and trends that we think inherently must continue will turn out to be an artifact of the time period analyzed.
So when I buy a vanguard index fund, is the market cap weighting causing me to buy high and sell low in the “short term” based on peaks and troughs in various s stocks? This seems less efficient.
Well, you’re certainly not selling low because you’re buying. But are you buying more of what has gone up recently than you would have a few months ago? Yes. But I don’t think there’s anything bad happening there. Obviously yesterday is the best day to buy anything, with today being the second best day!
This was extremely helpful. I really like the theory of fundamental indexes, but I’ve been disappointed in their historical performance. I’m curious to hear your thoughts regarding PRFZ, the fundamental small cap RAFI fund. As of market open 3-27-20 PRFZ is trailing the Russell 2000 in all time frames except for ‘since inception’, where it’s beating the Russell 2000 8.04% to 6.79%, as shown in the link below. Could fundamental indexing be a superior strategy for small cap investing?
https://www.invesco.com/us/financial-products/etfs/product-detail?productId=PRFZ&ticker=PRFZ&title=powershares-ftse-rafi-us-1500-small-mid-portfolio
They’re just a value tilt, and value has underperformed the last decade plus, so I would expect relative poor performance.
With reversion to the mean, wouldn’t you expect better performance?
I was referring to the last decade. Yes, I expect value to do better over the next decade, no matter whether you get it from “fundamental indexing” or by some other method. No guarantee though, my crystal ball is cloudy as always.