I will discuss fundamental investing on this site, so I can chronicle some of my adventures in stock picking. The vast majority of my marketable securities portfolio (modest though it is) is indexed. The majority of that money is in market capitalization weighted index funds (in retirement vehicles).
Yet, I am convinced of the merits of value investing and the persistence of the value factor. I will plan to post more on this in the future, with some links to better sources than this blog. If you can’t wait, you might go ahead start perusing:
The gap between value stocks and the market cap weighted indexes in foreign funds is larger than in the US and seems historically large. These markets also appear to be cheap relative to history and the U.S. Many would point out this is probably because these indexes are more heavily weighted to financial, mining and energy companies (which may all be facing obsolescence, or at least decreased margins due to US fracking and other technological advances) and the indexes are devoid of FAANGS/tech darlings.
I acknowledge these risks, but that is life as a value investor. As Joel Greenblatt of Gotham Asset Management says, “buy it cheap and something good might happen.” An example of this is what happened with U.S. banks over the last several years. They were cheap for good reason: they were going to be regulated into oblivion, rates would never move, and they would never earn their cost of capital. The narrative has since changed and the stocks have roared back.
I am also primarily exposed to U.S. assets, so foreign stocks should provide me with (modest) diversification benefits. So, I have decided to allocate some additional money to foreign value funds. As part of this process, I have been comparing some of the available foreign value ETFs.
Blackrock’s iShares MSCI EAFE Value ETF (ticker: EFV) was the first option I considered, as it has a relatively long track record and is offered with no trading fee at Fidelity. It has a .40% expense ratio; so $4 per thousand, per year. The turnover ratio is reasonably low for this fund at 28%. The index tracked by the ETF uses price-to-book, forward price-earnings, and dividend yield (equally weighted) as the proxies for value.
Feel free to check out the index methodologies in more detail on the MSCI website. It is a good idea to check these periodically. For example, I thought that EFV used the good old price-book as the primary factor, but I was surprised to find that it has been updated to use a multiple ratio approach. This method is becoming more fashionable as it is thought to be less likely to be influence by aberrations in one metric or ratio.
There is some additional nuance due to the growth versus value characterization (i.e., the growth metrics used feed into whether something is “value” as the definition is based partially on the growth inverse definition). https://www.msci.com/index-methodology. The index is designed to cover half of the investable market so you’re going to have a lot of stocks that either both growth and value or neither strongly growth or value. This provides a somewhat weak exposure to “cheap” stocks. It also sort of assumes that lack of growth is a positive as far as characterizing a stock as value. As Warren Buffett has said, growth and value are linked (definitely not opposites). This is a big drawback of this index (and all of the traditional growth versus value indexes). I note that IVAL and the RAFI indexes don’t use this method (of trying to split the market into halves). The index also only includes companies in “developed” markets, there is no emerging markets exposure.
Another option that I considered was the Vanguard International High Dividend Yield ETF (VYMI). This fund is relatively new and clocks in with a .25% expense ratio. The fund is an “ex-US” fund and includes some emerging markets exposure. High dividend yield is basically a weak value factor proxy. If you are curious about this assertion, you might check out: https://alphaarchitect.com/2015/03/24/investing-in-high-dividend-yield-stocks-is-a-sucker-bet/, or several posts (with source material) on http://www.mebfaber.com.
I also think there is tons of capital chasing the high dividend yield strategy right now, given the popularity of dividend investing and low interest rates around the world on competing income investments.
Massive loads of capital are the mortal enemy of outsized returns. The turnover for VYMI should be very low (as it capitalization weights the higher yielding members of the universe of stocks). It is listed at 6% currently. The US equivalent fund (VYM) has been around for a while and also has similar turnover, so it seems like this is likely to generally persist. I also believe that absent a change in management philosophy, Vanguard would be likely to bring the expense ratio down as the ETF gathers assets. It would be a compelling option in some environments (e.g., where dividend yield stocks were hated due to a spike in interest rates). [Vanguard has recently announced that it plans to launch U.S. factor funds in 2018. If Vanguard launches international versions of these funds, they may prove strong competitors in the future. Stay tuned for a post on this Vanguard announcement.]
WisdomTree also offers a number of potential options. Disclosure: I own shares in in WETF. I am planning a post on this in the very near future. I considered the WisdomTree International Equity fund (DWM). This fund has an expense ratio of .48% and (very low) turnover of of 14%. The stated exposure is ex-U.S. I did note some China exposure in the portfolio. All else being equal, I would prefer one ex-US fund versus two funds (developed and emerging markets) to make things simpler, limit decision fatigue, and the number of chances for decision errors (definitely a topic for future posts). This fund is weighted by dividends, so it should provide a value factor exposure. I think it suffers from some of the same drawbacks as VYMI due to the dividend focus and of course the expense ratio is higher than some other options. I do like the weighting mechanism. Dividends are real cash, so they can’t be faked for that long. They also don’t usually bounce around that much (unlike reported earnings). I also like that this method gives you exposure to every stock in the universe (even the fast growing ones) your weight will just move in accordance with the dividend. So, for a U.S. example you would have a big and growing allocation to AAPL even though they are growing relatively quickly (which would tend to make them be classified as a growth stock in the traditional growth value split, such as employed by EFV). The value factor exposure would be relatively week but you will be overweight more as price diverged from the fundamental (dividends in this case). Also, it just makes conceptual sense to be rooted in a fundamental and I would likely find it easier to hold on in a drawdown for that season.
Finally, I considered a couple of Charles Schwab ETFs that track Research Affiliates indexes. The Fundamental International Large Company Index (FNDF) and the Fundamental Emerging Markets Large Company Index (FNDE). The expense ratio for FNDE is .39% with 14% turnover. FNDF has 11% turnover and an expense ratio of only .25%. These ETFs track the Russell version of the RAFI indexes, which basically use shareholder yield (dividends and buybacks, in this version) and leverage-adjusted sales to determine weighting. For the same reasons as the dividend weighting, these should provide a fairly dynamic value factor exposure. The allocation to companies will move with their economic size and the result will be that the index will trade against the market capitalization deviations from the economic size. The factor exposure should be fairly week as the indexes are so diversified and growth stocks will also be held but in underweights if their growth results in rich valuations I also like the fact that the metrics used in these RAFI indexes (as opposed to the Powershares versions) are likely to be more difficult to manipulate (versus net income, ebit, or even book value), given reputedly inferior accounting internationally. Shareholder yield is based on cash transactions and the revenue line is less easily manipulated, in the absence of outright fraud, than other figures located down the income statement.
Finally, I considered IVAL from ValueShares. This ETF is based on an algorithm, described on http://www.alphaarchitect.com, designed to provide exposure to cheap stocks, based on enterprise value to earnings before interest and taxes, which are then screened for economic viability and returns on capital as a “quality” proxy. It is essentially a permutation of Joel Greenblatt’s Magic Formula. The expense ratio is higher than the other options at .79% and the turnover ratio will be high (currently listed at 119%). I think one concern about applying this strategy internationally is the issues with accounting internationally. Mr. Greenblatt has stated that his firm has to do a lot of work to “clean up” and standardize the numbers internationally. I also note the very strong country/market concentration in the fund. Probably a result of the limited (50) number of stocks included. Finally, high turnover seems likely to be more expensive internationally given higher trading and market impact costs versus the U.S. A strong benefit of this fund (and a differentiator from the other options) is that the exposure to the cheap stocks offered is very strong/concentrated. I really like the methodology and the use of enterprise value, which reflects leverage. I own some of the twin U.S. focused ETF (QVAL).
In the end, I decided the RAFI/Schwab ETFs are the best options. I am a big believer in the fact that costs and taxes can be known, controlled, and are very important. Morningstar and Vanguard are both on record with research indicating that costs may be the only known predictive metric for future fund performance. The .25% expense ratio for FNDF is simply fantastic. I hope you are as surprised with the cost of this option as I am (as your reward for reading). For a long time, EFV was pretty much the only game in town, though it is still not a bad option.
If I invest a couple thousand dollars in FNDF, however, I will easily earn back the ~$5 trading fee at Fidelity (via the lower expense ratio compared to EFV) in a year. I plan to hold much longer than that, as activity is the enemy of returns, so the savings will accumulate and compound over time.
I do think that I will use a standard market capitalization weighted index for emerging markets exposure for now, as these markets overall are still just dirt cheap (IEMG or VWO – though note one has South Korea included as a developing market and the other does not, which materially impacts valuation). I also think that the accounting concerns are likely even more pronounced in these markets, so I would prefer to let the “wisdom of the crowd” (including insiders and locals) discern which numbers are credible and should be valued more richly (WisdomTree has an interesting ex-government sponsored enterprise suite that could be compelling over time; a truer version of the “message of the markets”). Most importantly, the expense ratio gap (and potential market impacts from higher turnover) are likely most pronounced in emerging markets, given their lesser liquidity and developing capital markets infrastructure.
Thus, FNDF is my choice for best international value ETF. The low turnover and expense ratio, the fundamental proxies of value chosen by RAFI, which I think are probably uniquely equipped to perform in the international context are very compelling. I hope you will note that my analysis did not consider either past returns or back test results. I did consider the long-term premiums associated with the value factor in the U.S. and abroad, but the selection of the vehicle was based upon the process used and the impact that process is likely to have on expenses (including turnover) and future returns.
What other options did I miss? Are you aware of any potentially better ETFs offering exposure to the value factor in international stocks?