Wednesday, December 30, 2015

Reader Question: Small cap and Value Tilts

I received a thoughtful question from reader A.J. concerning how to index:
“Hello. I love your website.

I've read dozens of books on investing: Paul Merriman, Malkiel, Ferri, Ellis, Bogle, Swedroe, etc. All of these guys disagree on how to index! It is confusing. Some of them wholeheartedly believe in value and small stocks. I've seen their research, it seems very solid.

On the other hand, how can anyone predict the future?

In Bogle's book, He says that ‘if someone wants to tilt, it would be reasonable to do 85% total stock market fund, 10% value stocks, 5% small stocks.’

So that is what I've done with my US and International holdings. Can I get your thoughts on this? Thanks.”
Thanks for the kind words. Without knowing more about your financial life, I can’t advise you directly, but I can tell you how I view my own portfolio.

The truth is that we just don’t have enough historical stock information to make confident judgements about fine differences in stock allocations. I’ve chosen an allocation for my stocks, but I make no claim that it is optimal. As you say, “how can anyone predict the future?”

The main thing I do is stick with my allocation. I doubt it would have made much difference if I had used slightly different percentages. But what can make a big difference is tinkering. When investors make changes to their allocations, they are often shifting from an asset class that has recently performed poorly to an asset class that has recently performed well. Their reasons may sound smart, but they are really just selling low and buying high.

As long as investors are adequately diversified and have an appropriate allocation to fixed income, I doubt that fine differences among small cap tilt and value tilt percentages will make much difference. Much more important are costs and potential losses due to tinkering.

Good luck, A.J.


  1. Over the last 88 years, there has been a value premium of 4.8% according to this review below, and also a much smaller small cap premium. There have been long periods when the value (and small cap) premium disappears, and of course, there is no guarantee that it will continue in the future. However it does seem reasonable to tilt to small and value provided one understands that there will be periods of under performance and that one is able live with this tracking error and not sell out at the wrong time thus locking in losses. A good way of doing this is by using small value funds (the value premium is more predominant in small cap stocks) in US and international markets, perhaps 1/4 to 1/3 of ones equity allocation in those markets.

    However my take on this subject is that is does add another layer of complexity, and with also having to deal with the tracking error, most investors are probably better off to stick with broad market ETFs.

    1. @Grant: I've chosen to use VBR (Vanguard's U.S. small cap value ETF) for 20% of my portfolio, but I certainly understand those who choose to ignore small cap and value tilts.

  2. The following exchange is reproduced to remove broken links:

    ----- BHCh December 30, 2015 at 1:18 PM

    - You are heavily overweight in Canada and US.

    Not sure I follow the logic that one should invest where he spends time. It is quite likely that ones income/purchasing ability is somewhat correlated to his home markets, which should really be an argument for diversifying into other countries. There are other considerations, such as currency risk and taxation though.

    - I have also been overweight in the US over the last few years (only for a different reason) and it has worked out very well.

    Now US stocks seem a bit expensive and I have just went overweight into emergent markets instead and underweight in the US. Will also increase target fraction for Canada over the next year or so. Not saying it's the right thing to do... This clearly constitutes what you call "tinkering".

    ----- Michael James December 30, 2015 at 2:25 PM

    @BHCh: Having my investments overweight in the countries where I spend my money makes sense to me. The cost of living in these countries will have a long-term correlation with their stock markets.

    I wish you luck with your tinkering, but I'm not optimistic about it being a net positive for you over the long run.

    ----- BHCh December 30, 2015 at 3:50 PM

    @Michael: What is wrong with the following logic:

    - Cost of living = inflation.
    - Markets don't perform well (in real terms) when the inflation is high.
    - Therefore cost of living has a negative long-term correlation with the stock market.

    ----- Michael James December 30, 2015 at 4:01 PM

    @BHCh: The problem is the second point. Higher inflation increases the gap between nominal and real returns, but nominal returns tend to be higher when inflation is higher.

    Businesses sell the items that make up the basket of goods that are used to determine inflation. Higher prices mean more business revenue. Business profits tend to track inflation plus real growth in the economy. Over the long-term, stock prices track business profits. Of course, over the short term, stock price movements can diverge wildly from business profits, but not over the long term.

    ----- BHChDecember 30, 2015 at 4:27 PM

    @Michael - There are other factors, such as cost of salaries and cost of borrowing by the business.

    Benjamin Graham's "Intelligent Investor" (revised edition, commentary by Jason Zweig) demonstrates relationship between inflation and stock prices (Figure 2-1). Looking at US stock market between 1926 and 2002, when inflation was above 6%, stocks stank. Stock market lost money in eight years out of the 14 in which inflation exceeded 6% and the average return for those 14 years was a measly 2.6%. During deflationary periods stockmarket performs even worse - losing up to 43% of the value.

    ----- Michael JamesDecember 30, 2015 at 4:40 PM

    @BHCh: This is all focused on short time periods, typically year-over-year correlations. I don't care much if stocks and inflation swing in different directions in 1-5 year periods if it tends to balance out after 10 years. When we look at 25-year periods, stocks rarely lose out to inflation. Short-term volatility measures don't concern me much. My main financial risk is that my investments will suffer a sustained (multi-decade) decline compared to my expenses. Being over-weight in the countries where I spend my money reduces this risk somewhat.

    The following post digs into how stock volatility measures vary over time:

    ----- BHCh December 30, 2015 at 5:04 PM

    @Michael: the post on volatility is very interesting.