Welcome to Part 3 of the “Investing With Style: Opposites Attract” series. In part 1, I focused on the US stock market. In part 2, I focused on the International stock market. I will turn my focus towards a globally diversified portfolio in this last edition.
I got most of what I wanted to say out in parts 1 & 2, so no more disagreements with Mr. Bogle here. Instead of going through every table with comments, I am going to present them one after another with a few exceptions. After the tables and charts I will talk a little bit about the Momentum and Value investment philosophy for myself.
As a reminder, here are the definitions for the following tables:
- TSM = 50% US Total Stock Market (Represented by VTSMX) // 50% Total International Stock Market (Represented by the MSCI ACWI ex-USA Index from Jan 1995 through April 1996. VGTSX from May 1996 through December 2016)
- LG = 50% US Large Cap Growth (Represented by VIGRX) // 50% International Large Cap Growth (Represented by VWIGX)
- SG = 50% US Small Cap Growth (Represented by MSSGX) // 50% International Small Cap Growth (Represented by the Fama-French Global ex-US Small Cap Growth Factor Index)
- LV = 50% US Large Cap Value (Represented by DFLVX) // 50% International Large Cap Value (Represented by DFIVX)
- SV = 50% US Small Cap Value (Represented by DFSVX) // 50% International Small Cap Value (Represented by DISVX)
- LM = 50% US Large/Mid Cap Momentum (Represented by MSCI Momentum Index until May 2013 (Minus 15bps for Fund Fees). MTUM ETF from May 2013 – December 2016) // 50% International Large/Mid Cap Momentum (Fama French Global ex-US Momentum Factor from Jan 1995 through May 1995 (Minus 30bps for Fund Fees). MSCI World ex-US Momentum Index from June 1995 to Jan 2015 (Minus 30bps for Fund Fees). IMTM ETF from Feb 2015 through Dec 2016.)
Note: If you are looking for a deeper discussion on the following tables and charts, please refer back to parts one and two in this series. Links are at the top.
That Small Value run from 1995-1999 and the Large Growth run from 2000-2008 are brutal periods to hold through…
Okay, so no big surprise from the above charts. We discussed all of the challenges of picking one style in the first two pieces and we also discussed the valid argument for adding Momentum and Value together.
Now the truth is that I think everyone’s starting point when investing should be something similar to a low-cost, market-cap weighted, 3-fund portfolio: 1/3 Total US Stock Market + 1/3 Total International Stock Market + 1/3 Total Bond Market. I think this holds true for advisors working with clients as well. This is where I started my investment journey years ago, and I think it is the logical place for everyone to start.
Through experience and thousands of hours of reading and research, I have realized that some different investment strategies make a lot sense in my opinion. I strongly believe that Value and Momentum are great compliments to the core market exposure that I started with.
Some of the reasons for this strong belief (Which you absolutely need if you are to deviate from the 3-Fund portfolio) in Value and Momentum are:
- Over a centuries worth of evidence and out-of sample testing for both strategies (Value Bibliography // Momentum Bibliography)
- The strategies have performed well on their own, but complement each other even better as they have shown to be uncorrelated over long periods of time
- They make economic sense to me
- They provide a hedge against 100% market exposure (i.e. Japan over the last 30 years)
You might ask why I chose to split to Momentum and Value portfolios into a 60/40 mix? Great question. I do not have an answer. I think I am so used to reading about 60/40 portfolios that when I started putting all of the data into Excel, I subconsciously chose 60/40 style mixes.
For my own portfolio, I actually go with an even 50/50 mix for a couple reasons:
- Its easy to remember
- It fully hedges my bet on Value and Momentum
- It has historically had higher risk-adjusted returns compared to a 40LM/60 SV portfolio
- It reduces tracking error when one style is out of favor
- That small of a difference doesn’t really matter…
As you can see, there was not much of a difference between a 40/60 and 50/50 split over the last 22 years:
In fact, there was not that big of a difference no matter what allocation you chose over the last 22 years:
For me, I place a high importance on potential behavioral conflicts when it comes to my portfolio construction and investment strategy. The tables above and below are a great example of this. No “red” (aka Extreme) cells when the allocation is 50/50. The 50/50 allocation doesn’t have the best or worst returns. It doesn’t have the highest or lowest volatility. It has a strong Sharpe ratio. Its “worst” 3-Year period over the last 22 years is something I am comfortable sitting through, etc.
So, do I think a 100% Global Small Cap Value portfolio will provide the highest annual returns when I look back 30 years from now? I do actually, but I also know that sticking with that portfolio would potentially be a great challenge, to the point where I do not want to take the chance of making a behavioral mistake. I would rather hedge my bet with an uncorrelated style and take some potential returns off the table in exchange for lower tracking error. I don’t want to have an extremely frustrating portfolio, as I think the markets will do enough of that on its own to me.