Buy, Hold, Rebalance a Globally-Diversified Portfolio 2017

When I think about it, I am impressed with how different 2017 feels compared to when I started seriously learning about investing in 2003. Instead of only reading about it in few books mostly read by finance nerds, nowadays nearly every robo-advisor out there uses a globally-diversified mix of low-cost ETFs to build their portfolios. What used to be a relatively quiet alternative to buying 4-star active funds is now becoming the default choice.

We’ve seen from the Callan Investment Returns Table that the best-performing asset classes constantly change from year to year. In a industry magazine called Investments & Wealth Monitor, there was an article titled Why Global Asset Allocation Still Makes Sense by Anthony Davidow. (Found via AllAboutAlpha.)

Here’s an illustration of how a globally-diversified portfolio has outperformed. Below is a graphic from the article comparing a 100% S&P 500 portfolio, and 60/40 S&P 500/US Agg Bond portfolio, and a “globally diversified portfolio” using historical data from January 1, 2001 to December 31, 2016. Index values are used directly as opposed to actual ETFs or funds. The portfolio are rebalanced annually back to target asset allocation.

globaldiv

Their “diversified portfolio” had a rather finely-diced list of asset class ingredients:

  • 18% S&P 500 (US Large-Cap)
  • 10% Russell 2000 (US Small-Cap)
  • 3% S&P US REIT
  • 12% MSCI EAFE (International Developed)
  • 8% MSCI EAFE Small Cap
  • 8% MSCI Emerging Markets
  • 2% S&P Global Ex US REIT
  • 1% Barclays US Treasury
  • 1% Barclays Agency
  • 6% Barclays Securitized
  • 2% Barclays US Credit
  • 4% Barclays Global Agg EX USD
  • 9% Barclays VLI High Yield
  • 6% Barclays EM
  • 2% S&P GSCI Precious Metals
  • 1% S&P GSCI Energy
  • 1% S&P GSCI Industrial Metals
  • 1% S&P GSCI Agricultural
  • 5% Barclays US Treasury 3–7 Year

I do wish this portfolio was a bit more simple and easy to replicate. However, if you take a step back, you could simplify this asset allocation into the following:

  • 56% Global Stocks (50% US/50% Non-US)
  • 5% Global REIT (60% US/40% Non-US)
  • 34% Global Bonds (70% US/30% Non-US)
  • 5% Commodities

Now, we can’t necessarily expect a global portfolio to always outperform. One thing is usually doing better than another thing you own. Most recently, US stocks have outperformed International stocks quite significantly. Here’s an explanation from the article about the “free lunch” of diversification:

Diversification strategies do not guarantee capture of profits or protection against losses in any market environment, but they have been shown over time to provide a smoother ride. Rather than bearing the brunt of the 2000 Tech Wreck and the 2008 Great Recession, the diversified portfolio provided cushioning under the large market drop and was able recoup losses and grow over time.

Comments

  1. Whats interesting to me is that over this 16 year timeframe from 1/1/2001 to 12/31/2016, average annual returns have been tepid, even though we are currently at all-time highs in the US Markets.

    Based on the data from the chart, the 16 year annual return has been:

    Diversified Portfolio: 6.66%
    S&P500: 5.35%

    • It’s all a matter of perspective. If the chart started in 2009 it would look like we were at all time market highs. Add in the zero return many earned from 2000-2009 by the time the markets bottoms and the results are less impressive. It really bolsters the argument for a globally diversified portfolio, because we all lack perspective on how our current investing climate will look through a historical lens.

    • Performance of buy-and-hold strategies rely heavily on start-date. To reduce dependency on start-date, diversify your portfolio even further to include not just stocks and bonds, but commodities and currencies. Also, have an exit point for each investment. Exit points keep you out of the pain of multi-year downtrends and sharp crashes.

      See here for a simple example on implementing an exit point in your investments…https://www.earlytorise.com/short-term-investing/

  2. What Vanguard funds do you think would match this simplified allocation?
    VTSMX/VGTSX
    VGSIX/VNQI
    VBMFX/VTIBX
    VGPMX?

    How do you think this compares to your allocation?
    http://www.mymoneyblog.com/early-retirement-portfolio-income-2017-q1.html

  3. Perhaps I missed it, but how did they determine those asset class weights?

    What makes investing fascinating for me is that we can have the same investments, but could have vastly different results, depending on the weighting of those investments, the years we started investing, timing of cashflows etc. I think that even passive investing requires a lot of active input…

    For example, an investor with an overallocation ( or any allocation) in emerging markets in 2007 has not done as well relative to the investor who bought the VTSAX/VTSMX a decade ago. However, the investor in emerging markets fro 1994 has done better than the investor in VTSAX/VTSMX.

    I read a lot from people who have different strategies than me, and it seems like many are focusing too much on US equities ( myself included). I wonder if the performance of the past decade is to blame.

    • For emerging markets I meant 2000 not 1994..

    • I don’t know how they came up with those asset class weights, or more importantly *when*. If you build something today, then it is easy to backtest a mix that would do well. I was happier to see when you simplify the overall asset allocation, it wasn’t ridiculous or too far from the global marktet-cap-weight.

      • Yes there’s many different sets of rules that produce a profit in a backtest.

        In my experience, backtested results mirror actual results pretty well – except for the feelings part. Hint: Like dieting, many methods work but only a few can stick to it.

    • Passive investing does not exist. Index funds have rules for what instruments to include, how to weight them and how often to adjust weightings and composition.

      Little do investors, and even the advisors selling the funds, know that indexes are very much active.

      Investors try their best to implement passive behavior on top of the funds they invest in, but by nature they cannot. They fall prey to all of the behavioral biases that hurt performance and essentially guarantee that even if they invest in a free mutual index fund, they cannot beat the index itself.

  4. Aurelien Windenberger says:

    I would argue that any asset class allocation needs to consider current valuation of each asset compared to their long term average valuation, and other assets.

    So looking at a simple portfolio of US stocks and Int Stocks split 50/50 as a base valuation. If the US stocks are trading at a CAPE that is 50% above average and the Int Stocks are trading at an average CAPE, then adjust the split in favor of the Int Stocks somewhat, say maybe 40/60 or 30/70.

    That way you favor assets that are more likely to deliver higher expected returns, while still maintaining diversification.

  5. I really wish this analysis was averaged out over multiple start/dates spanning years or even decades. Very rarely do bonds beat stocks. Since bond yields are essentially capped at a lower bound of 0% (which we’re only slightly higher than now) there is no way you’ll see a repeat of performance for bonds. I don’t know very much about international investments, but I can’t help wonder if the same applies

    This isn’t to say bonds (and international stocks) won’t beat U.S. stocks over some time frame. Stocks are much more risky. But you should be diversifying to reduce risk, not to beat a 100% allocation of U.S. stocks.

    • I like where your head’s at.

      Historically, stock index funds produce MAR ratios (CAGR / Max Loss) of 0.10-0.30. An index with an historical 7% CAGR and a 50% Max Loss produces a MAR of 0.14. This, to me, represents a very poor risk/reward.

      I believe everyone might be better served coming up with their own objective (or bliss) function for their investment performance. Some prefer only absolute returns while others care more about a high MAR regardless of high returns.

      Not all high MARs are good though. Consider your savings account which earns an infinite MAR, but produces next to zero gross returns (negative when you consider fees, taxes, inflation and opportunity cost).

Speak Your Mind

*