“Passive” Investing: Theory and Practice in a Global Market

//, Active and Passive Investing/“Passive” Investing: Theory and Practice in a Global Market

“Passive” Investing: Theory and Practice in a Global Market

By |2017-08-18T17:11:58+00:00June 14th, 2017|Guest Posts, Active and Passive Investing|15 Comments

Purely passive investing is theoretically plausible, but practically impossible. That said, the practical implementations can often be “good enough.”

As a theoretical index investor, you deploy capital, take a long snooze, and wake up some day to consume your portfolio.

Unfortunately, the world doesn’t work like. Allocations change, life happens, and as we cover in this blog post, there are really no indexes that are truly passive.

For example, recently the index giant FTSE Russell proposed to exclude the popular social media app, Snapchat, from its index.  In fact, in their memo any company with no voting rights (like Snap) could be excluded from their indexes. This is something that other large index providers have also discussed. The knee jerk reaction might be, “Who cares about Snapchat?” Well, there is a “small” firm called Alphabet (e.g. “Google” as most know it), a stock representing over 1% of the S&P 500, and their shareholders have effectively no voting rights.(1) So perhaps this is a big deal? Who knows.

This recent example is a good anecdote, but there are arguably even bigger issues. Consider index investors invested in many Emerging Market ETF’s, like Vanguard’s VWO. These investors may notice that they don’t own Chinese internet juggernauts Alibaba and Baidu, either. (2) But if index investors think they can go back to sleep after finding a fund like the MSCI Emerging Market ETF (ticker: EEM), that does include Chinese overseas listings like Alibaba, perhaps they should be made aware of MSCI’s most recent proposal. MSCI will likely continue to largely ignore the $7 trillion dollar China A shares market comprising most Chinese domestic companies. Instead of the near 40% of their emerging markets index holding A shares, China could maintain its more modest 28-29% weighting in the MSCI Emerging Markets Index.

The below pie charts are where we now are likely to stand with MSCI:

If this all seems pretty arbitrary, well, that’s because the process is fairly arbitrary.  But a natural question is whether or not these details matter for the passive investor? Are these passive approximations good enough?

Let’s take a step back with a little theory before we address this question.

The Theoretical Global Market Portfolio

My story begins a long long time ago (well, 50+ years ago) in a land far far away (OK, Hyde Park, at the University of Chicago) when basic Modern Portfolio Theory (MPT) was still in its infancy. The theory argued that a rational investor who cared about minimizing volatility while maximizing returns should own the so-called “global market portfolio.” Of course, a major assumption (of many) underlying all of this is that markets are efficient — a highly contentious assumption.

The challenge is that the global market portfolio theoretically includes every risk asset, including everything from your human capital to Afghanistan Light-Rail Bonds.(3)These probably don’t exist, but you get the point. Setting aside incredibly esoteric assets that probably don’t have much weight in the theoretical global market portfolio, we will on the investable universe. The research suggests there is a 91 trillion dollar investable universe (as calculated by Doeswijk, et al 2014) and the portfolio looks something like the following:

Perhaps surprisingly, only 40% is tied up in public/private equity and roughly 29% is in government bonds. To simplify the chart even further, you have ~40% equity, ~30% government bonds, and ~30% “other.” Dumping 70% into stocks and government bonds is not exactly appealing given the historically high equity valuations and the current low yields on government bonds. So in theory this allocation makes sense, but in practice this might be questionable. Luckily, theory and practice don’t need to clash.

Gene Fama told the Chicago Booth Magazine on his 50th anniversary at Booth(4) that the best advice he ever got, from Harry Roberts, was that,

You do empirical work to learn from the data…but no hypothesis that you ever test is strictly true…

Likewise, although we need perfect capital markets to truly test the hypotheses stemming from Modern Portfolio Theory (MPT), the proxies for the global market portfolio — often constructed by CRSP, S&P, MSCI or FTSE — can be considered “close enough”(5). Consider by analogy (used by Fama) that true predictions coming from the laws of motion require a perfect vacuum: Does an anvil falling from the Earth’s sky, instead of an anvil falling through a vacuum, change your reaction to step aside? Likewise, an inability to replicate the perfect global market portfolio doesn’t mean we shouldn’t attempt to achieve this goal via indexing.

The Global Market Portfolio in Practice

Let’s start with the equity vs fixed income mix. The standard 60/40 split between equities and fixed income for a moderately risk adverse investor still seems an “adequate” start. An investor can adjust from there according to his/her risk tolerance and cash flow needs. Although past performance isn’t likely to match future returns with current low fixed income yields, Vanguard does give a nice range for a global portfolio’s average historical returns and volatility.

But peeling back the onion a bit further, what should allocations look like within the equity bucket? In other words, that does the global market portfolio theory say about our international stock exposure? Again, we start with research from Vanguard:

International equity represents over half of the global equity portfolio (~52%), and should theoretically represent around half of an investor’s equity exposure if they are trying to allocate to the global market portfolio. A 52% international equities as measured by MSCI feels high and judging from other investors who have only 21% invested internationally, I’m not the only one with a home bias. So many of us aren’t following the theoretical advice presented by the professors primarily hailing from the University of Chicago. Of course, if we blindly followed the advice of the global market portfolio allocations, then back in 1989 more of your holdings would have been held in Japan than in the US.(6) Any investor with equity allocations based on global market weights would have been shot following the epic Japanese equity market fallout in the ensuing 3 years. (they’ve probably been beat up the past few years as well, as US equity has dominated while international allocations have been lackluster).

Allocate 52% to International Equities? Or is there Wiggle Room?

In theory we should have our equity book roughly 52% allocated to international markets. But is that an iron-clad rule? Perhaps not. Adding to the argument against a full theoretical allocation to international equities, a team from AQR shows that the volatility of currencies can swamp the diversification benefits. Correlations between currencies and US equities is just too unstable, and often positive, as illustrated by the following AQR exhibit.

Vanguard also takes look at the question of optimal international diversification. While Vanguard recognizes that the diversification benefits of international equities are real, they suggest that anywhere between 20-40% is “adequate.” Vanguard buttresses this conclusion by showing that most of the benefits of international equity diversification dissipates quickly. So perhaps the “theoretical” allocation isn’t a hard and fast rule?

Concluding Thoughts

There are theories and then there are practical realities. Theoretically I should have 1 basis point allocated to Afghani Light-Rail-Bonds, but practically accessing that paper probably isn’t worth it for my portfolio (or my life expectancy!)

In this short piece, I highlight the concept of the global market portfolio, identify some of the recommendations stemming from this theoretical construct, and discuss some arguments for why we may not need to consider this theory the gospel. The general takeaway is that we often invest in markets in a theoretically sub-optimal way, but in a way that is adequate from a practical standpoint. We also focused on equity allocations. Fixed income can get trickier. The 17.4 trillion dollars of bonds represented by the Bloomberg Barclays Aggregate Index is less than half of the investable $39.1 trillion universe of US domestic bonds (as reported by Guggenheim). Is the AGG index good enough? Or do we need to dig deep into the 39.1 trillion dollars to maintain our tie with the global market portfolio’s theory? Possibly, but perhaps not. We can punt that discussion to another article.

In the end, investors should focus on what they can measure and control with more certainty, namely fees and taxes. Luckily, that task doesn’t take a PhD (or a high priced investment advisor). Most ETF’s are already structured to minimize capital gains and have relatively low fees. Tax loss harvesting strategies and maximizing the tax deferral nature of 401Ks and IRAs are also straightforward. So forget about optimizing your portfolio to match a global market portfolio construct and go back to sleep. You probably won’t miss much during your slumber.

  • The views and opinions expressed herein are those of the author and do not necessarily reflect the views of Alpha Architect, its affiliates or its employees. Our full disclosures are available here. Definitions of common statistics used in our analysis are available here (towards the bottom).
  • Join thousands of other readers and subscribe to our blog.
  • This site provides NO information on our value ETFs or our momentum ETFs. Please refer to this site.

References   [ + ]

1. source
2. Alibaba, at $350bn, is China’s 2nd largest company by market cap and 10% of it’s investable market!
3. Fama, Eugene F., and Miller Merton H., 1972, The theory of finance, New York: Holt, Rinehart & Winston., pages 20-21
4. “Father of Modern Finance”, Chicago Booth Magazine, Fall 2013
5. Fama, Eugene F., and Miller Merton H., 1972, The theory of finance, New York: Holt, Rinehart & Winston., page 22
6. the scale to which depends on your source

About the Author:

Jonathan Seed
Jonathan Seed began his career at Franklin Resources where he was an Assistant Portfolio Manager for their then quantitative asset arm, Franklin Asset Management Systems. There, he helped build value biased equity portfolios. After graduating with honors from the University of Chicago Booth School of Business, he began a 20 year career on Wall Street focused on fixed income, with an emphasis on structured products. He started in fixed income research before switching to institutional sales, leaving Credit Suisse as Managing Director in 2009 for RBS Securities and leaving the industry altogether in 2014, after which he started Seed Wealth Management, Inc., a Registered Investment Advisor incorporated in the state of Illinois. Visit www.seedwealthmgmt.com for a full summary of our approach.


  1. Hijo de la Luna June 14, 2017 at 3:28 pm

    Interesting that the AQR paper as far as I understand (went through it very quickly) suggests that it’s better to hedge the exposure to foreign stocks. Is it something you agree with? i.e. are IVAL and IMOM (and the part in VMOT exposed to foreing stocks) currency hedged? If not, why?

    • Wesley Gray, PhD June 14, 2017 at 3:35 pm

      We think there is room for debate on the question of whether or not to hedge FX.
      Our index construction does not involve FX hedging. If investors feel strongly about hedging FX this is something they can easily achieve on their own accord.

      • Hijo de la Luna June 14, 2017 at 3:50 pm

        What would be the arguments in favour of not hedging FX? I understand the logic of not hedging in GEM since there seems to be a correlation between the USDX and the US stock market, so when you bet through relative momentum that the US outperform the rest of the world, you’re betting that the USDX goes up – and viceversa. But if you have a stationary allocation, with a part to US and a part to non US, what’s the advantage of being permanently exposed to the risk of fluctuations in the non US currency for the part of non US stocks?

        • Wesley Gray, PhD June 14, 2017 at 4:17 pm

          We don’t find any difference in our particular context, but there is the added benefit of lower transaction costs, less complexity, and marginal diversification benefits.
          Long story short, our analysis, in our context, is different than that of AQR.

        • Jonathan Seed June 14, 2017 at 6:02 pm

          I’ll add that as a practical purpose, hedged international stock ETFs (e.g., HAWX and HEFA at 36bps) are more expensive than their non-hedged alternatives (e.g., IXUS and VXUS at 11bps). Many Int’l fixed income funds do hedge currency risk, though.

          Also, eye-balling the AQR rolling currency correlation, it seems to line up with the intuitive hunch that if the US was suffering more in isolation (like the 1980 Volcker, early-90s debt and early 2000’s recessions), the correlation is most often negative. Of course, 3 data points doesn’t prove much.

          • Hijo de la Luna June 14, 2017 at 10:43 pm

            yes the cost of the ETF is one of the reasons why I don’t hedge now (besides the fee there’s the additional cost due to interest rate differentials – though in theory that could go both ways https://etfus.deutscheam.com/US/EN/deutsche-x-trackers-blog/currency-hedging-costs-or-benefits) However, although I asked the q from the point a view of a US investor to make it more relevant to others, in my case I will retire in Euroland so in order not have too much currency risk I now overweigh euro-denominated stocks significantly relative to their market cap weight – something I am quite confortable with seeing the high valuations of US stocks. It’s true that people like Meb Faber have been suggesting for the past several years to overweigh non US stocks whilst the US has continued to significantly outperform, but momentum (besides valuations) seems to be now turning in favour of Europe.

  2. Aaron Brask
    Aaron Brask June 15, 2017 at 3:30 pm

    Nice article. I believe the free-float adjustments made by major index providers was another business rather than client-centric decision. It mitigated liquidity bottlenecks for behemoth indices but effectively reduced exposure to the owner-operator factor.

  3. Jonathan Seed June 15, 2017 at 4:40 pm

    I wish I had this https://papers.ssrn.com/sol3/papers.cfm?abstract_id=2978509 before I wrote this, but alas. At least their conclusions compliment mine: the 50/50 equity/bond portfolio outperforms their global market portfolio from both an absolute and a sharpe ratio standpoint. In otherwords, it’s close enough!

  4. John Butters June 16, 2017 at 4:39 am

    Thanks — interesting. Holding Japan at market weight was not so terrible if you compare the drawdown to those of 1973-74, 2000-2003 and 2007-2009. You make a good point about business risk but this is why upfront investor education is crucial for passive investing. Re. currency, from a UK perspective I tend to suggest that people be unhedged because the USD has a large weighting in a global index and tends to go up in bad times (e.g. 2 Sep 2008 – 20 Nov 2008 MSCI World fell 29% in GBP but 42% in USD). Many thanks again for an interesting post.

    • Jonathan Seed June 16, 2017 at 8:51 am

      The “flight to safety” aspect of the dollar point is a good one. And how much did your clients really suffer post-Brexit? Yes, the GBP got clobbered and it would have been wonderful to have perfect foresight and be 100% USD, but your clients consume in GBP (outside of the occasional trip overseas and some imported cheese). I also agree with you regarding upfront investor education, but it’s hard.

    • Hijo de la Luna June 16, 2017 at 1:47 pm

      concerning your remark on currency, please correct me if I’m wrong but if you’re a UK investor your country makes up approx. 6% of MSCI world. So if you have an allocation in shares similar to the global market portfolio, you have more than 90% in a currency other than GBP. Is this what you would suggest?

  5. Hijo de la Luna June 16, 2017 at 11:30 am

    I have been thinking a bit about the global market portfolio and the more I think about it, the less I understand it. It seems to me that the totality of investors in the world are different people who will each use their money in the currency of their own country, thus for an investor in a small country, e.g. a Swiss or UK investor, to own something close to a global market portfolio (currency unhedged) would seem to be madness to me.
    The other thing that doesn’t make sense to me is the parallel with physics (the falling anvil). The law of gravity to which you refer is a fundamental law of nature; the ‘laws’ of finance are not like the laws of physics, or if a comparison with physics really has to be made, it ought to be with something messy like tribology (i.e. friction, which is the product of many complex and not wholly understood concomitant processes). I am saying this here though the thing the puzzles me most in this context is people talking of momentum in finance as if it were a law in the same way that momentum in physics is. There was a trader (Sekyota) who even tried to write a paper in physics…

    • Jonathan Seed June 16, 2017 at 3:51 pm

      I’m no physicist (and stole the anvil analogy from my old Fama, Miller book), so am cautious to defend equating the law of motion with MPT. Before the it became law, though, the law of motion was a theory that had to be tested. Likewise CAPM. Neither had a perfect setting to test the theories, but eventually the theory of motion became a law. Not so much CAPM. Given the best theory backing the momentum factor is behavioral, I’m pretty sure it will never become law either. But boy, try getting the courage to dump your brick and mortar retail stocks today or finally biting the bullet and buying Amazon. I sure want to get paid a bit more for doing so.

  6. Sebastien Hitier June 16, 2017 at 2:31 pm

    The point made by readers about currency is crucial. Going back to MPT, investors with a different numéraire will see different returns and higher volatilities for foreign assets. So they will see different efficient frontiers. Foreign denominated short term bonds sure look more risky to you than to their local investor.

    • Jonathan Seed June 16, 2017 at 3:36 pm

      Very valid point, Sebastien, and one likely shared by the folks at AQR. This paper (https://papers.ssrn.com/sol3/papers.cfm?abstract_id=1385022) by MSCI-Barra, on the other hand, that hedging can often create more risk. I think it’s also important to note that within equities, many sales and profits are generated overseas (over 50% in the S&P and I read somewhere near 80% in Spain), so hedging and bending over backwards for direct international exposure may be overkill. To your point, I’m not sure if you can make a credible case not to hedge short term foreign denominated bonds — especially now that the majority of them trade at an under 1% yield!

Leave A Comment