Homerism (and notes)

2 Flares Twitter 1 Facebook 1 Filament.io 2 Flares ×

I’m rereading The Four Pillars Of Investing by William Bernstein.

This is a terrific book, that I highly recommend everyone interested in investing to read at least once.

Bernstein is a remarkable mind who is able to take complex economic history and theory and calculations and distill them all down into a very readable and entertaining narrative that actually helps you come up with an investment plan suitable for most contingencies.

And some of the concepts are so simple that their brilliance makes you wonder why you hadn’t thought of them yourself.

One of the tenets of behavioral finance is that we humans are biased to favor our own backyard. We all have tendencies to think that our home countries are uniquely structured and just. And we’re more comfortable pouring our own money into the familiar (when perhaps we shouldn’t be).

And you see evidence of this bias everywhere. Despite the fact that the US makes up only 50% of the world economy, people will routinely argue that they achieve enough international diversification just by investing in the whole US stock market due to the presence of multinational corporations and the foreign sales of these companies. This is despite the fact that there is an imperfect correlation between foreign and US indices, and thus a diversification benefit (at least historically) in owning both.

I myself am somewhat guilty of this in that I invest 60% of my equities domestically and only 40% internationally which is not in keeping with the structure of the whole world market. My justification for this slight favoring of domestic equities is that they are cheaper. But who am I to deny that I am also guilty of the home court bias?

People will also commonly refer to the unique infrastructural, and regulatory advantages of America over the rest of the world. And I am not so sure that there is so much evidence for this going forward.

But the point is this. A common graph you will see in investing is this one.

This purports to show the growth of one dollar in the US stock market over its history. And it is used to justify the magic of compound interest. (Which is in fact a very magical thing.)

But the point the Bernstein makes is that when we focus on the historical returns of this particular stock market, we are guilty of two biases: selection bias and survivorship bias.

Selection (or sampling) bias occurs when your study is flawed because the population studied has already become skewed by the method of recruitment.

The classic example of this was the polling data from the 1948 election between Truman and Dewey. Because the novel telephone was used to collect the polling data, The polling population was skewed towards the wealthy who had the resources to own phones. This made it seem as though the more conservative candidate was winning, when fact he was not.

Survivorship bias occurs when you study a population that gets smaller throughout the course of observation. Because the final population has been whittled down, it becomes skewed towards the traits of survival.

The classic example of this is in studies of mutual funds, where you look at the performance of current mutual funds over the last 15 years. In doing so you’re ignoring all of the funds that have gone out of existence during the time period of your study prior to your starting of the study. So there is a an invisible portion of your population that you unwittingly exclude from your analysis. You are, in fact, only looking at the winners.

The point being that when we look at the incredible performance of the American Stockmarket of the past couple hundred years, we are unwittingly selecting a winner.

I mean, how would a dollar invested in the Prussian stock market look right about now?

In a way this is quite similar to looking at Warren Buffett’s Record of success investing in the stock market and concluding that stock market investing is incredibly profitable.


Investing is easy


As Bernstein points out, when the American stock market started out it was a frontier market (even more unstable than what we would call an emerging market these days.)

It was indistinguishable at that time from any number of other markets whose subsequent track record was quite a bit less successful, an in some cases calamitous.
And just because there’s been tremendous success in the past for America, this does not mean that it was inevitable, nor that future outsized success is inevitable.

America could be just like the lucky active investor who has beaten the broad market year after year through good fortune.  Research suggests you should be very wary of betting on that investor.  Past performance does not guarantee future success (and may even increase the probability of future failure.)

And the take-home message is this I think:

  • When we look at data, we must be very careful about drawing broad conclusions. We often have blind spots and biases.


  • It’s probably a good idea for us to mix in more foreign assets into our portfolios, than feels comfortable.

And before I sign off a couple notes.

1. Check out this site.


It’s a coming brokerage that will offer fee free trading. If they include ETFs this is pretty exciting. I’m on their waitlist as of yesterday.

2. Here’s a new manufactured spending opportunity which I’m thinking about looking into. See if it has any appeal for you. (It’s chief appeal for me is that it involves no Walmart.)


2 Flares Twitter 1 Facebook 1 Filament.io 2 Flares ×

5 Responses to “Homerism (and notes)”

  1. Robert May 19, 2014 at 11:35 am #


    Interesting and thought-provoking post! You touch on issues that many investors are not aware of, yet with profound implications.

    Selection and survivorship bias are indeed major problems; anchoring is another behavioral factor affecting how we view US vs. ROW (rest of world) performance. However, I think the US does indeed have inherent advantages in resources, population, culture and government that are not likely to change anytime soon. (Well, I could suggest a few ways such change is happening, but you wouldn’t agree with me on those being negatives, so I’ll leave the politics aside from this discussion. 😉 ). Just as the rich get richer, not everything is mean-reverting, and I think the US vs. ROW falls in that bucket. As the report I’m about to cite said, if you line up 5 people off the street alongside a worldclass sprinter and have them race, you can predict who’ll win (no guarantee, to be sure; the sprinter might get hit by a bullet or something, but it is by far the wisest bet). You might not have been able to predict which would be fastest when they were infants, but now you can. Past performance **can** be a predictor in some cases. Assuming that US stock performance since the 1700s is due to random good luck and thus betting instead on the ROW is like betting against the world class sprinter, if in fact there are specific reasons for the US’ success and these are persistent.

    I used to think I needed to diversify more into international, but I have pulled back–well below your 40%. Reasons: (a) there is indeed a large exposure to international economies through investment in US-listed multinational corporations; (b) the correlation between emerging and developed markets has been increasing for many years, so the diversification benefit from buying ROW equities has been decreasing; (c) the correlation between emerging markets and developed markets has become especially strong (near 100%) during developed market crises, such as we experienced in 2008. Thus, the diversification doesn’t reduce portfolio volatility when you need it most. (d) US returns have consistently been higher than the ROW in aggregate.

    I also have changed how I think about risk. Conventionally, people talk about volatility and equate that to risk. But there are many other kinds of risk (including the risk that you’ll lose all your money, or the risk that you won’t achieve the rate of return necessary to achieve your long term goals). While markets don’t reward risk that can be avoided through diversification, markets do generally reward risk. However, sometimes too many people bid on risky assets resulting in risk premia that aren’t enough to compensate the investor for the increased risk taken. Thus, for example, people overbid growth stocks and undervalue value stocks. It is sexy to go for the homerun stocks. But as Jeremy Siegel made clear in one of his books, the best performers in the US longterm were boring stocks like tobacco, because they weren’t overbid by investors. I think the same thing goes on with emerging markets; they are overbid relative to the real risk involved, so investors are undercompensated for the risk and volatility they incur.

    The reason you usually find for adding international stocks is that the imperfect correlation between domestic and international markets results in a portfolio with reduced risk for a given rate of return (or higher return for a given level of risk). However, risk in this case means volatility. If you are a longterm investor with the intestinal fortitude to not exit the market during downturns, how much do you care about volatility vs. longterm rate of return? What if the asset class you are adding (international stocks) is overpriced because many others are trying to purchase it for the same reasons you are, or because they make the false assumption that high growth rate economies necessarily result in high growth rate stock markets? If international stocks reduced your portfolio volatility but also reduced your returns, would you still be interested? And what if the volatility reduction disappeared in times of crisis, when you most want it?

    One of the most influential documents to modify my thinking, and one I cannot recommend highly enough for you to read (it is a quick read), is the annually published yearbook from Credit Suisse. I came across it a couple years ago and now try to read it annually. In the 2014 edition there are good discussions of the issues noted above, with graphs showing the changing correlations between emerging and developed markets over time. One of the most interesting parts of the yearbook are the comparisons between major markets by nationality. You can see what the US has done compared to the rest of the world in aggregate. Studying this a couple years ago made me believe that investing in US stocks was best for me. I may someday buy indices for S. Africa and Australia, both resource countries, since they have consistently outperformed the US. Otherwise, I’m not seeing much benefit in investing in the rest of the world. I still do have some international funds (about 10%), but not nearly what I used to. I’m not sure why I still hang onto those, except perhaps the pressure of conventional thinking! Here’s a link to the Credit Suisse yearbook for 2014: https://publications.credit-suisse.com/tasks/render/file/?fileID=0E0A3525-EA60-2750-71CE20B5D14A7818.

    I love their data-centered approach and rational analysis; I think you will too. I’d like very much to hear your thoughts on why you should or should not keep a 40% international allocation (or even higher as you seem to be advocating in this blog) after you digest the Credit Suisse report.

    • Miles Dividend M.D. May 19, 2014 at 3:46 pm #

      Robert, thanks for the link. I don’t have time to look at it right now but will check it out later tonight. I’m excited to do so.

      I Have a couple impressions about your comments which were very interesting. The first is that the way someone structures their portfolio really tells you a lot about them.

      I see the chief advantage of the US since its inception as it’s plentiful resources, and geographic isolation, that has allowed the homeland and industrial complex to remain pretty much undisturbed for centuries.

      We also have an excellent political system with peaceful transitions of power.

      And there’s no reason to believe that our geographic isolation will not continue to be our literal moat going forward.

      But I don’t think that the US has outperformed the rest of the world’s developed countries by all that much in the past 40 or 50 years.

      Furthermore, if past is prologue, one would’ve been better off having 40% international allocation then any lower amount of allocation to the international developed markets over the same time period.(By better off I mean your CAGR would’ve been higher. )

      Which gets to my second thought. The remains imperfect correlation between international and US markets. And decreasing standard deviation is not just a matter of sleeping well at night.

      By combining two uncorrelated assets and decreasing the standard deviation of the portfolio one increases the CAGR (ie actual return) of the portfolio, even if the arithmetic return looks less impressive on paper. And since you can’t Bank the arithmetic return, you’re better off with a higher CAGR.

      The final point is that by most metrics the US stock market looks overvalued relative to international and especially emerging markets at this point in time. This suggests that if one wants to time the market, now might be the right time to increase your international exposure(If you’re into strategic asset allocation.)

      The coolest part though is that by making our portfolios we can create an asset that perfectly reflects our own Way of seeing the world. It’s kind of like taking a personality test don’t you think?


      • Robert May 19, 2014 at 3:59 pm #

        Your last point (undervalued markets in some other parts of the world) is the one that I can probably most closely agree with. But with a great deal of uncertainty. Is Europe cheap because it is in a cyclical downturn about to reverse, or because it has long-term problems with no end in sight? That is the perennial question about assets that have declined in price, isn’t it?!

        I’ll be interested to hear if your thinking has changed on CAGR and last 40-50 years etc. after you read the report. It was eye-opening for me, and written by a research group with an international perspective, of course (Credit Suisse).

      • Robert May 23, 2014 at 9:54 am #

        Alexi, have you had a chance to read the Credit Suisse report yet? I’m interested if it changed your thinking in any way, and if so, how. I’ve already told you how it (an earlier version, actually) changed my thinking, and it is OK if your preference for a large international stock weighting remains, but I’m curious if it has.

        • Miles Dividend M.D. May 23, 2014 at 9:58 pm #


          I got through emerging markets and am enjoying the analysis. I have a feeling I haven’t gotten to the meat of the argument yet so I am witholding judgement.


Leave a Reply

Visit Us On TwitterVisit Us On FacebookVisit Us On Google Plus