Blog #44 Humpty Dumpty Markets

I want to revisit an aspect of global equity diversification that I haven’t covered in my previous posts. This is the fact that diversification will not protect you from losses if the global economy tanks. As the nursery rhyme says, “all the king’s horses and all the king’s men could not put humpty dumpty back together again.” Market risk is what investors bear when they hold a diversified portfolio. Just as a rising tide lifts all boats, a crashing tide sinks all boats!

That is the bad news.  The good news is that global diversification may still offer better protection against downside risk compared to investing locally. To see whether there is any truth in this, I will address two questions in this blog. Question #1: Does a globally diversified portfolio fall more or less than individual markets on average?  Question #2: when the dust settles and markets recover, does a globally diversified portfolio recover more or less than individual markets on average? Does it recover sooner or later?

To answer these questions, we will probe the data scientifically. In what follows, I will share with you the findings of an evidence-based study which examines downside risk across 16 stock markets over a long period of time. These findings come from supervised research done by senior NUS undergrads reading Personal Finance and Wealth Management, a course I taught in NUS Business School for many years.

Before I discuss the results, here are the background facts:

  • The study analyzed the total returns of 16 markets based on local MSCI indices.
  • Returns for individual markets are in local currency units (e.g., Australian dollars for the Australian market, Singapore dollar for the Singapore market and so on).
  • A global portfolio is formed by equally weighting international markets.
  • To study the effects of diversification, we take perspective of an investor based in each of the 16 countries (say, Singapore). We compare what this investor earns from investing only in Singapore stocks with what he could have earned by holding a globally diversified portfolio of 15 markets (all markets except Singapore). Let’s call this global portfolio, G.
  • To ensure that we are making an apples with apples comparison, the returns for G are converted to Singapore dollars (i.e., we assume that the Singapore investor does not hedge against currency risks). This is a realistic assumption for most people. We repeat this exercise for the other 15 markets.
  • We adjust all returns using the respective country’s CPI, so as to focus on real returns.
  • The sample period for the most recent study is December 1969 to June 2017, a period of 48 years.
  • The main question we ask is: how does G perform when a particular market is doing badly?  We define “badly” in a number of ways: first, as worst monthly return over the sample period, secondly as the 1% conditional value-at-risk or 1% CVAR and thirdly, as the 5% CVAR. In general, the x% CVAR is simply the average of the worst x% returns.

Here’s a quick pictorial recap of some of the above points:

#1 We study how G performs when a local market is doing badly.


#2 We study a total of 16 markets.


#3 We compare the average returns of each individual market with the returns of G, the globally diversified portfolio.


#4 We want to see whether global diversification really protects against the downside risk of individual markets.



I present five key results from the study/

Result #1. Over the short term (one-month time horizon), G outperforms individual markets in most cases. 


Result #2. G has lower downside risk (smaller 5% CVAR) than any individual market


How to read this graphic: The vertical axis plots the 5% CVAR for G; the horizontal axis plots the 5% CVAR for individual markets (shown as dots on the plot). Numbers on each axis are negative as expected since we are talking about bad months. For both axes, numbers are more negative nearer the origin. Therefore, if diversification protects against worse case scenarios of individual markets, all the dots should be on left of the diagonal line. This is what we find.

Result #3 G has smaller 1% CVAR than any individual market 

This result is qualitatively similar to result #2.


Result #4. As time passes, G’s performance gap over individual markets widens

The graphic below shows that after 80 months, the 5% CVAR for G becomes positive, while the average CVAR for individual markets continue to be negative. Moral of the story: global diversification pays (eventually)!



Result #5 The diversification story is the same when we look at worst returns instead of CVAR.


In short, the evidence shows that diversification is good in the short run, and gets better in the long run as markets shake off the bad news and respond to the pull of the global economic engine.

You would think that investors are smart enough to know all this. WRONG! All over the world, investors are infected with a “home bias”, the tendency to believe that its best to own just local stocks (see next graphic). If this describes you, now is as good a time as any to kick the home bias habit.





Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s