When home and hot biases collide

Investor home bias prevails in markets around the world. So says a 1997 IMF survey of cross-border equity holdings in 29 countries, the first study that applied econometric cross-sectional analysis to bilateral portfolio holdings within a large sample of countries.

A 2004 IMF working paper compared the original survey’s observed percentages of domestic equity holdings with an optimal percentage predicted by the Capital Asset Pricing Model (CAPM).

There are valid reasons to deviate from CAPM-implied weights, but the survey provides a point of comparison. The authors consider the difference between actual weights and CAPM weights to be a descriptive measure of home bias. The numbers are striking.

In another paper, “Investor Diversification and International Equity Markets,” Kenneth French and James Poterba conclude, “non-trivial risk reduction is available from cross-border holdings.”

They also ask what set of expected returns would explain each country’s observed pattern of domestic holdings versus that of a value-weight strategy. They found home bias among U.S., Japanese, and U.K. investors implied higher return expectations of 90, 250, and 400 basis points in their respective home markets.

Also, an investor’s human capital is likely far more correlated with domestic than international markets, so the diversification benefits of investing globally are even more pronounced in total wealth. Despite this, the authors contend investors hold nearly all their wealth in domestic assets.

Literature has failed to develop a generally accepted explanation. Possible causes often cited include institutional constraints (e.g., the now-abolished foreign content limit in Canada), as well as taxes and higher transaction costs.

French and Poterba note while transaction costs may be higher in some markets, these cost differentials should lead investors to the most liquid markets, not their own domestic markets. Further, since all shares must be owned by someone (i.e., there are no orphaned stocks), differences in transaction costs should be reflected in variations in expected returns.

Recently, the home bias debate has pitted local investors’ informational advantage against behavioural issues. Investors tend to regard the potential performance of their own markets more optimistically. The French and Poterba paper documented that phenomenon, also known as relative return optimism, but this effect is also evident in research on market expectations of fund managers, as well as behavioural studies.

Information asymmetry is another behaviour-related explanation. It proposes an investor’s perception of informational advantage contributes to home bias, even though being close to the source offers no real benefit. Similar to relative return optimism, even sophisticated investors believe in the advantage of information asymmetry. A survey of fund managers in Germany by Torben Luetje and Lukas Menkhoff revealed the more they believe they’re better informed than foreign investors, the more they invest in domestic assets. Yet, these managers neither forecast home stock market developments better, nor rely on data sources that are only available locally.

New developments and data suggest a general reduction in equity home bias. Plausible explanations include globalization, free trade growth, the Internet, the rise of emerging markets, and increased mutual fund investment.

Leaving home

Some advisors structure portfolios to benefit from global diversification. Yet diversification neither assures a profit nor eliminates the risk of market loss. Two reasons for their partial immunity to home bias are:

• A solid grounding in modern finance; and
• An equilibrium view of markets.

Modern portfolio theory helps advisors recognize the benefits of diversification. Also, a belief in the efficiency of capital markets offsets the perception of information asymmetry that traps investors into high allocations of domestic securities.

There is more information in the price of a security than anyone possesses individually, regardless of where they reside. With an equilibrium view, you don’t care how much information you have about foreign securities. Instead, you care about how much information all market participants collectively have.

Even though the benefits of global diversification are undeniable, when the local market outperforms, clients could criticize portfolios designed with less domestic exposure. At this point, home bias will collide with an even more potent behavioural influence: hot bias, meaning a desire for clients to invest disproportionately in whatever has been doing well. Hot bias added to home bias can be like pouring gasoline onto a fire.

The resulting explosion of emotions can lead even savvy investors to abandon thoughtfully constructed portfolios and, according to Wall Street Journal writer Roger Lowenstein, conclude, “You can lead a happy investment life without leaving home.”

As a sophisticated investor, even Lowenstein fell prey to the collision of home bias and hot bias, as the 10-year annualized return for U.S. equities in 1997 was 18%, versus a return over the same period for non-U.S. equities of 6%.

Unfortunately for readers who acted on this idea, the following decade was marked by an annualized return of 6% of the S&P 500 Index, compared to 9% for the MSCI World ex U.S. Index. A concentrated portfolio of U.S. stocks would have underperformed a globally diversified portfolio while exposing investors to significantly more risk, despite the perceived safety of investing at home.

Collision in Canada

Canadian investors recently endured a similar collision of home and hot bias.

The chart (see “Five-year annualized return difference,” right) shows the annualized return difference between Canadian and U.S. equities for rolling five-year periods from 1961 to 2012. For nearly 20 years, there was never a five-year period when Canada outperformed the U.S., yet Canadian investors maintained portfolios tilted towards Canada.

But from 2003 to mid-2012, the five-year performance spread reversed, with Canadian equities outpacing the U.S. As the hot bias fuel poured onto the home bias fire, the global trend toward less home bias didn’t hold here.

If Canadian investors didn’t reduce their home bias during a long span of sub-performance relative to the U.S. (1984-2002), why would they diversify away from home when stocks were soaring?

Canada has had a great run relative to the U.S. But all good things come to an end, and in 2012, Canadian equities underperformed the U.S. over a five-year period for the first time since 2003. Is this the beginning of another hard lesson?

Severely home biased investors should learn from the past three decades and consider committing a larger share of their equity portfolios to non-Canadian markets.

Home bias survey

County	Actual Percentage of Domestic Equity Held	CAPM Optimal Percentage of Domestic Equity Held	Home Bias (Actual − Optimal)
Australia	77.96	1.46	76.50
Austria	65.82	0.20	65.62
Belgium	70.37	0.67	69.70
Canada	70.39	2.72	67.67
Denmark	71.90	0.60	71.30
Finland	88.44	0.39	88.05
France	79.75	0.42	79.33
Ireland	14.33	0.21	14.12
Italy	78.31	2.12	76.19
Japan	89.38	13.33	76.05
Malaysia	96.61	0.44	96.15
Netherlands	59.72	3.00	56.72
New Zealand	68.57	0.16	68.41
Norway	67.19	0.28	66.91
Portugal	91.92	0.29	91.63
Singapore	72.06	0.48	71.58
Spain	86.38	1.41	84.97
Sweden	69.55	1.37	68.18
United Kingdom	69.37	11.43	57.94
United States	85.45	55.20	30.25

Source: Hamid Faruqee, Shujing Li, and Isabel K. Yan, “The Determinants of International Portfolio Holdings and Home Bias” (IMF working paper, February 2004).

Brad Steiman is director, head of Canadian Financial Advisor Services, and vice president of Dimensional Fund Advisors Canada ULC.