NEW HAVEN, Conn. - Conventional wisdom says international stock investing adds diversification to the total equity portfolio. Well, it turns out it depends on when you look.
A new paper by three Yale University professors shows that benefits of global diversification - a way to reduce risk - come and go over time. And right now, correlations among international equity markets are near their highest point in 150 years, rivaling the peak reached during the Great Depression. High correlations mean diversification benefits are at or near their lowest point. What's more, the paper found half of the benefits of diversification are due to the number of world markets available today - a factor not generally considered. (The other half is from low correlations.)
A conundrum
For U.S. pension funds, with an average 12.9% of their assets invested in overseas stocks, this creates a conundrum. With U.S. investors earning far higher returns from domestic stocks than from overseas equities for the past decade, the major argument for international exposure has shifted to achieving greater diversification in the total stock portfolio. (Last year, the Morgan Stanley Capital International Europe Australasia Far East index's gross return was -21.1% v. -11.9% for the Standard & Poor's 500 index.)
Losing patience
The problem is that diversification benefits haven't delivered. Plan sponsors are "losing their patience," said Jay Kloepfer, director of capital market research at Callan Associates Inc., San Francisco, adding no Callan client has yet pulled the plug on international investing.
Increasingly, investors have been questioning whether the diversification benefits still exist. Average correlations among developed markets have risen sharply in recent times. They now are at 0.87, measured on a rolling five-year basis, driven by the global technology stock rise of the late 1990s. That compares with a 30-year historical average of 0.61.
The bottom line is that there's both good news and bad news from periods of globalization in the world economy, according to the paper written by William N. Goetzmann and K. Geert Rouwenhorst, professors at Yale University's School of Management, New Haven, Conn., and Lingfeng Li, a Yale University economist.
The good news is when world economies integrate, the available investment opportunities increase. The bad news is that diversification benefits increasingly rely upon investing in emerging markets - which have been performing dismally through much of the 1990s, although they have been rebounding lately.
What the academics have done is look at the benefits of diversification over a far longer period than the usual 20 or 30 years used by consultants.
Looking back over the past 150 years, they find that correlations are highest during periods of international financial integration, such as during the late 19th and late 20th centuries. The correlation pattern during the period mimics a vast "U" shape when graphed, revealing diversification benefits are more easily obtained during some periods than in others.
"Despite the limitations of our data, we find that international equity correlations change dramatically through time, with peaks in the late 19th century, the Great Depression and the late 20th century," the authors wrote in "Long-Term Global Market Correlations."
For example, during the two decades following World War II, low correlations existed among major national markets, the paper noted. But, during the last 20 years, when investment opportunities expanded greatly, those diversification benefits shrank, forcing investors to turn to emerging markets for risk reduction.
But high correlations during troubled periods, such as the 1929 crash, raise questions as to "whether diversification works when it is most needed," the paper said.
"The Crash of 1929 thus not only surprised investors by its magnitude, but also by its international breadth," the paper added.
Number of markets
While many experts have noted the importance of changing world market correlations, the authors also conclude that just as crucial is "the number, range and variety of markets that emerged or re-emerged in the last quarter of the 20th century."
What is striking is how much the opportunity set increased from the perspective of the U.S. investor, from only five markets around 1900 to about 50 a century later. Some markets, such as Czechoslovakia, had dropped out of the set just before World War II, only to reappear decades later.
The paper starts with a universe of four major markets for which total return data is available since 1872 - France, Germany, the United Kingdom and the United States. Investing on an equal-weighted basis in only those four major markets reduces risk by 30%, while investing equally across all available markets currently can reduce portfolio risk by about 65% of the volatility of a single market.
The upshot, the paper found, is that half of international diversification benefits are due to the increasing number of world markets available to investors, while half is due to lower average correlations.
Yet investors do not generally equal-weight countries. That would require them investing as much in, say, Sri Lanka as in Japan.
On a capitalization-weighted basis, risk reduction drops. Using the four core countries only, risk reduction falls to 20%; using a cap-weighted portfolio consisting of all available countries, risk reduction is 45%.
What's more, using cap-weighted data, "the diversification benefits are not dramatically less in the 1990s than they are in the 1970s," the authors wrote.
Thus, rising average correlations from emerging markets is not that significant because correlations from some of these markets are only marginal when the portfolio is viewed from a cap-weighted perspective, they wrote. Risk reduction obtained from emerging markets "has actually been roughly the same over the past 25 years," they concluded.
Experts agree
International experts praised the paper. "I think their basic point is right," said John Y. Campbell, managing partner, research, at Arrowstreet Capital LP, Boston, and Otto Eckstein Professor of Applied Economics at Harvard University, Cambridge, Mass. "There has been a big runup in correlations" over the last 50 years, and particularly the past 20 years, he said.
However, the paper exaggerates correlations by looking at dollar-based returns instead of examining returns on an unhedged basis. Using dollar-based returns adds a common factor, because international stocks tend to move together when the dollar moves up or down.
Mr. Campbell also said he would have preferred the paper to examine returns from the perspective of different home countries, and not just that of the U.S. investor.
But "these are quibbles about (the) magnitude" of correlations, and do not take away from the paper's central argument, he said.
Ken Kroner, managing director and research officer for Barclays Global Investors' global advanced active division, San Francisco, also praised the paper but worried it would be misinterpreted. "My fear is that people will walk away ... with the feeling that international diversification doesn't work," he said.
In reality, the paper shows that international investing provides 65% risk reduction over the last five years. "That's huge," Mr. Kroner said. BGI's own analysis shows diversification benefits should lead to U.S. pension funds increasing their allocations to international stocks by two to four percentage points, he added.
