The Investment Fund for Foundations: www.tiff.orgUS vs. Foreign Stocks:
The Arguments for Equal Weighting
First Principles
The Multi-Asset Fund (MAF), under normal circumstances, allocates 50% of its capital to a diversified portfolio of stocks. Trustees rarely question the fact that the fund invests half of its capital in equities: given its primary objective of earning a long-term real return of at least 5% per annum, net of expenses, a substantial allocation to stocks is virtually required. However, trustees periodically ask why the fund normally allocates fully as much to non-US stock markets (25% of fund assets) as it does to the US stock market (also 25% of fund assets). Indeed, this question has been asked so frequently of late that we have decided to answer it in writing.
Statistical Rationale
Allocations to non-US stocks have typically been based on a statistical analysis of US and non-US stock returns. Non-US stocks are generally assumed to provide a return at least equal to that of US stocks (an assumption that we consider reasonable); thus allocating capital to foreign stocks neither hurts nor helps expected returns. However, on the risk side of the equation, the less-than-perfect correlation between US and non-US stocks reduces expected portfolio volatility. This reduction in expected portfolio volatility without a decrease in expected return is often referred to as the "free lunch" provided by diversification.
Why Now?
Before delving further into the rationale underlying MAF's allocation to non-US stocks, it is worth considering why trustees are currently so curious about this issue. We believe there are at least three reasons. First, the typical US-based eleemosynary institution allocates approximately 70% of its portfolio to stocks, with approximately 10% allocated to non-US equities. Non-US stocks thus represent approximately 15% of the stock portfolio. MAF's equal allocations to foreign and US stocks thus create the potential for short-term embarrassment for those trustees who have dared to be different. Of course, the greatest discomfort typically arises when one is not only alone, but when one appears in hindsight to have been wrong. From the fund's inception on March 31, 1995, to December 31, 2002, non-US stocks as represented by MSCI's All Country World Free ex US index have produced an annualized return (in US$) of only 0.28% compared to an annualized return of 8.68% for the Wilshire 5000 index of US stocks. To make matters worse, the correlation between US and non-US stocks has been quite high of late. As shown in the graph below, the correlation has increased to over 0.8, or closer to the maximum correlation of 1.0 where one would expect to derive no diversification benefit. This latter point is troubling: most trustees know not to extrapolate past returns, but persistently high correlations seem quite possible given increased globalization of many industries. While it will become abundantly clear that MAF's allocation to non-US stocks is not based on statistical notions of diversification, we note that correlations have been nearly this high before (e.g., during the oil shock of the early 1970s), are quite volatile, and can fall quite rapidly.
Data depict the trailing 3-year correlation between the monthly total returns of the S&P 500 and EAFE indices. EAFE results are dollar-denominated, i.e., they reflect currency gains or losses experienced by US dollar-based investors. Sources: Standard & Poors and Morgan Stanley Capital International.
Opportunity Set
The primary reason that, under normal circumstances, 50% of MAF's equities are allocated to foreign markets is that this allocation mirrors the opportunity set. Just as an active investor in US stocks should not arbitrarily ignore specific industries, an investor should not arbitrarily ignore specific geographical sectors of the stock market. As of December 31, 2002, non-US markets represented approximately half of the world's stock market capitalization. By allocating only 15% of its stock portfolio to non-US markets, the typical US-based endowment effectively narrows its opportunity set. All else equal, the broader an endowment's opportunity set, the higher its expected return.
False Presumption
Many trustees may respond to the above by stating that all else is not equal. They posit patriotically, if somewhat myopically, that the US way of doing business is clearly superior to that of foreign countries. Examples certainly abound in foreign countries of structural barriers to success — the 35-hour workweek in France, the Japanese people's apparent resistance to rapid change — yet, as recently as the late 1980s, US business schools exalted the Japanese way of doing business. Furthermore, as recent abuses involving sell-side analysts and corporate honchos have reminded us, the US is not free of flaws that permit misallocation of capital. Indeed, some foreign investors perceive that the US stock market is too flawed to warrant investment. Nevertheless, the correct question is not whether structural flaws or other issues exist but whether or not current prices adequately discount those issues. Just as there is a difference between a good company and a good stock, so too are there differences between sound economic systems and sensible means of participating in them.
Leverage to Change
Somewhat counter-intuitively, the existence of structural barriers can lead to future investment opportunities. To name just one example, stock buybacks, a standard value creation tactic in the US, were illegal in Japan until only recently. By investing wisely in markets experiencing regulatory and cultural changes, endowments may increase returns substantially.
Arbitrage
The case for maximizing the opportunity set is particularly compelling when an endowment employs global equity managers. Because global managers research companies without regard to domicile, they are well prepared to identify and exploit price discrepancies created by, among other things, varying national and regional levels of fear and greed. This summer (2002) MAF added a second global equity manager, Delaware International, to more fully exploit such cross-border arbitrage opportunities. Over half of MAF's stock exposure will now be managed under a global mandate.
Currency Risks
International investing unquestionably involves incremental currency risk. All other things equal, a US-based investor in foreign stocks sacrifices return when the dollar is appreciating against foreign currencies. Of course, the opposite is true: when the US dollar is falling against foreign currencies, US-based investors who have allocated capital to foreign markets are rewarded. Over investment horizons appropriate to eligible members, most of which are perpetual institutions, such currency swings tend to be self-correcting and do not impact meaningfully long-term returns.
Relative Efficiency
A final argument in favor of investing in non-US stock markets is that active managers generally find it easier to add value than in the US. For example, since each fund's inception on May 31, 1994, through December 31, 2002, TIFF's International Equity Fund had outperformed its passive benchmark (the MSCI All Country World Free ex US) by over three times the annual margin of outperformance enjoyed by the US Equity Fund versus its benchmark of the Wilshire 5000 (see current and standardized performance on the performance section of the TIFF Website. Any endowment engaging in active management should favor rather than avoid those markets in which active managers are likely to meet with success.
Performance data reflect past results; future returns could be quite different. Investment return and principal value will fluctuate with market conditions, and a member's shares, when redeemed, may be worth more or less than their original cost. Total return assumes reinvestment of dividends. Investments in some of the TIFF mutual funds entail risks not associated with funds that invest solely in the United States. Foreign stocks may involve greater volatility; political, economic, and currency risks; and differences in accounting methods.
The MSCI All Country World Free ex US index comprises primarily large cap non-US stocks; the Wilshire 5000 represents substantially all publicly traded US stocks, weighted by capitalization. One cannot invest directly in an index.
This information is authorized for use when preceded or accompanied by a prospectus and the most recent quarterly report for the TIFF mutual funds. The prospectus contains more information about ongoing fees and expenses arising from an investment in the funds.
