One of the most revered principles of investing is that it pays to have a diversified portfolio. This precept was championed by Harry Markowitz in the early 1950s when he demonstrated that investors can reduce overall portfolio risk by adding risky assets that are not perfectly correlated with each other.1 In his words “diversification is the only free lunch in investing.”
While the initial application of Modern Portfolio Theory (MPT) was for domestic portfolios, academics extended the analysis to include international equities in the 1970s.2 This occurred during a period when international equities outperformed U.S. stocks and returns were not highly correlated.
Global investing became more prevalent in the 1980s and 1990s as countries relaxed restrictions on international capital flows and emerging economies regained access to international capital markets. By the turn of this century, few practitioners questioned the benefits of international diversification. Moreover, some advisors recommended that U.S. investors make strategic allocations to international and emerging markets of as much as one half of their overall equity allocation based on market capitalizations.
During the 2008 Global Financial Crisis (GFC), however, the benefits of diversification came under scrutiny when virtually all equity markets sold off in tandem. This caused some observers to assert that diversification benefits are least when you need them the most.
More recently, skepticism about international diversification has increased owing to two important developments. First, the U.S. stock market has outperformed non-U.S. markets materially in U.S. dollar terms – by a factor of 2.5 times – since the GFC (Figure 1). Second, over the past two decades the correlation between U.S. and international equities has increased significantly.
Accordingly, researchers have been re-examining how, if at all, the case for international diversification has changed. In this paper, we cite two recent studies that offer differing perspectives about this issue. We next investigate the reasons that U.S. equities have outperformed so extensively over the past decade and what it may portend for the future. This is followed by a discussion of whether international allocations for individual investors should be different than for institutional investors.
Our main conclusion is there are important benefits of international diversification, but they have diminished over the past two decades as correlations have increased. Therefore, for diversified portfolios, we are comfortable maintaining a strategic allocation to non-U.S. equities of up to approximately 20%. Also, within these portfolios, we do not favor making a large tactical tilt to international equities now, as the outcome may partly hinge on how tech stocks fare versus cyclical stocks over the coming decade, for which it is too early to be confident.
We begin by reviewing the benefits of diversification. Figure 2 presents data on returns and correlations for U.S. and non-U.S. equities for periods of market turbulence over the past two decades. They include the tech bubble in the early 2000s, the Global Financial Crisis from 2007-2009, the European debt crisis from 2010-2012, and the Coronavirus Pandemic in 2020. The data show both the returns that occurred as well as the correlations between domestic and international equities.
| Equity return | tech bust (2/2000 -2/2003) | great financial crisis (10/2007-3/2009) | european crisis (2/2011-4/2012) | covid shock (2/2020-current) |
|---|---|---|---|---|
| U.S. (S&P 500) | -37% | -50% | 12.2% | 15% |
| International (ACWI ex. U.S.) | -42% | -55% | -6.5% | 16% |
| Correlation (U.S. vs. International) | 0.85 | 0.90 | 0.93 | 0.95 |
These experiences have given rise to the perception that international diversification does not protect investors when they need it most. But is this perception justified when one looks beyond the crisis period?
A recent paper by Rodney Sullivan reaffirms the case for diversification domestically and internationally.3 The thrust of Sullivan’s article is that claims that “diversification fails when needed most” are not substantiated by the data. His analysis for a domestic portfolio of U.S. stocks and bonds (60:40) shows that over the past 75 years, a balanced portfolio helped to dampen the magnitude of peak-to-trough drawdowns during each of the 16 U.S. recessions that occurred. Sullivan also notes that the recovery of balanced portfolios has been at least as rapid as that of a stock-only portfolio.
Sullivan also analyzes results for globally diversified (long-only) portfolios since 1994, which is the common starting point for the nine asset classes included in the analysis. His findings reaffirm that more diversified portfolios not only provided downside protection; they also experienced quicker recoveries than less-diversified portfolios. The conclusion Sullivan draws is that, “Although a global diversified portfolio has not eliminated the risk of declines in portfolios in the past, it has been shown to provide both meaningful downside protection while also being an efficient portfolio.”
By comparison, a paper by Ford Donohue published by the CFA Institute argues that the evolution of the global economy in the recent decades has altered the relationship between U.S. and international stocks.4 Donohue’s analysis considers data for the period from January 1970 to June 2019 for portfolios that are rebalanced monthly. Over this 50-year span, he finds that while international equities slightly underperformed their U.S. counterparts and were more volatile, the inclusion of international equities improved the overall risk-adjusted returns of most portfolios. This reaffirms the traditional case for diversifying internationally.
However, the results change materially when data are examined for rolling 10-year and 20-year periods. The reason is correlations between U.S. and international equities have increased significantly -- to 0.86 over the past two decades from an average of 0.50 previously (Figure 3). The main impact of this change is the benefits of international diversification are lower today than they were previously.
Donohue finds that investors can now allocate up to 20% of their portfolios to international equities without increasing overall portfolio volatility. Beyond this, any increased allocation to non-U.S. equities entails a tradeoff between risk and return. For example, when returns for U.S. and international equities are assumed to be the same going forward, the model Donohue constructed recommends a strategic allocation to international equities anywhere between 0% and 20%, which is well below the 20% to 40% range that some consultants recommend. In this regard, Donohue’s findings are consistent with our own view that the strategic allocation for international equities should be about 20% of equity holdings within diversified portfolios.
Outperformance of U.S. Equities
While the increased correlations in the past two decades are widely attributed to globalization and the creation of the euro-zone, the substantial outperformance of U.S. equities relative to non-U.S. equities for more than a decade now is an enigma to many.
The most common explanation is that valuations, as measured by one year forward P/E multiples, are currently considerably higher in the U.S. than for non-U.S. equities. Some observers attribute this to record low interest rates in the U.S. and to the increasing dominance of a select group of U.S. technology companies – the so-called FAANGS. This is not the full story, however, as earnings for U.S. companies have grown more rapidly than their foreign counterparts, and valuations for non-U.S. equities are also elevated relative to their means (Figure 4).
Asset managers that invest in international equities argue that the relative performance of U.S. and non-U.S. equities tends to move in cycles, and they believe international and emerging market equities are poised to outperform before long. This perspective is presented in a report by Russell Investments titled “Is Global Diversification Still Worth It?”5 The study acknowledges that it has been a tough sell to convince Americans to invest abroad over the past decade due to subpar international returns and increased correlations. Yet, Russell Investments’ long-term forecasts suggest brighter days are ahead for non-U.S. equities: They call for non-U.S. equity markets to deliver annualized returns of 11% compared to just 5.5% for U.S. equities. If so, it would imply mean reversion from the past decade.
A recent Fort Washington report by Jamie Wilhelm and Dan Gibson titled “Growth…A Component of Value” offers a different perspective. Wilhelm and Gibson contend the primary reason for the prolonged growth leadership since 2007 has been that the growth segment includes a higher share of companies whose fundamentals have benefited from technological disruption, while low U.S. interest rates and slow global growth are secondary drivers. In their view, a sustainable rotation into value hinges on a dismantling of these three drivers, which carries considerable uncertainty. Rather than overweight or underweight growth stocks significantly, they consider growth to be a component of value that broadens the investable universe and improves the odds of generating attractive returns.
We contend the same is true for domestic and international investing. Namely, part of the outperformance of U.S. equities since 2008 reflects the greater weight of technology in the U.S. stock market and the dominance of U.S. tech companies globally (Figure 5). While there are indications that cyclical stocks could outperform in the near term, we are unsure how long this will last. Rather than make a large tactical overweight to international equities, we believe it is appropriate to include international names in domestic portfolios to expand the opportunity set and enhance the ability to outperform. This reflects our view that returns for U.S. and international equities are likely to be closer this decade than the 2008-2020 period, but we do not expect the recent trend will be reversed.
| region | annual return (15-Year) | tech weight | cyclicals weight |
|---|---|---|---|
| U.S. (S&P 500) | 9.9% | 28% | 37% |
| World ex-U.S. (MSCI ACWI) | 5.4% | 13% | 56% |
| MSCI EAFE* | 5.0% | 8% | 54% |
| MSCI Emerging Markets | 7.0% | 20% | 53% |
Individual Investors Versus Institutional Investors
First, we believe a common misconception is that the greatest differentiator between retail and institutional investors is the market value of the respective portfolios. In our view, size is a key factor, but it does not justify a different international allocation.
Historically, larger portfolios have had greater access to global markets. Institutional investors, such as pension plans, university endowments, and sovereign wealth funds, have the expertise and reach to build globally diversified portfolios. However, in today’s investment landscape with mutual funds and ETFs that invest in international and emerging markets, individual investors are presented with a similar set of investment opportunities.
If so, why is there still more of a home country bias in retail portfolios? If it is not a question of access, are individual investors making the right call by continuing to allocate less capital to international markets?
One explanation is individual investors generally have a range of investment objectives that may not be defined precisely, or that change over time. Also, because they reside in the United States, they may view international investments as riskier. By comparison, institutional investors have more stable, clearly defined objectives and characteristics that help mitigate personal biases.
In our view, the principal factor that explains the different allocations is the investment time horizon. Institutional investors typically manage portfolios with a long-term horizon because pension plans and endowments are enduring. Individual investors, however, may be skeptical about waiting another 10+ years for the business cycle to change, especially as they age.
It is important to understand the relationship between the time horizon and risk. The reason: an investor’s ability to bear risk is influenced by a variety of factors, but arguably none more impactful than time. That is a key reason why we believe it is appropriate for individual investors to review their strategic allocations periodically and to change them as they age.
Considering the relationship between risk tolerance and investment time horizon, one can understand why institutional and retail investors have different international exposures. That said, it is not just university endowments or pension funds that derive benefits from international diversification cited by Sullivan and Donohue. Retirees who depend on their portfolios to fund day-to-day spending also benefit from lower drawdowns in the face of global shocks when their portfolios are diversified internationally.
Investment Conclusions
First, there are important benefits from diversifying internationally, but they have lessened over the past two decades as correlations between U.S. and non-U.S. equities have increased. Accordingly, we continue to recommend a strategic allocation to non-U.S. equities of up to approximately 20% rather than 40%-50% based on market capitalizations globally.
Second, even though international and emerging market equities have underperformed for the past decade, we do not favor making a large tactical tilt in their favor at this time. In our view, differences in valuations based on P/E multiples do not fully explain what has happened, and there is no compelling reason to expect non-U.S. equities to outperform materially in the future. That said, we favor including them in domestic portfolios because they enhance the ability to outperform a U.S. benchmark by expanding the opportunity set.
Finally, we understand why investors are frustrated about the underperformance of non-U.S. stocks. Although we do not know when that will end, it is important to understand that returns can vary considerably over longer periods. As Jim Rogers observed, "bottoms in the investment world don't end with four-year lows; they end with 10- or 15-year lows." In the end, we believe that a globally diversified approach, with a strategic bias to the U.S., will offer investors the best likelihood of achieving their long-term goals.
4 “International Equities: Diversification and Its Discontents,” Enterprising Investor, CFA Institute, November 19, 2019. Ford Donohue, CFA, is affiliated with Homrich Berg.
