Close

    Maintaining the Flow: Will Foreign Stocks Help Your Retirement Portfolio?

    by Carl M. Hubbard

    Maintaining The Flow: Will Foreign Stocks Help Your Retirement Portfolio? Splash image

    Until recently, the ability of internationally diversified portfolios to outperform U.S.-only portfolios after adjusting for risk has been a foregone conclusion. But recent published research on international diversification questions the risk-reduction effects of international stocks.

    This issue has important implications for all individual investors. But it uniquely affects retirees who are living off of their retirement savings because of the impact it may have on the amount of money they can safely “spend” each year.

    For retirees, the important issue in this discussion is whether investing in foreign stocks can improve their portfolio performance enough to allow them to safely withdraw more from their retirement portfolio each year.

    The issue has taken on even more importance due to the recent severe bear market in the U.S., as retirees have been searching for ways to diversify stock holdings and smooth portfolio returns without severely lowering them. The objective of our research was to calculate portfolio “success” rates of retirement portfolios with and without foreign stocks in the portfolio for a range of withdrawal rates, and to evaluate the impact of international diversification on the ability of those portfolios to sustain withdrawals each year without exhausting the portfolio prematurely.

    Do Foreign Stocks Behave Differently?

    In theory, adding foreign stocks to a U.S. stock portfolio should reduce portfolio risk because past studies have shown that foreign stocks do not perfectly “correlate” with U.S. stocks—they do not always move up or down by the same amount or necessarily even in the same direction.

    In recent years, however, the stability and trend of the correlation between U.S. stock returns and foreign stock returns have become matters of controversy. Multinational diversification of large U.S. companies and inter-market contagion may reduce the benefits of international diversification that were observed in data from earlier decades. Several studies have reported returns among U.S. and foreign stocks that are more closely correlated than prior studies indicated.

    Our study began with our own look at how well the markets correlate. We looked at the correlation of monthly total returns (after conversion to U.S. dollars) of MSCI’s EAFE (Morgan Stanley Capital International’s Europe, Australasia and Far East) index, the most commonly used index of foreign stocks, and monthly returns of the S&P (Standard & Poor’s) 500 index, from January 1970 through July 2001. We found the correlations to be relatively low (0.55, where 1.0 indicates perfect correlation). Thus, on the surface, the data we examined suggest that international diversification should decrease portfolio risk and, therefore, may increase sustainable withdrawal rates from retirement portfolios.

    Varying Impact Over Time

    Why would these studies show different results?

    A chronological review of the findings on the benefits of international equity diversification demonstrates the varying impact of international equities on U.S. portfolios over time:

    • When foreign stock markets were outperforming the U.S. stock market in the 1950–1980 era, studies demonstrated the risk reduction and higher returns from international diversification.

    • The more recent studies that rely on mid-1980s to 1997 stock market returns show little benefit from international diversification.
    The recent reversal in research findings in the literature is not surprising when one simply compares the average returns and risk (as measured by volatility) of the relevant stock indexes. From January 1970 through December 1985, the average monthly total return of MSCI’s EAFE index exceeded the average monthly return of the S&P 500 by 20 basis points, while the risk (volatility, as measured by standard deviation) of the EAFE index was just a few basis points higher. Therefore, over this time period, adding international stocks to a U.S.-based portfolio would have provided greater returns, with just a little added risk.

    More recent data tell a different story. From January 1986 through December 2000, the average monthly returns of the S&P 500 exceeded the average EAFE index monthly returns by 32 basis points, while at the same time the risk was actually less than that of the international stock index (by 71 basis points). Adding international stocks to a U.S.-based portfolio over this time period would have lowered the overall portfolio’s return, and it would have increased portfolio risk. Thus, the long-term benefits of international equity diversification in U.S. portfolios may have disappeared in the late 1980s.

    How We Address the Problem

    How would the addition of foreign stocks benefit a retiree’s portfolio?

    For investors who are living off of their retirement savings, a key issue is their withdrawal rate: How much money can they withdraw annually from their investment portfolio without prematurely exhausting the portfolio?

    A safe withdrawal rate is a function of a portfolio’s total return as well as the sequence of returns that is earned by the portfolio over the payout period. The sequence of returns is important because money is being withdrawn each year. When assets are being withdrawn periodically, portfolios with low returns in the early years of withdrawal run a greater risk of exhausting assets prematurely than other portfolios that earn the same total return over the entire period but with their low returns earned in the latter years of withdrawal.

    To determine a safe withdrawal rate, one approach that is commonly used by financial planners and advisers (and which we have written about previously, see “Retirement Savings: Choosing a Rate That Is Sustainable” by Philip L. Cooley, Carl M. Hubbard and Daniel T. Walz, in the February 1998 AAII Journal) is to examine the “success rate” of various portfolio compositions.

    To see the real benefits of international diversification, we examined portfolio “success rates” for various portfolios—both with and without foreign holdings—over thousands of simulated market scenarios, using various rates of withdrawal.

    A “successful” portfolio is one that can maintain a positive portfolio value—in other words, not run out of money—despite the specified annual withdrawal amount over the specified time period based on the simulated market scenario—the pattern of returns of the various portfolio components over the time period.

    Our study simulated 1,000 market scenarios for each portfolio examined, based on the historical risk-and-return patterns and relationships of the various asset categories from 1970 to 2001.

    A portfolio’s “success rate” was measured by the percentage of all simulated market scenarios in which the portfolio was able to make the specified annual withdrawal over the specified payout period without running out of portfolio value. We examined payout periods of 15, 20, 25 and 30 years, and annual withdrawals rates of 4% to 10%, both without and with annual inflation adjustments.

    The simulated return data were derived from distributions of the following historical security returns data:

    • U.S. stock returns: S&P 500
    • Foreign stock returns: the MSCI EAFE index of equity returns (gross of dividends) converted to U.S. dollars.
    • Bond returns: High-grade corporate bonds calculated from the Salomon Brothers long-term high-grade bond index
    Because our analysis addresses retirement portfolios, we excluded consideration of the international stock indexes that include returns from stock markets in emerging or developing nations. While there are investment strategies that would include such markets, stock markets in emerging economies are too risky, in our opinion, and not appropriate investments for retirement portfolios.

    We examined various portfolio asset allocations and compared U.S.-only portfolios to those that were internationally diversified; for the portfolios that were diversified internationally, we used the EAFE index as a fixed 25% allocation (based on a recommendation in a prior study on the benefits of international diversification). The portfolio asset allocations that we examined were:

    • 100% S&P 500 vs. 75% S&P 500/25% EAFE;
    • 75% S&P 500/25% high-grade corporate bonds vs. 50% S&P 500/25% EAFE/25% bonds;
    • 50% S&P 500/50% bonds vs. 25% S&P 500/25% EAFE/50% bonds;
    • 25% S&P 500/75% bonds vs. 25% EAFE/75% bonds.
    Monthly rebalancing to maintain the desired asset allocation was assumed throughout all simulations.

    What We Learned

    Table 1 reports portfolio success rates for 1,000 simulated payout periods per withdrawal rate for each asset allocation and for each length of payout period.

    The table is divided into four sections by asset allocation. Within each asset allocation, 100% U.S.-security portfolio success rates are compared with success rates of portfolios composed of 75% U.S. securities and 25% foreign (EAFE) stocks.

    For example, in the first section of Table 1, the average portfolio success rate for portfolios consisting of 100% U.S. stocks, in which 7% of the initial portfolio value is withdrawn each year for 25 years, is 91%; that compares to a success rate of 92% for the foreign-diversified portfolio. When the payout period is extended to 30 years at a 7% annual withdrawal rate, the average success rates drop to 89% for both the 100% U.S. stock portfolio and the 75% U.S./25% EAFE stock portfolios.

    An examination of Table 1 suggests that international diversification by investing in the EAFE index increases portfolio success rates primarily in the equity-heavy portfolios by one to three percentage points.

    Overall, retirees who plan to withdraw fixed monthly amounts based on annual withdrawal rates of 5% to 9% of the initial values of mostly stock portfolios could benefit only modestly from international diversification.

    For retirees with substantial bond allocations, higher withdrawal rates of 9% to 10% and longer payout periods, the inclusion of the EAFE index lowered the average success rates by as much as four percentage points. At the higher withdrawal rates, the higher average returns and lower volatility of U.S. stocks appear to offset the beneficial effects of the lower correlations between foreign and U.S. stock returns.

    The success rates reported in Table 1 suggest that retirees who plan to make high, fixed withdrawals from portfolios that are at least 50% bonds might wish to forgo international diversification.

    TABLE 1. U.S.-Only vs. International Diversification: Portfolio Success Rates with Fixed Withdrawals

    Table indicates the percentage of simulations in which the portfolio was able to support all payouts over the time period and not run out of money prematurely.

    Payout
    Period
    Annual Withdrawal Rate as a Percentage of Initial Portfolio Value
    4% 5% 6% 7% 8% 9% 10%
    100% U.S. Stocks vs. 75% U.S. Stocks/25% International Stocks
    15 yrs 100 vs. 100 100 vs. 100 99 vs. 100 98 vs. 98 95 vs. 96 89 vs. 91 82 vs.83
    20 yrs 100 vs. 100 99 vs. 99 97 vs. 98 94 vs. 96 88 vs. 89 80 vs. 82 71 vs. 71
    25 yrs 100 vs. 100 98 vs. 99 96 vs. 97 91 vs. 92 83 vs. 85 74 vs. 76 63 vs. 63
    30 yrs 100 vs. 99 98 vs. 99 95 vs. 96 89 vs. 89 79 vs. 82 71 vs. 72 58 vs. 59
    75% U.S. Stocks/25% Bonds vs. 50% U.S. Stocks/25% International Stocks/25% Bonds
    15 yrs 100 vs. 100 100 vs. 100 100 vs. 100 100 vs. 99 97 vs. 98 94 vs. 94 86 vs. 87
    20 yrs 100 vs. 100 100 vs. 100 99 vs. 99 97 vs. 98 92 vs. 93 84 vs. 85 72 vs. 75
    25 yrs 100 vs. 100 100 vs. 100 98 vs. 99 95 vs. 95 88 vs. 88 77 vs. 79 63 vs. 65
    30 yrs 100 vs. 100 99 vs. 100 98 vs. 98 93 vs. 93 84 vs. 85 72 vs. 74 56 vs. 58
    50% U.S. Stocks/50% Bonds vs. 25% U.S. Stocks/25% International Stocks/50% Bonds
    15 yrs 100 vs. 100 100 vs. 100 100 vs. 100 100 vs. 100 99 vs. 99 95 vs. 96 89 vs. 90
    20 yrs 100 vs. 100 100 vs. 100 100 vs. 100 98 vs. 99 94 vs. 95 87 vs. 87 73 vs. 72
    25 yrs 100 vs. 100 100 vs. 100 99 vs. 99 96 vs. 97 90 vs. 91 79 vs. 78 62 vs. 60
    30 yrs 100 vs. 100 100 vs. 100 99 vs. 99 94 vs. 95 86 vs. 86 73 vs. 70 56 vs. 52
    25% U.S. Stocks/75% Bonds vs. 25% International Stocks/75% Bonds
    15 yrs 100 vs. 100 100 vs. 100 100 vs. 100 100 vs. 100 99 vs. 99 96 vs. 96 87 vs. 88
    20 yrs 100 vs. 100 100 vs. 100 100 vs. 100 99 vs. 99 95 vs. 95 84 vs. 85 70 vs. 66
    25 yrs 100 vs. 100 100 vs. 100 100 vs. 100 96 vs. 97 88 vs. 88 75 vs. 72 54 vs. 53
    30 yrs 100 vs. 100 100 vs. 100 99 vs. 99 94 vs. 94 84 vs. 83 68 vs. 65 46 vs. 43

    Adding Inflation to the Equation

    Table 2 reports average portfolio success rates over simulated payout periods where the retiree is assumed to withdraw inflation-adjusted amounts based on the initial withdrawal rates at the top of the table.

    The portfolio success rates in Table 2 suggest that allocating 25% of the total value of the portfolio to the EAFE index has little effect on the success rates of inflation-adjusted monthly withdrawals. Including the EAFE index increases portfolio success rates by one to three percentage points at initial withdrawal rates of 4% to 6% through shorter payout periods and when equities are 50% or greater of the portfolios.

    Foreign diversification decreases portfolio success rates by one to four percentage points at initial withdrawal rates of 6% or higher through the longer 25- to 30-year payout periods.

    The success rates in Table 2 indicate that international diversification can be used advantageously by some retirees and not by others.

    • Retirees who plan shorter payout periods and inflation-adjusted initial withdrawals of 7% or less are likely to benefit from international diversification.

    • Retirees who plan to make inflation-adjusted withdrawals over 25 or 30 years from portfolios with some allocation to bonds are not likely to increase their expected portfolio success rates by international diversification.

    TABLE 2. Portfolio Success Rates with Inflation-Adjusted Withdrawals

    Table indicates the percentage of simulations in which the portfolio was able to support all payouts over the time period and not run out of money prematurely.

    Payout
    Period
    Annual Withdrawal Rate as a Percentage of Initial Portfolio Value
    4% 5% 6% 7% 8% 9% 10%
    100% U.S. Stocks vs. 75% U.S. Stocks/25% International Stocks
    15 yrs 99 vs. 100 97 vs. 98 92 vs. 94 85 vs. 86 76 vs. 76 64 vs. 62 50 vs. 49
    20 yrs 96 vs. 98 90 vs. 92 81 vs. 82 69 vs. 70 55 vs. 54 43 vs. 42 32 vs. 30
    25 yrs 93 vs. 94 83 vs. 86 70 vs. 72 56 vs. 55 44 vs. 42 33 vs. 30 22 vs. 20
    30 yrs 88 vs. 89 77 vs. 78 62 vs. 64 50 vs. 46 36 vs. 35 25 vs. 24 17 vs. 14
    75% U.S. Stocks/25% Bonds vs. 50% U.S. Stocks/25% International Stocks/25% Bonds
    15 yrs 100 vs. 100 99 vs. 100 95 vs. 97 88 vs. 90 77 vs. 77 63 vs. 62 51 vs. 47
    20 yrs 98 vs. 100 94 vs. 95 85 vs. 85 70 vs. 69 56 vs. 55 41 vs. 37 26 vs. 23
    25 yrs 96 vs. 97 87 vs. 87 72 vs. 71 57 vs. 56 42 vs. 38 26 vs. 23 14 vs. 14
    30 yrs 93 vs. 93 78 vs. 79 64 vs. 63 49 vs. 46 32 vs. 29 17 vs. 17 9 vs. 9
    50% U.S. Stocks/50% Bonds vs. 25% U.S. Stocks/25% International Stocks/50% Bonds
    15 yrs 100 vs. 100 100 vs. 100 98 vs. 98 92 vs. 93 80 vs. 81 62 vs. 62 43 vs. 40
    20 yrs 100 vs. 100 97 vs. 97 87 vs. 88 71 vs. 71 49 vs. 49 30 vs. 28 17 vs. 13
    25 yrs 98 vs. 98 90 vs. 91 73 vs. 74 51 vs. 50 31 vs. 28 17 vs. 14 8 vs. 5
    30 yrs 95 vs. 96 82 vs. 82 60 vs. 60 39 vs. 37 20 vs. 19 10 vs. 7 5 vs. 3
    25% U.S. Stocks/75% Bonds vs. 25% International Stocks/75% Bonds
    15 yrs 100 vs. 100 100 vs. 100 99 vs. 98 92 vs. 91 78 vs. 76 57 vs. 54 34 vs. 31
    20 yrs 100 vs. 100 97 vs. 96 87 vs. 84 66 vs. 64 40 vs. 38 21 vs. 19 10 vs. 9
    25 yrs 99 vs. 97 89 vs. 86 69 vs. 65 42 vs. 39 21 vs. 19 9 vs. 7 3 vs. 3
    30 yrs 95 vs. 94 78 vs. 76 52 vs. 48 28 vs. 25 12 vs. 10 4 vs. 4 2 vs. 1

    Maintaining a Steady Flow

    A 30-year perspective on international diversification suggests that U.S. retirees should view foreign stocks as a sector in the stock market in the same manner that healthcare stocks, financial stocks or energy stocks represent sectors. The inclusion of stocks from a sector in a portfolio depends on the investor’s expectations regarding that sector’s expected return and risk and his or her diversification objectives.

    There were years, especially the 1970s, when the stocks in the EAFE index performed very well in comparison with the S&P 500. However, in the 1990s, U.S. stocks dominated EAFE portfolios on a return/risk basis.

    It is clear from our analysis and from previous literature that international diversification has not been a panacea that can be relied on to offset U.S. bear markets.

    Nevertheless, the low correlation of EAFE returns with those of U.S. stocks, and EAFE’s occasional superior returns, imply merit in international diversification for longer payout periods. In planning for the future, retirees who invest primarily (at least 50% of their portfolios) in stocks are likely to benefit modestly from including a 25% allocation to international stocks.

    However, you should not expect dramatic results by adding international stocks to your retirement portfolio—it will not, for instance, allow you to increase your withdrawal rate substantially while maintaining the same probability of success in terms of ensuring the portfolio supports all payouts over your planned withdrawal period.

    The actual net benefit of international diversification to a retiree depends on the investments chosen and the actual return and risk of the foreign stocks relative to U.S. stocks in a U.S. portfolio over the actual time period that the retiree is invested.

    Because international diversification may be accomplished easily and inexpensively through the purchase of shares of mutual funds that invest in EAFE stocks and others, retirees with equity-heavy portfolios should give serious consideration to internationalizing their portfolios.

       More on Retirement Spending
    Visit the AAII.com archives to access these past AAII Journal articles on retirement spending plans. Links are provided in the on-line version of this article in the AAII Journal area, found under Publications. Or, click on Search, go to Advanced Keyword Search and select Financial Planning/Retirement Issues from the category box.

    “Retiree Stock Allocation Recommendations: Do You Fit the ‘Mold?’” by William Reichenstein, February 2004

    “Can They Maintain Their Lifestyle? Retired Couple Seeks Assurance,” by Loyd J. Stegent, August 2003

    “Will Their Retirement Assets Last? Analyzing the Chances for Success,” April 2003 and “Will Their Portfolio Last? The Rest of the Story,” July 2003, by Dennis Stearns

    “Bear Market Strategies: Watch the Spending, Hold the Stocks,” by T. Rowe Price Associates, May 2003

    “Retirement Savings: Choosing a Withdrawal Rate That Is Sustainable,” by Carl M. Hubbard, Philip L. Cooley, and Daniel T. Walz, February 1998


    Philip L. Cooley, Carl M. Hubbard and Daniel T. Walz are professors of business administration at Trinity University in San Antonio, Texas.

    This article is based on and excerpted from an article that originally appeared in the January 2003 issue of the Journal of Financial Planning, published by the Financial Planning Association. Reprinted with permission by the Financial Planning Association, Journal of Financial Planning, January 2003, “Does International Diversification Increase the Sustainable Withdrawal Rates From Retirement Portfolios?” by Philip L. Cooley, Carl M. Hubbard and Daniel T. Walz. All rights reserved.

→ Carl M. Hubbard