International Diversification in Retirement Accounts

by Frank Armstrong - Friday, 07 March, 2003

International diversification is an issue that should have been settled a long time ago. However, many investors believe that cross-border diversification increases risk. Now, new support from academics provide new evidence of its value as a risk reducer.

Does International Diversification Increase the Sustainable Withdrawal Rates from Retirement Portfolios?, by Philip L. Cooley, Ph.D.; Carl M. Hubbard, Ph.D.; and Daniel T. Walz, Ph.D. confirms that international investment provides an important diversification benefit, and that there is no clear trend toward increasing correlation between foreign developed markets and the U.S. This data has long been available to professional advisors, but the myth of increasing correlations simply refuses to die.

The paper finds that international diversification in the form of incremental additions of the Morgan Stanley Europe, Australia, Far East Index (EAFE) increases the average survival chances for retiree portfolios.

A few minutes of playing with any Monte Carlo simulator should demonstrate the value of risk reduction in a portfolio taxed with systematic withdrawals. As I demonstrated in my paper “Investing During Retirement” a portfolio with an assumed 10% rate of return and 6% withdrawal rates was likely to experience less than a 1% failure rate over 30 years at a 10% standard deviation. But, the failure rate increased to over 23% at a 20% standard deviation.

My paper held rate of return fixed while varying standard deviation. Their previous and current papers allow both rate of return and standard deviation to vary, an important distinction. Nevertheless, my takeaway from my exercise is that any reduction in portfolio volatility increases the probability of a successful outcome for the retiree.

That being the case, the optimum retirement portfolio will have the lowest volatility possible for any assumed expected rate of return. The logical extension to this insight is that any diversifier that serves the purpose should be considered. Why limit the process to EAFE? We can reduce the portfolio risk by further diversification into small and value stocks on a global basis. We have identified nine equity asset categories that serve the purpose of reducing risk while increasing rates of return. Asset classes such as International Small, International Small Value, and International Value have higher expected returns than EAFE, and lower correlation to our domestic assets.

Additionally, a reduction of the average duration of the bond portfolio reduces total portfolio risk without compromising average returns.Cumulatively, these modifications substantially reduce risk, enhancing the survivability of retiree accounts.

To the extent that emerging markets/developing countries (or any other asset or asset classes) offer risk reduction to the portfolio without compromising expected returns they should be considered. This approach is consistent with Modern Portfolio Theory, required under ERISA, the Restatement of Trusts, and the new Uniform State Prudent Man Rules. So, I was very disappointed that the paper finds that Emerging Markets are too risky for retiree portfolios without supplying any justification or testing results.

Asset class diversification with the resulting reduction in volatility is a key concept in developing retirement portfolios with a high probability of success. While I’m glad to see the support for international investing, I’m sorry that the paper limits their tests to a simple S&P 500 – EAFE example, and I disagree with the conclusion on emerging markets as unproved. We believe that they will provide a measurable diversification enhancement, and that an inclusion of at least a small weighting is prudent and conservative.