Morningstar Advisor - Spring 2008 - (Page 32) Spotlight Adding the Income Dimension By Peng Chen and James X. Xiong1 A new framework allows advisors to fully examine a portfolio’s sustainable income levels and the risks of it coming up short. The large number of soon-to-be retiring baby boomers. The paradigm shift away from defined benefit pensions to defined 0.12 contribution pensions. Long life expectancies. 0.10 Uncertainty surrounding Social Security 0.08 benefits. No wonder retirees face such tough 0.06 decisions about retirement spending. 0.04 0.02 Many retirees who must rely on their own 0.00 personal savings in retirement seek out help from financial advisors to manage their portfolio in the retirement-income phase. And those advisors traditionally have developed asset-allocation portfolios by constructing efficient portfolios for various risk levels based on mean variance efficient frontier (MVO).2 Depending on a retiree’s risk tolerance, the advisor chooses one of the efficient portfolios in which to place the client’s savings. After that, both advisor and client hope for the best. asset allocations, its effectiveness in retirement-income planning is inherently limited. To more effectively evaluate the risk-reward trade-off of retirement-income patterns generated by different portfolios, we developed a new retirement-income efficient frontier framework to complement the traditional MVO framework.3 In this new setting, we can fully examine a portfolio’s sustainable income levels and the risks of it coming up short. We can also evaluate the roles of various asset classes and investment strategies, such as principal-protected equity-linked CD products and variable annuities with lifetime guarantee minimum withdrawal benefits (VA+GMWB). A New Retirement Framework The Traditional Mean-Variance Frontier Expected Return 12% Aggressive 80/20 Conservative Moderate 40/60 Moderate 60/40 Conservative 20% Stocks/80% Bonds 8 4 5 0% Standard Deviation 10 15 20 We think there is a better way. The key consideration is the recognition that retirement investors care more about income than about portfolio returns. But MVO only considers a portfolio’s risk-return trade-off in terms of returns; it does not consider the risk-return trade-off of the retiree’s main concern—a portfolio’s ability to generate sustainable income streams. Although MVO is widely accepted in academic and finance circles as the golden standard for developing A traditional efficient frontier consisting of stocks and bonds only considers the risk-return trade-off in portfolio returns, not the portfolio’s ability to generate sustainable income levels. In addition, as our earlier research found, it does not consider one of the major risks that investors face during retirement—longevity. On the other hand, the downside protections offered in some new investment strategies— such as principal-protected strategies and VAs with lifetime GMWB—imply a skewed income and return distribution in the corresponding combined portfolios. Our new efficient frontier framework is able to account for these considerations, because the new retirement-income efficient frontier is generated using a Monte Carlo simulation technique.4 Hence, we can model these unique distribution properties in the analysis and not be restricted to the traditional mean and variance measures. We believe that the combination of the traditional MVO and the new retirement-income efficient frontier is much more effective in evaluating retirementincome portfolios. 0.00 0.05 0.10 0.15 0.20 The new retirement-income efficient frontier is established using average sustainable income versus shortfall income level against the target income level. The average sustainable income level is the annual income that can be sustained at the 50th percentile over the target horizon (e.g., 30 years). More precisely, it is the median sustainable income level. 32 Morningstar Advisor Spring 2008
For optimal viewing of this digital publication, please enable JavaScript and then refresh the page. If you would like to try to load the digital publication without using Flash Player detection, please click here.