Capital market expectations, asset allocation, and safe withdrawal rates

Most retirement withdrawal rate studies are either based on historical data or use a particular assumption about portfolio returns unique to the study in question. But planners may have their own capital market expectations for future returns from stocks, bonds, and other assets they deem suitable for their clients’ portfolios. These uniquely personal expectations may or may not bear resemblance to those used for making retirement withdrawal rate guidelines. The objective here is to provide a general framework for thinking about how to estimate sustainable withdrawal rates and appropriate asset allocations for clients based on one’s capital market expectations, as well as other inputs about the client including the planning horizon, tolerance for exhausting wealth, and personal concerns about holding riskier assets. The study also tests the sensitivity of various assumptions for the recommended withdrawal rates and asset allocations, and finds that these assumptions are very important. Another common feature of existing studies is to focus on an optimal asset allocation, which is expected either to minimize the probability of failure for a given withdrawal rate, or to maximize the withdrawal rate for a given probability of failure. Retirement withdrawal rate studies are known in this regard for lending support to stock allocations in excess of 50 percent. This study shows that usually there are a wide range of asset allocations which can be expected to perform nearly as well as the optimal allocation, and that lower stock allocations are indeed justifiable in many cases