An investment portfolio should provide a modest return while exposing the investor to an age-appropriate level of risk (or volatility). What a portfolio can't do is make up for years of inadequate contributions (i.e., underfunding). When portfolios are put together with that inappropriate objective in mind, they eventually blow up, and it's not pretty.

In light of the recent market implosion, clients may be anxious to make up for lost time (and returns). When their needs in retirement are unrealistic, their portfolios cannot support them. However, the composition of their portfolios can mitigate the blow.

This article compares four different retirement portfolios: an all-bond portfolio, a typical two-asset balanced portfolio, a multi-asset portfolio and an all-stock portfolio. The performance of each portfolio is tested during the distribution phase of an investor's life cycle, when money is being withdrawn during retirement.

An investment portfolio typically comprises two broad asset classes: equity and fixed income. Other important asset classes are real estate and commodities. These are sometimes referred to as alternatives. In this article, alternatives will be treated as an "equity-like" asset.

Most portfolios represent a blend of individual asset classes. A common blend is 60% stock and 40% fixed income. This particular mix is referred to as a balanced fund (or balanced portfolio). The multi-asset portfolio consists of seven assets equally weighted (large U.S. stock, small U.S. stock, non-U.S. stock, real estate, commodities, bonds, cash). Today, building a portfolio with only two assets is comparable to a diet of meat and potatoes. Since there are more asset classes than there were 40 years ago, a diversified portfolio should include more than two asset classes.

DIFFERENT PERSPECTIVES

Financial planners rightly think of the mechanics of distribution in terms of withdrawal rates and cost of living increases (COLAs). Individual investors think of retirement funding in dollar terms, such as "I need to withdraw $40,000 each year from my retirement account with a 3% annual increase." These are different perspectives. This analysis will take the investor's perspective to demonstrate what happens when an irresistible force (the annual income demands of a retiree) meets a movable object (an underfunded portfolio).

This analysis of various distribution portfolios is based on a $400,000 balance at the start of retirement. The annual cost of living adjustment is 3%. There are four different withdrawal rates: 5%, 7.5%, 10% and 12%. The 5% withdrawal rate is associated with a $20,000 annual retirement income; the 7.5% rate with a $30,000 annual retirement income; the 10% rate with a $40,000 annual retirement income; and the 12.5% with a $50,000 annual retirement income. These are unusually high withdrawal rates, but we are assuming that the cash flow demand of the retiree is the driving factor, not a safe withdrawal rate. No doubt, you've met prospective clients who think this way.

The study analyzes each portfolio's survival rate over the fifteen 25-year rolling periods between 1970 and 2008. Each of the four portfolios survived all fifteen 25-year periods, assuming a 5% withdrawal rate, or $20,000 per year with a 3% COLA (see "Survival Stats," on page 98). If, however, the retiree initially withdraws $30,000 (which equates to a 7.5% initial withdrawal rate), the survival rate of the all-bond portfolio and the all-stock portfolio slips to 80%. The 60% large U.S. stock/40% bond portfolio survived 93% of the time, while the multi-asset portfolio survived all 15 historical 25-year periods in this scenario. The multi-asset portfolio consists of seven assets equally weighted (large U.S. stock, small U.S. stock, non-U.S. stock, real estate, commodities, bonds, cash).

We have our first glimpse of what an all-bond or all-stock retirement portfolio can't do-sustain a 7.5% withdrawal rate with a 100% probability of survival. If a retiree needs $40,000 per year from a $400,000 portfolio (a 10% initial withdrawal rate), the outcome is even uglier. The all-bond portfolio never survived any of the 15 historical 25-year periods. The 60/40 portfolio survived 60% of the time, and the all-stock portfolio survived 66% of the time. The multi-asset portfolio was the most durable portfolio, but it still only had a 93% success ratio.

A $50,000 withdrawal each year (equivalent to a 12.5% withdrawal rate) crushed all four portfolios. Interestingly, the all-stock portfolio had a higher success rate than the multi-asset portfolio. This is due to the erratic nature of an all-stock portfolio; when it's good, it's great and when it's bad, it's awful.

TIMING IS EVERYTHING

As shown in "The 5% Solution," on page 100, the all-stock portfolio has dramatically different ending account balances based on the particular 25-year period, given a 5% withdrawal rate. For example, over the period from 1970 to 1994, it ended with a $1.7 million balance, whereas the period from 1975 to 1999 produced a $13.7 million ending balance. The ending balances in the other three portfolios are much more consistent. An all-stock portfolio is extremely sensitive to the timing of returns, which makes it a poor choice for a retirement portfolio. Retirees should not be betting their retirement on the timing of returns.