The year 2008 is shaping up to be a nasty year in which to retire. From their high on Oct. 9, 2007 to September 9, 2008, two of the three major Russell Indexes—the Russell 1000 and 3000—dropped to within a fraction, plus or minus, of negative 20%, which most investors define as a bear market. The Russell 2000 did better during that time period, falling by only 14.79%.

Soon-to-be retirees who saw the market take huge bites out of their investment portfolios in just 11 months' time must surely be confused by the wide divergence of opinion on the appropriate exposure to equities going forward. While some call for lower exposures, it is common to hear recommendations for exposures of 50% and higher. Both approaches have good rationales; equities, over the long term, should keep pace with inflation, and therefore help counter longevity risk from a different angle. So a natural question is: How should retirees configure their portfolios for the long term, while not exposing themselves to inappropriate risk?

Aggressive portfolios are inherently risky, and an overexposure to equities at the point of retirement could put an investor at risk of going broke later, especially if the investor happens to retire into a down market. The risk is more acute for those without a pension or other savings to fall back on.

Retirees, therefore, must consider two conflicting elements of risk—the risk of outliving the portfolio because it shrank in a down market and the risk of losing pace with long-term inflation.

It's time we turned conventional wisdom on its head. A common argument is that portfolios with conservative asset allocations may not provide sufficient investment returns in retirement. Even if true, this argument does not mandate that investors should be invested aggressively over the entire retirement period. Knowing this, many suggest that retirees' portfolios start more aggressively early in retirement and then taper to become more conservative as they age.

But this could mean taking the most investment risk precisely when it is most perilous to do so. Counter to conventional wisdom, longevity risk can be reduced, on average, by planning for a more conservative portfolio early in retirement and a more aggressive portfolio later.

The Lesson of 2000-2002

The bear market years of 2000, 2001 and 2002 offer a good example. Consider investors with modest spending goals who encounter such a bear market at the start of retirement (see "Fighting the Bear, Round One," at right). This hypothetical scenario, based on index returns, shows how an investor who begins retirement with an asset allocation of 32% stocks and 68% bonds, and who then moves to a more aggressive 60/40 allocation after 10 years may be better protected against longevity risk than an investor with a 60/40 allocation all 30 years. We used indexes as a proxy for investing in the market.

On the other hand, if a bear market were to occur later in retirement, the adverse effect of holding a more aggressive portfolio at that time is reduced (see "Fighting the Bear, Round Two," on page 149). In fact, longevity risk tends to decline over time along with one's declining life expectancy. The greatest danger exists in the very first year of retirement.

Investors are free to change the allocations of their investment portfolio at any time, although there may be some transaction costs and tax considerations associated with this. Advisors, knowing that bear markets can irreparably damage a portfolio long term, can help risk-averse clients fund a more secure retirement by suggesting a conservative allocation to equities at the beginning of retirement, when longevity risk is at its greatest.

The Volatility Bow Wave Effect

It is well documented that people are living longer. This puts pressure on an advisor to design a portfolio that should last 20, 25, even 30 years or more after a client leaves the workplace. For retirees concerned about fully funding their golden years, the proper asset allocation depends not so much on their tolerance for investment risk, but their tolerance for longevity risk. What most retirees really care about is not going broke.

When a ship plows through water, it creates a bow wave, a V-shaped curl of raw power. While mesmerizing to watch, naval architects attempt to reduce a vessel's bow wave because it can sap energy and reduce fuel economy. A boat with a large draft and blunt bow produces a large bow wave, while vessels that plane over the surface create a smaller one and are thus more efficient.

The volatility of investment returns is like a bow wave that can sap energy from a retiree's portfolio. In terms of its effect on longevity risk, this volatility bow wave naturally tends to be bigger early in retirement than later. Fortunately, however, it can be managed by means of asset allocation. Reducing the volatility of return at the beginning of a portfolio's distribution period by allocating assets more conservatively helps reduce the bow wave effect, enabling the portfolio to "plane over" rough periods in capital markets.

Why is the volatility bow wave particularly worrisome during the retirement years? Volatility is more bearable in the accumulation phase when the investor has more options available, such as increasing the savings contribution rate or working longer to push back the retirement date. It is much less bearable in the distribution—or decumulation—phase, when these options are not likely to be available. Further, the impact of volatility is lessened during the accumulation phase because the dollar-cost averaging (DCA) that occurs as a result of making periodic investments throughout the accumulation phase has beneficial effects to the investor.

Unfortunately, during the decumulation phase, the often-promoted benefits of DCA disappear and, in fact, work against the investor when cash is flowing out of the portfolio. (Dollar-cost averaging does not assure a profit or prevent a loss in declining markets, and you should consider your ability to continue investing during low price levels.) It turns out that the sequence of investment returns is important in the decumulation phase. As a result, retirees face the greatest amount of "sequence of returns" risk early in retirement, when subpar performance can jeopardize their future standard of living.

Of course, the proper asset allocation depends on a number of factors, including the investor's tolerance for risk, spending and bequest goals, life expectancy, etc. However, when longevity risk becomes the primary concern, it may be that a more conservative portfolio is in order during the early retirement years, in order to get through the riskiest period. Later, it may be safe to move to a more aggressive portfolio.

Implementation

Such a strategy can be implemented in a number of different ways. For example, when an "income bridge" approach is used, one pool of money may be invested to fund the first 10 years of retirement and another pool of money to fund the years after that. Even more funding pools may be used under what could be called a laddering approach.

In either case, any particular pool could be allocated 100% to a single asset class, such as 100% cash or 100% equity, for that matter. However, the overall portfolio allocation should always take into account the client's tolerance for longevity risk and investment risk. It may serve advisors well to challenge conventional wisdom by reconfiguring their clients' portfolios to protect them against late-stage risks. Retirees have but one retirement to plan for, and there are no "do-overs." Reducing one's exposure to equities early in retirement may increase the likelihood of funding the golden years and preserving wealth.

Richard K. Fullmer, CFA, is a senior portfolio strategist for Russell Investments in Tacoma, Wash.