The panic of 2008 has struck a mighty blow to many investors’ retirement planning. Strategic asset allocation, an investment style widely used by investment professionals, failed badly in protecting investors against risk in many cases. Now, some advisers are taking another look at an investment approach called “liability-driven investing,” which is often used in managing defined benefit pension plans. Can this investment style also work for individuals planning for retirement?
Let’s understand how these approaches differ. Strategic asset allocation focuses on the asset side of one’s balance sheet: How much do I have and how much can it be expected to grow? It attempts to answer the latter question by looking at the expected returns and risks of various asset classes, then combining asset classes in proportions that are supposed to match an investor’s goals and risk tolerance. At least in theory, this should maximize the investor’s assets at any given time in the future.
Liability-driven investing, in contrast, looks at specific needs to be funded at specific times in the future. Here’s a simple example to illustrate how the approaches differ.
Suppose Joe Prudent, having turned 66, decides to retire. Recently widowed, he lives a modest lifestyle. He enjoys several hobbies and likes to travel, especially to visit his grandchildren.
Joe just finished making his last mortgage payment; he has no other debt. Joe’s assets include $100,000 in a brokerage account and $650,000 in an IRA he rolled over from his 401k plan at work, the result of his contributions and employer matches over the past 25 years.
To maintain his lifestyle, Joe figures he needs about $4,700 a month in income. His Social Security starts at $2,200 a month. For the remaining $2,500, Joe must look to his financial assets. How should he invest his assets to meet those expenses?
A typical asset-focused plan would apply a spending rate to Joe’s investable assets to determine how much can be safely withdrawn against the asset pool. Most advisers today would suggest limiting withdrawals to 4 percent of capital annually. Joe’s assets total $750,000. The 4 percent spending rule tells Joe he can withdraw $30,000 the first year, or $2,500 a month, just enough to meet his goal.
Using Strategic Asset Allocation, Joe’s adviser constructs an asset allocation plan with an expected return of 7 percent using a mix of stocks and bonds. This should be sufficient to meet the 4 percent withdrawal, 1 percent in management expenses and 2 percent growth. Under this plan, Joe’s assets should last indefinitely, growing the principal at 2 percent annually to offset the effects of inflation. All well and good—let’s call this Joe’s Plan A.
There being no room for error, a bad year early in Plan A can doom it to failure. Suppose Joe started Plan A at the end of 2007, when credit was easy and the equities markets were at historical highs. Even very conservatively managed accounts lost 20 percent in value during 2008. If Joe had followed Plan A in 2008, he would have begun 2009 with only $570,000 left—$750,000 less $150,000 in investment losses, less $30,000 in withdrawals. Joe now has three options:
• Continue the $2,500 monthly withdrawals, raising his spending rate to 5.2 percent of capital and jeopardizing future growth of the portfolio;
• Seek out higher returns by increasing the portion held in riskier assets; or
• Reduce withdrawals to 4 percent of the new total, or $1,900 a month, forcing Joe to alter his lifestyle to lower his spending by $600 a month.
Now let’s look at Plan B, a liability-based approach.
Retirement spending needs can be viewed as a stream of one-year liabilities, starting at $30,000 a year, or $2,500 a month. Joe, who is in good health, knows he could easily live into his 90s or beyond. Plan B is intended to avoid the risk that one or more bad investment years (as in Plan A) will threaten his ability to meet those future liabilities.
One way Joe can ensure he meets his retirement spending liabilities is through use of a fixed immediate annuity (not to be confused with a variable annuity, a very different investment product of doubtful benefit in retirement planning). The annuity lets Joe transfer the risk of investment failure to an insurance company during his retirement years in much the same way he used life insurance to insure his family against the loss of his earning power.
To generate lifetime payments of $2,500 a month, Joe finds he’d have to commit about $400,000 of his IRA to a retirement annuity, leaving him with $350,000 of investable assets. Joe is free now to invest his remaining assets without worrying that investment failure will impinge on his lifestyle. A 20 percent investment loss during the first year of Joe’s retirement under Plan B would reduce his nonannuity assets to $280,000—unpleasant, but it wouldn’t compromise his ability to meet his spending needs.
Then, under Plan B, reasonable investment gains in Joe’s nonannuity portfolio can be allowed to accrue, enabling Joe to wait out major downturns in the stock market. At an average investment return of 7 percent, the nonannuity assets should double in value every 10 years. Of course, Joe will have to begin required minimum distributions from his remaining IRA at age 70.5, but he doesn’t have to spend them other than paying income taxes on the distributions. The net distribution can be added to his taxable investment portfolio to help it grow.
I’m not advocating that any investors use immediate annuities in their retirement income planning—only illustrating a principal. It’s possible to generate reliable periodic sums of cash using bonds, for example, without giving up a portion one’s principal, as Joe must do using the annuity approach.
Liability-driven investing can relieve investors of anxiety about meeting retirement income needs and assure them that they won’t run out of money.
