Retiring early? You may not want to bet your life on it.

But for plan sponsors, funding could improve

With the collapse in the global stock markets, many investors — staring at the wreckage of their 401(k) plans and their declining savings and concerned about the viability of their defined benefit plans — are quietly postponing their retirement dreams. In the midst of all this bad news there might be a silver lining for participants and pension plan sponsors. Some scientific data suggest that the later a person retires the longer they will live.

While some workers might welcome the opportunity to live to see their children happily married or their grandchildren graduate, others may only see the extra drudgery of longer employment. It may be some consolation to those of us who delay retirement to remember that the actions of optimistic folk who are “dying to retire early” actually translates into “retiring to die early.”

The scientific basis for the theory is largely rooted in a study co-authored by Shan P. Tsai, manager, epidemiology of Shell Health Services of Shell Oil Co.* The study, titled “Age at retirement and long term survival of an industrial population,” was published Oct. 21, 2005, in BJM, a publication of the British Medical Association. The study looked at Shell employees who retired at age 55, 60 and 65, between 1973 and 2003. It covered approximately 3,500 people over the 30-year period. Basically, the mortality rate of those who retired at 55 (adjusted for health factors) was twice that of their still-working colleagues. The mortality rate was not gender neutral however. The study found the early deaths are 80% more likely to happen to men. Further, Shell could not identify any specific causes for these observable facts.

From the pension plan funding standpoint, these findings offer some perverse conclusions. An earlier start date for a pension payout schedule might suggest more payments overall and higher pension costs over the life of a retiree — but the study found the opposite result. The early retirement of a 55-year-old worker is actually cheaper for the fund. Costs are lower because early retirement means a lower monthly pension payout (as a result of shorter tenure) and fewer payments overall (as a result of early death). Conversely, the same 55-year-old worker delaying retirement for five more years — and thereby cutting his mortality rate by 50% — will not only live longer, but also do so at a full pension.

Because of the extra expense of full pension obligations, pension plans will necessarily require a larger pool of assets. However, plans will then have at least five more years of potential investment growth to offset the increase in each retiree’s cost. This delay will also reduce immediate liquidity needs. But whether it will be more expensive, cost neutral, or actually lessen the funding burden is subject to the success of the investment strategy employed. Right now, the investment return outlook for any asset class over the next five years does not inspire confidence, so we can expect actuaries to require higher funding levels.

If the theory is valid, this will require a major rethinking of everyone’s career aspirations. Early retirement has long been the holy grail of many workers looking forward to a lengthy, contented and low-stress lifestyle in their twilight years. If the new theory is widely accepted, many men may prefer to work longer — and live longer — which will also give them more time to build a larger retirement nest egg. Later retirement may be more palatable to workers who already are aware that life spans have been significantly extended by better health practices and that a longer working career will become increasingly normal.

Shell conclusions backed up by other studies

Beyond the pension funding arena, the implications for many governments’ fiscal policies, social security entitlements and the like are considerable. The Shell conclusions were work-group specific, but they have been corroborated by a number of international general population studies. Many of those were in the European Union, which faces a huge financial shortfall in social services. EU governments welcome these studies as a way to justify eliminating the mandatory retirement age of 55 in the hope of saving money. Ironically, following the Shell study findings, it will have quite the opposite effect — and actually will increase social costs — as their citizens will then live longer and consume more state resources.

A study titled “Timing to retire — timing to die?” by Hilke Brockmann, professor of sociology at Jacobs University, Bremen, Germany, and Rolf Muller of Bremen University, was presented at the American Sociological Association meeting last July 31 in Boston. The study had a much larger sample than Shell, covering approximately 120,000 German nationals over a 14-year period. The authors came to the same conclusion as Shell in identifying the same higher mortality rates attached to the 55-year age retirement decision. The authors did cite a possible interpretation of the causes by noting a correlation between the higher mortality rates and reduced earning capacities. They also came up with their own startling observation concerning a group not covered in the Shell study — those retiring even earlier, at 51 to 55 years, had the highest mortality of any group studied.

Finally, a Greek study by Christine Bamia, Antonia Trichopoulou and Dimitrios Trichopoulos** — “Age at Retirement and Mortality in a General Population Sample,” published in the March 1, 2008, American Journal of Epidemiology Advance — covered approximately 17,000 people over a 12-year period. This report corroborated the other studies in demonstrating the increased mortality risk of early retirement of people with otherwise good health, and offered its own causal factors: “deterioration of economic status” and severe depression. Of course, that exactly describes the circumstances and reaction of anyone today watching a retirement nest egg implode. Better to stay working, it seems — because both your economic welfare and actual life depend on it.

* The other co-authors are Judy K. Wendt, senior epidemiologist; Robin P Donnelly, director, health services; Geert de Jong, senior health adviser; Farah S. Ahmed, epidemiology research associate. All are with Shell Health Services, Shell Oil Co., Houston, except Mr. de Jong, who is with Shell International, The Hague, Netherlands.

** Ms. Bamia and Ms. Trichopoulou are professors in the Department of Hygiene and Epidemiology, University of Athens Medical School in Athens, Greece; Mr. Trichopoulos is the Vincent L. Gregory Professor of Cancer Prevention in the Department of Epidemiology, Harvard School of Public Health, Boston.

John Woods Conlin is principal of Conlin Associates, Chadds Ford, Pa., a consulting firm that advises money managers in the institutional market on business and product development, marketing and other areas.