WC Archive

Sanderson and Scherbov refer to a "standardized age" model, using 2000 as the reference year. If a 30-year-old in 2000 could expect to live another 50 years (assuming no changes to health care), and a 40-year-old in 2025 could expect to live another 50 years, then the future 40-year-old will have a "standardized age" of 30. Because overall life expectancy increases, even as the median "standard" age rises, the median "standardized age" can stay flat or even decline, sometimes dramatically. The United States, for example, had a median age of about 35 in 2000 (hence a median "standardized age" of 35 then, as well); by 2100, assuming that lifespan increases at the same measured pace as now, the median American will be about 47 years old -- but the median "standardized age" will be about 29. That is, the typical 47-year-old citizen of the US in 2100 could expect to have at least the same number of years left as a 29-year-old did in 2000.

So what does this mean?

Among other things, it means that social policies based on estimates of life expectancy -- most notably social security and pension plans -- have to take into account more than an aging populace: they have to account for the fact that the populace will have more healthy, productive years as well. Sanderson and Scherbov argue that steady, gradual increases in the retirement age -- 2 more months per year for the US, 3 more months per year for nations like Germany with a higher median age -- could easily solve any "crisis" such plans might be facing down the road. With such a change, reflecting the ongoing increase to healthy lifespan, the dependency ratio (the number of retirees per worker) falls from 0.5 to 0.2 by 2050 in the United States, and from over 0.6 to under 0.4 in Germany by 2050 (Japan is even more dramatic -- 0.9, nearly one retiree per worker, drops to 0.5 by 2050).

Although proposals to raise the official retirement age abound, most are much more radical than what Sanderson and Scherbov suggest would be necessary.

But beyond current policy debates, this model also has the potential to shift the way we think about aging. The current system, which measures by average number of years lived, is subtly backwards-looking, triggering thoughts of what life was like all those years ago, emphasizing deeds already accomplished. The prospective age metric, conversely, is inherently forward-looking, making people think about what might happen over all those years to come, what they might choose to do with the increasing time they have left.

As interesting as the argument and the graphs linked here may be, they nonetheless feel to me as if they're missing a crucial piece of data: the ratio of number of years added to life expectancy to number of years lived. This would probably vary by age cohort, but would still be an interesting measure. So far, the ratio is relatively low, but slowly growing; if it ever gets to 1.0 (or greater), then lifespan is increasing at least as fast as we age -- look out!