According to the Centers for Disease Control and Prevention, life expectancy at birth in the United States was 78.8 years in 2013 — 76.4 years for men, 81.2 years for women. But I have good news. Those statistics don’t mean what you probably think they mean.
In fact, an American child born in 2013 will most likely live six or more years longer than those averages: boys into their early 80s, girls into their late 80s.
Any life expectancy calculation is based on a series of probabilities: If you’re currently 21, what are your odds of living to 22; if you live to 22, how likely are you to make it to 23; and so on. In the most commonly cited life expectancy statistics, which actuaries call period life expectancy, the death rates at each age are held fixed over time. That is, a baby girl born Jan. 1, 2016, who lives to see 2076 is assigned a probability of dying in 2076 that equals the probability of a woman who is currently 60 years old dying next year.
Of course, the assumption that 60-year-olds in 2076 will be no healthier than 60-year-olds today is unlikely to be correct. But actuaries don’t use it because they’re idiots; they use it because it’s useful for comparing current health conditions across countries. If babies in Country A and Country B are likely to have about the same life span, but currently old people are sicker in Country A than in Country B, this method will produce a lower reported life expectancy for Country A. What this life expectancy figure tells you is, that when you average all the young people and the old people, people in Country B will tend to live longer.
So this statistic is useful for measuring the health of a country’s inhabitants, but it’s not useful if what you want to know is how long your new child will live. For that, you need to look at cohort life expectancy, a statistic that adjusts for the fact that death rates tend to decline over time as health and safety improve. According to the Social Security Administration, that’s 83.1 years for boys born in the United States in 2015, and 86.8 years for girls.
But wait, I have more good news: Those estimates are probably not optimistic enough. The Technical Panel on Assumptions and Methods established by the Social Security Advisory Board, an independent government agency that advises Social Security’s trustees on matters including actuarial assumptions, says Social Security is systematically underestimating future declines in mortality rates, and therefore underestimating the likely life spans of young Americans. (Demographic estimates, of course, are necessary to project future revenues and expenses for the program.)
In the long run, the Social Security Administration assumes the “mortality improvement rate” will be 0.71 percent — that is, the odds of dying at a given age will fall, on average, that much each year. Remember, that’s a rate of relative improvement, and at most ages the odds of dying are low, so the numbers won’t move around very much. If a person your age had a 1 percent chance of dying in a given year, the S.S.A.’s assumption of 0.71 percent annual mortality improvement would imply a 0.9921 percent chance of death the following year for people a year younger than you — just a little bit safer.
The technical panel thinks 0.71 percent isn’t high enough, and that Social Security should assume an improvement rate of 1 percent a year, which better reflects past improvements in health. Over time, that small difference would compound into a meaningful increase in life expectancy. By 2090, this change in assumptions would add two and a half years to life expectancy, relative to the assumptions currently used by Social Security.
That is, there is a very good chance today’s newborn boys will live to be about 85, and newborn girls will approach 90, on average.
Of course, you could just say the Social Security Administration is being conservative in forecasting an unknowable future — further improvements in life expectancy are likely to depend on unknown and unforeseen improvements in medical technology. But one quirk of Social Security is that a piece of obvious good news (People will live longer than we thought!) is bad news from the narrow perspective of paying for retirement benefits (The government will have to pay benefits longer!).
A higher life expectancy estimate is optimistic for the human condition, but pessimistic for the Social Security Trust Fund. So perhaps it’s not surprising the trustees of Social Security have resisted prior recommendations to raise life expectancy assumptions.
A succession of six technical panels established by the Social Security Advisory Board — in 1995, 1999, 2003, 2007, 2011 and this year — have argued that Social Security was assuming unrealistic mortality rate improvements. Eventually, the Congressional Budget Office started tweaking Social Security’s assumptions to adjust life expectancy upward.
“We felt at C.B.O. a few years ago that we just couldn’t stick with the S.S.A. projections anymore, given where the experts seemed to be,” said Doug Elmendorf, who until recently served as director of the office.
That’s a modest piece of bad news for the federal budget — lower mortality rates may worsen the shortfall in Social Security by 11 percent over the next 75 years, with nearly all of the increased deficit emerging after 2040. But the need to close that gap should be a small price to pay for the upside of living a bit longer.