JT and I talked many times about pensions and the unfortunate arithmetic behind most defined-benefit pensions. We keep seeing issues arise with the pensions in places like Detroit, or now Illinois. To reiterate, I have no issue with the defined-benefit pension plan in theory. The issues I have focus on the management, or should I say mismanagement, of the pension plan. The mismanagement really falls into two categories: investment of pension fund assets bordering on fraud and, a seemingly willful ignorance about the changes in life spans for people (in this case plan recipients).
In other posts I talked about the ludicrous assumptions made regarding investment returns that states and local areas utilize to justify smaller investments into retirement plans. Changing the assumed return from 7% to 8% reduces the necessary contribution and results in a savings to be spent elsewhere, or maybe given as tax relief (just a little bit of humor there for you).
The second issue is where I am focusing right now though, because it seems overlooked, at least in this context. We regularly see newspaper stories about life expectancies increasing, and that more people are living longer, so why do we not think this will impact pensions. There was a really great post by the Office for National Statistics in England recently (here). The whole link is worthwhile reading, but the pension discussion is at the bottom of the page when you scroll down. The relevant details are that when first introduced state pensions needed to last an average of nine years in England. In 2011 they needed to last 20 years.
The key number to focus on is life expectancy at age 65, not life expectancy at birth. (I am using 65 as a general proxy for retirement age here. I do recognize that it is variable.) The link shows circumstances in England, but what about the U.S.? I grabbed data from the CDC (they keep the life expectancy data for the US).
What does the graph show? Compared to 1950, an individual making it to age 65 in 2010 would expect to live 5 more years on average. That is 5 more years of paying out a pension. Too long a time gap for comparison. That’s fair. A person at age 65 in 2000 expected to live 17.6 more years. That is 17.6 years of paying out a defined-benefit pension. In 2010 the expectation at age 65 increased to 19.1 years. Those aged 55 in 2000, when they reached age 65 in 2010, could expect another 1.5 years of living, and would need another 1.5 years of pension paid to them.
This is not meant to bash defined-benefit only. Most people do not accurately anticipate their lifespan after retirement and to be honest those who die earlier than anticipated are not the financial problem here. There is a real risk of people in all types of pension outliving their pension assets in retirement. The pace of healthcare innovation, the improvements in lifestyle (smoking, etc.), all add to the longevity of individuals. We need to get better about communicating these ideas to people in terms they understand so they can plan better.
States and local governments that offer defined-benefit pensions need a crash course in this as well, and soon. They are bad enough at the investment assumptions that they need to better evaluate the lifespans of employees covered under these plans before everywhere resembles a Detroit or an Illinois.