The problem with pensions

Scott Adams reveals a solution to save a struggling company's pension plan.

I'll respond to Honest Partisan's request for comment and start by saying the New York Times article is less about pensions and more about human failure.

Part of the reason for the problems with defined benefit pension plans is that raising future benefits is appealing to both union leaders and management. Management gets agreement with its workers in a manner that doesn't impact the current year's budget. Union leaders get to tell their workers "Look at this benefit I got you! Now, re-elect me to this cushy job with its six-figure salary and boatloads of power." (The description of the actions of public employee unions is particularly sickening.

The moral hazard issue [that companies invest more aggressively because the Pension Benefit Guaranty Corporation guarantees pension benefits] certainly isn't unique to pensions. It's the same reason why people don't buy flood insurance for homes in a flood plain; they know the government will pick up the tab for repairs.

The larger problem is simple: politicians, lobbyists, union leaders, and MBAs are meddling in work that should be left to actuaries. There is nothing inherently problematic with a defined benefit pension plan. You can accurately estimate the present value of all promised benefits, and you can accurately estimate how your plan's assets will grow to fund these benefits. To the extent you later believe assumptions are wrong, you come up with new assumptions, and balance the equations again.

Consider the following example:

Company A offers a defined benefit pension plan with an annual benefit equal to 62.14% of final average salary for an employee that works 40 years.

Company B offers company matching to a 401(k) plan, up to 6% of salary.

A person aged 25 starts work at each company, for a salary of $40,000. Salaries are increased 5% each year. These people retire at age 65.

Both the company and the 401(k) plan earn a 6% return.

The cost of a $1,000 annual annuity for a person aged 65 is $11,000.

The worker for Company B, getting the 6% match, will accumulate $825,713 from that match, good for an annuity of $75,065.

Company A estimates the employee will retire at age 65 with an average salary of $120,800 and a pension obligation of $75,065 a year. The company chooses to pay for the obligation by setting aside a portion of the employee's salary each year. The amount to be set aside is exactly 6% of salary.

In other words, there's no difference just because one plan is defined benefit and the other defined contribution. The difference comes from the design of defined benefit pension plans. A typical plan is a percent of average salary (career or last 10 years) times years of service. Most plans don't vest until someone has worked 5 years. I worked at one company with a DB plan, but left before 5 years, so I get nothing. (That allows higher benefits for those who stay, of course, and it's all calculated in designing the plan.)

The article is entirely wrong in saying that without pensions, people may be forced to work until death. There's nothing in a defined benefit pension plan that can't be imitated with a defined contribution plan rolled into a life annuity. Without someone else calling the shots, some people may make bad decisions, of course. They may not save enough, or they may not annuitize when they should. But it's less likely that an insurance company's annuity will fail than a pension plan will be taken over by the government.

To the extent that pension accounting isn't accurate, due to the design of the laws, they should be changed. And PBGC should be charging a greater premium. Everything is legal, of course. Again, these laws were designed by politicians and lobbyists.

Pension plans have been on their way out since ERISA. The regulation makes it more costly to maintain a DB plan than a DC plan, and few workers expect to stay in one job long enough to make a DB plan attractive.