The C.D. Howe Institute published a well-written report on Target-Benefit Pension Plans that explains how they differ from traditional Defined-Benefit (DB) and Defined-Contribution (DC) plans. Target-benefit plans solve a number of the problems with DB and DC plans, but they have some serious challenges as well.
Defined-Benefit (DB) pension plans push all of the risk onto employers who have to provide predictable benefits. Employers must shoulder the risk that investments may perform poorly, forcing them to make large contributions.
One problem with some DB plans is that they use unrealistic assumptions about future returns to reduce today’s contributions. This can lead to chronic under-funding. Another problem with some DB plans is they use unrealistic actuarial information. Effectively, they assume people will die younger than they actually will. This leads more under-funding problems.
Defined-Contribution (DC) pension plans push the risk from employers to employees. The employers know exactly how much they will have to contribute each year, but employees cannot predict how their savings will grow, and certainly cannot predict their future pension payments.
Even worse, DC plans create a risk that wasn’t present with DB plans: longevity risk. A given employee cannot predict how long he or she will live and must either underspend or take a chance on not living past the average lifespan. With DB plans, this risk gets diversified away, but with DC plans, employees must choose between frugality early in retirement or potentially running out of money late in retirement.
Target-Benefit Plans eliminate many of the disadvantages of DB and DC plans. Employer contributions are predictable within certain ranges. Whether benefits rise by more or less than inflation is determined by how well investments perform. So, employees take on some risk, but longevity risk is eliminated.
A big concern for Target-Benefit Plans is potential influence by employers or employees on how plans are run. If competent, well-meaning people choose investments and choose actuarial tables, then Target-Benefit Plans look like a great solution to difficult problems.
However, employers and employees would have strong motivations to influence how these new pension plans are run. If an employer can get the pension plan to use unrealistically high investment return assumptions, the pension plan will be chronically underfunded. This saves the employer a lot of money, and employee benefits will fail to keep up with inflation. Similarly, employees could get benefits rising faster than inflation if investment return assumptions are too low.
Another way employers and employees could influence the pension plan is through actuarial assumptions. Employers would prefer to assume we all die as young as possible to reduce the size of their contributions, and leave employees with benefits trailing inflation. Employees would prefer to use actuarial tables that assume long lives to increase required employers contributions and increase employee benefits.
Another risk is the growth of expensive administration for the pension plan. With DB plans, expensive administration leads directly to employers having to make larger contributions. With Target-Benefit Plans, the cost of bloated administration comes out of employee benefits. Usually, employers are in a better position to control administrative costs than employees are.
I don’t think it is overly difficult for competent independent people to handle investment return assumptions and actuarial tables fairly. The problem is that there are billions of dollars at stake, and employers and employees will be very strongly motivated to influence how Target-Benefit Plans are run.