As has been reported on the BBC and elsewhere, my employer has suggested it will close its defined benefit pension schemes to new pension accrual by existing members.
There is a 60-day consultation period coming up, during which the company will discuss the proposals with some elected representatives from each of the IBM pension plans. I am a candidate in this election to represent those people already on the IBM defined contribution pension scheme, known internally as the M plan. For those unaware of the functioning of pension schemes I shall now explain some of the terminology.
A defined benefit, or final salary, pension scheme promises that during retirement you will receive a fraction of your final salary for each year of service with the company. For the sake of argument I will talk about DB schemes in which this fraction is 1/60. These schemes usually require that you retire at a certain age, commonly somewhere between 60 and 65; if you retire earlier than that your pension is reduced. The market rate of this deduction is roughly 6% less pension per year of retirement before 65. The sum of all this is that if you work for the same employer under such a DB scheme for your entire working life and retire at 65 you will get a pension of 2/3 your last year’s salary.
The other sort of pension is a defined contribution, or money purchase, scheme. With these schemes both the employee and the employer pay agreed proportions of the employees salary into a shares fund. When the employee wants his pension he can use whatever the total value of that fund has reached to purchase an annuity from an insurance company. Whatever the annuities which cost that much pay out is his pension. Some typical values for the contributions are that the company will pay in 10% of the employees salary and the employee will pay in 5% each year.
Behind the scenes both are run in a very similar way. With a defined benefit scheme the employer invests some money each year in the stock market and then sells however much they need to sell to buy annuities for their employees as they retire.
Companies prefer to use defined contribution schemes for a number of reasons. Firstly, the risk in the scheme is placed on the employee: if the stock market plummets just as the employee retires it is the employees problem; whereas in a defined benefit scheme it is the employer who must find the money which has just been lost. Secondly, defined benefit schemes have the problem that the cost to the company of providing that benefit is hidden from the employees who are benefiting from it.
It is this second aspect which I think is more important. To provide enough funds for the above-described defined benefit pension scheme, the company needs to invest about 35% of each person’s salary each year. This means that a defined-benefit pensioned employee costs the company about 25% more than is apparent from their salary, whereas a defined-contribution pensioned employee can easily see what their total compensation is. Having a mix of defined-benefit and defined-contribution employees is a problem for an organisation because it is difficult to compare their compensation fairly.
So then, there are good reasons to prefer to have your employees on a defined-contribution pension scheme. If you want to move them across you should take the opportunity to ensure that you are paying each of them in proportion to their performance. If you have been taking their DB pension benefits into account while deciding their salary you need to increase their salary by 15-20% – if the employee wishes they can put this extra money into the DC scheme and they should be no worse off when they retire. If you are in the unfortunate position that you have been paying people the same irrespective of their pension scheme, then you obviously want to keep their salaries constant: this will annoy those who calculate they have lost 15-20% of their compensation, and I doubt they will agree to count themselves lucky that they were overcompensated previously.
I hope the above goes some way to communicating what I think is a just outcome of closing the DB schemes.
I also want to say a bit about what a good DC scheme looks like. One of the best I can find is that offered to IBM employees in the US.
Defined contribution schemes effectively say to the employees: “Your pension is in the stock market: you’d better make sure it’s in the right places so you don’t lose it as it’s your problem.” This is at best an annoyance to the employee, but in the case of some plans it borders on insulting. From the other side.
If you are going to say the above, you need to ensure that your employees at least have a fighting chance to protect their money. They need to have as broad a choice as possible of places to put it, they need to be able to move it rapidly, and they need access to advice on where and when they should move their money. The IBM US plan provides access to well over 200 very different funds, actions requests to move money on the same day that the request is made and gives employees access to financial advice.
I would very much like the scheme I am in to provide me those facilities, but it does not. As you can read in the plan brochure I linked above I have a choice of 4 funds, but since each of those tracks the entire market and performs in pretty-much the same way that is as good as no choice at all. I can move money between these funds, but I can only do so once per month: I have to fill out a paper form and then post it, and it takes up to 5 days for the changes to take effect. If I want financial advice I have to find it and pay for it myself.
Phil
This is a pretty good description of how pensions work, but you have missed some important points.
Firstly the cost of DB changes with your age – your 35% figure might be true for say a 45 year old but is too high for a younger person. Also it varies a lot depending with what increases the pension is due to receive after you retire and what benefits are due to any remaining dependents after your death eg a 50% pension for a wife outliving a husband. It is also very dependant on what assumptions you make in order to work out how much needs to be saved, some sets of (reasonable) assumptions might reduce the cost to the employer to much lower than the figure you suggest. In summary, while DB is usually more expensive than DC the difference is not as strightforward as you suggest.
Another point is that the two types of scheme do not operate in the same way after retirement. DB pension schemes do not in general purchase annuities for retiring members as they retire. The cost of buying annuities is almost always higher than the amount DB pension schemes put aside to pay pensions. This is partly due to insurers having stricter reserving requirements and needing to make profits. Also there is an important difference in the investing of the money. DB pension schemes get expert investment advice, whereas individuals in DC schemes are reliant on their own investment choices – which do you think usually result in higher returns over the long run.
This leads me to my main point (admittedly not that quickly). You state that a “good” DC scheme offers 200 funds to members. You could not be more wrong. Even an investment savvy person with high intelligence will struggle to choose from this many funds effectively, this why 60-80% of members in DC funds end up in the default option. Actually around 10 funds is as many as a person can reasonable choose between. Also many of the more varied types of fund are not likely to be suitable for investors with small pots. I suggest you read the excellent book Nudge by Thaler & Sunstein for further explanation. Actually a better way of protecting peoples money is to have a well designed default fund – ie something better than lifestyle, but that’s another discussion.
You also suggest the fund offer financial advice – fair point but this is incredibly expensive and not feasible for all employees. Incidently if you want to find an IFA Unbiased.co.uk is a good place to look.