The soaring cost of old ageThe real problem with pensions

PAYING for pensions is like one of those never-ending historical wars; a confusing series of small battles and skirmishes that can obscure the long-term trend. The latest conflict is in Britain where university lecturers are indulging in strike action over changes to their future benefits.

Let us start by making the long-term trends clear.

1. People are living longer and retirement ages have not kept pace. This increases the cost of paying pensions 

2 Interest rates and bond yields have fallen. This increases the cost of generating an income from a given pension pot

3. Private sector employers have reacted to this cost by closing their defined benefit (DB) schemes (which link pensions to salaries) and switching to defined contribution (DC) schemes (which simply generate a savings pot)

British universities have reacted in a similar way; they are proposing switching future benefits to a DC basis. To avoid confusion, this means that past benefits will be unaltered; if you are 50, and have worked for 25 years, you will still have 25 years of DB benefits. But since pensions are deferred pay, it does mean that the total benefits of academics are being cut so one can see why they are upset.

But there is still plenty of confusion, as this piece in the Independent illustrates all too well (to cite just one example, in a piece about workplace benefits, it quotes OECD numbers on state-pension replacement rates). There are three big areas where the debate gets muddled.

1. Investment risk. If there is a pension fund, then there is investment risk regardless of whether this is a DB or a DC scheme. The difference is on whom the risk falls. In a DC scheme, it does fall on the employee. In a DB scheme, it rests largely on the employer. But in a sector heavily funded by the public sector that could mean the taxpayer. 

2. Accounting. The real cost of pensions can't be measured in cash flow terms: how much is being paid out this year, as opposed to the contributions being put in. They are a long-term commitment in which one must work out the cost of future benefits, allowing for longevity, inflation etc. These future payments must then be discounted at some rate to get to a present value.

This column has always argued for the use of a bond yield as the discount rate. That is because pensions are a debt which must be paid. The problem is that low bond yields have forced up the present value of future benefits and widened deficits. The unions in the university case argue this is too conservative and that one can reasonably expect higher investment returns. But this rather contradicts another element of their case. On the one hand, they are saying that DC pensions are too risky for employees because the markets might not deliver; on the other hand, they are saying the markets will be fine so the employer should keep promising DB.

In the US, public pension schemes do assume a high rate of return on their investments and they are in a mess, with a $4trn deficit. In one school district I visited, the entire budget increase was eaten up by higher pensions payments. 

The true test of a pension cost is "what would it cost to get rid of it". Insurance companies will take over pension schemes but when they do, they use a bond yield as their discount rate. This buyout basis makes deficits look bigger. 

3. With public pensions, the rich tend to subsidise the poor. They are also run on a pay-as-you-go basis with today's workers paying the pensions of current retirees. What you put in is not what you get out. With public pensions, the rich tend to subsidise the poor. They are also run on a pay-as-you-go basis with today's workers paying the pensions of current retirees. But in a DC scheme, contributions are very important. Yes, returns matter a lot. But the real reason that DC pensions are lower is that total contributions are smaller; that is why employers are switching after all. In the US, some employers make no contribution at all. In Britain, matching is fairly common. still, the ONS reckons that total contributions averaged 21% of payroll in British DB schemes and just 4% in DC.

That is the big issue; not investment risk and not management costs. As it happens, the university scheme is offering a fairly generous 13.25% from the employers. But that is still a lot less than they might be expected to contribute to bring the DB scheme back into balance.

So the real issue for workers is this; how much is the employer contributing? And the same is true in a sector heavily funded by taxpayers? If the scheme requires more money where will it come from? Higher taxpayer grants? Higher student fees (which will lead to more taxpayer support if the fees are ulitmately unpaid)? Or worse services?