The government published its response to its response to the consultation on what to do about all the small pension pots that will proliferate once auto-enrolment starts a couple of weeks ago. It committed the government to a implementing a solution based on “pot follows member”. When you change job, a small pot built up in your previous employer’s scheme will be automatically transferred to your new employer’s scheme unless you say otherwise.
This was not the solution favoured by the TUC or other consumer groups. We issued a joint statement with Which? and Age UK identifying a range of problems. We were not the only critics. Only around one in five respondents backed pot follows member.
One big issue is the practicality of pot follows member. People do not always move smoothly from job to job. The previous employer’s scheme may well have severe difficulties working out where they are meant to move the pot.
While there may be ways round this administrative challenge, the principled objection remains. There is a real risk that people will have their pots moved from a good scheme to one that is not as good. Under any transfer system there will need to be a ban on some transfers when people leave schemes that have some special features, such as access to favourable annuity rates or risk sharing. No-one should have an automatic transfer from a DB scheme to a DC scheme for example.
But those exceptions will exempt only a tiny proportion of potential auto-transfers. The real issue is whether people should have pensions auto-transferred from a good quality, low charging DC scheme to one that is demonstrably inferior.
To be fair the government’s consultation recognises this issue and says that they need to work through it.
But – as I argued in an earlier post – this raises some profound questions about the structure of current pensions. The evidence in the rest of the consultation paper does not suggest that the full implications of this have been recognised.
The fundamental intellectual flaw in the consultation paper seems to me to be a belief that there is some kind of consumer friendly functioning market in DC pensions. The test of policy approaches it uses is how much they minimise the number of pots. The belief is that this will reduce costs in the pension system, which will in turn be passed on to consumers. The best approach therefore is automatically one that ends up with the smallest number of pension pots.
The paper rules out some approaches because they would lead to “market distortion”. It only models transfers into an aggregator of small pots of less that £2,000 because this is the maximum size the pensions industry could contemplate.
But it is very obvious that the market for auto-enrolment products is not one that automatically produces the best outcome for consumers.
Auto-enrolment pensions are sold to employers. Their prime concern will be buy a product that makes them legally compliant and costs them the least in time and money in integrating auto-enrolment requirements with their existing pay-roll and pensions arrangements. In particular there are no monetary incentives for employers to choose a low-cost pension. They don’t pay the charges, their employees do.
Of course many employers are carefully considering the quality of the pension and giving this a great deal of care and attention, but there are no market forces driving this. But the smaller the employer the less likely it is that they will have the desire or competence to make any kind of judgement about the quality of the pension they use. They lack the pensions managers working for bigger companies, some of which I know have taken this decision very seriously.
More transparency about charges is necessary – and even though Labour and others are right to raise hidden transaction charges as well, as Craig argues – the NAPF work in this area is still likely to be an important step. The entry of new low cost schemes and products into the auto-enrolment market is also welcome, though doubts remain about whether some are offering loss-leader pricing to secure business among big early employers.
But these do not add up to a functioning market that will drive down charges in the long term for all and ensure that any efficiencies gained from pot consolidation get passed on to savers.
So how can we protect consumers in a pot follows member model?
There seems to be a number of possible approaches open to the government:
- ensure that all pensions are of reasonable quality by using regulation – in particular the government’s much mentioned but never implemented power to cap charges. This would be met with howls of protest from the industry, though would be good for consumers. This though would still only limit the potential losses not eliminate them (unless the regulation was extremely strict). There is an interesting debate to be had about trade-offs here.
- not transferring pensions where there is a risk of significant detriment – but that will prevent pot consolidation, and lumber good pension schemes with the costs of looking after small pots.
- accept that some people will lose out, but argue that the overall benefits of pot consolidation make this a price worth paying. But while arguments about the greater good have power (unless you are one of the individuals who lose out), the weakness here is that there is no guarantee that these benefits will be passed on.
So we have a choice of a politically unacceptable solution or two approaches with major flaws.
I’m glad it’s not my choice.
But perhaps there is a fourth option, though I have not detected any appetite for it in government. While I’d still prefer a straight aggregator approach, it might be one way through that does not require what all governments dislike, a u-turn.
This would be a hybrid solution between pot follows member and an aggregator approach. Pot follows member when the receiving scheme can show that it is of sufficient quality to receive auto-transfers. This would not eliminate detriment but ensure that it was not substantial. Schemes would probably need to apply to the Pensions Regulator for a quality mark of some kind that would make them eligible for auto-transfer.
If the new employer’s scheme did not meet this mark – or if there isn’t a smooth transition to a new job – the old employer would transfer the pot to an aggregator. In this system it would make sense to have only one aggregator – and NEST is the obvious candidate, as long as NEST’s other members did not have to pay for the costs of running small pot transfers. This would almost certainly mean having a higher limit for small pots than £2,000.
(Disclosure: I am a member trustee of NEST, but these views are mine and I’ve not discussed this option with NEST colleagues.)