The long-term costs of more choice in auto-enrolment

Pre-budget rumours are a British political tradition.

This year’s include the possibility of introducing pension choice alongside automatic enrolment.

Swirling around in debates about ‘a pot for life’ are arguments to unpick the basis of auto-enrolment’s success by having workers actively choose their own pension, rather than the current system where the employer chooses.

Superficially attractive, the long-term unintended consequences of this change for retirement incomes could be significant, including undermining the government’s push for greater pension fund investment in illiquid assets.

One of the many benefits of auto-enrolment’s design has been to restrict marketing costs. Pension choice – by creating a retail market – immediately brings advertising to the fore.

Since scale drives net returns, schemes will be forced to compete on brand – expect lots of adverts on the sides of buses and taxis and on TV, paid for, in the end, by scheme members.

The Financial Conduct Authority, in its study of non-workplace pensions, found that in practice the majority of individuals who bought pensions chose on brand advertising. This, as the regulator points out, is a consequence of how hard it was for most people to evaluate complex pension products.

International examples

There are several examples from other countries of the negative impacts of increased choice.

In Mexico, for instance, when pension choice was introduced for workplace schemes, 40% of workers switched into schemes with both higher costs and lower returns.

In Sweden, as Nobel prize winner Richard Thaler has famously noted, a huge effort at significant cost to taxpayers to persuade Swedes to choose their own fund for part of their pension savings had a perverse effect. Individuals did make a choice but never subsequently reviewed it, even where there was overwhelming evidence that they should do so.

Pension choice could also, depending on design, lead to higher opt-outs as automatic inclusion in the employer’s selected scheme is replaced by a new process: employers having to ascertain from employees the retail scheme to which they want to belong and then send the contributions there.

This could load further costs and complexity onto employers’ HR departments, and of course most employers in the UK are small businesses without large functions of this kind. A key insight of the successful design of auto-enrolment was paying attention to potential unintended consequences of policy decisions on employer support for workplace pensions.

Increasing risks

If pension choice comes to mean the right to transfer auto-enrolment pots into retail non-workplace pensions, the risk is greater still. These funds are on average more expensive, sometimes dramatically so, as the FCA has highlighted.

Providers of non-workplace pensions say that the operation of the market prevents abuses. The FCA has pointed out that there is no difference in the findings of their analysis of consumer behaviour in the non-workplace market and the previous findings of the Office of Fair Trading (OFT) in 2013 with respect to workplace pensions.

The OFT report led to charge caps and the requirement for independent governance committees for workplace pensions.

It is also worth remembering that the last time employees were encouraged to switch out of occupational pensions, in the late 1980s and early 1990s, there was a mis-selling crisis that undermined confidence in saving.

Productive finance

Pension choice also runs counter to the government’s widely lauded objective of increasing investment in illiquid assets, also called ‘productive finance’.

First, as above, the need to divert resources to advertising and marketing means that there will be less available for investment and illiquid assets are more expensive allocations.

Nor are members likely to exert a countervailing pressure: investment returns aren’t a big factor for very large numbers of people who consolidate savings – as recent research published by the People’s Pension demonstrates.

Second, schemes will have to retain more liquidity as a hedge against an outflow of assets through the exercise of pension choice.

Third, pension choice has the potential – if the government opens automatic enrolment up to non-workplace pensions – of undermining the consolidation of small pensions into larger asset pools, which is the stated aim of government policy.

Making choice work

For pension choice to work in the worker’s interest, the choice would have to be between a set of scaled-up schemes that demonstrably deliver value for money, operate with a legal duty to put the pension first, and whose advertising costs are capped. This would mean a system that is carefully designed with a limited set of choices, to avoid the poor outcomes associated with markets where information asymmetries between buyer and seller are vast.

Even then, if the government’s priority is to increase UK pension fund investments in unlisted assets, then undermining the kind of institutional investment arrangements that facilitate significant illiquid allocations would be odd.