I published yesterday the OECD’s latest recommendations for the design of DC pension plans

I was surprised that they were so radical, though looking back at their “Pensions at a glance” paper last year, perhaps I shouldn’t have been. The paper called for

 “bold discretionary measures in a financially unbalanced pension scheme”.

Which is about as outspoken as a pension paper gets. Today I want to look at the parts of the paper I highlighted, not least because I got three calls from journalists about the blog, clearly the OECD has moved stealthily!

DC pension plans that are coherent with their long-term purpose and role in the pension system.

What do UK pension plans set out to do within the pension system? There is currently little coherence here. In practice , the majority – by pot size – are too poorly funded to be much more than a windfall payment, perhaps easing people into part time work and acting as a bridge to the arrival of the state pension. At the other extreme, SIPPS are used as a store of wealth, used to provide liquidity in estate planning. What there is not – and this is likely to change – is any urgency from the Government to produce what the OECD call coherence across the accumulation and pay-out phases.

The recommendation that there is coherence is explored in the 7th recommendation, the OECD’s emphasis is that DC plans are able to fulfil their role in meeting adequacy objectives. There are no obvious adequacy objectives in UK policy, though the PLSA’s Retirement Living Standards are generally held to command political consent. It is ironic that Government’s pension policy demands accountability from the private sector but offers no targets for the success of DC pension reforms. Steve Webb did embark on such a task in the early years of auto-enrolment but what he initiated was not followed through.

Coherence in the pay-out phase is recommended to  “provide some level of lifetime income as a default for the pay-out phase”. Lifetime income can be provided by annuities with guaranteed payments or by non-guaranteed arrangements where longevity risk is pooled among participants.

This is a very radical view which has so far got very little traction in the UK. The idea of a non-guaranteed income in retirement is of course coherent with the accumulation of savings which is sporadic , depending on the nature of work and of the employer’s DC contribution promise. The way many of us live our lives is with a great deal of uncertainty about how and what we get paid and – at least for the payment of the pension pot, I don’t see why total certainty is expected or required. Clearly the OECD see it the same way.

What the OECD do see as critical is that we return to the concept of a pension as an insurance against our living too long. Recommendation 8 of the paper is that DC pensions facilitate the regular monitoring and management of longevity risk.

I do not know any DC schemes that make it their business to either monitor or manage the longevity risk of their membership. The future sees this task being taken on by the Trustees of Royal Mail’s Corporate (CDC) Pension Plan, but that is as far as that goes.

The experience of monitoring and managing longevity risk is currently in the hands of the insurers, who are currently mostly exercised in the pricing of bulk annuity buy outs of DB liabilities. I would expect insurers to see the responsibility of a DC plan to manage and monitor the longevity risk pooled by the partners as key to their value proposition. But so far, this has not been the case. I say so far – I suspect this is changing.

The OECD’s paper is not in itself, a turning point, more a weather report. I do not think that any of the phrases in red would have appeared in these recommendations had there not been a “change in the weather”.

Nowhere is this weather changing faster than in Australia, where the management and monitoring of longevity is now a part of retirement modelling for its DC superannuation system. Only a few days ago, the Australian Actuaries Institute published a 41 page Information Note on this subject. As Jim Hennington noted in a comment on the OECD recommendations,