It’s difficult to go anywhere on planet pensions at the moment without hearing people extol the virtues of scale. I seem to be the only sceptic on this. (I’d better upfront declare my conflict of interest as our clients include a fair proportion of small schemes.)
To get scale, you need to merge schemes from different employers and my scepticism starts from knowledge of the already existing multi-employer schemes operating on sector bases. My experience of a number of schemes like this is not good. Governance frameworks look cumbersome, expensive, unresponsive and not representative of either employers or members. Employers in particular feel estranged from the scheme bureaucracy which seems to them to be out of their control. Administration services are patchy – at times dreadful, and actuarial services looks rather copperplate – no innovations spring out and disenfranchised employers often feel they need separate actuarial advice so savings at the scheme level are eaten up.
The area where pooling seems most advantageous is in investments and the pooling of LGPS schemes might bring some lessons. The idea of a small number of combined investment funds for pension schemes has some real attractions. To some extent this already happens as small schemes almost universally invest through pooled funds. But individual schemes still take advice on which funds to choose and there is a myriad of investment managers to choose from. Say we had 3 national investment pools – an individual scheme could give its assets to one pool leaving the managers of the pool to allocate assets and choose managers. But this looks very much like fiduciary management. So if the fiduciary managers address concerns raised around transparency and conflicts and then win lots of business, we’d get to the same place?
In administration, the scope to generate savings seems limited. We have almost 6,000 pension schemes in the UK and my guess is that this will cover lots more than 6,000 different benefit structures! Many schemes have a variety of benefit structures within them – different pension ages, pension increases, partner benefits, death benefits, leaver benefits, some contracted in, some out, some with on-top benefits like bridging pensions, funeral grants, additional lump sums – and some so dripping with historical complications that a member asking for a retirement quote can generate 2 days of work involving the administrators, scheme actuary, pensions manager – and the ex-administrator now working in your research department who’s the only one who really understood the special agreement reached with members in 1983. This may not be ideal but it is the reality in many schemes. Now, administering 5 different benefit structures in 1 scheme is only marginally cheaper than administering 5 schemes – the real admin savings come from harmonising benefits and to some extent, erasing a scheme’s history.
An example of a scheme that exploits these potential admin savings well is the PPF. On entry to the PPF, a member’s benefits are standardised to meet the PPF template; pre-97 increases disappear, GMPs are equalised the PPF way and administration can be standardised. This is easy to do as it’s within a legal framework that allows the reduction of member benefits.
This is where my real cynicism creeps in. I can’t help thinking that calls for scheme consolidation mask a real call for legal backup to a standardisation of benefits which might just involve some collateral damage to things like pension increases – all in the name of efficiency of course!
I’m willing to be persuaded – but for the moment, I view the case for consolidation as unproven and needing to be treated with extreme caution.
Hilary Salt is a Founder of First Actuarial and one of the Pension Plowman’s bosses!
She’s our brains trust. When she’s not being brainy, she supports Man Utd:- one of the reasons she’s called @redactuary on twitter.
This article first appeared in Professional Pensions here http://www.professionalpensions.com/professional-pensions/opinion/3003980/is-consolidation-really-the-way-forward