The small pots solution: does it pass muster for members?

Though he probably wouldn’t be best pleased to hear it, my mental model of the auto-enrolled member who has a proliferation of small deferred pots – or would have, if it weren’t for a sustained nagging campaign on the part of his mother – is my elder son.  

Now on his fourth auto-enrolment eligible job since graduation, and his fifth different address over the same period, he has many fine qualities, but staying on top of “life admin” isn’t among them. (Indeed, I have a strong suspicion that he ended up acquiring a small pot on the occasion of his first enrolment because he never got around to even opening his information pack before the opt-out window expired.)

So when I read the recent Small Pots Delivery Group Report, a key question in my mind was this: how well would the proposed solution work for a member like him? And what will the member experience be?

Hunt the member

The first area of interest is the issue of data matching.  For consolidation at scale to be feasible at all, a pension scheme holding a small pot needs to be able to identify with a sufficient degree of confidence whether the relevant member has a pot with one or more consolidators already. So accurate matching of the member to the records held by the authorised consolidators will be vital.  Whether this happens in practice will depend on the quality of the data supplied to the Small Pots Data Platform, which will be created to facilitate matching.

The recommendations from the Delivery Group were for a two-level approach:

• Primary data (which must be shared with the Platform) – first name, surname, date of birth, and NI number, where available
• Supplementary data (provided where possible) – mobile number, email address, and previous addresses and postcodes

The possibility of creating a “unique identifier” for each member specifically for the purposes of consolidation was also considered, but ultimately not recommended. Now, this is a little puzzling, since members who are eligible for auto-enrolment should in the vast majority of cases already have a unique identifier (their NI number), and it is already mandatory for employers to supply this information to the pension scheme as part of the “jobholder information” when a member is auto-enrolled.  Unlike (eg) a personal email address for the member, there is no discretion for a pension scheme to tell an employer that it does not require this information. 

There will, of course, be some cases where the employee does not have (or does not supply) an NI number on starting work, and this may account for some of the gaps, though in most cases, those ought to be filled where schemes are diligent in pursuing employers for the information. HMRC will expect the employer to obtain and record an employee’s NI number, and will trace and supply it if it is missing from submitted payroll data, so the employer shouldn’t remain in ignorance for very long. 

One also can’t rule out errors, and the tragi-comic case of Mr S is a cautionary tale in this respect. (For those unfamiliar with it, Mr S was auto-enrolled into NEST by successive employers, first with an incorrect date of birth and then with an incorrect NI number. On the second occasion, he was unable to exercise his right to opt out because he was unable either to supply the erroneous information to NEST as NEST required under its processes, or to get his employer to correct the mistake, and in desperation finally reported himself to NEST as dead in an attempt to obtain a refund of his contributions.) 

Enabling, and requiring, schemes to validate NI number data received as part of the auto-enrolment process against DWP records – ideally via an automated batch-checking process, akin to that operated by the Student Loans Company – would be a significant step forward in improving the likelihood of accurate matching (and would save any future Mr S significant stress, into the bargain).

As regards the other suggested matching data, for the cohort of members in scope, recorded data for mobile numbers and email or postal addresses may well be out-of-date. Younger workers in particular may end up needing to move location when changing employment in a difficult job market, and are more likely to be renting, which generally means less stable accommodation. It’s also not particularly likely that previous address information will be held by a scheme with an active member pot set up under auto-enrolment.  So it’s at least questionable how useful these supplementary items will be in matching members across schemes.  Securing the presence of a (correct) NI number seems much more likely to be the key to successful matching.

“Hello? Anybody there?”

The other key area where the suggested consolidation process may find it stubs its toe a little against the hard reality of the typical member with multiple small deferred pots relates to the method and content for member comms. The suggestion is that the pension scheme looking to move a qualifying small deferred pot (less than £1K; no contributions for at least 12 months) to a consolidator will need to write to the member to let them know that consolidation is on the cards. 

The theory is that this will give the member the opportunity to opt out of consolidation, and also to express a view as to the consolidator vehicle to be used (where the member doesn’t yet have an assigned default consolidator). However, it seems more than a little doubtful that a member who is sufficiently disengaged with the whole pensions scene to have ended up with multiple small deferred pots will suddenly have a “road to Damascus” moment of enlightenment and start to grip the problem for themselves in any meaningful sense. 

Indeed, the ability to opt out without also requiring the member to transfer the pot elsewhere could – if the member doesn’t understand very clearly what is happening – lead to an opt-out as a panicky knee-jerk reaction.  Bear in mind that this is a pot dating from at the very least a year back, or possibly much further into the past, given the millions of legacy small pots which have accumulated since 2012.  The member has moved on, and may well have forgotten all about it.

And that’s assuming that the letter even reaches the member.  Reverting briefly to my elder son: remember those five different addresses since uni. Not a single postal redirection put in place (despite sage parental advice), and address records updated “when he got round to it” (don’t ask).  If it just relies on its existing records, his last auto-enrolment scheme but one would in all likelihood be writing to him three addresses back. The current tenants at that address will almost certainly just bin the letter, leaving both the member and the scheme none the wiser.  (I will leave on one side the risk of such misaddressed comms being exploited for identity theft.)

In today’s world, members in the target cohort are most likely to respond to digital means of communication, and ideally via multiple mechanisms: think, SMS and email, rather than either/or, with links through to carefully constructed explanatory materials on the scheme’s website.  Hard copy could be used as a fall-back, but is unlikely to be particularly effective as a mechanism for reaching (in all senses of that word) these members. 

Again, encouraging schemes to collect and periodically validate mobile numbers and email addresses will be crucial.  Whilst the Delivery Group thought it would be going too far to make provision of email address data mandatory at the point of auto-enrolment, how about making provision of either a personal email address or a personal mobile phone number mandatory? From a contact details perspective, these are the modern equivalents of the postal address and landline numbers of yesteryear, and it doesn’t seem so very unreasonable for a scheme to require them to be provided.

ETA: 2030

Finally, however, none of this is destined to arrive until at least 2030.  This will be a little too late to assist my son, who – in response to yet more maternal nagging – is currently buckling down to request yet another transfer of his deferred pension from his last employer’s scheme into his new employer’s scheme.  (The real irony being that the new employer uses the same auto-enrolment provider as his last but one employer, and the transfer-out from that scheme only completed a bare 6 months ago.)  For those without a friendly contact in the pensions industry, I suspect automated consolidation cannot come fast enough.