Think data accuracy, not just data

Data, data, data, data - it really is a buzz word in the pensions world at the moment. What with GDPR knocking at the door, the Pensions Regulator (tPR) requirements for common and conditional data in pension Scheme Returns and GMP reconciliations already through the door and pensions dashboards coming up the drive, it is a topic we aren’t going to escape any time soon.

Accurate data is a heart

Without it, the running of a pension scheme could suffer. Get a blockage and problems could occur that may be catastrophic. Incorrect benefits valued and paid, additional funding concerns and increased costs for employers…

Accurate data gives pension trustees a solid foundation that is key to funding discussions with the employer and ensuring liability reduction exercises (such as pension buy-ins or buy-outs) are more cost effective and more timely.

Previously, focus for pension trustees has been on gaps in data but it also really needs to be on ensuring data already held is correct. So, how do trustees know if their data is less than accurate?

The clues are out there…

Clue number 1: the results from your Guaranteed Minimum Pension (GMP) reconciliation. Members appear on HMRC records but not on your scheme administrator’s (or vice versa). Pension schemes with complex histories and benefit structures will more than likely have issues.

Clue number 2: your pension scheme has a history of different administrators. Inaccuracies caused when data isn’t transitioned correctly or scheme knowledge is not documented and lost.

Clue number 3: your pension scheme has seen previous merger or acquisition activity. Corporate transactions affecting a pension scheme tend to bring with them changes to benefit structures, new categories of member, benefit guarantees etc. They can all too easily contribute to errors in scheme or member data.

It isn’t all the administrator’s fault

Data errors don’t always occur because the pension scheme administrator has been poor at record keeping. Yes, transitions between administrators in the past have contributed to the problem – which we are seeing in the results of GMP reconciliation etc - but, in my experience, it is usually a combination of errors.

‘At source’ errors are quite commonplace - for example, incorrect pensionable salaries provided by the employer to the administrator. Including allowances that are not pensionable is a great example - or forgetting to include an allowance that is pensionable. Issues with part-time and maternity service and pay are another classic. The list goes on…..

Once diagnosed, what’s the treatment?

This really depends on the issues found and the cost benefit analysis of rectification or acceptance of the data as it is. Ongoing prevention to stay fit and healthy could be regular independent data audits – check-ups if you like. At the very least, your scheme actuary can give a professional view on data quality as part of pension scheme valuation process by looking at changes year on year etc.

Tighter controls for administration transitions are a must. Training and documented procedures for all data processors should include data awareness. The big picture is important - all those involved in a scheme (including the pension trustees) need to know what data is provided by whom and when, how it is or will be used in the future and the implications if something goes wrong.

It really is time for pension trustees to make a permanent ‘lifestyle’ change and get their scheme data into shape. I genuinely believe small, steady and regular changes will make a difference!

 

 

Back to opinions