An announcement from accountancy giant PwC that two thirds of the schemes it deals with in company takeovers have significant deficiencies in their membership records, left me feeling a tad under-whelmed.
Put bluntly, the quality of historical pension scheme data held electronically is a disgrace and the industry, and not a few trustees, have known this for some time. Perhaps the only truly surprising point of note is that nobody has taken any steps to meaningfully address the issue.
In the internet age when most of us have a PC on our desk, at least one in our home and a Blackberry, i-Pod or PSP in our pocket, we have seen a considerably slower pace of development in IT services for administering final salary pension schemes. It is however vital that we do not under-estimate the importance of these data deficiencies.
We need to keep sight of the fact that the purpose of a pension scheme is to pay the right amount, to the right person, at the right time. Simple! You would have thought so! Yet inaccurate or incomplete records can prevent this being achieved.
It is not uncommon when taking over a scheme to find member details incomplete, past service entitlements not properly recorded, even benefits being retained for members who have died! We also still find ourselves frequently being offered expensive paper based solutions for exercises which really should be relatively simple to complete assuming the data was available electronically.
Importantly this same data is used by actuaries to estimate scheme funding costs. It’s not unreasonable to assume that dodgy data will give rise to equally dodgy estimates. Given the myriad risks already faced by trustees of final salary pension schemes, why have so many been happy to date to accept an additional risk which they could eliminate?
The reality is scheme administration has never really been afforded the priority it deserves unless or until it causes things to go seriously wrong.
For of the larger consultancies record keeping and administration have been areas of low margin work, which need to be offered to help secure the much more lucrative consulting and actuarial work. This attitude has been transmitted to clients who also perceive administration to be a low value service, where he who quotes lowest wins most. As in all spheres of life you tend to get what you pay for.
This has undoubtedly led to a culture of make do and mend where administration providers are unwilling to raise issues which require solutions they believe their clients will be unwilling to pay for. Unfortunately this is unlikely to change as we see mounting pressure on administrative costs as more and more schemes close to new members or future accrual. The Pensions Advisory Service estimates that around a third of all complaints received last year were directly related to poor administration, a significant cost for many schemes.
As noted above administration and the quality of records held electronically is an industry problem, and a problem which is being exacerbated by the attitudes taken by both providers and clients. Many schemes will have been running for over 20 years and probably began with paper records. These records may or may not have been transferred to what has now become an obsolete system or to another supplier who will have worked on the data provided. Data will have been input manually and short cuts were often taken in recording historical data. The benefits for deferred members are often particularly poor. It is rare in my experience to see data audited at the point of transfer to ensure its accuracy, due to the perceived cost of doing so. There is clearly a potentially greater cost in not addressing these issues.
An existing administration supplier is often unwilling or unprepared to come clean about data issues as they perceive that it may reflect badly upon them but this is to miss out on an excellent opportunity to put the house in order. The reality is that the data is the data and all parties carry a degree of responsibility for the point we have now reached. Trustees have a responsibility for the administration of their scheme and need to focus on the quality of delivery. There are significant financial implications in terms of data inaccuracies, manual interventions to correct benefits paid out or the additional cost of recalculating benefits once errors have been identified. It is not uncommon to witness organisations who spend as much money and staffing resources on clean up exercises as they do on day to day administration.
If there are inadequacies in the data these need to be addressed and seeking to apportion blame isn’t going to help. The fear of our “blame” culture is one of the main factors preventing the issue of poor data being addressed – to everyone’s detriment, including scheme members who may end up being paid too little or too much..
It is also important to recognise that cost is unlikely to be driven down in the long term by continuing to bury our heads in the sand. More effective use of technology is needed to help manage administration costs, but the old computing adage still holds true – “rubbish in, rubbish out”. If the underlying data is incorrect or incomplete then technology can be of little assistance.
As an industry we and our clients are all culpable for ignoring this problem, and even where data issues are highlighted they often aren’t addressed due to the associated costs.
It is the trustees that are ultimately responsible for the scheme administration and accountable to the members. The Pensions Regulator code of practice on Internal Controls identifies a number of areas of risk that trustees must assess and manage. Poor record keeping is identified as one such risk. Hopefully this will place the quality of data much higher on trustees and their advisers agenda.
Published in Financial Solutions July 2006