PTL's Colin Richardson argues the benefits of cleansing data and what can be improved by doing so.
Although it doesn't sound as exciting as investments, data audits help save money - so that makes it more interesting! If pension schemes pay benefits with incorrect data, then the comeback certainly won't be dull. Other decisions based on incorrect or incomplete data can be fairly catastrophic too and if you've ever seen a scheme try to unwind past benefit inaccuracies you won't need convincing of this.
When it comes to ensuring pension scheme data is as complete as possible, the Pensions Regulator requirements are a push, but the real incentives for Trustees and employers are far greater than that.
Haven't we cleaned data already?
Some schemes have, but with varying success. How many data gaps still remain in your latest assessments? If no data audit has been undertaken recently the needs are strong. Furthermore, new SORP pensions auditing requirements are set to become stronger in many areas, for example, knowing details of old annuity policies made by schemes. Key reasons to audit data include:
- administering the Scheme in accordance with the Rules is as basic a trustee duty as there is: without complete data this cannot be done
- the information will be needed in the future for members who may not draw their pensions for some time yet. Data cleansing is a case of "now or later" so the time and cost should be seen in that context
- when members retire, or other events trigger benefit payments, it is very inefficient to need data on each occasion rather than a bulk exercise in advance.
- funding and strategy assessments become more accurate and useful with correct data.
- funding contributions and liabilities in financial disclosure may be artificially too high with incomplete data
- the cost of securing benefits with insurance policies becomes cheaper with more comprehensive data
- incorrect benefits could be paid or insured. Unwinding this would be unwieldy and very expensive.
- matching cash flows within investment strategy is more difficult and more approximate with incomplete data
- missing members may appear out of the woodwork unless data is complete
So what are the key areas?
Many Schemes share similar issues. The most common ones are:
- lack of information on which members have dependents and the details of the associated contingent dependents' benefits
- incorrect or unknown splits of deferred pensions for members who left service many years ago
- incorrect or unchecked State Second Pension benefits including Guaranteed Minimum Pensions (GMPs).
- incorrect total deferred pension
- unknown "pre-commutation" pension at retirement for current pensioners which may be needed to calculate future dependents pensions
- unknown or incomplete information on member contributions, and interest due if applicable
- unknown dates of birth for some pensioners
- unknown deaths leading to pensions being paid where they should have ceased
What can the gain be from improving data?
Suddenly a bulk buyout may be affordable, as the cost of insuring benefits is lower if data is more complete. For example, knowing dependents details may reduce the cost of insuring current pensioners by up to 5 per cent or so. This is no small difference and can alter the position of affordability or timescales required as, in essence, the insurer has to assume the missing data details and will be conservative in terms of how many dependents there are and their ages.
There is an opportunity cost if poor data jeopardises or loses a good potential deal as delays can mean that a market window is missed. Data cleansing can take many months, so is best to be undertaken before you are ready to undertake the buyout.
Written by Colin Richardson, Client Director at PTL.
