New research has revealed error rates of up to 50% in the auto enrolment data submitted to pension providers, resulting in increased costs and administration for employers, payroll firms and pension providers.
Fintech specialist firm pensionsync analysed data representing contributions to over 10,000 schemes and found that data sent on behalf of employers to their pension providers had a 50% error rate.
Common submission errors included:
- Contribution amounts that are too high or low
- Contributions made for workers who do not belong to the scheme or have opted out
- Wrong pension scheme identifiers
- Inaccurate postcodes
- Incorrect pay period dates
One particular area of pension administration confusion highlighted by the report relates to contribution errors resulting from employers or their agents incorrectly believing a pension scheme operates on a Relief at Source basis rather than Net Pay, and vice versa, and so confusing gross and net figures.
According to the study’s conclusions, such high error rates suggest the need for greater attention to be paid to data accuracy by pension administrators or employers themselves.
Speaking at a House of Lords industry roundtable hosted by pensionsync chair and former pensions minister Ros Altmann, authors of the research emphasised that this does not mean all auto enrolment data contains data that is 50% inaccurate. Altmann also praised the “fantastic success” of the scheme, with opt-out rates far lower than expected.
However, the requirement to correct and resubmit the data adds time to pension administration, resulting in increased costs for employers, payroll bureaus, bookkeepers and pension providers.
Speaking on the issue Altmann said the industry needs to discuss the data issue before they become “another legacy issue” that needed to be dealt with. “We don’t want people to get the message that they can’t trust their pensions,” continued Altmann.
Following the significant issues caused by inaccurate records in legacy systems, many in the industry were hoping for a fresh start with auto enrolment. Unfortunately, a lack of regulatory checks, common standards around data and inaccurate record keeping have led to the current state of pensions data.
Neil Esslemont, head of industry liaison at The Pensions Regulator stated that the watchdog’s “top priority” is to have data improvement plans to check on legacy data and how it is being corrected.
However, as flagged by the research, the current regulations only require the regulator to check employers are paying into a scheme - not the accuracy of information.
The CIPP surveyed their members on auto enrolment, and flagged several major issues. These included the fact there was nowhere to go for clear advice on the scheme, too much choice in the industry and no standardisation of terminology.
At the roundtable, payroll and pensions experts admitted that there were no “silver bullets” for fixing the issue. One suggestion was for pension providers to capture salary data to ensure the right contributions are being paid, while another was for the industry to form a working group to agree a set of standardised terms all providers could work to.