Rise of the machines
Written by Laura Blows
Laura Blows examines the growing interest in data quality for pension scheme administration and the tools helping to achieve this
It's been said that 'the meek shall inherit the Earth'. They may be still waiting for that, but in the world of pensions the quiet, oft-forgotten back-office skills of the administrators are finally having their day.
And the reason why admin teams are stepping out and grabbing the attention of both trustees and sponsors? An increasing interest from all parties in pension scheme data, particularly the quality of that data.
This focus can be attributed to increased regulatory pressures in some areas of Europe. For instance in the UK, its regulatory body The Pensions Regulator has set a target for schemes to aim for 100% compliance on new common data (data that is necessary and applicable to all members) created after June 2010 and 95% on legacy common data by December 2012.
In Italy, data quality has also increased in importance, says Martino Braico, senior manager of pensions funds at Italian-based third-party-administrator (TPA) Previnet. This he attributes to the requirement for calculations in relation to payments of benefits to be strictly recorded, as external supervisory authorities on a peri- odical basis require precise details from pension schemes about members and contribution distribution across the country.
While pressure from external bodies may have increased the need for good data in some countries, in others it was a more 'organic' process. Girish Menezes, principal for ACS' global benefits administration highlights the Netherlands
as an example of a country that has noticed the improvements in information and the reduction of risk by improving data quality, so has naturally looked to adapt techniques and processes.
The growth of DC provision also throws up its own host of data problems, in particular with keeping track on contributions. “With DB, you are getting money on a fund level and investing on a fund level, but paying it out on a member level. With DC you are taking small micro-payments for individual members and investing for members in different asset classes so keeping track of that money in, money out is hugely complex. If that data is incorrect it could spiral out of control,” Menezes explains.
Also shining a spotlight on data quality was the recent financial crisis. As schemes were exposed to market volatility and the realisa-tion that asset classes were more closely correlated than previously expected, trustees and sponsors looked to derisking strategies, such as buyouts, buy-ins, enhanced transfer values (ETVs) and pension increase exchanges (PIEs) to protect against such turbulent times.
“In the past a TPA would have said that sorting out data would
cost x amount and the next thing it would be shelved as too expensive. But now as schemes look at derisking programmes it has exposed gaps in their data to achieve this. As the benefit of derisking is so much greater than the cost of sorting out the data, schemes have been willing to tackle their data issues, rather than simply tidying up their data for the sake of it,” Muse Advisory director and co-founder Avgi Gregory explains.
However, many of these derisking techniques were unaffordable or unachievable with the insufficient quality data held by some schemes.
A common data issue for schemes is not keeping up to date with basic information such as changes to names or addresses. However Gregory has noticed many of her clients experiencing problems with GMP conciliation, discretionary increases and deserters.
DB schemes also suffer from legacy issues, as historically they were not expected to hold as much information. “However as DB schemes close it is no longer sufficient to wait until a member retires to check their data. So the information checking that should take place over decades is now being done at the same time,” Menezes explains.
The complexity of managing data is further compounded by the advent of international and pan-European pension provision, made easier by EU regulations such as
the IORP directive. Data-wise, this increases the risk of missing data, language issues and keeping up-to-date with individual countries' rules and regulations.
In order to understand the quality of a scheme's data, conducting
an analysis is required. Aon Hewitt commercial director for benefits administration Colin Hamilton says there is a chance the scheme may be pleasantly surprised by this analysis, but it is more likely to find gaps in information, meaning data cleansing is required.
A large part of this data cleansing involves going through old paper-based records. There are tools available for this, for example to scan microfiche and extract the data, along with tracing services for deserters. Menezes says its company provides text recognition software within its scanning solution so that the words scanned turn into a text document, which can then be searched, rather than just an image. Translation software is also becoming more beneficial as international pension provision increases.
A lot of this cleansing can be conducted through automated processes, but human involvement is still inescapable, as there will also be some issues with data that require manual intervention. For instance information from clients may be placed in the wrong order, so that file can be manually solved rather than rejected.
Previnet is also using its software to help deal with different legislations for schemes it administers outside of Italy. Braico explains: “Schemes within Italy may need a certain level of information, while schemes from another country may require additional or less information. Our in-house system enables us to set up different requirements for different countries. So the biggest challenge for us is ensuring we stay up to date with different countries' regulations.”
Putting the onus on the individual member to keep their records up
to date is also improving data quality. They are encouraged to do so through online self-service that enables them to update their own records. The internet is hardly a new phenomenon, but its use in this ‘web 2.0’ way is still taking some time, particularly for DB schemes, Gregory notes.
The increased focus on data protection has also played a part in improving records, Hamilton adds. Through more scanning of information (“as it's easier to have security on a computer than a filing cabinet”) the move away from paper has improved the security and by default the quality, he explains.
It is not only member details, or even contributions records, being improved by technology, but also fund administration. Dublin-based Trinity Fund Administration MD John McCan has noticed an increased demand for analytics on a more regular basis, which has only been possible through improved software developments. Along with cyclical reporting of funds, usually once a month, there is a move towards more frequent reporting, on a weekly, or even daily basis.
As well as regular reports, fund information can also be accessed online by investors using a secure web-portal.
Despite the growing use of technology to assist with pension administration, the pace of change is not even throughout Europe.
For instance, according to Braico, the Netherlands is very far ahead
in terms of pension administration automation and in Italy paper processes for pension fund administration is being phased out. However, in his experience French pension scheme administration is still a largely paper-based manual processes.
It may be progressing at different speeds, but the importance of good quality data and the technology to achieve this is gaining awareness across European pension schemes. Administration is finally demanding recognition as the backbone to the rest of the pension scheme.
Written by Laura Blows