What do I mean when I talk about “investment product data quality”?

I got asked a question recently by a reader of this blog – in summary they wanted my opinion on what I really meant when I talked about data quality, specifically with reference to investment product master data.

You might think the question is a bit simplistic – I would disagree entirely – as my own personal experiences are that people have really diverse views on what quality actually means when they talk about data.

I covered this point off in a whitepaper published last year and I will re-produce the core points here….

I believe that data quality can be best described under three headings: Consistency, Timeliness and Accuracy, as follows:

Consistency : too often we see printed fund fact sheets containing information that contradicts material available on the company’s own website. Even more prevalent is the inconsistency often observed between an investment manager’s in-house data and data made available to the various data and technology vendors as well as third parties who publish/use that data externally e.g. the Morningstar, Lipper or eASE databases.

There is a regulatory onus on the investment manager to ensure that product information they put into the public domain is consistent, timely and accurate. An interesting segue here is the recent settlement between Standard Life and the FSA in the UK. In summary the FSA investigation concluded that:

• marketing material regarding the Fund was not ‘clear, fair and not misleading’;
• despite the majority of the Fund being invested in Floating Rate Notes by July 2007, marketing material issued by SLAL referred to the Fund as being wholly invested in cash;
• there were no adequate systems or controls in place to ensure that marketing material issued accurately reflected the investment strategy for the Fund;
• customers were therefore misled as to the true nature of the investments held by the Fund and as a result, they were given misleading information on the risk of capital losses; and
• as the Fund was intended primarily for the investment of pensions, it was considered appropriate for individuals approaching retirement and as such, the capital risk associated with an investment was of great importance.

The complete notice makes very interesting reading and I would recommend you take time out to read it– we can conclude here that within Standard Life there were systems that clearly flagged a set of holdings as being e.g. floating rate notes, yet by the time that data ended up in the public domain the classification of those same assets got mapped to “Cash”, which as the FSA found, was wholly misleading. This is an excellent example of the kind of costs that an asset manager is exposed to when they have inconsistency between internal and external data – the cost in this case was not simply the fine of £2.45 Million but also the £102.7 Million that Standard Life paid into the fund to compensate for the losses incurred in January 2009 when the problem came to the surface.

Timeliness: how much confidence would you have in a fund provider who offers you sales support documentation that is 6 weeks old? What about 8 weeks old, worse still what about 3 months old? Yet this is often the accepted norm in our industry.

In the internet age you expect documentation to be completely up-to-date. The whole purpose of an investor micro-site should be to provide the latest market position for the product – not a copy of the data published at the end of last quarter.

In the asset management industry the accepted norm is that printed documentation should be updated to at least the last quarter end and ideally to last month end. Yet many organizations struggle to get their data to the printer before the end of month.

In most cases the data is available on business day one or two – so why the delay? Is it efficiency, attention to detail, or something more endemic? There should not be a monumental struggle to get information to market (print or online) by business day five in any organization, yet very few organizations can manage this.

Accuracy: the greatest issue of all is accuracy, yet it is probably the easiest to resolve. Inaccurate data can not only damage an organization’s reputation but also their balance sheet. The negative impact of publishing incorrect data is significant and immediate.

There is simply no excuse for publishing inaccurate or ambiguous information. Not only will clients not accept this but the regulators also take a very dim view on the subject.

Most investment managers believe their data is accurate by the time it gets into the public domain, although the behind the scenes processes to getting there are often extremely costly, manual, labor-intensive, prone to error and wholly inefficient.

Finally, some would argue that ‘Security’ directly relates to data quality, my own view is that data security is a subject matter in itself…we can cover that another day.