Understanding Solvency II – Is your data “accurate, complete and appropriate”?

Regarded as the “Basel of the insurance world,” Solvency II will impose new obligations on insurers and reinsurers operating in Europe. And with 2012 fast approaching, it is just two years until C-Day (compliance day). January 2014 is the latest amended date from the Financial Services Authority (FSA). If this deadline isn’t in their diaries already, insurers and reinsurers will have a Solvency II shaped New Year’s resolution come January 1, 2012.

Solvency II fundamentally reviews the capital adequacy regime for the European insurance industry. It aims to establish a revised set of EU-wide capital requirements and risk management standards that will replace the current solvency mandates.

Meeting these new requirements won’t be possible without the enterprise risk management frameworks that will demand clean and timely data for portfolio valuation that is capable of standing up to close scrutiny. The data-centric connotations for an overhaul of this stature are huge from both a collection and reporting standpoint.

The recent Solvency II: Internal Mode Approval Process (IMAP) Thematic Review Findings report from the FSA highlights serious concerns regarding insurers’ capabilities to deliver accurate data under Solvency II. According to the report, “Few firms provided sufficient evidence to show that data used in their internal model was accurate, complete and appropriate.”

But what exactly does it mean for data to be “accurate, complete and appropriate”?

Let’s start with appropriateness. The language of the regulator states that insurance companies must ensure that their data is “suitable for the intended purpose.”  In this case, “appropriate” is an industry specific term. Insurers must prove they are able to cover any and all claims made. In short, their data must be able to demonstrate the ability to cover risks associated with potential claims.

Accuracy is a measure of the confidence that can be placed in the data. It must therefore be error free, consistent, and demonstrably so. Firms need to prove that they are using and applying this data to their undertakings. Sounds like a simple enough ask, but the fact of the matter is that for many firms, data is just not up to the required standards.

And finally, completeness. A complete data set will include a combination of sufficient historic data which gives insurers the full picture of risk, and the necessary quantities of risk data.

Here at Asset Control, we’ve had a similar mantra that reflects the data requirements for Solvency II compliance; focusing on providing financial organizations with accurate, accessible and actionable data. But however you interpret the language of Solvency II, it’s not going to go anywhere and it looks set to radicalize the way the insurance space record and report on their data.

Making a leap into the void more of a walk in the park…

(The accidental CTO)

I remember when the first ATM cards came about and there were two extreme types of users. First, there was the young Friday night reveler who marveled at the accessibility, convenience, ‘always-on’ (before ‘always-on’ was even coined as a phrase) and generally ‘cool’ aspect of this money-on-demand facility. They shared cards, used them until they ran out and showed off their new-found sophistication to their friends.

Then there were the older, established people, like my parents, who felt it was an uncontrolled, unsecure, provocative feature that bordered on recklessness on the part of the banks. The whole sorry business would lead to accounts being drained mysteriously, armies of middle class people shuffling around like bankrupt zombies, and the fall of democracy and the fee-world as we know it. Needless to say my parents continued to live under their own roof – at least until that dreadful mix-up over the ‘herb garden’ in the backyard, but that’s a story for another day!

Between these two extremes lay the majority of people who saw the convenience factor, but worried about the potential for fraud or theft, and therefore started to tip-toe their way towards adoption. This middle-ground group understood the pros and cons, and importantly, lobbied for enhancements and changes that closed the loopholes and narrowed the risks. Arising out of this are technology changes that maximize the benefits and obfuscate the negatives.

When I look at cloud computing today I see a similar pattern. We have those who have embraced and pushed the boundaries by removing the shackles of discreet boxes or on-site packages, by taking advantage of the huge reach, the elastic capabilities and the ‘pay for what you use’ psyche.

When one looks at music and video storage streaming and sharing, accessibility through tablets and smartphones, and heart-stopping innovation and creativity in those who have allowed the vast internet to act as a canvas, it is very impressive. And at this point, most of the middle-ground equivalents have also gotten comfortable. The last frontier – the final step, I believe – sits with those who, like a novice parachutist, are standing in the doorway of a plane at 2000 feet, and who simply have to make just one step …

In today’s world, in my world of data management, these are the institutions who just cannot contemplate the idea of letting go of their data beyond their physical firewall. Rather than be paralyzed by fear, I think we have to make a choice to accept that the cloud is here to stay and look to repeat the revolutionary success it has had in other industries.

We should advocate the elasticity and multi-locational, super-redundancy of the global network of data centers to cover the fact that with some foresight, the same set of data can be found in more than one place. We should celebrate the database technologies that allow replication, mirroring, gridding, and splitting. We should marvel at the processing technologies that transfer huge volumes of data in small packets so that every second of uptime is maximized. We should revel in the new security technologies that protect both stored and inflight data, and we should look for ways to capture snapshots synchronized on-site and stored in the cloud, bitemporal (time and transaction) data management to recreate moments in time…at any time, both for the data itself and also for the provenance of that data.

As software solution providers, and as an industry, we need to provide solutions that utilize these things, embrace innovation and are optimized for use in such a world. This way, we can perhaps create the right atmosphere and a safety net that will allow the last step to be a baby step and not a giant leap.