Solvency 2 - Could data quality undermine your calculations

Page 1

AN EXAXE WHITEPAPER

Solvency II Could data quality undermine your calculations? A lesson from the bankers’ experience of Basel II

Authors:

Tom Murray / Kathryn Desmond

Published: 3rd June 2010

Exaxe Ltd. Unit 5A Sandyford Business Centre Sandyford Dublin 18 Phone: +353-1-2999-100 Fax: +353-1-2999-199 E-mail: tom.murray@exaxe.com


Solvency II—Could data quality undermine your calculations?

AN EXAXE WHITEPAPER

Executive Summary Over the next 3 years, Solvency II is the key regulatory requirement facing the Life and Pensions industry. Due by 31st October 2012, the changes are designed to strengthen protection for policyholders by ensuring that insurers and reinsurers fully quantify the risks inherent in their businesses and that they allocate enough capital to cover those risks. Drawing on the Basel II process in banking capital adequacy regulation, it applies the same three pillar approach: Pillar I—Risk quantification and capital adequacy Pillar II—System of governance Pillar III—Reporting requirements It is pertinent, therefore, to look at the experience of the banks in complying with Basel II in order to see what issues they encountered and whether Insurers can learn anything from that process. While many publications have focused on the algorithms for assessing risk and calculating solvency ratios, almost no reference has been made to the experience of the banks, particularly to their key problem in meeting Basel II standards; namely the quality of their data. Large amounts of effort and budget had to be expended by the banks to bring data up to the standard where it could be relied upon .Insurers may well face this issue too. No matter how good the algorithm is for their internal solvency calculation model, and even if the Insurer has decided to use the standard model, the old adage still applies: Garbage in—Garbage out. If the data being fed into the model is wrong, then the result will be wrong and will distort the assessment of the business’s risk position and consequent capital requirements.

High Data

Accurate

Sufficient

Quality

Internal Model

Capital Reserves

In parallel with the development of the internal calculation model, Insurers should be running assessments of their data in order to establish the level of confidence they have in it and the extent of the need for data-cleansing operations to bring it up to the quality level required. Getting the solvency capital requirements wrong could endanger the business; if the resultant capital ratio is found to be insufficient to meet its commitments, it could be shut down or go under; if the company is retaining too high a level of capital to cover future risks, it will stifle growth.

Data quality is an essential determinant of the reliability of the Solvency II calculations. Insurers need to ascertain that their valuations are correct, as bad data input into their internal models will give unreliable results that could leave a company, and its directors, dangerously exposed in terms of capital requirements.

2

Data quality was one of the primary difficulties in assessing capital adequacy in Basel II and Insurers may well suffer the same problems in Solvency II


Solvency II—Could data quality undermine your calculations?

AN EXAXE WHITEPAPER

Solvency II requirements At its simplest, Solvency II is designed to ensure adequate capital reserves in Insurers and Reinsurers in order to meet future obligations to policyholders. This involves an assessment of their current business levels and an extrapolation of future payments to policy holders based on standard actuarial rules. These required amounts then have to be fed into either an approved internal model, or the SolvencyII standard model, in order to derive the capital requirements of the business. Companies also have to ensure that the model is being used in day-to-day business management and decision making, a process referred to in the regulations as the Use Test.

Accuracy of the risk calculations is vital for assessing a company’s solvency position

The primary focus of companies to date has been to develop and test their Pillar I internal model in order to gain FSA approval for the approach.

With the dry-run test of the models planned by the FSA for June-Nov 2010, the FSA are issuing warnings that companies not participating in the dry-run are risking not having their model approved by 31st October 2012. If the model is not approved by then, the Insurers would be forced to use the standard model in calculating Solvency II capital. Projects to cover Pillar II and Pillar III are underway, although generally of a lower priority than the Pillar I phase. This is the good news, as the amount of time required to achieve compliance with Solvency II should not be underestimated. However, companies need to look beyond what is laid out for compliance to check whether the organisation as a whole is ready for Solvency II. Having the prescribed procedures is not sufficient on its own to make a business compliant. Even the best run companies have areas that are not as up-to-date as they would like and it is dangerous to assume that getting the model ready and the new procedures in place will make achieving compliance a straight-forward job when October 2012 comes around. Key questions need to be asked as to the readiness of the organisation. In particular, what are the possible issues that may surface later to cause delays in achieving compliance. The later these are found, the more expensive and risky they will be to resolve. And when one looks at the consequences of non-compliance in the new regulatory order, it is clear that companies need to be very sure that the scope of their Solvency II programme is sufficiently broad.

3

Projects are underway for the 3 pillars of Solvency II, but this may not be sufficient to ensure that the end result accurately reflects the solvency position


Solvency II—Could data quality undermine your calculations?

AN EXAXE WHITEPAPER

Consequences of insufficient reserves In the current economic climate, regulation is only going to get more intense. And regulators, smarting from what is seen as their failure in the banking crisis, are looking to restore their reputations by acting tough and fast. This is an imperative for them as national governments are looking at restructuring and/or replacing regulators with supra-national bodies. We can expect extremely strict interpretation of the Solvency II regulations and immediate enforcement from October 31st, 2012. This means that regulators will be looking for accountability in the Life and Pension firms they regulate, not just at the level of the nominated responsible officer but across the whole board of directors. Insurers will have a difficult time getting their models passed by the regulator and should expect to have to do considerable revision to the models after the dry run in 2010. This will take them very close to the start date, prior to getting approval for their chosen model.

The Regulator Strikes The born-again zeal of regulators across Europe was demonstrated in Ireland. After much criticism of his performance the regulator was replaced by a new non-Irish candidate. Within weeks of taking office, he put Quinn Insurance, one of Ireland’s largest insurers, into administration due to deficiencies in

Therefore, executives, and non-executive directors, need to ensure that everything is being done at all levels of the organisation to comply with the regulations. This will require them to make a thorough examination of the approach being taken to implement Solvency II and to take all reasonable steps to ensure that those responsible have covered all the bases on this issue. These steps include the carrying out and acceptance of the ORSA, (the Own Risk and Solvency Assessment). The ORSA should enable board members to control their businesses and will provide them with an early indicator of the firms solvency position, as the insurer may well breach its economic capital target level before it breaches its regulatory capital requirements.

its claims provisions— a sign of

We do not intend to go into the details of the issues surrounding the creation of the modthings to come! els providing the ORSA in this paper. Instead we wish to focus on other issues that could effect the output of the ORSA. It is essential that the ORSA is accurate as otherwise decisions taken by the board will be badly informed. This would leave the board and company exposed, despite the huge amount of work and testing that has been done on the ORSA. Directors need to start looking beyond the specific area of the pillar projects to see what is being done within the organisation to ensure the detection and resolution of these issues, prior to the full-scale testing of the Solvency II programme. So where better to look for an early sight of these issues than at the banking sector, who have recently undergone a similar experience in trying to achieve compliance with Basel II?

4

After taking a lot of flak for their role in the banking crisis, regulators are looking to restore their reputations by tough enforcement of new regulations, such as Solvency II


Solvency II—Could data quality undermine your calculations?

AN EXAXE WHITEPAPER

Banking experience Not many people have looked beyond the immediate requirements of the Solvency II project to see if there are issues out there that may trip them up. Yet the banking sectors programmes for meeting the Basel II deadlines provide excellent templates for the way Insurers should be tackling Solvency II and the key issues faced by the banks can be usefully studied in order to pick up early on any of those issues that would be relevant to the insurance / reinsurance sector. Beyond the capital adequacy calculations, the banks found that they had a number of problems which caused them to have to spend significant amounts of time and money to rectify in order to achieve compliance. Of these issues, the most important was the fact that the quality of their data was insufficiently reliable to provide what they needed. One of the key questions faced by the banks was “how can you be sure that the data you are inputting into your model is correct?” Without that confidence, the results could only be described as tentative, at best. As a result of this, banks were unable to be sure that they had the basic information needed to be able to assess the capital adequacy ratios their business required. This was a severe blow, which for many came quite late in the day, when they had already felt that the worst of the project was over. To resolve this issue, the banks had no alternative other than to undertake extensive data-cleansing operations, which were carried out so close to the deadline that they were high—risk, expensive projects. Thus, alongside projects to comply with Pillars I, II & III, Insurers need to be looking at their overall data quality to ensure that inputs into the internal models are sufficiently accurate to give them the confidence they need in the model output. Data issues can’t wait as they are critical, they will invalidate the Own Risk Solvency Assessment, and they are time-consuming to resolve. Without resolution, all the work that companies are doing in the dry runs and the further finessing of the models will be utterly wasted. And it is the board who will be held responsible if the capital reserves are inadequate for the level of business risk held by the company. We would recommend that those companies which have not begun a data quality assessment immediately set up a task force to examine this issue. Data issues can be costly and time-consuming to resolve and the earlier the analysis is started, the cheaper the solution will be.

5

Insurers are too engaged in the Pillar I modelling issues to cast an eye over the bank’s Basel II experiences and learn from them


Solvency II—Could data quality undermine your calculations?

AN EXAXE WHITEPAPER

How good is your data? The problem with assessing the quality of your data is the classic problem of the refrigerator light bulb. How can one know that the light goes off when the door of the refrigerator is closed, because as soon as you open the door to check, the light will come on.? Similarly with data, valuations programs can run, but without external verification companies have to work on the assumption that the data being fed in and the calculations being used are correct. This is not sufficient for anyone who has to stand over the official solvency figures produced when questioned by the regulator. The obvious answer is to commission a second suite of programs for performing valuation calculations. With a comparative valuation being carried out, discrepancies are quick to spot and investigate. This will flush out the problems within the data and the valuation calculations and bring a high degree of certainty to the figures being input into the solvency calculation models. A second cross-checking valuations suite can be created by a new internal team who are not currently involved in the valuations process. This independence is necessary to ensure that existing problems in the data are not automatically accommodated and that existing calculations problems are not replicated within the suite, thereby destroying its validity. External validation suites must be truly independent if we are to be sure that cross-checking of data and calculations will be effective. An alternative approach would be to purchase an external valuations system. As the need for certainty about the data is so high and it will be an ongoing process, this could be a more effective solution. This approach would not put a strain on the internal actuarial teams at a busy time and would also be more acceptable to the regulator as an independent verification of the data being input into the risk models. Improving data quality is a key component in demonstrating that the company is endeavouring to fulfil its obligations under the Solvency II programme to retain the correct level of reserves for the risk that it is holding. Independent verification of that data gives a level of comfort to both the company and the regulator that the policyholders’ interests are being kept at the forefront of the company’s decision making process. Demonstrably clean data is an essential component of an accurate solvency calculation. Directors need to be able to satisfy themselves, and the regulators, that all reasonable efforts have been made to produce an accurate reserve calculation.

6

Many companies remain exposed to bad data issues as they have no means of automatically and independently verifying that the data is correct


Solvency II—Could data quality undermine your calculations?

AN EXAXE WHITEPAPER

Concluding Summary Solvency II poses many issues for the Life and Pensions Sector. However, outside the specific requirements laid down by the regulators there remain other issues within the organisation that could impact the programme and hinder the achievement of compliance. Regulators are increasingly keen to come down heavily on companies which endanger their customers financial security by running a business over-exposed to risk. After taking a lot of criticism for their failure to police the banks adequately, regulators in most countries are keen to justify their position by cracking down hard on any insurance firms that they feel are in breach of the Solvency II rules. Looking at the banking sector; one of the greatest issues they faced while working towards Basel II compliance was the quality of their own data. The fact that their data was not of sufficiently reliable accuracy meant that the output from their risk modelling was flawed and that any decision making based on that output was also flawed. Learning from the banks experiences by initiating early data quality assessments will help companies to establish a high confidence level in their data and will enable them to build up an accurate assessment of their solvency position.

Therefore, the banks had to undertake high-risk, last-minute projects to clean up their data in order to gain any confidence level in the inputs to their risk modelling. Once this was done, the banks were in a position to assess their capital adequacy ratio and plan their reserves accordingly. Life Insurers can profit from the experience of the banks by moving to assess and improve their data quality at an earlier stage in the cycle, rather than waiting until it becomes a proven problem. This is a cheaper and better way to resolve the issue and achieve confidence that their company can assess accurately its business risk and can plan to have the required levels of reserves to achieve compliance with Solvency II. Cross-checking of the data used for valuations is key to achieving confidence in your data. From this, all else flows. Accurate cross-checking will require new suites of valuation programmes to cover the entirety of the company’s policy bank. This could either be developed internally by actuaries unconnected with the current system used for valuations, or it could be purchased in from outside. External systems have the benefit of built-in independence and so should automatically deliver unbiased cross-checking of the data and the calculations. This independence would be significant in terms of demonstrating that a company is determined to get the most accurate Solvency measurements by putting the cleanest possible data into the ORSA (Own Risk Solvency Assessment) model. Using independent systems to cross-check valuation calculations is the most reliable way to increase confidence in the quality of the data that forms the bedrock of all required compliance calculations.

7

Insurers need to beware of overfocusing on the calculations to the detriment of the data to be input into the calculations. Remember the old adage—’Garbage in / Garbage out’.


Solvency II—Could data quality undermine your calculations?

Tom Murray

Kathryn Desmond

Tom is Chief Operation Officer of Exaxe with primary responsibility of overseeing product development. Tom has extensive experience of the Life and Pensions industry and has been directing the Exaxe global product set for 11 years

Kathryn is Business Development Manager with Exaxe. She holds responsibility for the marketing of Exaxe products worldwide. Kathryn’s background is in Capital Markets and Life and Pensions Consultancy and she has been with Exaxe for 10 years.

Email: tom.murray@exaxe.com

Email: Kathryn.desmond@exaxe.com

About Exaxe Established in 1997, Exaxe helps Life and Pensions companies launch new products faster, administer post retirement products more efficiently and respond with greater flexibility to the marketplace. With offices in Ireland, the UK and the Netherlands, we provide leading edge; front, middle and back office solutions specifically for life and pensions. Exaxe’s Illustrate Plus provides best of breed illustrations and calculations technology without a major IT systems overhaul. It is a component based solutions which more effectively manages; product development, quotations and illustrations, channel distribution and commissions management, administration, etc. Illustrate Plus is proven and is in use in a wide range of client organisations. “Exaxe” is a registered business name of Exaxe Limited, company registration number 222246 Registered Office: The Concourse Building, Beacon Court, Sandyford, Dublin 18.

8

AN EXAXE WHITEPAPER


Issuu converts static files into: digital portfolios, online yearbooks, online catalogs, digital photo albums and more. Sign up and create your flipbook.