Interview with Prof. Matthias Müller-Reichart
Solvency II as conditio sine qua non in the insurance industry
RiskNET [Editor-in-chief]14.04.2016, 08:00
After years of preparatory work, a new, uniform supervisory system for the European insurance industry came into existence on 1st January, 2016. Solvency II represents a new quantitative and qualitative supervisory regime with a totally changed framework. It will have a long-term influence on both business models and risk management in the insurance industry. Solvency II harmonizes the rules for 28 EU states with different markets, which previously had specific national supervisory systems. Implementing the new regulations will be a very complex task for companies, which is not made any easier by the historically low interest level at the launch of Solvency II. We spoke to Prof. Matthias Müller-Reichart, Dean of Wiesbaden Business School, about the objectives, challenges and outstanding issues relating to the new regulatory framework.
Let's start by reviewing the original objectives of Solvency II. One of the key primary objectives was to protect policyholders and other stakeholders against insolvency of individual insurance companies. A "prudential framework" was also intended to ensure the effective functioning of the insurance industry in Europe. Thirdly, there was the issue of "levelling the playing field", i.e. identical rules for identical risks. A further focus was to create incentive systems for insurers to introduce new and more effective methods and systems for capital calculation and control. What grade would you give for achievement or non-achievement of these objectives? What do you think are the major problem areas?
Matthias Müller-Reichart: Insolvency protection has been significantly improved by the introduction of Solvency II, particularly the focus on risk management in Pillar 2. By strengthening the governance functions and introducing the obligation to produce a meaningful ORSA report, insurance companies are now obliged to engage in continuous "self-assessment", which increases the transparency of possible existential threats, both internally and externally. In view of multiple conditional risks, such as the sustained low interest situation combined with capital market weakness, consumption dynamics and weak provision promoted by quantitative easing, an increase in administration costs caused by regulation, and a high elderly dependency ratio due to demographic trends, insolvency risks in the insurance industry have risen considerably – and Solvency II with its quantitative and qualitative action parameters has arrived right on time. In this sense, I would give the achievement of insolvency protection objectives an A or B grade. The stated objective of a single European domestic market of course requires a collective roll-out of Solvency II in all 28 EU states to create a "prudential framework", but the illusion of uniform standards in all EU countries cannot be maintained. Since Solvency II is an approach based on principles, national implementation is likely to vary considerably. At best, we can currently talk about an implementation based on the lowest common denominator – in that sense I'd give a grade D for achievement of this lofty objective to date.
As long as a regulatory instrument like Solvency II also has to satisfy economic policy objectives, it will never be able to "level the playing field". Since we started the risk calibration for Solvency II back in 2002 in the Financial Services Action Plan, risk quantification has always been a moving target and it remains so today. The first revision of the Solvency II approach will come in 2018 – by then European government bonds will irrationally be viewed as risk-free, property as high risk (25 percent stress in the standard approach) and shares as explosively risky (up to 59 percent stress for shares outside the EU and OECD with maximum equity adjustment). As there are no "identical risks", Solvency II is unable to respond with "identical rules" – therefore in the pursuit of the primary objective of insolvency protection they have moved away from the original objective of risk adequacy. Overall, I can give achievement of the "level playing field" objective a satisfactory grade.
Ultimately, Solvency II was conceived as a control system to make risk capital allocation more efficient. With the exception of companies that have developed an internal full model, this objective could largely be missed. A European standard approach is not suitable for company control; it is merely an instrument for risk capital calculation in extreme stress scenarios. Insurance companies that choose the European standard model to calculate their SCR therefore need to map their company control systems using an independent model. To that extent, apart from internal full models I would rate the achievement of the objective of Solvency II providing a control instrument as adequate to deficient.
Matthias Müller-Reichart: The development period for Solvency II clearly got out of hand and this may initially be a source of irritation, but it is important not to forget that our work was interrupted by a financial and state debt crisis that began in 2008. It is ultimately thanks to the EIOPA that we finally managed to bring Solvency II to the "go live" state in 2016. If all those academics and professionals with concerns had managed to assert their – certainly justified – objections to these new insurance regulations, we would have had to wait even more years for Solvency II to be implemented. Despite the 15-year development period and the billions in investment needed by the market, taken as a whole I believe that Solvency II's cost/benefit ratio is acceptable as these new regulations represent a paradigm shift and an epochal change in the reality of insurance industry business models. Risk-based regulation is the only way to counter the universe of new business model risks or – as the Chairman of Munich RE, von Bomhard, puts it – the accumulation of risk. In addition to the necessary capital resources, the required changes realized in Pillar 2 and 3 will make the insurance industry's business models more resilient today and in the future.
Matthias Müller-Reichart: Who can make a judgement about the relevance of risks? Solvency II certainly includes all existing risks in the insurance industry business model, but it is inherent in the system that these regulations do not cover new risks emerging. In terms of its view of risk, Solvency II is subject to the problem of the time stability hypothesis, in that only known risks are captured and extrapolated into the future. However, disruptive changes to the business model in the insurance industry (we need only think of the consequences of digitalization, the Internet of Things or Industry 4.0) will bring about new, as yet unknown, risks that can only be incorporated through continuous reform of Solvency II. To that extent, Solvency II remains a moving target, and shares the fate of any risk analysis model in that it has a retarding effect.
The politically motivated low interest environment that has persisted for some years is one of the biggest challenges for many life insurers. We are seeing the revenge of minimum interest guarantees that were promised many years or decades ago. The consistent solvency balance sheet across the market under Solvency II creates transparency here and relentlessly brings to light the actual challenges. Is it not quite simply the case that political intervention in market mechanisms has transformed a macroeconomic and political risk into a microeconomic risk that is threatening the basis of life insurers' business?
Matthias Müller-Reichart: In exaggerated terms, life insurance companies and – as a direct consequence – all pension savers are collateral damage from a well-intentioned but poorly implemented European project. With its quantitative easing programme, the European Central Bank – whose decisions have long since lost all semblance of independence – is attempting to maintain the illusion of European unity, at the expense of savers in the European states with solid economies. By financing states with no macroeconomic justification, we are creating artificial growth and, when combined with low unemployment and wage rises, producing a feel-good climate in Germany – but at the same time we are storing up a huge problem of neglected pension provision. Current pension provision systems are unable to compensate for this threat of precarious living conditions in the future, as the low interest has eliminated the basis of their business. The insurance industry has created its contractual basis over decades, trusting that there would be stability in financial policy – but these assumptions have been unilaterally revised in favour of the European ideal.
Reading through the 2016 Insurance Supervision Act, which implemented the directive 2009/138/EC (Solvency II), it is apparent that legislators have used a very narrow and very German definition of the concept of risk. They do not define risks as potential variations from targets and plans, as is normal practice internationally, in ISO 31000 for example. In actual fact, potential opportunities (upside risk) are explicitly excluded. Is it not true that the biggest risks are a failure to recognize opportunities and sleeping through changes in the market? How is appropriate stochastic risk modelling even possible if I only look at one side of the coin?
Matthias Müller-Reichart: The new 2016 Insurance Supervision Act follows a purely German tradition of equating risks exclusively with hazard (downside risk). This perspective began back in 1998 in the Corporate Sector Supervision and Transparency Act (KonTraG) and was also reflected in the Minimum Requirements for Risk Management (MaRisk) in the financial services industry. Now that the new Insurance Supervision Act is in force, the Federal Financial Supervisory Authority (BaFin) has suspended MaRisk, which means that this hazard-based perspective now underpins the Insurance Supervision Act.
However, for a company an exclusively hazard-based business policy would be disastrous as, in terms of results, ignoring opportunities leads at best to stagnation and economic mediocrity, and in the long-term to a silent and unspectacular market exit. Risk management should weaken a company's weaknesses, but also strengthen its strengths, with a focus on taking opportunities and not on avoiding hazard. Current risk modelling models use a Monte Carlo simulation to analyse the entire distribution of results in the form of hazard (failure to meet a target value) and opportunity (exceeding a target value).
Let's look at the standard formula. The parameters and the structure of the standard formula have been repeatedly adjusted over the years depending on prevailing interests. More than some of the calculation logic and parameters are based on erroneous assumptions and incorrect computational logic (for example, look at the study by Stefan Mittnik from Ludwig-Maximilian University (LMU) in Munich: "Solvency II Calibrations: Where Curiosity Meets Spuriosity"). Overall the standard formula is an extremely complicated construct that only a few experts understand.
Matthias Müller-Reichart: Over the various stages of the QIS studies, the standard method was on the one hand refined and on the other hand politically exploited. All the adjustments to the standard formula have certainly increased the complexity, but the only difference is that we are now mistaken at a higher mathematical level. Our colleague Mittnik did a great job of elucidating the mathematical weaknesses of the standard method, with numerous incorrect assumptions (the assumption of volatility for property risk based on the British property market) rounding off the weaknesses of the data set. However, no shining intellect has yet come up with the magic formula for correct and risk-appropriate identification of all risks that affect an insurance company. As a result, despite all of the justified criticism it is right to start with a highly risk-averse calculation formula that has been evaluated in numerous studies and then to continuously review it.
With a one-year evaluation period and an "optimistic risk measure" value at risk, can we obtain reliable predictions about an insurer's financial stability?
Matthias Müller-Reichart: A definite no – but how do you define the reliability of a prediction? Of course, the incoherent value at risk measure in a single-period perspective is an invitation to make incorrect predictions, but could companies actually implement the uncertainty and complexity of more mathematically correct models? Companies lack the representative number of events necessary for a coherent tail value at risk, and for multi-period analyses we have to go to the trouble of producing cash value comparisons with unknown discounting factors that create further risk. Quite honestly, a nice mathematically stringent modelling world is impossible because of the nature of reality.
Is sufficient emphasis given to the proportionality principle? Or do Solvency II and standard methods follow a "one size fits all" logic?
Matthias Müller-Reichart: Our national supervision will take account of the proportionality principle. However, we need to say goodbye to the idea of linear proportionality (otherwise the ORSA report for an insurance company with gross contribution receipts of 15 million Euro would take up half a page) and move towards a degressive perspective. While Solvency II is based on principles, each individual insurance company has an obligation to claim its own proportionality through clear agreement and consultation with the BaFin. Of course, national supervisory authorities have overall responsibility for interpretation of the Solvency II returns submitted (Pillar 2 and 3) but the insurance companies retain the risk of, and also the opportunity for, interpretation. In discussions with national supervisory authorities, the proportionality principle will develop and will give the regulatory mechanism a level of flexibility appropriate to the market conditions.
Should a capital investment strategy (and a product strategy) be controlled by regulatory constructs? I'm thinking particularly of the politically defined zero risk for government bonds.
Matthias Müller-Reichart: The principles that Solvency II is based on should, in fact must, also apply to capital investment. Regulation of insurance companies is based on trust in maintenance of the ex-ante insurance contributions paid in and capital investment must also be subject to this objective. Under Solvency II, the "prudent person principle" applies to capital investment, which means that the capital investment must reflect security, profitability, liquidity and quality (see § 124 I of the new Insurance Supervision Act) – these are exactly the attributes that must be the guidelines for asset management by insurance companies.
Does a politically defined zero risk for government bonds – and therefore alleviation of state debt at the expense of the security of insurance company claims – contradict the consumer protection objective mentioned previously?
Matthias Müller-Reichart: The politically motivated zero risk assumption for European government bonds is intended to make the insurance industry the financier for European states. If we were to now assume that these government bonds have a higher default risk, your argument would be correct. But if we consider the fact that the ECB prevents European states from being insolvent, making the no-bail out clause an absurdity, increased involvement in European government bonds primarily means a deterioration in average returns in insurance companies' asset management. This predominantly affects shareholders' profitability objectives and only results in non-payment of insurance company claims to a limited extent. To this extent, although the zero risk assumption for European government bonds leads to a huge returns problem in terms of the non-insurance result, it does not necessarily bring about an increase in the probability of insurance company insolvency. I believe that consumer protection is provided, while there is pressure on insurance product and share certificate returns.
Is the relevance of intangible assets – such as the value of a brand – sufficiently covered in Solvency II?
Matthias Müller-Reichart: Intangible assets remain largely excluded from the solvency balance sheet – for example derivative goodwill may not be recorded at all. This perspective is due to the extreme risk aversion pursued by Solvency II – any value that can take on dimensions that cannot be replicated in the market tends to be excluded. Thus, a solvency balance sheet under Solvency II is characterised by a high degree of risk aversion and intangible assets will tend to be ignored.
The delegated act of 17th January 2015 ran to around 800 pages. The Insurance Supervision Act has added further regulation. How do you assess the risk of excessive complexity and bureaucratisation of regulation and supervision?
Matthias Müller-Reichart: Legislators are using excessive complexity and bureaucracy to try to avert any cases of liability. In that sense, these elements are protective mechanisms put in place by the regulator – of course to the detriment of comprehensibility and therefore practical acceptance. A regulatory system cannot be represented on three pages like a key information document under MiFID II, but the current volume goes way beyond the limits of human attention and receptiveness. The consequence of this complexity is a need to concentrate on key information and thus the risk of being unable to utilise marginal or secondary information for the purpose of economic optimization. At this point, we should praise the massive support provided by the German Insurance Association (GDV), which consistently summarizes Solvency II information in an appropriate form for the industry and target group.
Matthias Müller-Reichart: I believe that the real benefit of Solvency II is in Pillar 2 – the obligation to put in place high-quality risk management supported by various governance functions. By implementing this pillar, insurance companies' risk and corporate culture will have to adapt to the regulations. Thus, not only do the insurance company's key functions have to be "fit and proper", the entire corporate culture has to demonstrate its fitness when it comes to risk management. The introduction of Solvency II has been worthwhile for this regulatory quantum leap alone.
How do you assess the risk of regulators' fear of their own responsibility – what is known as "regulatory forbearance"?
Matthias Müller-Reichart: As I have already alluded to, when the decision to implement Solvency II was taken this regulatory forbearance hung over the development process for the new regulations like a "deus ex machina". A regulatory system must be liability proof and therefore virtually perfect – a demand that regulators are unable to satisfy due to the emergence of new risks. Therefore, they develop a system in which compliance to the best of one's knowledge and belief does not allow any inference of liability.
What do you think are currently the biggest practical issues with the implementation of Solvency II?
Matthias Müller-Reichart: Solvency II is a big issue for insurance companies – but there is an ability and willingness to learn. I believe that German insurance companies will have few problems meeting the solvency capital requirements against the backdrop of a 16-year transitional period and the use of volatility adjustments along with accrual and interest transitionals. Production of the ORSA and establishment of the government functions should be no problem for the companies either. I can see a real issue in compliance with the reporting system under Pillar 3. Here companies still make insufficient use of the possible synergy effects of BaFin, EIOPA and ECB reporting, and this still has an overwhelming impact on reporting obligations for most insurance companies. When, in addition to annual Solvency Financial Condition Reports (SFCRs), Regular Supervisory Reports (RSR), ORSA reports and ECB reports, the quarterly Quantitative Reporting Templates (QRTs) and the various quarterly balance sheets (trade balance sheet in compliance with BiIMoG and BilREG, tax balance sheet, solvency balance sheet, IAS/IFRS balance sheet, US GAAP balance sheet) have to be produced, there is little remaining scope for underwriting and policy acquisition.
[The interview was conducted by Frank Romeike / editor in chief of RISK MANAGER and FIRM board member]
Prof. Matthias Müller-Reichart is the Dean of the Wiesbaden Business School (Economics department), leader of the Insurance and Financial Management course and holder of the chair in risk management at the Hochschule RheinMain in Wiesbaden. After studying business management and Catholic theology, he worked as an assistant director and head of corporate planning for the German Generali insurance group. Completing his doctorate, teaching at the University of Munich and starting his post-doctoral lecture qualifications motivated him to move into research and teaching, and he was appointed at the Hochschule RheinMain in 2001. In over 115 publications and as the leader or lecturer at numerous conferences, Matthias Müller-Reichart has addressed a wide range of interdisciplinary areas of risk research. In addition to his academic career, he is a consultant to various international direct and re-insurance companies and is the owner of a retail business that has belonged to his family for over 100 years.
Kommentare zu diesem Beitrag
Future Outlook and Emerging Risks
RiskNET [Editor-in-chief]07.06.2018, 10:47
Artificial Intelligence (AI) refers to the ability of a computer program to think and learn like a human and its snowballing impact on society is undeniable. AI applications already pervade many...
Tensions and Turbulence Ahead
RiskNET [Editor-in-chief]14.05.2018, 09:59
In 2017, the global political risk landscape was turbulent, and is likely to remain so over the next 12 months. The Political Risk Map 2018, published by Marsh and based on BMI Research findings,...
Decision making in risk management
RiskNET [Editor-in-chief]29.03.2018, 14:10
Our world is in a state of upheaval – whether due to war and terrorism, the increasing number of environmental disasters or because of cyber risks and economic distortions. The risk map is growing as...
Compliance index model
RiskNET [Editor-in-chief]23.03.2018, 13:54
Compliance measures result in costs but do they also have a benefit? It's difficult to assess the success of measures. It tends to arise – if at all – after a delay and the contribution made to...