Challenges to discounting liabilities for Solvency II

Quite how difficult a challenge policy makers are facing was made apparent at a high-level roundtable organised by the Centre for European Policy Studies (CEPS) in June, where academics, policy makers and industry representatives came together to debate the future of Solvency II.

The measures

Going into the trialogue negotiations there were three main measures under discussion, all of which discount liabilities. Their names and structure have been slightly altered since QIS5 but most texts refer to them as:

  • A Matching Premium (MP) – reduces the amount of capital that needs to be held for long-term liabilities that are ‘matched’ by a portfolio of assets with similar duration and cash flows. This is to be a permanent measure that can only be applied to specific assets and liabilities that meet a strict set of criteria.
  • A Counter Cyclical Premium (CCP) – reduces the discount rate used to value liabilities in times of excessive market volatility. This temporary measure will apply to all liabilities in the portfolio (other than those benefiting from a matching premium). The CCP will be introduced by EIOPA.
  • An extrapolation of the risk-free interest rate where market data is no longer considered deep, liquid and transparent.

Broadly speaking, the debate on the subject addressed the appropriateness of the principle of discounting liabilities and how to apply the discount.

Solvency II Matching Premium

The proposed Matching Premium will apply only to ‘matched’ assets and liabilities that meet a strict set of criteria. For example, the asset portfolio must be constructed of high quality assets (generally BB rated bonds or above) that are held to maturity and managed separately, or possibly in a ring-fenced portfolio. The liabilities must have no option of surrender at a cost to the insurer, and both assets and liabilities must have highly predictable cash flows. By holding assets to maturity, it is argued, the insurer is not exposed to market volatility (spread risk) because it does not have to sell them until they mature. Olav Jones, Deputy Director General, Insurance Europe, was strongly supportive the idea. “We need to recognise that where companies can hold-to-maturity they are faced with default risk rather than spread risk,” he said. “It is vital that Solvency II measures the actual risks and does not charge capital and create volatility for risks to which the companies are not actually exposed.” Mr Jones also noted that he was not aware of a single academic study that supported the idea that spread risk volatility was linked to default, further supporting the case that default risk and spread risk must be separated. Others questioned the very principle of discounting liabilities and warned that the thinking was flawed. Dr Enrico Perotti of the University of Amsterdam acknowledged that the insurance industry was facing a ‘perfect storm’ in relation to long-term guarantee products. On the one hand interest rates are being kept artificially low by loose credit and monetary easing, while on the other, some asset prices may be depressed because of poor financial conditions. These circumstances were putting pressure on insurers’ ability to honour their existing commitments on long-term guarantee products as well as potentially constraining their ability to offer new products. But Dr Perotti called into question the entire concept of reducing liabilities to adjust for excessive market volatility on the grounds that it was an economically unsound practice. What’s more, “In reducing liabilities there is an implicit assumption that an insurer will be unable to meet its obligation, which clearly is not the intention.” He said that there is not a single academic who will back the argument that discounting liabilities is a sound economic principle. And while agreeing that under certain conditions a matching premium could be applied, he argued that this is a short cut and should be treated as an exception, adding, “To extend this single case to a principle was economic nonsense.”

From Matching Premium to Counter Cyclical Premium

The wider application of discounting liabilities is being explored in the proposals to introduce a Counter Cyclical Premium (CCP) in times of stressed economic conditions. The premium will be applied to the risk-free term structure used in the valuation of liabilities. Currently it is proposed that EIOPA will decide when to apply and withdraw the CCP. Irrespective of the principle of discounting liabilities, there are a number of problems with this proposal. Participants noted that it could be difficult for EIOPA, or any other regulatory agency for that matter, to announce that the markets were in stressed conditions. The political pressure would be enormous. From a technical point of view, it would be difficult to determine exactly what constitutes stressed market conditions and when conditions have returned to normal.

Automatic or judgement based measures?

There was also debate on whether the CCP should be introduced using a set formula or using judgment. The challenge here is to design a tool that offers sufficient predictability but isn’t so predictable that firms can price it into their products. Some expressed reservation about the use of any automatic, formula-based macro-prudential tools, noting that all such tools are backward looking and that only human judgment provides forward looking insight. The argument for using automatic application of the CCP (using a formula or predefined set of triggers) is that it will provide stability and predictability for insurers. Mr Jones, explained that without predictability it would be impossible for firms to use the tool in a meaningful way and it will render the CCP useless. “This lack of predictability will likely force firms to exit these areas of business, because their stress testing will show the massive problems caused by the temporary periods of distressed markets, and unless they can also estimate how the CCP could help, they will have to ignore it,” he said. “We can already see a trend of some firms moving away from guarantee products towards unit-linked and short-term simpler products in anticipation of the Solvency II rules.” However, Gabriel Bernardino, Chairman, EIOPA, said that Solvency II has to deal with a new economic reality for long-term guarantee products. “It is completely unfair to say that Solvency II will kill the long-term products in the life insurance sector. It is not Solvency II, it is the economy. It’s the reality of the long-term yields that are low right now. Even if we didn’t have Solvency II it would still be a challenge.”

Macro-prudential effects of the CCP

The macro-prudential impact of the measures under proposal was also subject to much discussion. Some were cautious about the unwanted effects of the CCP. They noted that it should not become a general regime because that could create concentration risk as insurers seek assets that will benefit most from the regime. While Dr Perotti argued that applying counter cyclical measures to the insurance but not to the banking sector could lead to risk transfer between the two. A further criticism of the CCP was that it lacked symmetry and would only act to reduce liabilities in the bad times. To make the CCP a more efficient macro-prudential tool it should be applied when asset prices were considered too high as well, so insurers could build up capital buffers. This would make it more compatible with banking regulation and would reduce the potential of risk transfer.

Adjusting the asset side of the balance sheet

Challenges to discounting liabilities for Solvency II

An alternative to reducing the value of liabilities to counteract excessive volatility would be to make adjustments to the asset side of the balance sheet. National supervisors could also be allowed some flexibility in determining the amount of regulatory capital required when market consistent valuation was putting a strain on the firm’s capital. Solvency II already contains a Pillar I equity dampener, which is a symmetric adjustment mechanism that can be applied to the capital charge on equities to guard against volatility in equity prices. There are also other mechanisms that can be applied by supervisors if an insurer breaches the higher of the two capital requirements – the Solvency Capital Requirement (SCR). David Revelin, Deputy Head of International Insurance Affairs Division, ACP, told participants, “Solvency II has already built into it a ‘Pillar II dampener’ which allows supervisors to extend their recovery period in case of a breach of the SCR from nine months up to a maximum of possibly thirty months in crisis situations.” In addition, an SCR breach will not have to be immediately disclosed, which will leave some time for companies to solve the situation with their supervisor beforehand. “Companies will only need to disclose a breach of the SCR publicly, either after two months if a realistic recovery plan has not been submitted, or after six months if a realistic recovery plan has been submitted but the SCR is still in breach after that period.”

Politically charged science

There was, however, some consensus on the fact that the proposals on long-term guarantee products are highly political and that they are being championed by different member states to the benefit of their respective insurance markets. Broadly speaking the measures and the countries they are said to serve are:

  • Matching Premium – UK and Spain;
  • Counter Cyclical Premium – France and Italy;
  • Extrapolation – Germany.

The subject remains charged, which is adding to the difficulties in the Omnibus II trialogue negotiations. Yet it was noted that the distinctions between the interests of each of the member states were not so cut and dry. Industries in countries other than those listed above would also benefit from these measures.

Lack of understanding

In general it was agreed (and to an extent evident in the discussion) that there was a lack of understanding of the different measures being proposed. This confusion is in no small part due to a the different terminology and the fact that the issue is relatively new. “This concept was developed in March 2011 and first articulated in the Level 2 text in November,” Mr Jones said. “While Insurance Europe has been pushing for these measures for a year, some parts of the industry only started looking at it in January 2012, and this was properly made public in March, after the vote in the ECON committee.”


The debate is a welcome engagement with some topics that have received little public airing so far. Dr Rym Ayadi, Senior Fellow at CEPS and member of the IRSG of EIOPA, who chaired the roundtable, expressed concerns that the proposed measures could reflect poorly on the entire directive. “Because of the deviation from market consistent valuation, introduced by the Matching Premium and the CCP, Solvency II is under threat of losing its essence, which would certainly weaken its credibility. A prompt revision of these deviations is not too late to bring back the European solvency framework on track.” The proposed assumptions used in the treatment of long-term guarantee products affect not only insurers’ balance sheets today, but also their ability to withstand market volatility over time and meet their commitments to policyholders. The long duration of the contracts means that small changes made now can have a big impact in the future. The contrast between the minutiae of the technical details under discussion and the potential huge adverse effects on the retirement prospects of so many policyholders can hardly be grasped. In fact it is not grasped: that is why this debate is receiving so little attention outside of the Solvency II community.


1 thought on “Challenges to discounting liabilities for Solvency II

Comments are closed.