THE TREASURY COMMITTEE INQUIRY INTO EU INSURANCE REGULATION
WHAT HAVE WE LEARNED SO FAR?
By Timothy Ellis and Steven McEwan
In September 2016, the Treasury Committee of the House of Commons (the Committee) launched an inquiry into EU insurance regulation, to supplement its work on the relationship that the UK might seek to have with the EU following Brexit.
The inquiry has four main aims:
to consider the options for the UK insurance industry that are created by the decision to leave the EU;
to assess any impact of Solvency II on the competitiveness of the UK insurance industry;
to examine the impact of Solvency II on the role of insurance in meeting the needs of UK customers and the wider UK business economy; and
to assess any learning for regulators and industry from the introduction of Solvency II.
Since the end of 2016 the Committee has received submissions of written evidence from a variety of market participants and interested parties. During January and February of 2017, it has also held three sessions with market participants to allow for questions to be asked by the Committee.
Most recently, on 22 February 2017, the Committee heard evidence from:
Sam Woods (Deputy Governor for Prudential Regulation and Chief Executive Officer, Prudential Regulation Authority (PRA));
Victoria Saporta (Executive Director, Prudential Policy, Bank of England); and
David Belsham (External Member of the PRA board),
to seek the views of the PRA on Solvency II (the PRA Evidence).
Subsequently, the Committee also published on 15 March 2017 a letter from Sam Woods to the Committee dated 3 March 2017 in which Sam Woods responded to additional points that the PRA had been asked to clarify in writing (the SW Letter).
In written evidence to the Committee the Association of British Insurers (the ABI) identified a number of areas where they believe the PRA has restrictively interpreted Solvency II. In the SW Letter, the PRA responded to these points and noted that there were some areas where they were open to reform and prepared to work with the industry to try and amend how Solvency II works, in particular the internal model amendment process; recalculation of the transitional measure on technical provisions (TMTP); external audit of the Solvency Financial Condition Report; notification of longevity risk transfers and the use of legal entity identifiers. The SW Letter also responded to some of the ABI's other concerns and identified those where the PRA did not agree with the ABI's categorisation of the issues.
Given that one of the key themes that has emerged from the written evidence and oral testimony before the Committee was how insurers view the way in which the PRA have interpreted and applied Solvency II in the UK, the PRA Evidence and the SW Letter were an opportunity for the PRA to provide its response to these comments. It is quite rare that we get to see in the open the
dialogue between a financial regulator and market participants so it has been interesting to see these exchanges as part of the inquiry.
What is clear from the PRA's responses to the Committee is that the PRA are aware of a number of shortfalls in how Solvency II operates, and they are not ambivalent to the impact that these issues have on firms, but the PRA also sees limitations on what it can do to help.
In light of PRA's responses, we thought it would be good opportunity to review some of the issues that the inquiry has considered to date and to consider what these issues might mean for Solvency II and insurance regulation in the UK in the future.
THE RISK MARGIN WHAT IS IT AND WHAT ISSUES HAVE UK INSURERS FACED UNDER SOLVENCY II?
Solvency II requires that insurers calculate and hold assets to cover a risk margin, in addition to a best estimate calculation of their insurance liabilities. The risk margin is one of the most common issues the Committee has received evidence on to date.
The theoretical purpose of the risk margin is that it is an amount that, when added to the insurer's best estimate calculation of its insurance liabilities, means the resulting technical provisions are equivalent to the amount the insurer would have to pay to another insurer (a Transferee) to take over and meet the relevant insurance liabilities. The risk margin therefore essentially acts as the premium that would have to be paid to dispose of the insurer's insurance business.
The risk margin is calculated by determining the cost of providing an amount of eligible own funds equal to the Solvency Capital Requirement (SCR) that would be necessary to support the insurance liabilities transferred over their lifetime.
The rate that is used in calculating the cost of providing the eligible own funds is called the Costof-Capital rate and this is specified in the Solvency II Delegated Regulation, being 6%. This Costof-Capital rate is the same for all EU insurers.
The risk margin calculation is done for each future year over the expected life of the insurer's insurance business, and then a present value is calculated using the risk-free interest rate curve.
While this approach is theoretically sound, in practice it has led to serious concerns for insurers as small changes in the risk-free interest rate can cause large changes in the overall risk margin due to the period of time over which it is calculated and the multiple ways in which interest rates affect the calculation. We consider this issue in more detail below.
In the evidence that has been submitted to the Committee a number of issues have been identified with the risk margin. We highlight two of these issues:
1. Longevity risk is not considered to be a hedgeable risk as part of the risk margin calculation
The risk margin calculation and the associated assumptions that form part of that calculation do not recognise the ability of an insurer to enter into longevity reinsurance to hedge some of its longevity risk, prior to any transfer of the insurance liabilities to a Transferee. If this were permitted, this would help UK insurers who provide long-term insurance products such as annuities, which are exposed to longevity risk. Longevity risk is the risk that the insurer's policyholders will live longer than expected.
This is an issue that was raised by a number of the respondents to the Committee and addressed in detail in the evidence given by the ABI.
The ABI and a UK insurer have been in discussions with the PRA since the end of 2016 to try and encourage them to address this issue. Written evidence submitted to the Committee shows that two suggestions have been put forward to the PRA as to how this issue could be addressed:
the Solvency II Delegated Regulation could be amended to permit longevity risk to be a hedgeable risk in the calculation of the risk margin (the Amendment Solution); or
in the calculation of the risk margin, insurers could be permitted to make an assumption in relation to the management actions they could take to reduce longevity risk. This assumption would be that the board of the insurer would seek longevity reinsurance to cover its liabilities in certain defined future circumstances (the Management Action Solution).
The Amendment Solution is not within the gift of the PRA, as this would require the EU Commission to amend the Solvency II Delegated Regulation, but it has been suggested by the ABI that the PRA could permit the Management Action Solution without the need for the PRA to obtain approval from the EU or EIOPA.
On behalf of the PRA, Sam Woods stated that while they was sympathetic to the issue, the PRA did not feel comfortable with the Management Action Solution because there is no guarantee that firms would not use this approach in other aspects of Solvency II (for example, to reduce their capital requirement) once it had been permitted in the calculation of the risk margin. Further, Sam Woods considers that if the PRA introduced a local solution in the UK, then this may undermine the ability to amend the risk margin at a European level, so that this matter is amended appropriately across Europe. Sam Woods was also keen to point out that it is in the interests of the UK to try and solve this problem at a European level, due to the importance of Solvency II as a piece of global insurance regulation.
In light of the response from Sam Woods in the PRA Evidence and the SW Letter it seems unlikely that the PRA will permit firms to address longevity risk in the risk margin calculation by using the Management Action Solution. As a result, firms will have to hope that this issue is addressed by EIOPA as part of its review of Solvency II. EIOPA is due to produce its findings to the EU Commission towards the end of 2017.
Whilst this is a matter of insurance regulation that may be thought only to impact life insurers, the issues surrounding the risk margin can be seen as having a direct impact on consumers also. The impact of the risk margin (and in particular longevity risk within the calculation) is one of the reasons that many UK life insurers have considered whether they ought to withdraw from the annuity market in the UK. If this were to happen competition in the UK annuity market would surely be impacted.
2. The risk margin is overly sensitive to changes in interest rates and the Cost-ofCapital rate is very high at 6%
A number of the submissions to the Committee have highlighted the sensitivity of the risk margin calculation to changes in interest rates.
One UK insurer focused on pension insurance buy-ins and buy-outs for defined benefit pension schemes, estimated in its written evidence to the Committee that a 1 basis point movement in gilt yields can cause a 45 basis point movement in the risk margin, due to the various levels of interest rate sensitivity built into the calculation.
The PRA is aware of this issue. David Rule, Executive Director of Insurance Supervision at the PRA said the following in a recent speech to the ABI:
"The interest rate sensitivity arises because lower interest rates both reduce the discount rate applied to the future capital requirements and lead to higher future capital requirements. UK life firms with long-dated annuity liabilities are more affected than most other EU insurers. This is primarily a UK issue."
Both David Rule, in his speech to the ABI, and Sam Woods before the Committee have stated that to some extent, they consider that UK insurers are protected from interest rate sensitivity by the use of the TMTP.
In respect of insurance business that was in force prior to 1 January 2016 (Pre-2016 Business), an approval from the PRA to use the TMTP allows firms to apply a deduction in the calculation of their technical provisions.
The TMTP deduction corresponds to the difference between:
(a) the technical provisions of the firm (taking account of the benefit of risk-mitigation techniques such as reinsurance) in respect of Pre-2016 Business calculated in accordance with Solvency II; and
(b) the technical provisions of the firm (taking account of the benefit of risk-mitigation techniques) in respect of Pre-2016 Business calculated in accordance with the solvency regime that applied to firms prior to Solvency II coming into effect, which in the case of UK firms would be the solvency regime that previously applied under INSPRU and GENPRU.
Solvency II requires that the TMTP to a firm's technical provisions has to "taper" in a linear manner so that the deduction is 100% during 2016 (the first year of Solvency II) and will reduce to 0% during 2032.
The TMTP was included in Solvency II as a means of smoothing the introduction of the changes to liability valuation brought about by Solvency II, in particular the introduction of the risk margin.
A firm which can use the TMTP is generally able to remove the impact of the introduction of the risk margin on its Pre-2016 Business. It is for this reason that the PRA sees the TMTP as assisting insurers in coping with issues surrounding the calculation of the risk margin. As noted above however, the TMTP only applies to Pre-2016 Business and the benefit of the TMTP has to taper off in each year to 2032, so as a mitigant to the issues that insurers are experiencing with the risk margin, the TMTP can only do so much to alleviate the impact of issues.
Further, given current interest rates, the 6% rate for the Cost-of-Capital is thought to be high. This is likely to be one of the areas that EIOPA will focus on as part of its review of Solvency II, as the
Cost-of-Capital rate was set at a time when market conditions were quite different to how they are today.
As a result of the PRA's responses to these points, it seems likely that firms will have to wait until these issues with the risk margin are addressed at a European wide level, which in reality is not likely to be before the end of 2018.
THE MATCHING ADJUSTMENT HAS THE PRA INTERPRETED SOLVENCY II RESTRICTIVELY?
Under Solvency II firms are able to obtain a benefit in calculating their best estimate insurance liabilities where they hold assets with similar cashflow characteristics to these liabilities. This is thought to be appropriate because where a firm holds long-dated assets to maturity it is not exposed to the risk of changing spreads on those assets. As a result, firms are permitted, subject to approval from the PRA, to apply a matching adjustment (MA) in the calculation of the best estimate portion of their technical provisions.
The MA allows the cashflows associated with a firm's insurance liabilities to be discounted at a rate in excess of the risk-free rate interest rate by taking account the yield on the matching assets. The MA can result in a significantly lower total value of insurance liabilities for the firm.
One of the issues that firms have encountered with the MA is that it requires the cashflows of qualifying assets to be "fixed", not merely predictable to a high degree.
This means that if firms want to include assets that do not have fixed cashflows (such as equity release mortgages (ERMs), callable bonds or floating rate assets) in their MA portfolios they have to restructure these assets so that they produce a fixed cashflow. This has led to firms creating internal securitisations so that they can structure the assets with variable cashflows in a manner so as to make them eligible for the MA.
In evidence submitted to the Committee, one UK insurer estimated it spends in excess of 3 million per year in maintaining their MA structures and it also spent over 3 million setting up a solution to make ERMs eligible for the MA.
As a result of this cost, and the added complexity that comes with having to manage an internal securitisation and other matters to ensure compliance with the MA, a number of market participants have suggested that the PRA should adopt a more principles-based approach to the MA.
A principles-based approach might reflect the fact that if a firm holds a large amount of ERM assets, then whilst each ERM on its own may not be sufficiently fixed as to amount and timing of its cashflow, it might be possible to say across the total portfolio of ERM assets held by a firm they would in aggregate to produce a "fixed" cashflow. An appropriate "haircut" could be applied in order to minimise any uncertainty.
Sam Woods was quite clear as part of the PRA Evidence that his view was that the PRA have a duty to faithfully interpret the words and intention of Solvency II (i.e. cashflows have to be "fixed", not "almost certain") and he believes that the PRA have done this in reviewing applications to use the MA.
Sam Woods also emphasised that he considers the PRA has taken a "reasonably flexible" view of how to address the concept of fixity of cashflows. By way of example he highlighted the fact that the PRA have allowed firms to pair floating rate assets with derivatives to allow the requirement for fixity to be met. The Solvency II Directive and Delegated Regulation do not mention the ability to meet the fixity requirement in this way (and nor do they refer to the
possibility of restructuring assets such as ERMs so as to make them eligible for the MA) so in this respect the PRA can be seen to be wanting to be helpful to UK insurers.
We would also note that the PRA have been helpful to UK insurers using the MA in other respects too. For example, they have allowed for MA securitisations to be classified as "intraentity" transactions so that the benefit of the MA is not lost at group level when calculating the group capital requirement. As a result these interpretations, it is not hard to see why the PRA thinks they have been flexible in how they have applied the MA.
In relation to what insurers might be able to expect from the PRA post-Brexit, Sam Woods said the following before the Committee:
"On the matching adjustment specifically, if I had complete control of it, I would probably do something slightly different. I would want to take account of unexpected defaults and I would probably be a tiny bit more flexible around this notion of "fixed"."
Whilst this is unlikely to give much comfort to insurers (especially given the uncertain nature of what insurance regulation following Brexit might look like) this response does at least show that, if the MA or a concept like it is to be retained following Brexit, there may be more flexibility from the PRA when it isn't so concerned with how its interpretation of a piece of EU legislation will be considered by EIOPA and the EU Commission.
In the SW Letter, Sam Woods also highlighted that whilst compared to the liquidity premium under ICAS the eligibility criteria are more restrictive, the benefit that firms can obtain from the MA is greater than that which was permitted under ICAS. As a result of this, it seems that the PRA are of the view that more rigid eligibility criteria are not necessarily inappropriate due to the potential increased benefit that firms can derive from the MA as compared to the liquidity premium.
LONGEVITY REINSURANCE TRANSACTIONS IS APPROVAL FROM THE PRA REQUIRED?
At the beginning of 2016, the PRA sent a letter to UK insurers addressing the longevity reinsurance market (see our previous coverage of this letter here).
As part of this letter, the PRA expressed a concern that UK insurers were using longevity reinsurance for reasons other than genuine risk transfer and reminded firms to consider counterparty default risk and concentration risk as part of any longevity reinsurance transaction. In addition, the PRA said that it expected to be notified of any proposed longevity reinsurance transaction as soon as possible.
At the end of 2016 this letter was followed up with a supervisory statement which addressed the same concerns.
The ABI noted to the Committee that the PRA's request to be notified of any longevity reinsurance amounts to a requirement for the PRA to approve these transactions. The ABI consider that any such approval goes beyond the requirements of Solvency II and adds unnecessary delay and complications to undertaking these sorts of transactions.
In the PRA Evidence, Sam Woods stressed that the PRA has not sought to introduce a requirement for it to approve longevity reinsurance transactions, but the PRA does want to know what transactions are being undertaken in the market and for this reason has introduced the requirement to be notified.
We understand that some insurers have been required to provide substantial information about proposed transactions as part of the notification to the PRA. As such, regardless of whether this is an "approval" or a "notification", this additional requirement (which is not contemplated by
Solvency II) is clearly another matter that insurers (and reinsurers) have to be mindful of when entering into longevity reinsurance transactions.
In the SW Letter, Sam Woods clarified this position further and stressed that the PRA has introduced this notification requirement so as to monitor the build-up of concentration risk. Sam Woods also adopted a more conciliatory tone and stated that the PRA is happy to work with the industry so as to reduce any undue burden. He revealed that the PRA intends to conduct a review of firms' approaches counterparty risk management and once this is completed the PRA may review its current approach to these transactions.
The inquiry has provided a useful opportunity for market participants to come together to consider what the UK insurance landscape might look like follow Brexit. What is clear from the evidence submitted to date is that there is not a significant appetite to abandon Solvency II and start again with a new insurance regime in the UK, especially considering the cost and effort firms put into complying with Solvency II.
It has also been interesting to see that a number of insurers have suggested that they would like to see more flexibility and support from the PRA in how certain aspects of Solvency II have been implemented. This issue, that Solvency II has to be supervised at a national level by regulators who may be more or less willing to help firms (where they have discretion), is always one of the potential issues with EEA-wide harmonising legislation.
What the inquiry does potentially suggest though is that the PRA may be willing (depending on what form insurance regulation takes following Brexit) to be slightly more flexible in its approach to regulation in the future than it considers it is currently permitted to be. However, we would remind insurers who are hoping for greater PRA flexibility that to the extent that the UK wants to remain equivalent for the purposes of Solvency II, there will still be limits to how far the PRA can go to be flexible and still ensure that the UK remains an equivalent jurisdiction.
Insurance Partner, London T +44 20 7296 2972 firstname.lastname@example.org
Insurance Associate, London T +44 20 7296 5723 email@example.com