Good afternoon, and thank you for inviting me to speak today.
A year ago, my colleague Anna Sweeney spoke at this conference about the important role the insurance sector plays in serving the real economyfootnote . At the time, the immediate issues around the Covid-19 pandemic were obviously still front and centre. Anna talked about the important role the sector can play in helping the economy pool and transfer risk, and avoid protection gaps. We had just had another demonstration of the benefits of having a sufficiently strong and resilient insurance sector, able to absorb unexpected shocks and help policyholders do likewise.
Today I would like to talk about one of our current insurance priorities, which is our work – in close collaboration with HM Treasury – on the Solvency II Review (‘the review’). This is an important opportunity for us to deal with some key areas of Solvency II which we know can be improved, while maintaining a regime which is robust in protecting policyholders, is well-respected internationally and which befits the UK’s position as one of the leading international insurance centres. A regime which allows insurers to continue to play the key role Anna described – underpinning sustainable growth, financing long-term investment, and helping policyholders deal with the unexpected.
The Government’s Call for Evidencefootnote  for the review launched last October, and set out three objectives. These are:
- to spur a vibrant, innovative, and internationally competitive insurance sector;
- to protect policyholders and ensure the safety and soundness of firms; and
- to support insurance firms to provide long-term capital to underpin growth.
I want to focus today on how we see these three objectives as being mutually supportive, and why we think there is an opportunity to craft a package of reforms which will advance all three of them together. We are still working with the Government to develop specific proposals for discussion and consultation, but I’ll illustrate with some examples how we think such a package could be achieved. And as we construct a set of reforms, I will also highlight the importance of a comprehensive assessment of its costs and benefits.
Progress on the Review
First let me give a short progress update. The Call for Evidence prompted many thoughtful submissions from industry, the actuarial profession, and other stakeholders. We have worked closely with HM Treasury to consider these responses, which were summarised in the government’s response in Julyfootnote .
Many of the themes aligned with the Prudential Regulation Authority’s (PRA) own priorities for reform – such as the risk margin, matching adjustment (‘MA’), improving scope for judgement and proportionality, and streamlining processes and reporting.
The PRA has since launched a further two-part information-gathering exercisefootnote . The first request in July was a ‘Quantitative Impact Study’ (‘QIS’); the second, in August, a qualitative questionnaire to collect more free-form responses. This exercise will give us valuable information to help us and the government assess potential reforms. It will help us consider a wide range of policy options and allow us to carry out a careful analysis of their likely impact and effectiveness. And from these options a reform package will be chosen and consulted upon. The QIS exercise is thorough – and it needs to be – and we are grateful to those who have been spending their summer getting ready to respond.
When regulators run impact exercises such as this, we inevitably have to select particular scenarios against which we ask firms for data. It is perhaps natural that people start to infer that these represent particular policy choices that we have already made, or that the scenarios represent bookends for options we are considering. This is not the case here. The QIS will give us the data we will need to assess a wide range of policy options, beyond the specific scenarios tested and in different combinations, which we can then narrow down for consultation.
The objectives of the Solvency II Review are complementary
In looking at the three objectives for the review, it would be easy to assume that they are in conflict with each other. For example, some might argue that increasing innovation and competitiveness, or boosting long-term investment, can only be delivered through cuts to capital requirements. In other words, delivering one objective requires compromising on another.
We don’t take that view. Indeed, trading off the objectives could be actively counterproductive. Policyholder protection and maintaining safety and soundness of firms are core to the statutory objectives the PRA has been given by Parliament. But I would argue they also provide the necessary foundations to build sustainable solutions to achieve the other review objectives. Materially reducing safety and soundness, for example through significant cuts in capital requirements for particular assets in an attempt to target the investment objective, could lead to more insurance failures which would damage policyholders and damage confidence in UK insurers and our market. Ultimately a less stable industry would be less well equipped to support investment in the real economy over the longer term.
Rather, we should consider potential reforms in the round. We need to understand how different reforms might interact, and choose a set of options which best complement each other and together achieve all three objectives.
We will need the QIS data to help us assess these options in detail, before we can reach a settled view. We do not yet have that. But for me, there are some clear outlines emerging about how a set of reforms might deliver a more streamlined and efficient regime, support the sector’s ability to provide long-term productive and green finance, and preserve safety and soundness and policyholder protection. I will illustrate this with some examples, taking each objective in turn.
A vibrant, innovative and competitive sector
Solvency II was originally designed to operate uniformly across 28 different EU Member States. This led to a large number of prescriptive rules and processes, designed to promote a common approach. But they are too onerous. Compliance costs, for example around internal models, the MA, and reporting, are unnecessarily high as a result. And because many of these requirements were hard-coded in law, the PRA has to date had very little power to deviate from them.
We now have the opportunity to look again at some of these areas to streamline the regime, making it fit better with the UK market and allow us to apply the sort of judgement we use elsewhere in our supervision. This aim was supported in the Call for Evidence responses. The government has also been looking at this issue more broadly through its Future Regulatory Framework review.
To take one example – we think there is scope to streamline processes around firms’ use of the MA.
It is right that there are robust controls and supervisory monitoring around the use of the MA. As we have highlighted on many previous occasions since Solvency II was introduced, the MA confers a significant benefit to firms, allowing upfront booking of profits that would otherwise only be recognised over time as they were earned. These profits can then be invested for further growth, or distributed to shareholders. It plays a valuable role by insulating the sector from temporary market volatility. But if cashflows are not well-matched, or if there is insufficient allowance for the risk in the assets that insurers retain, there is a real risk of too much profit being recognised upfront. That is a very important issue for us to monitor.
But some of the processes currently baked into the MA regime are overly complex and restrictive. For example, at the moment to invest in any new asset class, as well as firms having to demonstrate cashflow fixity, issues around valuation and internal rating methodologies are also often reviewed. This provides useful supervisory information, but may impede the ability of firms to move quickly in the market and invest dynamically. We think there is scope to make significant improvement to the upfront MA approval process to address some of these issues and help firms move more quickly, without compromising on the core MA requirements. But to ensure policyholders are still adequately protected, we do need to be confident on the level of MA benefit firms are able to claim. I’ll talk more about our views on that later.
There are a number of other areas too where we think processes can be streamlined. For example, we have already consulted on some initial ‘quick wins’ to cut back some reporting requirements, and we plan to consult on further changes after further industry engagement. We also think there is scope to simplify the approval regime for internal models, again without compromising standards. We also plan to look at unnecessary requirements on branches of international insurers operating in the UK. And for new insurers, we are considering how to apply a more proportionate approach in firms’ early stages of growth to reduce barriers to entry.
Making the regime more administratively straightforward, and taking steps to improve market access, will help reduce costs and support the government’s objective to spur a vibrant, innovative, and internationally competitive insurance sector, and one which is able to respond more quickly to opportunities. The examples I have listed here illustrate how we can make a number of improvements in this area, without undermining policyholder protection.
Provision of long-term capital
Both the Government and the Bank of England (Bank) are keen to see how the financial sector can support increased investment in long-term productive assets and help support an orderly transition to a net-zero economy.
It’s worth noting that Solvency II in itself has relatively few hard constraints to investment. Indeed since it was introduced in 2016, both the size and the investment profile of the sector has changed substantially, showing how insurers are already contributing to this objective. In 2016, insurers held £257bn of assets in their MA portfolios; that figure has now grown to £322bn. When the Solvency II rules were being drafted, 85% of assets backing annuities were corporate or government bonds; this proportion has now reduced to 55%, as insurers have invested in a wider variety of asset classes. Last year, infrastructure was already the third most heavily invested asset class in firms’ MA portfoliosfootnote . Annuity writers’ holdings in assets we have identified as illiquid – such as infrastructure and green assets – have increased from £88bn in 2016 to £127bn this year. A diverse range of productive assets including agricultural mortgages, education loans, social housing, infrastructure assets such as ports, airports and hospitals, and more, can all be found in insurers’ MA portfolios today.
Nevertheless, the Call for Evidence responses highlighted that there are some features of the current MA regime, which make deployment of capital into less liquid asset classes more challenging. And which can incentivise firms to choose other asset classes instead.
I’ve already talked about how we think some of the MA processes can be streamlined to help facilitate new investment. There are two other further areas I’d highlight here.
First, we believe there is scope to make some targeted changes to widen MA asset eligibility criteria, again without undermining the core protections needed within the MA regime.
It is a fundamental principle of the MA that asset cashflows be certain and closely matched to liabilities. This ensures firms are genuinely free from market risks around early liquidation of assets, or reinvestment of maturity proceeds, and allows them to safely take upfront the benefit of the market compensation for those risks. But we recognise there are assets which are not currently eligible for the MA, or which are only eligible on unattractive terms, where we think the current treatment is too restrictive or penal and where some sensible changes could be made to address this. For example, assets where the issuer has optionality over the redemption date but pays inadequate compensation on early repayment. Firms have highlighted to us examples of some green projects where changes like this could help remove a barrier to investment. We are also carefully considering the submissions to the Call for Evidence which referenced areas where liability eligibility could be widened.
Second, we need to look carefully at the relative investment incentives within the MA regime. Empirical evidence suggests the regime – as currently structured – incentivises firms to invest in assets secured by existing property. Such assets can obviously play an important function, but may not support financing for new infrastructure or sustainable growth in the economy. Assets secured on existing property have grown in the MA portfolio since 2016, notwithstanding the increased investments in infrastructure assets I mentioned earlier. We would like to understand in more detail the relative incentives in the current regime between different asset types, and under potential policy reforms. We have asked questions in the QIS exercise to help shed more light on this. If the regime currently tilts incentives in a way which does not reflect the underlying risks that the assets carry, we would want to look at reforms to address this.
The purpose of these sorts of targeted changes to the MA would be to address any unnecessary constraints on firms’ ability to invest more widely. We want to ensure the regime is appropriately risk-based but also understand where it might disincentivise investments in assets which can help underpin sustainable growth in the economy. In doing so, we can further the review objective around long-term productive investment, but also ensure that the regime is safe and sound, and policyholders are protected.
Of course, the principal purpose of insurance regulation domestically and internationally is to ensure a safe and sound system, and in doing so contribute to the securing of an appropriate degree of protection for those who are, or may become policyholders. In our view, this is a necessary foundation to deliver the other review objectives. Policyholders need to have confidence in the long-term safety and soundness of the sector, particularly when they are dependent on firms in the long-term to provide significant parts of their long-term financial security such as their retirement income. And without the security of this long-term funding from policyholders, insurers cannot confidently invest productively and sustainably in the economy.
Let me focus briefly on two important areas of reform. First, we know that the Risk Margin is too volatile, and in the current environment it is too high. Partially in response to this, many firms have been incentivised to increase use of techniques such as offshore reinsurance to reduce the size of the risk margin and stabilise their balance sheet. This has resulted in a transfer of assets outside of the UK, an increase in counterparty credit exposure, and a rebalancing of UK annuity writers’ profiles towards credit risk and away from longevity risk. Stabilising the risk margin, and reducing it in current market conditions, can reduce these incentives and would be good for multiple objectives including policyholder protection. We’re collecting data in the QIS to allow us to consider different ways we could implement this in practice.
But also, as I mentioned earlier, we need to remain confident that the MA benefit available to firms is calibrated appropriately. The integrity of its value is critical to insurers’ balance sheets, policyholder protection and the investment choices that firms make. We see an appropriate calibration of the MA as a necessary foundation for reforms which widen the range of eligible assets to support the provision of long-term capital.
Over the last year, the PRA has highlighted publicly on a number of occasions the risk that the MA specification may not have kept up with the changing nature of the marketfootnote . The MA should only include the component of asset spreads that reflects compensation for risks to which firms are not exposed by virtue of being long-term investors. But it is possible that some of the returns which are currently treated as an illiquidity premium might, in fact, reflect compensation for variability around future credit losses. If so, in adverse scenarios, these profits might not materialise. Firms might be forced into recovery actions such as fire-sales of illiquid assets to manage solvency and meet policyholder commitments, reducing their ability to support sustainable long-term investment in the economy. As firms have invested in recent years in a much wider range of assets with different risk characteristics, the basis risk between these assets and the assets originally used to calibrate the MA has also increased.
So we think we need to look again at this issue, to be confident that the MA regime can safely support our ability to widen MA eligibility, encourage further expansion into new and innovative asset classes, and streamline our upfront review of firms’ MA applications.
Clearly the final position on both the risk margin and the matching adjustment will be important for firms’ balance sheets. As I have made clear, the QIS exercise will help us analyse a wide range of potential combinations of options beyond the scenarios tested. While we start from a position that a significant weakening of the regime is not needed to deliver the objectives of the Review, we are not trying to backsolve to a specific capital outcome and we will only form firm proposals once we can assess their likely net overall impact, drawing on the data from the QIS.
The importance of comprehensive Cost Benefit Analysis
Finally, let me say a few words about how the impact of potential reforms needs to be assessed against the three objectives of the review. Each reform cannot just be considered in isolation. We will need to recognise and consider carefully the potential interactions between them as a package.
It is tempting sometimes for commentators to try to measure the impact of regulatory changes purely by whether it delivers some sort of overall cut in regulatory requirements, and if so, how large. But of course, a rigorous analysis of costs and benefits needs to go beyond that if we are to properly assess the impact on all three of the policy objectives in the review.
As an example, there are a number of reasons why it is not obvious that loosening of capital requirements per se would necessarily increase productive investment.
- First, any analysis needs to consider whether and why insurers would invest differently to previous practice, if relative incentives are unaltered. It appears that the current regime provides strong incentives for insurers to invest in a small number of, particularly property-backed, assets. When considering the impact of reforms, analysis which claims insurers will shift their investment strategy needs to give an economic justification as to why they would choose to do this.
- Secondly, any analysis needs to consider where any released capital would be deployed. Firms may choose to run-off any excess capital by returning it to shareholders, by investing internally to support new business growth, or by purchasing more assets. Analysis needs to consider the various calls firms would have on any freed-up capital, and consider how firms might choose to balance those. It is not necessarily the case that all released capital would or could necessarily be deployed into the sort of productive assets the government wishes to encourage.
- Thirdly, the analysis needs to assess the impact on incentives and investment across the UK economy in aggregate, rather than the insurance sector in isolation. The Bank and the government are interested in the net change in productive investment across the system as a whole. Assessing the net effect of growth in one sector at the cost of potential shrinkage in another is complicated, but vital to present a complete picture of the effect of policy reform.
There are also other important issues to consider. The costs and benefits of regulatory changes are often borne by different stakeholders, with potentially different interests and time horizons. We are seeking to achieve a reform package which supports sustainable investment for the long-term, while maintaining the long-term resilience of the sector. Potential costs and benefits to firms can often be more immediately apparent and identifiable, whereas the benefits to policyholders and wider society from avoiding tail events such as major insurance failures can accrue over a longer horizon and be harder to quantify (although we can obviously look to events such as Equitable Life for a sober reminder of why this matters).
It is often challenging to try to assess the full costs and benefits of prudential regulatory changes for these sort of reasons. But any analysis needs to consider these issues if is to give a complete picture. For the Solvency II Review, we have designed our QIS exercise to give us the data we and the Government will need – both to develop policy proposals, and to help us assess the potential costs and benefits in a comprehensive way. Once we have had the opportunity to assess the QIS results we will use the results to inform our discussions with HM Treasury on proposed reforms for consultation, and on how any changes can best be implemented. As we always do, where the PRA consults on proposed changes to our requirements, we will also publish our assessment of the costs and benefits of our proposals and invite feedback.
In closing, I’ve explained today that we see the three objectives of the Solvency II Review as being mutually complementary. We need to consider the options for improvement and reform across all three, and work on a package which can best meet the objectives together in a sustainable way.
Assessing the dependencies within any package will be very important in order to do a full cost-benefit analysis. We need to identify where changes may be needed in one area to unlock the ability to make other reforms. Some changes can be positive for multiple objectives. Other options may not produce the benefits we or the government want to see.
There is more work to do to develop our proposals and assess their impact using the QIS data. But I believe that there is space to craft a meaningful package of reforms to improve the regime and achieve the three objectives of the review. A package which would streamline parts of the regime which are too onerous, support competitiveness, encourage more investment by insurers in infrastructure and green assets, and ensure the regime is tailored for the UK market. But also a package which does not compromise on the policyholder protection objective, or on the resilience the sector needs to invest in the economy for the long-term in a sustainable way.
We recognise there will be room for debate about how we and the Government should construct a set of reforms which best meets these objectives, and we want to make sure we hear a diverse range of views on these issues. We will use the data we are collecting to give us further insights, and we are also looking for other ways to have further dialogue and roundtable discussions with a range of interested stakeholders to help explore these issues further.
Working closely with the Government, we are determined to play our part in helping to deliver a successful review across all three objectives.
I am very grateful to Anthony Brown and Ruth Hendon for their assistance in preparing these remarks. I would also like to thank a number of other colleagues who provided comments, including Manuel Sales, Alan Sheppard, Francesca D’Urzo and Jemima Ayton.
Behind Corporate and Government bonds, and then Equity Release Mortgages