You can read the transcript below.
Marcus: Welcome to A&O Shearman's latest insurance-focused podcast. I'm Marcus Gwyer, a senior associate in our insurance team in London. I'm here with Kate McInerney, a partner in my team and Iona Misheva, a partner in our securitization team. Today we're going to be speaking about matching adjustment reform, a topic which has been the linchpin of prudential reforms in the UK. So why are we talking about the matching adjustment today?
Kate: Well, Marcus, let's start by just commenting very, very briefly on what the matching adjustment or MA is. In essence, all insurers have to value their liabilities by projecting into the future and applying a discount rate to obtain a value for those liabilities at the date of calculation; a present value. The higher the discount rate, the lower the present value of the liabilities and vice versa. The basic premise in insurance regulation is that the discount rate should be risk free and because yield rewards risk, so too should the discount rate be lower.
But the matching adjustment can be used by some insurers to (in general) increase the discount rate, so reducing the present value of some of their liabilities. These insurers can use the matching adjustment because the nature of those liabilities means that they can hold assets which produce cash flows that closely match them, meaning they're not exposed to the risk of having to sell those assets when experience is adverse. So in summary; capturing the matching adjustment results in a lower liability valuation in relation to eligible liabilities and therefore increases capital resources for firms which do capture it.
But it was seen as a bit rigid when it was first introduced in Solvency II, and in February 2022, the UK government announced a package of changes to insurance regulation, which, having previously been dictated by EU law in the form of Solvency II was positioned as a Brexit win. A central pillar of that package were changes to the matching adjustment rules, which together with the changes to the risk margin, were trumpeted as freeing up to 10 to 15% of capital held by life insurers so that they could invest in long-term productive assets such as infrastructure. And now, as of 6 June, we have the PRA’s final policy statement on matching adjustment, which reflects that liberalisation of the matching adjustment rules and which life insurers will be paying close attention to after a very long incubation period.
Marcus: Understood. So what are the key points coming out of the final policy statement?
Kate: The background here is that the government trumpeted changes to the matching adjustment regime as a tool for stimulating investment in long-term productive assets. In the immediate aftermath, there was a good deal of controversy as the government and the PRA (Prudential Regulation Authority) did seem to come to blows somewhat over the implementation of the policy. Now that's something of a distant memory, but the changes are less radical than the February 2022 announcements about massive capital releases foretold.
The changes of greatest impact are: firstly, broadening of asset and liability eligibility for MA, but really it's asset eligibility which is the key thing for us; second, the introduction of the attestation regime around the matching adjustment and additional reporting; and third the introduction of notching.
Looking first at asset eligibility, the core premise of the matching adjustment under EU Solvency II was that you had to have assets which produced fixed cash flows and were not capable of being changed by the issuer, except in limited circumstances. That essentially ruled out infrastructure finance assets relating to projects which were still in the construction phase, i.e., where the construction end date was uncertain. The rules now permit up to 10% of an insurer's matching adjustment benefit to come from assets which have highly predictable rather than fixed cash flows. This is seen as a critical change which will help unlock broader investment by insurers in green and good infrastructure. Now, highly predictable means contractually bounded, and that means you've got to have a contractual right to receive an ascertainable amount and an ascertainable time. And it also means that failure to pay that amount at the time must constitute a default. In other words, there can be flexibility in the Ts & Cs about what the investor is entitled to and when, but there can't be limitless discretion, and the obligation to pay has to have teeth.
On top of this, you've got to consider whether the cash flows do, in fact, create an appropriate source of matching for the relevant liabilities. That is to say, “Do they represent a material risk to the quality of matching?” and the PRA has specified five tests which insurers have to apply to determine whether an asset can be included in the MA portfolio as a highly predictable asset.
These tests are quantitative and technical in nature and they focus on the quality of matching in particular scenarios. Firms also need to consider whether the matching adjustment portfolio as a whole and each asset within it would meet the Prudent Person Principle, because that has been introduced as a new MA eligibility requirement for all assets, and not just assets which are going to be included in that new highly predictable bucket.
Now the prudent person principle, or PPP, is a longstanding Solvency II-derived rule that insurers must only invest in assets which have risks that they can properly identify, measure, monitor, manage, control and report and take into account in evaluating their solvency needs. So this shouldn't be new for insurers. But by making it making compliance with the PPP a condition of MA eligibility, there are now additional consequences that would flow from failures from insurers failing to comply, namely that they could lose part of the benefit of the matching adjustment, with the amount of the matching adjustment benefit reducing (potentially up to 100%) over the duration of non-compliance. Now that's potentially very serious given the importance of MA for many firms. Other important asset-side changes include the removal of the limit on sub-investment grade assets, but again this has to be understood in the context of the new PPP eligibility requirement, and insurers looking at sub-investment grade assets will note that the PRA's emphasis on ensuring that firms with exposures to those assets have adequate workout capabilities.
Finally, a brief word on notching. I mentioned that the matching adjustment impacts the discount rate that an insurer can use to value its liabilities. Obviously, a key question is “How much?” Since the rationale for matching adjustment is that insurers as investors are able to capture additional yield by holding assets to maturity, yield which comes from other factors (most obviously credit risk) has to be removed. This element is called the “fundamental spread”. The greater the fundamental spread, the lower the MA benefit. One of the changes that the PRA has introduced is to include a more risk-sensitive calculation of the fundamental spread by asking insurers to reflect rating notches, not just headline ratings in their determination. This will make insurers more sensitive to notching changes, but it is intended to ensure that they don't capture an inappropriate amount of MA benefit.
Marcus: Thanks for the recap on what's come out of the final policy statement. Iona, you're a partner in our securitization team and as such, someone whose career has been focused on creating fixed cash flows. What will the real-world impact of these changes to MA asset eligibility be for insurers?
Iona: Traditionally, insurers looking to benefit from MA have looked to invest in long-term assets with a very precise amortisation profile. That means looking at either a specific type of transaction, like a ground rent deal – where during the life of the lease there are fixed rental payments – or highly structured, bespoke deals involving assets like RMBS, which don't traditionally have a fixed pay down rate. The intent of the changes is to expand out that universe of transactions which can be MA eligible. However, this has to be balanced with policyholder protection. So, headroom for investment in HP assets is not huge and we expect insurers to be choosy about how they use up that headroom. We think there are two main problems that the changes will help to overcome. The first is the current inability to invest directly in certain asset classes because they don't produce fixed cash flows and the second is related: if an asset doesn't produce fixed cash flows, then the alternative is to restructure it so that it does. And by that we mean create a fixed senior cash flow and a junior variable cash flow.
Historically, this has required a lot of work, resulting in very bespoke deals. That adds both time and cost to the investment process. It's hoped that the changes will mitigate the need to highly structure investments to create bond type rigidly fixed cash flows. The result is that the insurer should be able to either invest directly into the relevant asset or reduce the amount of restructuring required to produce the desirable cash flow profile. So the investment process is simpler. And more cost effective and this is key, will hopefully allow insurers to invest in more off-the-shelf transactions rather than highly bespoke ones.
In terms of new investments, the obvious structures for insurers to now look at will be securitizations. The starting point for any securitization transaction is that cash flows are highly predictable. So theoretically any securitization could become MA eligible within the limits specified in the new regime. The next step will be to link the assets being securitized to those that help the UK economy and so feed into the government rationale for the changes. So while the government has made it clear that one of the reasons behind these changes is to channel insurance money into economic infrastructure, the changes are agnostic on the sector and purpose of the investment, so assets with similar cash flow characteristics will also meet the new requirements.
An example would be certain real estate assets with predictable cash flows. So schemes to build new housing, whether to rent or buy, (and especially shared ownership or social housing) become interesting and more generally development stage assets will be an area to watch, both from an infrastructure and a pure real estate lens. As Kate mentioned, another area to watch (and the government has also expressed their support for this) as the green and good projects which benefit both the economy and the environment. Overall, we think that there will be a ramp-up period during which firms will figure out in conjunction with the PRA, the best type of assets to fit within the HP framework. And so seeing the real world impact of the changes will take a little bit of time.
Marcus: So that's the asset side. Is anything changing in relation to the liabilities which are eligible for the matching adjustment?
Kate: Yes, Marcus, there has been some expansion there too, but the fundamental limitation that excludes policies which have future premium income remains in particular firms which have permission to do so, can now use the matching adjustment in relation to in payment income protection policies and the guaranteed portion of with-profits annuities. But this is generally seen as a much less material liberalisation than the changes to eligibility on the asset side.
Marcus: You also said that there were important changes relating to governance and reporting. Could you tell us about those?
Kate: I'd love to do that, Marcus. And it's a really important topic. The question that insurers will have to grapple with is how their senior management function holders (ie their CFOs) will be able to get comfortable attesting that they've taken an appropriate amount of matching adjustment benefit. This is no doubt going to be a challenge. Actuarial science is often described as more art than science. But we want to do that topic justice and so in our next podcast, I'll be joined by Claire Hayden, who is an executive director and we will be discussing this attestation regime and lessons learned from others. After that, a series of insurance-focused podcasts will continue, with topics including enforcement themes, AI and employment disputes and insurance.
Marcus: All sounds very interesting. Well, we'll look forward to that. Thanks again and goodbye.
