Tier 1 awaited after Solvency II arrival

Although Solvency II is not triggering changes to Fitch’s ratings of insurers, it is a credit positive for the European industry, according to Harish Gohil, managing director, EMEA insurance, Fitch Ratings, who discusses the framework’s impact and shares his expectations regarding Tier 1 issuance.

How is Solvency II affecting the amount of capital insurers need?

What Solvency II has done is increase capital requirements, through the introduction of the risk-based regime. However, the higher capital requirements do not necessarily lead to an increase in the capital actually held, mainly because most insurers were already holding capital well in excess of the previous Solvency I requirements.

One of the reasons would have been ratings; most of the larger insurers are rated and they tend to be rated in the higher rating categories and to achieve those ratings they always needed to be quite well capitalised.

Another important factor is that insurers have been preparing for Solvency II for several years. Even when the requirements were only just being developed and were uncertain, they were managing capital with that very much in mind, so we would say they were generally somewhat cautious in how they were managing their capital. What they certainly wouldn’t have wanted to do was get in a position where Solvency II arrives and suddenly they find that they need to go and raise additional capital.

There are clearly some exceptions. The most publicised is Delta Lloyd, which found itself in a situation where it had to actually go and raise capital externally. Also, smaller and medium-sized insurers may have been less well prepared for Solvency II, so it’s more likely that they may need to look to boost capital, by raising additional capital, or by finding other ways of covering or mitigating their capital requirements, such as reinsurance solutions, changing asset strategies, hedging, and so on.

As you mentioned, the larger insurers have been well capitalised and highly rated, but how does Solvency II affect the credit profile, or the inputs into the ratings, and the outlook for them?

From a ratings perspective, Solvency II overall is neutral, in that there haven’t been any ratings changes or any expected as a direct result of Solvency II.

On the other hand, overall we would say it’s been positive for the credit quality of the insurance sector because it has improved risk management. A big part of Solvency II is around risk management and governance, which as a result has improved across the industry as a whole. The risk-based framework is also a very big step forward from the old Solvency I regime. There are clearly many challenges with Solvency II, in terms of lack of consistency between different countries, even between different companies within a country. But overall it is a big step forward from what we had before, the old Solvency I regime.

On both of those points — improved risk management and the risk-based regime — Fitch believes the larger insurers were already developing those aspects, even before the advent of Solvency II. So, if you went back even as long ago as 10 years, especially the largest insurers across Europe were improving their risk management and governance, partly as a response to the previous crisis that the industry faced in around 2001/2002. That led to many changes in the industry, in terms of improvements in risk management and governance. Most of the larger insurers were also already focusing on some kind of economic capital metrics, and managing themselves to those. So, Solvency II is a further development of that, but it is not fundamentally new, for the largest insurers, anyway.

Would you have any general views on the quality of Solvency II disclosures and how you factor that into your credit analysis?

It’s early days in terms of Solvency II disclosure. Obviously Solvency II only came in on 1 January 2016 and there is no mandatory requirement for any public disclosures until 2017, so it will be around spring 2017 when we first see the mandatory disclosures from the insurers.

But of course all the larger insurers and medium-sized insurers have been disclosing their Solvency II ratios, and also providing some sensitivities on those Solvency II ratios in terms of how they might change in response to changes in credit spreads, equity market movements and other drivers. And that’s certainly very helpful for understanding the resilience of insurance companies’ solvency positions.

But at this stage there isn’t that much detail beyond this. For example, it is rare for companies to disclose what benefit they have taken for transitional relief on Solvency II requirements, and that will become more evident next year when we have the public disclosure, which would then make the Solvency II ratios more comparable. Right now, you have to be careful how you compare the reported Solvency II ratios because there are various differences behind those calculations, one of the main differences being around transitional benefits.

How will you go about assessing what ratings the Solvency II Tier 1 instruments should have?

In terms of the ratings criteria that will apply, they will be the same that we have applied for the last few years. Our criteria for rating debt capital instruments has evolved over the years, but we haven’t specifically made any changes in response to Solvency II as such. The difference will be in the instruments that are issued under Solvency II. We will apply the principles we have set out in the Fitch criteria to different instruments, instruments that may look quite different from what has been issued in the past, particularly around the Tier 1 instruments.

In terms of rating capital instruments, there are two aspects in our criteria that are key. One is the level of subordination, and what that implies for recovery prospects or loss severity in the event of the insurer being wound up or liquidated or resolved in some way. And the other aspect is what we call non-performance risk, by which we mean the risk of something like a coupon deferral or principal loss-absorption on a going-concern basis. So the first aspect, the level of subordination and recovery prospects, is more an end-game scenario, what happens at the end of a company’s life. The second aspect, risk of non-performance, is the risk of loss for an investor on the instrument on a going-concern basis, while the company is continuing to operate.

On Solvency II Tier 1 instruments, clearly there have been no such instruments issued yet, so all we have at the moment is the minimum Solvency II requirements. How they get rated in practice will very much depend on the actual detailed terms and conditions of the instruments. With that caveat in mind, I can nevertheless comment on what the likely considerations will be and how we might rate them in practice.

Under our criteria, we have an anchor rating and then notch down from that anchor rating to reflect the different characteristics of different instruments. On subordination, let’s say, if you take the example of a purely European insurance group — because how we rate the instruments would be different for a European group with a significant non-EU operation — operating under a Solvency II group solvency regime, we would typically notch down two for the level of subordination. That’s because a Tier 1 instrument would be deeply subordinated, so we are notching down two from the anchor rating of the holding company.

On the second part, the risk of non-performance, can you — taking into account your caveat about there not being any issuance yet — give any views?

For Solvency II Tier 1 it would be coupon cancellation we’d be talking about and in relation to that on Tier 1 there would be two types of triggers that would be key features. Firstly, Tier 1 instruments are required to have a mandatory trigger related to meeting the Solvency II Solvency Capital Requirement (SCR). That’s actually similar to Tier 2, the trigger itself.

However, on Tier 1 the key feature for us, from a Fitch criteria perspective, is the fully flexible coupon cancellation feature that a Tier 1 instrument would have to have — fully flexible, at management discretion, so that management has the unconditional right to cancel the coupon at any time. For us, that feature is the one that is key to how we would reflect the non-performance risk in our ratings. The key unknown at this stage is around how regulators might put pressure on management to exercise this feature, so whether it is possible that the regulators might be looking to force a coupon cancellation quite early on, before the company has actually got close to its SCR. On this point, we’ve been talking to regulators and other market participants to form a view. Our current assumption is that the regulators are likely to be less assertive than on the banking side. So, if we draw a comparison with how Fitch has rated bank Additional Tier 1 (AT1) instruments, this feature, the fully flexible coupon cancellation feature, has been seen as the most relevant for assessing non-performance risk, and this risk has been reflected by notching down three just for this particular feature. On the insurance side, our current view is that we are most likely to notch down two for the risk from this feature because we think the insurance regulators are less likely to force a coupon cancellation than on the banking side. It is only when we get actual Tier 1 issuances that our interpretation of this risk will firm up — as the experience develops in the market and it becomes clearer over time exactly how this fully flexible feature will work in practice.

And on the types of loss absorption — equity conversion or write-down, permanent or temporary — is there anything you can add?

On this particular question about whether we would make a distinction between equity conversion and permanent write-down or temporary write-down, the short answer is no. Under Fitch’s criteria the main consideration really is around the likelihood of the trigger being breached, so it’s not so much about what happens when the trigger is breached — so we treat each of those features the same — the key consideration is when the trigger comes into play. So say, for example, the loss absorption feature is linked to 100% of SCR, it’s that 100% of SCR that is most relevant for us, so that would feed into our assessment of the likelihood of this feature being triggered. But the feature itself — whether it is a temporary write-down, a permanent write-down or equity conversion — that doesn’t affect the rating, or at least it’s not typically expected to affect the rating.

You’ve already drawn some comparisons with the banking side. What other differences or similarities could there be between the Solvency II Tier 1 instruments we might see and the bank AT1 structures that have evolved?

There are some similarities, not least in terms of the investor base — we think the kind of investors that may buy Solvency II Tier 1 instruments are likely to be the same investors that are buying into bank AT1, so there is clearly a commonality there, which obviously then begs the question which you asked.

In terms of how we go about rating Solvency II Tier 1 instruments, we would certainty want to be ensuring we are consistent with how we rate the bank AT1 instruments, given some similarities, at a high level at least, around the coupon cancellation and other loss absorption features.

However, we recognise that there are actually significant differences between the two as well. The bank and insurance capital regimes are quite different, and that then affects how the instruments might behave in different circumstances. One simple difference there that I would maybe highlight is that on the bank side there is the concept of the Common Equity Tier 1 ratio as part of the overall bank capital regimes. There isn’t really a comparable ratio on the insurance side. You clearly have the Solvency II SCR requirement, but that’s it; there isn’t the same thing as a CET1 ratio and the buffers that you have to maintain in relation to that. That will then lead to differences in terms of how these Solvency II Tier 1 — Restricted Tier 1, or RT1, as they are being referred to — and the bank AT1 might behave in practice, and we would then reflect that in how we rate the instruments.

To what extent will the use of such instruments complement the existing capital structures of insurers, and help improve the overall financial stability of the sector?

There is certainly a lot of discussion in the market, with arrangers and insurers themselves, about the potential for RT1 as part of insurers’ capital structures. So we would certainly say there is potential for RT1 to complement the existing capital structures. It remains to be seen how important they actually will be in practice as part of the insurers’ capital structures.

There will be a market; the uncertainty is about how large, how significant the Solvency II RT1 asset class will become. For some insurers it may be a relatively important part, but not necessarily for others. For example, Axa has said very clearly that they don’t intend to issue RT1 instruments. They have legacy Tier 1, grandfathered Tier 1 instruments, but they don’t expect to issue any new-style RT1 instruments. As their current instruments come up for refinancing, they would issue Tier 2 rather than Tier 1. That’s a first very clear example where they don’t see a role for Tier 1 in their capital structure.

But, overall, it certainly has a role to play for the industry as a whole. For example, some insurers certainly feel that the right kind of RT1 instrument could be a way to manage and support their Solvency II ratios on the downside in stress scenarios.