05 December 2018
The first insurance restricted tier 1 (RT1) deal was issued in late 2017. Romain Miginiac highlights why investors should consider these instruments and how they differ from traditional bank AT1s.
The insurance RT1 market is a relatively new, but growing, part of the well-established USD 150 billion plus European insurance subordinated debt market. The first deal was issued in late 2017 and there are now seven available instruments (with approximately USD 3.5 billion outstanding), but these assets have experienced a few challenges. In this paper, we outline the risks of these instruments, consider how they compare to other capital securities within the financial sector and outline our outlook for the insurance RT1 market and the opportunities in terms of diversification as well as attractive yield that can be provided by these new instruments.
The solid fundamentals and strong debtholder-friendly track record of issuers should, in our view, mitigate any instrument-specific risks. Moreover, we expect forthcoming issuance to be gradual as old-style instruments are refinanced or instruments are issued on an opportunistic basis. Overall, we are constructive on the insurance sector and on the subordinated debt universe as we believe valuations are compelling and supported by strong issuer fundamentals. However, we remain highly selective regarding both issuers and specific instruments.
RT1 instruments are junior subordinated debt securities issued by insurers that can qualify as capital under current European insurance regulation (Solvency II). To qualify as Tier 1 capital, the instruments need to be perpetual with a minimum five-year non-call, have non-cumulative fully optional coupons and a contractual trigger to principal write-down or equity conversion.
Table 1 summarises the main features of the instruments and compares them to other capital securities in the financial sector.
Table 1: features of financials capital instruments
There are three key risks for investors: the loss of the principal investment (via a write-down or conversion); the non-payment of coupons; and extension risk:
- The risk that the principal investment would be lost potentially occurs when one of the contractual triggers is breached – usually 100% of solvency requirements – at which point the bonds are written-down or converted into equity.
- In general, coupons are only cancelled when solvency requirements are breached; however, this cancellation can be waived by the regulator if the payment of coupons can be shown not to threaten the solvency position further.
- To address these first two risks, our research suggests the average Solvency II ratio for European insurers is currently circa 200% (as at 31 December 2017) – twice the requirement and trigger levels – and in our view strongly mitigates the risk of principal loss and coupon cancellation (more than USD 350 billion aggregate excess capital to requirements) for our sample of aggregated data from individual insurers’ figures.
- Finally, extension risk is the risk of bonds not being called at the next call date by the issuer; meaning the bond price is likely to decrease as investors require a higher yield or spread to compensate for the potentially longer holding period.
Table 2 – Key risks and mitigants of investing in RT1s
The main risks and mitigants of both instruments are set out in table 2. In our view the risk of principal loss, or a breach of requirements, is similar for both instruments. While there is no specific trigger level on Tier 2 (T2) instruments, the trigger level on RT1s is relatively remote and we assume an insurer would either take remedial action to boost solvency ahead of any breach or would be in liquidation at this level; in short, we would expect close to zero recovery levels for both RT1s and T2s. On the other hand, there is a difference in coupon risk as RT1s have non-cumulative coupons, while they are cumulative for T2s; that said, the deferral / cancellation levels are the same for both instruments and, again, we would not expect the outcome to be differentiated. The main difference relates to extension risk. T2s are usually callable structures with a coupon-step up (so the issuer has an incentive to redeem), whereas RT1s are true perpetuals with no coupon step-up after the first call date. In our view this risk can be mitigated by observing the long track record of insurers calling bonds, yet it must be pointed out that RT1s are a new asset class and therefore this is will not actually be tested before 2027 (the first call date of an RT1). Overall we believe there is limited incremental risk in RT1s compared to Insurance Tier 2 instruments.
Again bank AT1s (Additional Tier 1s: securities and contingent convertible capital instruments, also known as CoCo bonds, which absorb losses when the capital of the issuing financial institution falls below a supervisor-determined level) have similar features to insurance RT1s (as shown in Tables 1 and 2) in that they are perpetual, have fully discretionary coupons and a contractual trigger; yet we believe the risk for investors is different. Firstly, banks have a bail-in regime where an instrument can be written down when an issuer reaches the “Point of Non Viability” (PONV) – this is at the discretion of the regulator when a bank is deemed failing or likely to fail; however, no such scheme exists for insurers. This reduces the risk of an early regulatory intervention to impose losses on creditors or force coupon cancellation. Secondly, there is no Minimum Distributable Amounts trigger – this is the amount of capital required for banks to be able to make coupon payments and is typically well above the trigger level. For European banks, this creates an incremental coupon risk that is not present in insurance RT1s. Overall, we believe the risk from RT1s is lower in our view compared to bank AT1s, as coupon skip risk (ie risk of coupons not being paid on the instrument) arises at very remote levels and there is no bail-in regime.
The RT1 market is currently very small – with less than USD 4 billion of RT1s outstanding in comparison to more than USD 150 billion of existing insurance subordinated debt (as at 25 November 2018). There are several reasons behind this. First, the Solvency II regime was introduced in 2016 and grandfathering provisions (in which the old rules continue to apply for a certain period of time) for legacy instruments run until 2026. As a result, the focus of issuers has been on adapting to new regulations and building common equity capital (Unrestricted Tier 1). Secondly, the Solvency II allowance for Tier 2 and Tier 3 debt is generous compared to banking regulation and issuers are allowed up to 50% of solvency requirements to be issued in Tier 2 and Tier 3 instruments which are often cheaper to fund than RT1. Chart 1 illustrates the capital structures of both banks and insurers. As insurers have Unrestricted Tier 1 (common equity capital) covering more 100% of requirements (151% average) and the allowance for Tier 2 and Tier 3 is high – the focus has been on issuing Tier 2 debt. Last but not least, leverage metrics (debt divided by total capital) are closely monitored by investors and rating agencies and this often compels insurers to hold larger equity buffers and caps the amount of debt that can be issued. For example, issuers with higher leverage (well above 30%) are often penalised by the equity markets (reflecting the higher cost of equity from accrued financial risk) and ratings agencies, who may lower their ratings. All in all, this suggests to us that issuing the maximum amount of subordinated debt is not necessarily efficient and that it is unlikely insurers will fill their subordinated debt allowance in comparison to banks, whose solvency is assessed purely on regulatory capital metrics and subordinated debt allowances are less generous, meaning banks are more incentivised to maximise their AT1 and Tier 2 entitlements.
Chart 1 – Banks and insurer’s illustrative capital structures
Insurers can have up to 20% of total Tier 1 capital in RT1s (including both restricted and unrestricted Tier 1). If insurers were to utilise their full allowance, we believe the RT1 market could reach USD 150 billion. A more realistic approach is to assume that insurers will refinance their existing grandfathering Tier 1s (issued under Solvency I and grandfathered until 2026) – which broadly accounts for around USD 50 billion. With this in mind, we estimate the market will likely reach USD 50–75 billion, reflecting our expectations regarding the refinancing of legacy T1 debt into RT1s plus additional opportunistic issuance. In short, we expect the issuance of RT1s to be very gradual given the grandfathering provisions until 2026.
Since the first RT1 (in a major currency) was issued in late 2017, performance has been weak; the total return of instruments over this time has been around -10%, although there are currently only seven instruments available. In our view, this underperformance has been driven by both weak markets and instrument-specific factors. Risk assets (equities and credit) have generally performed poorly in Europe since the beginning of the year and, like other subordinated debt, RT1 assets are sensitive to market sentiment. Moreover, these instruments were issued in late-2017/early-2018 which we now see as the high point of the market, meaning initial spreads were set at very low levels. This has led to a significant re-pricing of instruments – especially as the risk of the instrument being extended post first call date (extension risk) has increased materially, causing the price to decline further as investors require a higher premium for the risk.
The European insurance sector has undergone a transformation in recent years, which has improved the sector’s credit standing. First, insurers have been reducing their exposure to market risk in line with tougher regulation that does not incentivise risk-taking, while eroding margins from the low interest rate environment also prompted drastic changes to business models. For example, life insurers focused on reducing capital-intensive products, such as defined benefit pensions and guaranteed savings products, in favour of products where policyholders bear the investment risk. Yet, regulation has been a positive catalyst in our view; the Solvency II regime led to significant capital accumulation and insurers have become more conservative regarding capital management. Insurers’ own investment portfolios also remain predominantly invested in high quality bonds, despite a slight trend towards re-risking, which was mostly driven by the purchase of illiquid credit and alternatives to enhance yields and match duration of long-dated life insurance liabilities. Earnings have also been resilient despite the challenges of low interest rates – especially in the eurozone – with European insurers delivering an average Return on Equity of around 10% (as at 31 December 2017). Finally, we believe issuance should remain very manageable as solvency positions are strong and supply should mostly stem from the refinancing of existing grandfathering Tier 1 bonds, although it is unlikely that issuers will use up their full allowance for RT1.
Overall, RT1s have had a challenging start, as there has been a disconnection between valuations and fundamentals. However, we believe strong issuer fundamentals and a positive track record would mitigate the risks of the instruments and yields are currently very attractive on RT1 instruments. We remain constructive on subordinated debt as valuations are attractive. Yet, volatility in newly issued RT1s highlights the importance of instrument and issuer selection and the importance of having a well-diversified portfolio.
Source: GAM unless otherwise stated. The information in this document is given for information purposes only and does not qualify as investment advice. Opinions and assessments contained in this document may change and reflect the point of view of GAM in the current economic environment. No liability shall be accepted for the accuracy and completeness of the information. The mentioned financial instruments are provided for illustrative purposes only and shall not be considered as a direct offering, investment recommendation or investment advice. Past performance is no indicator of current or future trends.