In this article, we analyse the quality of capital of European insurers following the implementation of Solvency II. We fear that the relatively loose grandfathering rules associated with the reliance on soft elements of capital could lead to weakening risk profiles of European insurers.
Since the implementation of Solvency II on 1st January 2016, we have observed a capital optimisation trend among European Re/Insurers. This trend makes sense as the new regulatory framework allows European insurers to rely on soft elements of capital such as expected profit in future premiums (EPIFP), letters of credit issued by banks who, on average, have a lower rating than Re/Insurers and net deferred tax assets. But, paradoxically, due to the economic approach of Solvency II, Re/Insurers will likely increase their reliance on these soft elements of capital while possibly reducing their core capital at the same time. This works well in favour of equity holders as reflected by the increasing dividend pay-out ratios and share buy-back programmes we have observed in the last couple of years; but we fear that their credit quality could possibly deteriorate due to their prospective weak quality of capital. As a trader would say “go long insurer equities and short their bonds”.
Capital tiers under Solvency II include soft elements of capital that are volatile by nature
Solvency II, the new European regulatory framework for European insurers, was finally implemented on 1 January 2016. In essence, as opposed to Solvency I that was somewhat “archaic” but based on a prudent man approach, Solvency II aims at having an economic approach. In short, Solvency II compares the available capital with the required capital which is calculated either through internal models and/or a standard approach. Both internal models and the standard approach aim at assessing the sum at risk taken by an insurer, namely investment, underwriting, pricing, credit and operational risks. Pretty much like structured products in finance (a very good example could be found with CDOs (Credit Default Obligations)), the more the insurance company is diversified, the lower the required amount of capital is. Hence, over the past years, we have seen some insurers acquiring portfolios and/or companies as a means to increase their risk diversification factor allowing them to underwrite more with a relatively lower marginal amount of regulatory capital. A very good example could be illustrated by the recent announced acquisition of XL by Axa; Axa’s management mentioned that the new group post-acquisition Solvency II ratio of 190 % to 200 % will remain in line with its current Solvency II ratio (205 % at year-end 2017) thanks to the increasing risk diversification benefits of this relatively large acquisition.
Under Solvency II, there are three capital tiers. According to Norton Rose, the insurer’s regulatory capital is “divided into three 'tiers' based on both 'permanence' and 'loss absorbency' (Tier 1 being the highest quality). Tier 1 is also divided into 'restricted' and 'unrestricted' Tier 1. The rules impose limits on the amount of each capital tier that can be held to cover capital requirements with the aim of ensuring that the items will be available if needed to absorb any losses that might arise. This means that they need to be sufficient in amount, quality and liquidity to be available when the liabilities they are to cover arise. Items with a fixed duration, or a right to redeem early may not be available when needed. Similarly, obligations to pay distributions or interest will reduce the amount available to the insurer. The rules on 'tiering' are designed to reflect the existence of such features”. In summary, the Tier 1 capital layer is made of shareholders’ equity, expected profits included in future premiums (EPIFP) and Restricted Tier 1 (RT1) debt. Tier 2 is the sum of reinsurance covers, subordinated debt (Tier 2 debt) and letters of credit. Finally, the Tier 3 capital layer encompasses Tier 3 debt and net deferred tax assets.
It is our opinion that European Re/Insurers suffer from a relatively low quality of capital under Solvency II. Indeed, on the basis of an economic approach, EIOPA recognises as regulatory capital net deferred tax assets while this item is not formally recognised for European banks under their current regulatory framework for example. Additionally, letters of credit tend to be issued by players, typically banks, who on average suffer from a weaker credit quality as reflected by their lower ratings. In addition, while banks tend to be exposed to systematically risk, one could argue that insurers are far less exposed to systematic risk. Finally, the last but not the least point that strikes us is the recognition of future profits. While the inclusion of this item is certainly in line with the economic approach under Solvency II, it will likely add volatility in the Tier 1 capital layer considering the volatile nature of future profits and the possible inexactness of the assumptions when calculating this item. Additionally, the inclusion of EPIFP could cast a concern for bond holders as the RT1s rank pari passu with all other elements of the Tier 1 capital; but how can future profits be written down similarly to RT1s or shareholders’ equity?
The paradox of the capital tier definitions under Solvency II certainly relates to the anticipated capital optimisation of European insurers. There is no doubt that CEOs of European insurers have been struggling in making their companies appealing to investors considering the barrier to entry to the insurance sector, the relatively low returns of their stocks and the “boring” aspect of the insurance industry especially compared with more appealing and easier to understand sectors such as car manufacturers or even luxury. Despite the intrinsic resilience of the sector, wouldn’t the current definition of regulatory capital under Solvency potentially lead to higher default rates?
The capital limits imposed by EIOPA seem quite loose. In fact, the level of unrestricted Tier 1 capital of 80 % within the Tier 1 capital (unrestricted and restricted) seems to be quite high. But considering the inclusion of future profits in the unrestricted Tier 1 capital layer, one could argue that the constraint is quite loose in particular for insurers with long tail lines of business such as Savings and Pensions, which allow them to increase their capital base considerably compared with insurers with a short term liability duration. For example, we think that RSA, a pure UK based P&C insurer is penalised compared with Pru plc, a pure UK based Life insurer.
Of interest, while European regulators and rating agencies recognise subordinated debt as capital up to a certain limit, they do not give any credit to senior debt. Considering the reverse cycle of production insurers benefit from, the use of senior debt by European insurers should be quite low and essentially used for funding reasons such as working capital and means of funding for real estate acquisitions for examples. If the ultimate company is a pure holding company, the issuer may be tempted to issue senior debt at the holding company level and uses the proceeds to make a subordinated loan or potentially a rights issue. Recently, Vivat, the holding company of SRLEV, issued a senior debt and uses the proceeds to make a loan in the form of RT1 to its main operating entity, SRLEV. This type of funding has been relatively rare as the issuer “leverages” the proceeds from the senior bondholders; but they are become more and more popular …
Value Creation under Solvency II
To make it short, the value creation can stem from the business or capital management. In the current low interest rate environment, already optimised business mix for Life insurers, relatively slow GDP growth rate in Europe compared with fast growing economies such as those in Asia or Latin America, intensifying competition in the Primary P&C market and he continuing soft P&C Reinsurance market, it seems easier to optimise the capital structure instead.
One easy key indicator that is commonly used to quantify a company’s value creation is the difference between its Return On Equity (ROE) and its Weighted Average Cost of Capital (WACC). The higher the difference is, the more value the company creates to its shareholders. As mentioned above, we believe that in the current economic and regulatory environments, insurers have more incentives and fewer difficulties in further optimising their capital bases than increasing their overall profitability levels. From a capital management perspective, the low interest rate environment is quite positive as it allows companies to lower their cost of capital by issuing debt at a relatively low cost. In other words, we should have seen or should start seeing a rebalancing of their positions between shareholders’ equity and debt instruments.
If you were in the shoes of a CFO, you would aim at reducing the cost of debt. Since only subordinated debt are given equity credit from both EIOPA and rating agencies, the exercise would then consist of issuing Tier 3 then Tier 2 and then RT1 debt instruments. There is limited interest to issue Tier 3 debt instruments for two main reasons. Firstly, the use of Tier 3 capital layer is limited by nature because it is capped to 15 % of the insurer’s SCR and secondly, apart from Fitch, Tier 3 debt instruments are not recognised and hence credited in the rating agencies’ capital models. RT1s tend to be relatively expensive because of their loss absorption features on a going concern basis.
The P&C reinsurance market remains characterised by an excess of supply leading to a soft market; as a consequence, profitability has declined while terms and conditions of P&C reinsurance contracts have become in favour of the ceding companies. Additionally, in the low interest rate environment, insurers have been facing two majors concerns. Firstly, capital market reinsurance offers (e.g. cat bonds, ILS, collateralised reinsurance programmes, etc.) have surged because of the lower cost of these programmes compared with traditional reinsurance covers. In fact, the cost of these non-traditional reinsurance programmes is dependent on interest rates; the lower the rates, the lower the reinsurance programme. Secondly, the low interest rate environment has penalised reinsurers because of the lower net investment income. The recent nat cats that occurred in the third quarter of 2017 will further deteriorate the reinsurers’ 2017 full year earnings; their value creation is therefore expected to deteriorate in 2017. European reinsurers have recently suffered from lower ROEs due to (i) the continuing soft P&C reinsurance market and (ii) more importantly the recent losses from natural catastrophes in the USA. Hannover Re has kept a strong level of value creation thanks to its optimised capital structure under Solvency II as the company relies on soft elements of capital that are not accounted in its shareholders’ equity, enabling it to benefit from a higher ROE levels than its peers. European based Life players suffer from a relatively low profitability level essentially due to the impact of the low interest rate environment. Only Pru plc and L&G have managed to maintain a strong level of value creation to their shareholders; Pru plc has disclosed an APE margin above 50 % at the group level for the past years compared with c. 15 % for CNP and c. 30 % for the major European Composite insurers. Finally, among the European Composite insurers, Aviva, RSA and Unipol have the lowest levels of value creation due to their volatile earnings; Unipol suffers from a relatively high cost level as well due to its exposure to the Italian market. In contrast, Allianz and Talanx benefit form the most stable and highest level of value creation over the past years; this situations stems from their relatively low cost of capital while maintaining a relatively stable level of ROE. Finally, Axa, Mapfre and Generali have released relatively low levels of value creation due to the combination of a relatively high cost of capital compared with their peers and a relatively lean level of profitability level. Prospectively, we would expect a higher level of value creation for Generali as its management has taken drastic steps to improve the insurer’s Life business mix that could be further optimised; in contrast, we find that Axa’s Life business mix is already optimised and therefore we do not expect further improvements from that side.
European Re/Insurers benefit from too much capital under Solvency II
The graphs 1 and 2 are of interest. Indeed, it indicates the level of Unrestricted and Restricted Tier 1, Tier 2 and Tier 3 capital layers as a percentage of the company Solvency Capital requirements (SCR). EIOPA requires that the Tier 1 capital layer (made of the Restricted and the Unrestricted Tier 1 capital) is 50 % minimum. Most insurers enjoy a level well above this minimum level; this situation certainly stems from the inclusion of the future profits and the risk diversification benefits under Solvency II. The companies that disclosed low levels of capital tiers have either dealt with the situation (e.g. RSA issued a RT1 in 2017) or have been acquired (e.g. Delta Lloyd). We see NN Grop as the White Knight of Delta Lloyd since the latter suffered from a weak capital position from both a qualitative and a quantitative perspective despite its two rights issues. Others like Ethias had to reduce significantly their risks by selling their problematic legacy minimum guaranteed Life portfolios through several tenders offering a premium up to c. 25 % of the policy value.
We do not see the interest of keeping such a high amount of Tier 1 capital, especially considering the increasing pressure from equity holders. In fact, considering the relatively low level of value creation, management has been under pressure “to deliver more with less”. With the exception of insurers in fast growing economies, we see limited upside from the ones in matured markets and believe that most of the prospective increasing profitability level will likely stem from capital optimisation. Recently, numerous insurers have undergone significant capital optimisations in the form of share buy-backs (e.g. Munich Re, Allianz and Scor to a lesser extent) and/or increasing dividend pay-out ratios (e.g. Hannover Re). Despite not complying with Solvency II as a Swiss based insurer, we like to mention Zurich Insurance’s dividend pay-out ratio of 75 %, a level that we find unsustainable in the long term.
Importance of transitional rules under Solvency II
We have to admit that there is one caveat in our prospective outlook. Solvency II allows for transitional rules, which in essence let the insurers to smooth the implications of applying the Solvency II rules over time. We bring two good examples that illustrate our view. Perpetual subordinated bonds, typically those referred to as Legacy Tier 1 and Upper Tier 2 bonds, could be grandfathered as Tier 1 under the Solvency II grandfathering rules; these rules have a 10-year period and therefore could be used until 31 December 2025. By applying these grandfathering rules, some insures can “inflate” their solvency Tier 1 capital with constituents that do not necessarily fully comply with the aim of the Tier 1 capital, in particular the loss absorption feature. A second good example could be found in the transitional rules regarding the insurers’ reserves; this rule allows insurers not to book their reserves at their best estimates under Solvency from the first day of the implementation of Solvency II but to smooth over the next sixteen years the increase in reserves.
There is no doubt that these transitional rules weaken the quality of capital of European insurers as they tend to smooth the implementation of Solvency II. The two most important transitional rules that weaken the quality of capital are (i) the grandfathering rules for subordinated debt and (ii) the transitional rules on reserves. Therefore, from a theoretical point of view, the relatively high level of Solvency II ratios might deteriorate prospectively due to these transitional rules. Or, put it simple, the Solvency II are somehow inflated today and might possibly deteriorate over time.
While all European insurers use the grandfathering rules under Solvency II, only a few have applied the transitional rules on their reserves or the matching adjustment, two of the most important transitional rules under Solvency II. The rules on reserves have been particularly applied by those with relatively long tail businesses (typically the UK based Life insurers), Aegon and those who suffer from a relatively weak capital position (Ethias, Groupama). In the case of Aegon, we are surprised that the insurer’s regulator accepts the equivalence regime with the USA; we think that a full Solvency II implementation would be more appropriate for this insurer as two-thirds of its revenues from the USA. There can be some significant discrepancies between the US regulatory regime (RBC) and Solvency II. For example, longevity risk is not assed under the RBC while there are capital charges for this risk under Solvency II. Considering the regulatory arbitrage, most longevity risk transaction have been done through US based Re/Insurers.
The Solvency II ratios are meaningless. The quality of capital should be the key point of analysis
This situation leads us to think that the risk profile of European insurers is likely to deteriorate for the ones whose Solvency margins extensively rely on transitional rules. It is our opinion that the current Solvency Ii margins are somewhat meaningless. Our rational is based on the fact that they are difficult to compare considering the different assumptions used by each insurer. Additionally, while there is a general framework in the calculation of Solvency II ratios through internal models, each jurisdiction has the right to adapt the rules and potentially change them from the general framework.
Prospectively, we would expect European insurers to issue Tier 2 bonds predominantly. The capacity for Tier 3 bonds is relatively limited as the Tier 3 capital layer is capped to 15 % of the insurer’s SCR and they are not credited under the rating agency capital models. We have seen only a few issuances of RT1 bonds so far for two main reasons. Firstly, as highlighted above, European insurers do not need to strengthen their current Tier 1 capital layer. Secondly, due to the loss absorption feature of these bonds, their cost is higher than Tier 2 bonds that do not encompass this loss absorption feature.
Prospectively, we expect the continuing deterioration of the quality of capital of European Re/Insurers
Prospectively we expect additional share buy-back programmes from European Re/insurers as a means to improve the company profitability levels and optimise their solvency capital under Solvency II. A softer approach would consist of increasing the insurers’ dividend pay-out ratios to possibly “unsustainable” levels in the medium-to-long term. These capital management strategies aim at swaping the shareholder’s equity (hard capital) with soft but fully eligible elements of capital.
If this trend were to occur in 2018, it would be positive from an equity perspective as stocks will likely raise everything else being equal. From a creditor perspective, it will likely be negative considering the inherent volatility of the soft elements of capital under Solvency II.