Insurance roundtable: Life under Solvency II

Solvency II is finally becoming concrete. While insurers have long been preparing for its arrival, the final details of it and other international standards could yet affect their capital planning. How are market participants positioning themselves in light of this and the increased room for hybrids under the emerging framework? Neil Day sought the views of all sides of the market.

How do you anticipate your activity in 2014, in financial institutions in general and the insurance sector in particular?

Hervé Boiral, head of Euro credit, Amundi Asset Management: Our credit activity has been quite strong in 2013: although our funds were impacted by outflows in the first half of the year, they registered large inflows in the second half. Therefore we have been very active, both on the primary market to invest our subscriptions, and on the secondary market to adjust and optimise our portfolio positions. Even if we don’t anticipate 2014 being as rich and positive for credit as 2013, we should continue to see inflows in the asset class, especially on the lower spectrum of ratings, 5Bs and high yield.

The insurance sector represents a small part of the asset class, but has the benefit of being one of the less volatile among the financial sector. It also offers on average better ratings than classical banks, and is almost the only bucket where you can find subordinated bonds which are still investment grade, thus offering interesting spreads. For all these reasons, we should continue to closely monitor the insurance sector in 2014, as long as we are able to trade and find liquidity on primary or secondary issues.

Thomas Maxwell, investment director, Standard Life Investments: We remain constructive on credit going into 2014. The main themes remain supportive for our asset class. We expect a moderate growth environment in developed markets in 2014, with up to 3% growth in the US and 1% in the Eurozone. This was historically the best environment for investment grade credit. The improving growth picture in southern Europe supports a further normalisation in Spanish and Italian risk premiums. Central bank policy remains supportive. ECB crises management and disinflationary pressures in Europe will keep short term rates depressed for longer, leading to further allocations out of lower yielding bond markets into credit by institutional investors.

We also remain positive on insurance fundamentals in general and in particular see value in the Tier 2 subordinated segment. This being said, following the strong spread performance of the subordinated insurance sector in 2013, valuations are becoming increasingly tested and we expect performance to be increasingly driven idiosyncratically rather than systemically as macro tail risks subside. With this backdrop in mind, we have been selectively adding high quality credits such as Allianz and names where we see credible strategies in place to improve the overall credit quality of the institution, for example, Aviva and Generali. We have been funding this position through reducing names which lack clear catalysts for spread compression and/or see risks of deteriorating credit fundamentals as a consequence of rising risk appetite.

What are the key challenges the insurance sector is facing?

Karin Clemens, managing director and lead analytical manager, insurance, Western Europe, Standard & Poor’s: I would like to start with a brief review of 2013. We saw economic conditions improving, albeit slowly. In addition, our sovereign ratings in the euro-zone started to stabilise. These two factors contribute to increased stability of our rated insurers across EMEA. You can also see that from the distribution of the outlooks attached to our ratings: as of the beginning of December over 75% of our rated insurers in EMEA now have a stable outlook — that’s up from just about 70% at the start of 2013.

But it is also fair to say that one in seven insurers still remains on negative outlook. Looking ahead, we continue to see two key challenges.

The first are the prevailing low interest rates. Even though we expect a rise in interest rates over the next three years, this will be gradual and compared with historic levels rates remain low. As a result, investment earnings are limited. We therefore expect — particularly on the life insurance side — that insurers will continue to adapt their business models and product offerings. At the same time, we believe that it is really key for companies to maintain underwriting discipline in their non-life insurance businesses.

The second challenge is that there continues to be significant regulatory change. However, the focus is now shifting more to conduct-of-business regulation, which will likely influence companies’ future product and distribution strategies.

How do you assess the capital position of the global multi-line insurers?

Clemens, S&P: We consider that overall global multi-line insurers display a higher credit quality than other insurance groups. Their average rating is AA- compared with our global insurance rating universe where the average rating is in the A-range. We view the capital positions of global multi-line insurers as a rating strength, and for most of them we rate this specific factor, capital and earnings, as strong or very strong.

Are insurers in general adequately capitalised?

Charles de La Rochefoucauld, head of insurance coverage, Crédit Agricole CIB: In short, I would say that large insurance companies in Europe are generally already well capitalised, including within the new framework which is taking shape in the form of Solvency II. There might be some exceptions, though.

The overall requirement is increasing under Solvency II versus Solvency I — the SCR is between two and three times the Solvency I requirement on average across the industry based on QIS5 results — but large insurance companies started moving to economic balance sheet and risk-based capital management in line with the requirements of Solvency II some time ago.

It’s not only Solvency II that has been driving that trend, but also the markets. Investors — whether shareholders or debt investors — have been more and more favouring insurers that are well advanced in monitoring risks and shareholder value creation, having an economic balance sheet management, rather than a factor-based approach.

Rating agencies also — even though they are largely relying on factor-based models — make a qualitative assessment of the risk management capabilities of insurance companies.

So it’s a whole set of trends in the industry that have moved the insurance sector to economic balance sheet management, not only Solvency II regulation.

The new thing with Solvency II is that there is significant headroom for subordinated instruments within the Solvency Capital Requirement (SCR). Large insurance companies will progressively look into optimising their capital structure within this new framework and taking into account their rating targets.

Will the implementation of Solvency II have any rating implications?

Clemens, S&P: We believe that from a financial strength perspective a holistic and risk-based regulatory framework such as Solvency II can contribute to supporting the credit strength of the insurance sector. 2013 was a very important year, in particular the last quarter because, firstly, the European Commission announced that the implementation date has been delayed one last time, to early 2016. Secondly, the decision-making body finally settled on a package of measures including how to value long term guarantees under Solvency II. The agreement also includes a very long transitional period of 16 years for life liabilities — that gives the industry further time to adjust. Irrespective of the latest agreements, Solvency II has already strongly influenced insurers’ capital and risk management strategies.

We also recognise that the insurance industry as a whole has made huge strides in building up their own risk management. Solvency II has been one incentive for that, but companies really took on board the lessons learned during the financial crisis after September 11th 2001. We see the improvements in insurers’ risk management practices as a strong contributor to the fact that the insurance industry in Europe has fared relatively well across the most recent financial crisis.

Lotfi Elbarhdadi, director and analytical manager, insurance ratings, S&P: Solvency II has been viewed by us as something that really spurred insurers across Europe to improve their risk management practices, particularly the introduction of the Own Risk Insolvency Assessment (ORSA).

Among the most visible improvements are greater controls of risk appetite and better governance. This has been actually one of the main factors behind our broadly revised ERM (Enterprise Risk Management) scores for a number of insurers.

But to add more to what Karin mentioned with regards to this delay of Solvency II, it could probably prompt some insurers to reduce their efforts to develop their risk management frameworks, which might potentially negatively affect our views if it were to ultimately result in less attention towards ERM and less attention towards developing capital modelling tools in order to steer and manage the business.

How do you expect the SCR to compare with the rating agencies’ capital requirements?

Jozef Bala, head of debt management unit, Assicurazioni Generali: At this time key decisions on the Solvency II framework are being taken, preventing a definite answer to the question. Clarity on some fundamental aspects has not yet been achieved (for example, on the level of volatility adjustment, boundary of contracts, etc) so there are still question marks against both value and risk assessments.

More generally, rating agencies’ capital requirements rely on a set of assumptions and simplifications that is by far larger than that surrounding SCR calculation, and this makes the comparison quite complex. Yes, in some specific cases and companies the two valuations can differ substantially, but the expectation is that for global and diversified groups the two valuations will not be very different. There are also good reasons to expect further convergence in the future: in recent months rating agencies were active in understanding and monitoring both the standard formula and developments in internal models and frameworks in order to perform a “peer” comparison and even amend their model — also following pressure from the industry to fix some valuation issues.

Alik Hertel, head of group treasury, Talanx AG: According to our experience of comparing the SCR with capital requirements by rating agencies, we know that our internal models tend to show up lower requirements. This is due to the fact that diversification effects are adequately captured. This is according to our analysis the main driver of the difference between internal-based models and those of the rating agencies.

Marco Circelli, head of capital and treasury management, SCOR SE: Solvency II has its own definition of the capital requirements (the SCR and the Minimum Capital Requirement, or MCR) without reference to external ratings. As the SCR is a more economical way of assessing capital needs than the rating agencies’ methodologies, it is possible that markets will look more and more at the SCR rather than external ratings when assessing the solvency of insurers.

SCOR is already managed based on its internal model, i.e. on the basis of a Solvency II SCR. Our capital management policy based on the solvency ratio was presented during our recent Investor Day in September. Rating models are another important constraint, and we consider that our current solvency target is commensurate with our current rating level.

Hannes Bogner, member of the board of managing directors, UNIQA Insurance Group AG: A comparison is difficult because of the differences between the methodologies. Therefore for capital management purposes all capital models (Solvency I, Solvency II, internal, rating) are calculated, especially when funding activities are taken.

Lotfi Elbarhdadi, S&P: The comparison of the SCR with our model is an interesting one and is a question that has been raised many times. It is very difficult to make a direct comparison.

We can point to some differences. For example, we have regionally specific charges, whereas in general for the SCR charges are applied to all the insurers across Europe.

We can also point to some differences in terms of how the models are built. Our capital adequacy tries to assess the capital deficiency with regard to a target rating — this is different to how the SCR approaches the calculation of target capital and own funds. Our primary measure of own funds capital is the Total Adjusted Capital (TAC), which is a different measure from Solvency II.

But it’s very difficult to establish a direct link. There are significant differences with regard to diversification.

There are nevertheless more similarities for example between the requirements for approval of internal models under Solvency II and the strategic risk management component of our ERM assessment.

I would add that our new criteria have introduced new components to the overall measurement of capital, which are the representativeness of modelling and the risk position. The representativeness of modelling is a factor that is designed to adjust the result of our capital model — which will remain our key starting point for analysing insurers’ capital and earnings — if we believe that the model does not fully capture the risks specific to an insurer, while the risk position can be used to adjust for additional sources of volatility that are not captured in the model.

To give an example: for the global reinsurers, catastrophe exposure might not be completely or easily captured in the capital model, and in this case we would adjust the risk position to allow for that.

Can you expand upon the diversification issue?

Elbarhdadi, S&P: We include diversification, but the first thing we can say is that we do so more conservatively than in the SCR, be it the standard formula or in comparison to the many groups who publish their own economic capital valuations. We nevertheless include diversification benefit in our model in two ways.

We assess each risk factor in our capital models on the basis of confidence levels — for example, it’s 97.2% for BBB. Taking 97.2% for BBB (a five year cumulative default data, instead of a more onerous one year horizon) already allows for some implicit diversification benefit in the charges. We also allow for an explicit diversification benefit, which is based on diversification correlation matrices between risks. The maximum theoretical diversification benefit would be approximately 18% in our model. We apply a 50% haircut to the explicit diversification benefit in our model because of our cautious view on the diversification benefits in the tail.

Looking at other regulatory developments, will Global Systemically Important Insurers (GSII) designation and the Common Framework (ComFrame) for the Supervision of Internationally Active Insurance Groups have an impact on capital requirements?

La Rochefoucauld, CACIB: They could in theory be a factor, but they are not expected to have a significant impact.

Regarding the nine GSIIs that have been designated by the Financial Stability Board (FSB), the outcome is still unclear, but market expectations are that the respective capital add-ons shouldn’t change the picture too much.

On the ComFrame side, the new Insurance Capital Standard (ICS) applicable to approximately 40 internationally active insurance groups should not differ too much from the Solvency II SCR, so again, unlikely to really change the picture.

Does classification of insurers as GSIIs have an impact on the rating of them?

Clemens, S&P: We don’t see any immediate rating implications from the new G-SII designation. However, in the longer term it could have positive or negative effects, depending on how the insurers will ultimately respond to the new regulations. If, for example, the new regulation requires designated insurers to hold more capital, that could be a positive for ratings, all other things being equal. Another consequence could be higher costs of capital, which are generally a negative for ratings. It remains to be seen what the strategic implications — if any — might be.

It is important to stress that the insurance industry’s business model differs significantly from banking. While the difficulties faced by certain banks during the financial crisis included severe liquidity and funding issues, in our experience the insurance business model rarely gives rise to liquidity or refinancing concerns. We currently don’t impute extraordinary government support to any pure insurer or insurance group except for government-owned insurers. This reflects our perception of the relative importance of insurers compared with banks. In contrast, we classify banks by their systemic importance to recognise the likelihood of extraordinary government support. We currently reflect such support in the ratings on a material number of banks.

While insurers have long been anticipating the general shape of Solvency II, recently more details have been decided upon. For example, how will the reporting of insurers’ solvency positions be affected by the implementation of the 16-year transition period to existing businesses? Do you feel that you will need to report a “fully-loaded” solvency ratio?

Bala, Generali: Even if further time and analysis is needed for a final decision, we do not expect that the transitional measures will affect our Solvency ratio as we are not planning to use such option in the solvency assessment.

Hertel, Talanx: In the current version of our reporting we differentiate between a regulatory and economic view. The 16 year-transition period is first and foremost a regulatory topic. While the economic view will of course not change, the regulatory perspective will benefit.

However, it is premature to quantify this improvement as of today, as many important Level 2/3 details are yet to be determined. Moreover, these specifications or pre-assessment feedback from BaFin on our regulatory model may also contain additional/partly off-setting instructions.

Circelli, SCOR SE: The transitional measure essentially targets direct life insurers; therefore SCOR has no plans to apply it at this stage.

Bogner, UNIQA: This national option will be used for sure by life insurance companies with endowment contracts with guarantees (classical) and a low solvency ratio to achieve a release in best estimates. The question will be the comparability of Solvency ratios because of the different way of application of the transitional measures.

Do you think that some national regulators would be tempted to gold-plate the Solvency II rules? If yes, on which specific topics?

Bala, Generali: It is difficult to predict the reactions of national regulators. Anyway, we believe that the Solvency II legislative framework (Level 1 Directive, Level 2 Delegated Acts, Level 3 Guidelines) is quite comprehensive and it affects — with a good level of detail — most of the aspects of the insurance business. In the interests of increased transparency and the development of a level playing field across Europe, we hope that national regulators will not go beyond the detailed requirements foreseen in the Solvency II framework.

Hertel, Talanx: The Talanx has since several years been in the so-called pre-application phase. Over the years we have experienced not only a gold-plated Solvency II implementation, but also a platinum one. This is true for many aspects of either Pillar 1, Pillar 2 or Pillar 3. This experience is transnational, but not necessarily “equally strong”.

Circelli, SCOR: Yes, national regulators could be tempted to gold-plate Solvency II. The philosophy of Solvency II, though, is to ensure maximum harmonisation, so it should hopefully be limited. The areas where gold-plating could take place are in Pillars 2 and 3, e.g. over-prescriptive requirements on governance or additional reporting requirements (in theory limited to specific national items).

Bogner, UNIQA: We don´t have the signs pointing to such a development. We expect that the national FSAs will be more for a conservative interpretation than on the gold-plating side.

Do you anticipate much debate on the Level 2 measures? If so, on which topics?

Bala, Generali: Yes, we do expect an intense debate on Level 2 measures. The Omnibus II Directive represents an important step in the right direction, but a lot of work remains to be done on the technical details of the new regime. The main points are those related to the agreed Omnibus II text and, in particular, the so-called long term guarantees (volatility adjustment, matching adjustment). Another topic may be the calibration of long term investments (changes to SCR design or calibration in order to avoid disincentives for long term investments). Other areas could be contract boundaries and own funds (eligibility, tiering, grandfathering).

Hertel, Talanx: Currently we are faced with a threefold regulatory framework under Solvency II, which is labelled as Level 1, 2 and 3 requirements. The different levels are not only related to different details of the topics they treat; they also differ in chronology. For that reason, they are not perfectly consistent, which opens the door to the need for discussions and clarifications. This is also true for the Level 2 documents.

We sometimes really do miss a certain teleological perspective in regulatory discussions.

Circelli, SCOR: The regulators may intend to limit debate over Level 2 measures as much as possible, so as not to jeopardise the challenging timeline of Solvency II application on 1 January 2016. Hence, the regulators may want to limit debates to the same issues that have been debated over the past few years and which have already delayed the application of Solvency II, i.e. long term guarantees and long term investments. Indeed, insurance regulators should address the spillover impact on insurers of the policy choice of prolonging low interest rates. It would be a pity if the regulators did not use the remaining time for discussion to settle other important technical issues. For instance, rather than focusing purely on investment risks, this would be the time to prepare the rebalancing of life insurance from investment risk to biometric risk, by ensuring the reasonable calibration and design of the corresponding modules.

Bogner, UNIQA: We are closely following the debates on the interim measures and in particular here the long term guarantees and the risk free rates. We strongly believe that further discussions will be necessary to address this fundamental topic properly.

How is hybrid issuance likely to develop under Solvency II?

La Rochefoucauld, CACIB: The dust is now settling on Solvency II, which is certainly going to help people to take decisions, even if everything is not yet fixed. We therefore expect capital optimisation to take place and liability management will probably be on the agenda.

In terms of the calendar we are looking at, the European Parliament is expected to vote in early February. Then in May we have the Parliamentary elections, and transposition of Solvency II in each of the member countries is expected to take place early in Q1 2015. Level 2 discussions will take place around Q2, Q3 this year, which will lead to clarification of Tier 1, Tier 2, and new Tier 3, hybrid instruments’ characteristics.

What can we say about Tier 1 structures?

Michael Benyaya, DCM solutions, Crédit Agricole CIB: In the banking space the grandfathering rules have been very strong drivers for capital management initiatives, notably in the form of liability management exercises. As currently drafted, the grandfathering rule is much more lenient on the insurance side. Everything that has been issued under the Solvency I format will be grandfathered for 10 years within the limits of the Solvency II capital structure. This means that the grandfathering rules pose little risk to the total solvency position of insurance companies.

There is also another interesting item regarding these grandfathering rules, which is that the cut-off date will probably be very early 2015, in line with the publication of the Level 2 delegated act. This means that issuers, in theory, still have one year in which they can issue old-style Solvency I sub debt to target a grandfathering in Tier 1 under Solvency II — it’s a little odd and it will be interesting to see how issuers will approach this rule.

But I think that it’s fair to say that for the large issuers that are well established in the market the best practice will probably remain the Solvency II-compliant format, as we have seen recently with Axa sticking to the Solvency II format and not trying to play with these grandfathering rules.

More generally, I think that the overall intention of the regulator is to have a structure that will be similar to the one that exists in the banking sector. And effectively the current Level 2 text shows that the features of the new-style, Solvency II Tier 1 compliant instruments are very close to the bank Additional Tier 1 (AT1). So, in terms of overall structure, I think that the key structuring items are clearly defined.

We are still lacking details regarding the write-up mechanism because in terms of loss absorption the text requires that either we have conversion into equity, which is relatively straightforward, or a write-down of the nominal of the bond, but the question of the write-up mechanism is not addressed in the Level 2 text. That will be part of Level 3, and here we don’t know exactly how the regulator will write this rule.

Our hope is that the write-up mechanism will be simpler and more investor-friendly in the insurance space than the one existing in CRD IV, and it’s clear that insurance companies will probably wait to have the final Tier 1 rules before they can start issuing or considering issuing Tier 1 instruments.

Maybe just one last word on Tier 1 instruments: the capacity for hybrid Tier 1 is fairly limited in the Solvency II capital structure, but insurance companies may yet have an interest in this instrument to lower the overall cost of equity.

With the introduction of capital tiering under Solvency II, are investors likely to focus more on Tier 1 capital?

Bala, Generali: A key driver for investors is the clarity of the rules regarding the structure. Therefore investors currently prefer Tier 2 structures for two reasons: the first is the relative stability of the rules of the current Level 2 implementing measures; the second is the fact that the new Tier 2 de facto corresponds to the “Solvency I-style” Tier 1 and therefore the instrument is well known.

Going forward, investors are likely to focus also on Tier 1 capital as we saw in 2013 with the first bank Additional Tier 1 issuances. Nevertheless, we expect the Tier 1 capital market to be more volatile in terms of windows of opportunity.

Boiral, Amundi: We have always closely monitored subordinated debt, whether “old-style” Tier 1 bonds (when they were still fashionable), or new hybrid corporate debt, especially considering the large issuance of 2013. Usually, a new asset class will offer a premium over its fair value in order to attract new investors, a premium that tends to disappear with the development of the bucket. The Tier 1 capital debt of insurance companies, even if not so new, can offer attractive opportunities for our investors, and we stand ready to grab them in our credit funds. So, focus and be ready to invest!

How relevant are alternative sources of capital in your capital planning?

Bala, Generali: For the future, it is our intention to target a reduction in senior debt in favour of a higher percentage of subordinated instruments, while among subordinated tools there is the intention to rebalance the mix.

This strategy will allow us to achieve our capital targets while maintaining the overall quality of our capital position and reducing the overall cost of funding, thanks to a better mix of different debt capital instruments.

In terms of Insurance Linked Securities, we might consider this market as an alternative source of capital in respect to traditional reinsurance, if attractive.

We do not foresee modifying our base capital structure, which will be represented mainly by equity and subordinated tools. Alternative sources of capital could be used, but will remain marginal.

Hertel, Talanx: One of our key aims is to continuously monitor market developments and possibly increase the number of alternative sources of capital in view of the various regulatory and rating agency capital models. So any additional opportunity is regarded as beneficial.

Bogner, Uniqa: At the current time we are only working with traditional instruments (equity and hybrids — 30 non-call 10), and because of our capital position and the funding we did last year we are monitoring the further development of available instruments but do not anticipate any transaction in the near future.

Circelli, SCOR: Alternative sources of capital are very important in our capital planning. SCOR is an active issuer of insurance-linked securities. In 2013 SCOR issued a mortality bond providing the Group with $180m protection against extreme mortality events in the US. In addition, we are a regular issuer of national catastrophe bonds. Over the last few years we have also established our own nat cat fund in SCOR Global Investments, called Atropos. This fund is open to third parties and allows investors to invest in the reinsurance business through a financial instrument (i.e. Luxembourg SICAV SIF). Through our extensive experience and knowledge of the sector, these funds provide very attractive investment opportunities.

In addition, we are always exploring new, innovative capital sources. In December last year we announced an innovative and cost-efficient three year contingent capital facility, which takes the form of a guaranteed equity line, providing the Group with Eu200m coverage in case of extreme natural catastrophe or life events.

Elbarhdadi, S&P: We have argued that insurance linked securities (ILS) will continue to be a complementary product for traditional reinsurers, for example on the property catastrophe market.

The growth of such an asset class is very much dependent on macroeconomic conditions, and it is confined to certain risk. For example, the historically low interest rates we were mentioning earlier had an impact on this market, with investors turning to this kind of asset class for the higher returns it provides in comparison to more traditional asset classes that suffer from low interest rates. If the situation were to revert back to higher interest rates, investors’ behaviour might change and perhaps focus on more traditional asset classes, which could result in declining volumes for asset classes such as catastrophe bonds.

The new class of Tier 3 is introduced under Solvency II. How much clarity is there on what role this will play?

Benyaya, CACIB: Tier 3 is a fairly new concept in the insurance sector because it did not exist in the Solvency I text, where we had only dated and undated formats. Tier 3 is in essence short dated sub debt with deferral language in relation to compliance with the MCR. The capacity for Tier 3 items in the overall Solvency II capital structure will be limited to 15% of the SCR, according to the latest draft Level 2 text, but for some insurance companies it could make sense to look at this capacity, most notably to manage the interest coverage ratio. This measure is under pressure and it’s fair to assume that Tier 3 items will probably be less expensive, so for some companies it could make sense to look at this bucket in the overall capital structure.

Have insurers been active in the capital markets recently with related issuance and how do you anticipate your activity in 2014?

Bala, Generali: We were active in the subordinated euro market in 2012 with two Solvency II-compliant Tier 2 30 non-call 10 bonds: the first, in July, for refinancing purposes and the second, in December, whose proceeds were used to purchase minorities stakes in our CEE markets.

We decided to elect for the 30 non-call 10 structure in consideration of its “well known” and established structure in the euro market. Both the order books were characterised by a high degree of geographical diversification and with the domination of real money investors which contributed to the success of the deal.

Regarding our funding strategies for 2014, we already issued on 7 January a senior bond to refinance part of our total maturing senior bonds (Eu2.25bn). The issuance, for a total amount of Eu1.25bn, was a six year senior unsecured transaction. The deal was very satisfactory, both in terms of cost for the issuer, with a yield below 3%, and in terms of investor participation, with an order book amounting to Eu9.4bn from high quality investors.

Hertel, Talanx: The last subordinated bond issue was successfully concluded in the year 2012. By 2015 three of the Group’s subordinated bonds will have reached their first call dates. We will need to assess to what extent recapitalisation is going to be necessary. The current capital position of the Group is comfortable, so additional capital issues might be driven more by new investments of the Group in future.

Circelli, SCOR: Our new three year strategic plan, “Optimal Dynamics”, which was presented to our stakeholders in September 2013, includes no external equity or debt funding. On top of this, SCOR is currently very well capitalised. However, we are always actively managing our capital and liabilities, monitoring capital market activities and analysing opportunities in order to further optimise and enhance our capital efficiency.

Bogner, Uniqa: We did a rights issue and a hybrid debt transaction in 2013. No such activities are planned for 2014.

One of the recent capital trades from the insurance sector was Allianz’s perpetual non-call 10 issue. What did you make of this? Are similar transactions of interest to you?

Bala, Generali: We have carefully considered the structure issued by Allianz, and it is definitely a format that is of interest to us.

Hertel, Talanx: From our perspective this has been a very interesting deal, since it was the first perpetual insurance issue in Europe in a longer period and does show that the investors are again being receptive to such issues. Given the developments since 2008, the available capacity of dated issues has been widely used by the industry, including our Group, so there is sufficient unused capacity available on the perpetual side. At the same time, fully Solvency II-compliant perpetual bonds have still not been issued, also due to the ongoing regulatory discussion. Therefore Allianz’s issue is regarded as a good combination.

Circelli, SCOR: As previously stated, we don’t intend to raise debt for the time being. However, it is indeed very interesting to see that there is an appetite for this type of transaction in the euro market. SCOR likes perpetual debt and today is the largest issuer in the Swiss franc perpetual market.

Maxwell, Standard Life Investments: The Allianz deal is an interesting one as it is structurally subordinated to dated Tier 2 debt and perpetual in nature, yet it is likely to be treated within the Tier 2 capital bucket for Solvency II purposes. This raises questions about the rationale for Allianz issuing such a structure as opposed to the more familiar dated Tier 2 instruments.

In general, we are cautious on rating agency-driven structures given the uncertainty regarding the evolution of rating methodologies, especially when the bonds feature rating agency calls. A fundamentally important issue is the pricing of such instruments, as investors are often complacent in pricing certain structural features in strong credit markets. A good example of this — which may also be explained somewhat by scarcity — is the recent insurance dollar denominated deals marketed primarily to Asian investors, many of which were structurally weaker when compared with the equivalent euro or sterling denominated bonds from the same issuer yet commanded a premium valuation. Adverse structural features can increase the risk of moral hazard (i.e. issuers exercising right not to redeem bonds at first available date, or to call at par due to ratings/accounting/tax event) and hamper bond performance, particularly in times of stress.

As insurance regulation evolves we expect increasingly innovative capital instruments to emerge, including structures similar to the Allianz perpetual and indeed Solvency II-compliant Tier 1 bonds. Our interest in such deals will be subject to valuations and meeting our investment process.

Boiral, Amundi: This kind of issue can prove interesting, as long as it is fairly priced on the primary market and it is offering an “honest NIP” (new issue premium), and provided that the issue is well placed and not too large. The NIP is not only a reward for investors; it’s first of all a premium for not choosing the timing to invest in a specific issuer. It’s also a measure to compensate for the given level of interest rates at a certain point in time, imposed on the investor by the issuer. It’s a necessary cushion for both secondary market volatility and the bid-offer spread involved.

Unfortunately, the most recent subordinated deals have tended to be a little overpriced, even if primary books appeared to be very well oversubscribed. Secondary market performance may therefore be disappointing, or at least not in line with the risk taken on these deals: they can be very volatile and strongly affected should the credit market turn bearish. We are ready to take risk, but at a good price!

In the recent Allianz perpetual non-call 10 transaction, the perpetuity element was hardly reflected in the spread, based on where dated bonds were trading. Do you think that is normal?

Maxwell, Standard Life Investments: There are two key structural features which differentiate the recent Allianz perpetual with the existing dated Tier 2 bonds. Firstly, the new instrument is structurally subordinated and secondly it is perpetual in nature. The pricing of such features is often complicated given diverging views across the investor community regarding the premium required. From our perspective the premium we demand depends on how we view the overall credit quality of the company and our assessment of how the instrument’s structure impacts on bondholder risk. For example, assessing the probability of a bond not being called at the first available date and what the implications of a non-call would be on valuations. In general we would demand a lower premium for such features when we have a strong overall credit view and the issuer has demonstrated a consistent debt capital management track record. Looking historically, the majority of high quality insurers have honoured bondholder expectations and redeemed bonds at the first available date. However, with the market largely ignoring insurance call risk at present, it is an important dynamic to follow going forward, especially given the increased use of economically driven calls in European banks and expectation of generous grandfathering provisions for subordinated capital instruments under Solvency II.

Boiral, Amundi: In this kind of trade, the problem is less the perpetuity element than the conditions that may trigger the call. The idea is to evaluate the probability of the call being exercised, and then price the bond accordingly. As a good and instructive example, take the former Tier 1 bank bonds. At the time of the launch of these Tier 1 bonds, at the beginning of the 2000s, the market was deeply convinced that the bond would be called, therefore all issues were priced without taking into account the perpetuity. When it appeared that this would not automatically be the case, the asset class went through a deep correction. This is why we have to be aware and keep in mind that the call is an option, it’s not mandatory. Once this option is correctly priced, perpetuity is not a problem. But I am confident that most investors have learned the lesson, and are now very concerned with call options.

Vincent Hoarau, head of FIG syndicate, Crédit Agricole CIB: When we kicked-off the bookbuilding process for the Allianz perpetual non-call 10 in October last year, the outstanding Allianz 5.625% 10/17/42 dated issue callable in 2022 was trading at around 220bp over the interpolated swap rate (yield to call). Adding a few basis points for the curve adjustment, we came up with a premium of around 40bp for the perpetuity premium, including the new issue premium. We eventually priced the deal at mid-swaps plus 260bp. So you are right: that’s nothing!

But what is valid for Allianz is not necessarily valid for everyone. First of all, it was the first perpetual insurance subordinated deal since 2007. So there was a bit of price discovery around the key element of perpetuity, even if there was an old Allianz perpetual 4.375% 12/29/49 which we used as an anchor in the pricing rationale. But in the end you have to bear in mind that when you are dealing with a transaction for Allianz, either you get it right from the outset on pricing or you go nowhere. The liquidity situation was conducive when we proceeded, there are a lot of positives around the signature, including its outstanding credit profile, strong ratings, scarcity value, etc. So with the right IPTs — we started in the high 200s — and powerful traction from the outset you can build a strong book in a few minutes and leverage off that to compress the NIP and perpetuity premium to the minimum. So the perpetuity premium was not limited per se. We also ended up with a ridiculously low number because the overall new issue process went extraordinary well.

But I agree with Hervé, when you are fine with the call option, and when it is priced in correctly, the perpetuity extension is not a problem at all for investors. Elsewhere, when you look at a name like Allianz the different curves and instruments are really compressed and tend to converge further in a normalising market. At the moment, there is a spread differential of roughly 30bp (including the curve adjustment) between the Allianz 5.625% 10/17/42 and Allianz 4.75% 10/29/49. And this perpetuity premium has been stable since the pricing of Allianz perpetual.

How comparable is demand for undated subordinated instruments from insurers and newly established bank contingent capital trades in perpetual format?

Hoarau, CACIB: Looking at the placement of the Crédit Agricole AT1 in US dollars, we were somewhat astonished by the depth and the breadth of the distribution: a $24.5bn book, nearly 900 different orders and, more importantly, roughly 450 of $10m or smaller. So the granularity of the book was exceptional. Several conclusions can be drawn from that.

Everyone is red hot, ready to take duration risk, spread risk and risk with regards to the nature of the instrument because you can’t find such return on investment anywhere else. But in CoCos the decision to buy is firstly driven by the belief that the trigger will never be breached, while the instrument is offering an extraordinary coupon. For the rest, it’s all about relative value analysis.

CoCo investors are also convinced that issuers cannot fail in primary given the stock of Core Equity Tier 1 they need to accumulate in the coming years. So deals must perform. Therefore there is a lot of inflation in the numbers mentioned above and syndicates must be very careful when sizing transactions and fine tuning the final re-offer yield if they want to protect secondary performance.

I think investors are more sober when looking at subordinated trades issued by insurers. There is less fantasy when it comes to price discovery, the coupons are less irresistible, and the potential for immediate performance in the secondary market more moderate. So you will never see the same degree of delirium and inflation in order volumes. Hedge funds and fast money investors will think twice before getting involved. And when you look at the volatility in the secondary market, you can logically observe a greater stability in subordinated insurance paper.

But in the end, we all know that we are not talking about the same type of instruments. CoCos are loss-absorbing instruments and therefore attract some classical high yield and equity investors in addition to very sophisticated hedge funds that you will never find in an insurance perpetual book. Meanwhile lots of European investors, namely German, are still very reluctant to buy such complex and risky instruments as CoCos. German investors were instrumental in the return of Allianz in euro perpetual format, with 23% of the placement. And I am convinced that they would have offered the same type of traction to a non-domestic insurer. In contrast, they contributed only 2% of Barclays’ placement when it tested the AT1 market in euro format. But the bid from UK investors is crucial in the end. They took the lion’s share of the Allianz perpetual, with 40% of allocations, and drove demand in all the recent contingent capital trades in US dollars as well as in euros.

How interesting for issuers are the levels at which the insurance sector is trading given the compression in spreads?

Bala, Generali: Market conditions improved over the last two years and spreads are currently near their tightest valuations since the financial crisis, therefore the current cost of funding for Generali is at its lowest level since 2008. This is the reason why we decided to take advantage of the positive momentum on 7 January, issuing the six year senior bond. We expect moreover that in the coming months potential tapering in the US and the Asset Quality Review in Europe may drive interest rates higher and increase credit spreads, although we expect the market to remain relatively stable and open during 2014.

Hertel, Talanx: Spread levels are close to all-time lows, this is true. Nevertheless, in a low yield environment, they still look attractive for investors. Consequently, we expect them to remain receptive to new issues.

Circelli, SCOR: The levels are still attractive, with people being desperate for yield and searching for diversification. There is still not enough insurance paper in the market and a lot of investors see insurance as a great diversification for their portfolio. My sense is that they still prefer insurance compared with bank credit, especially when you look at the Tier 1 Basel III type of banking structures. Insurers have proven to be very solid counterparties, as clearly demonstrated during the financial crisis.

What do you think about the current spread situation in the financial sector?

Maxwell, Standard Life Investments: Looking at current valuations, we continue to see value in subordinated insurance paper, in particular, the Tier 2 segment of the capital structure, where we see value versus senior bonds, other financials and corporate hybrids. However, as a consequence of a strong spread performance in 2013 valuations are now less compelling and likely to be more driven by single-name stories. A key concern we will be assessing in 2014 will be signs of insurance credit fundamentals weakening, especially given tighter valuations. Furthermore, spread compression may to some extent be constrained by continued investor concerns regarding low interest rates and the impact of various incoming regulations (such as Solvency II; globally systemically important financial institutions; the common framework to analyse international insurance players) on insurer business profiles and capital requirements.

As European macroeconomic sentiment improves, we would expect this to support further spread compression between core and non-core issuers. However, given the diversity of business profiles exhibited across the insurance debt market we continue to focus on single-name stories where we see fundamental catalysts for credit quality improvements and an individual rationale to drive spread compression.

Boiral, Amundi: At the beginning of 2013, the average spread on the financial sector was much larger than on the industrial sector, by 15%. At the end of 2013, the compression trade has taken effect, and the premium is now closer to 10%. Still, we believe financial issuers remain interesting, considering the improvement of fundamentals for SIFI banks, largely under the pressure of the regulator. In the insurance sector, senior bonds may appear a bit expensive, even if they are an endangered species. We prefer to invest in subordinated securities offering spreads over 200bp with an investment grade rating compared with subordinated bank bonds, which are often cross-over or even high yield.

Looking at the convergence between core and non-core issuers, the development here will most likely be linked to what happens on the government bond market. If you expect Italian and Spanish rates to continue converging towards German rates, core and non-core issuers will also converge. Even if this is clearly Amundi’s view, we have to keep in mind that most of the path has already been covered. The idea is now to focus on the fundamentals of these issuers, independent of their nationality. So, yes, the compression trade between core/non-core issuers should continue in 2014, and on European insurers, too, but idiosyncratic risk could alter the global trend.

How do you balance benchmark sized issuance with private placements?

Bala, Generali: Our actual capital structure is already well diversified in this respect, in fact roughly 20% of our hybrid and subordinated debt has been privately placed. The future mix will be driven mainly by market conditions and by the specific interest of some investors in our debt issuances. However, consistent approach of the Group with fixed income investors will be a key driver for both the benchmark sized issuance and possible private placements success.

Hertel, Talanx: Talanx has established its capital markets footprint in the last 24 months and we see it as primarily beneficial at this stage to increase the number of benchmark sized transactions over time. However, under special circumstances we are also open to reviewing bespoke private placements, which we did, for example, in co-operation with Meiji Yasuda in 2010.

Circelli, SCOR: It depends on the use of proceeds. We don’t issue debt just for the sake of issuing debt. SCOR follows a thorough process with regard to such decisions, by analysing type, cost, size, investor perception etc within the framework of a well-defined governance process. In certain markets, such as Asian retail, it would be absolutely critical to enter the market with a nice benchmark transaction. Such benchmark transactions open doors to new and interesting financial markets.

Thomas mentioned recent insurance issuance in US dollars — how is demand for such subordinated paper evolving?

Hoarau, CACIB: We have not seen a lot of insurers in the primary market in general recently, including in US dollars, but my feeling is that the demand for US dollar capital instruments is going global for insurers, too. The time when Asian private banks were indispensable and provided a bedrock of demand for capital instruments in US dollar format is over. But the Asian bid remains crucial. The participation of Asian investors is just declining in relative terms because the bid from the UK and to a lesser extent US institutional investors is growing rapidly.

This is affecting pricing dynamics in US dollars. Historically, the US and UK investor bases are driven by spreads and relative value, while the Asian bid, dominated by private banks, remains driven by the absolute yield level, deal momentum and overall market conditions. This certainly explains why the Asian bid was fairly volatile in 2013 when yields were declining and market volatility was consistently present. But now that we are seeing more and more institutional investors and hedge funds involved in Asia, I think the demand and pricing dynamic will become more and more homogeneous.

So do you recommend going global and adopting the 144A and RegS format?

Hoarau, CACIB: No. For insurers looking at deeply subordinated instruments this topic has to be approached on a case by case basis taking into account how often they tap the market. The 144A format is very heavy in terms of documentation, costly and time consuming. Meanwhile, in non-vanilla products, you can only rely on a handful of US on-shore institutional investors for tickets in size.

Looking ahead, the issuance of insurance paper will remain limited compared with bank paper and the investment in 144A is not worthwhile for everyone. Together with Asia, Europe can satisfy everyone’s needs in US dollars. And above all, UK investors are essential because of their capacity to buy dollars as well as euros in size. So the answer to the question is not straightforward.

More generally, wherever investors are based around the globe, they don’t want to miss out on the success in the primary market of issuers with powerful name recognition, offering tempting yields and scarcity value, such as core European insurers — particularly when they can enjoy a robust performance in the secondary market.

Across the board, investors are also convinced that European markets are normalising and that European core insurers will never go bust. So when you buy the convergence story you have to buy subordinated instruments issued by insurance companies and offering two or three times the risk-free yield. So for borrowers the liquidity is there across formats. The choice of global or RegS-only format also depends on the aspiration of the issuer in terms of investor diversification.

Is the scarcity effect in the insurance sector playing a role in the relative value analysis? How do you assess fair value on insurers, what are the main criteria you are looking at? Are liquidity and issue size relevant in your investment decision process?

Maxwell, Standard Life Investments: Scarcity often explains some technically-driven anomalies exhibited in specific bonds, particularly within the senior space. However, the effect does not play a major role in our relative value analysis as the situation can easily change, for example as a consequence of capital management actions and new issuance. Our relative value analysis is predicated on a five factor investment process which we utilise: we identify the key drivers of the credit quality, what is changing, what is priced in by the market, why the market will change its expectations going forward, and finally highlighting the triggers. Furthermore, looking at the structure of debt instruments is paramount to identify embedded risks vis-à-vis each other and whether investors are adequately compensated for this risk. This approach allows comparisons across the capital structure, peers and other sectors.

Liquidity and minimum size certainly form an important component in our investment process as this can impact eligibility for benchmarks, fund mandates and bond performance. We apply a higher illiquidity premium for bonds that are sub-benchmark in size or where the issue size is too large as these unfavourable technicals can adversely impact the performance of bonds.

Boiral, Amundi: Insurance issuers’ scarcity explains why this sector appears expensive when it comes to senior bonds. Nonetheless, scarcity is not a sufficient reason for an asset manager to invest in rich paper — especially when an abundant primary season may quickly correct this shortage, as we have seen on utilities, for example. Luckily, the CDS market is less influenced by these technical effects, and allows investors to have a more objective view on the spreads. All-in-all, to assess a fair price on these insurance issuers, we will consider in the first place — of course —fundamentals, followed by the size of the issuance, ideally between Eu500m and Eu1.5bn, the level of the CDS, and then the competitors: firstly, among the peer sector, by activity and country, then we broaden the range of comparables. The last important factor to take into account is the syndicate’s quality, and we have learned that it can sometimes play a larger role, particularly on more complex, very large transactions.

Hoarau, CACIB: I couldn’t agree more. The incorrect sizing of new issues can be devastating in the secondary market. The quality of placement and the level of granularity of the book are also essential to protect performance. So yes, a strong syndicate is a key element, particularly when markets are shaky and issuers need dealers who are ready to put balance sheet to work to support the paper.