The Chairman (Mr J. R. Crispin, F.F.A.): Good evening and welcome. The format of tonight’s event is as follows: First, Andrew Chamberlain is going to give us an overview of his experience in front of the Treasury Select Committee (the TSC) last week. Then we have presentations from two working parties. Dick Rae is going to give an overview from the Retrospective of Solvency II Working Party, and then Andy Rogan is going to give a presentation on the Transitional Measures on Technical Provisions Working Party.
Firstly, then, it’s Andrew Chamberlain, Chairman of the Life Board.
Mr A. J. M. Chamberlain, F.I.A.: The opportunity to give evidence to the Treasury Select Committee on behalf of the IFoA is a relatively rare event. I recall that there have been appearances on behalf of the profession at a select committee but they were more about defending the profession’s role in unhappy situations. This is one of the few times where we have been able to give evidence to a select committee in the role of expert, and to try to give it a better insight into a matter.
First, I think that it is important to understand the role of the Treasury Select Committee. There are 19 select committees in the House of Commons that relate to various government departments and functions. Among those select committees, the Treasury is either the most, or the second most, influential. The Public Accounts Committee is also an influential committee.
As a consequence, it attracts some quite influential and expert MPs. The Committee posed questions including the following: Is Solvency II suitable for the UK? Would we be better off giving up Solvency II and doing something different? What is the importance of equivalence to UK firms?
Perhaps the most surprising question was: Should the Prudential Regulation Authority (PRA) have an objective on competition and encouraging competition in the industry? I had to think on my feet about this and came to the conclusion that it should, the reason being that, I think, the PRA has sometimes lost sight of the importance of what it does for the general public. I think that the competition objective that the Financial Conduct Authority has, and the Financial Services Authority (FSA) had, is quite instrumental in keeping that point of view in place. I would be interested to hear in the discussion from those who disagree with the line that I took. The panel was not unanimous on this: Jane Portas, from PricewaterhouseCoopers, thought it should not; I thought it should; and Phil Smart, from Klynveld Peat Marwick Goerdeler, went halfway between – maybe it should or maybe it should not.
I was impressed by one of the members in particular. Jacob Rees Mogg clearly had a good understanding of his brief on Solvency II, particularly in relation to the risk margin and matching adjustment and the constraints that they produce. He drew the analogy between the role of the risk margin and the fact that a pension scheme can be solvent on the regulator’s basis, but if he wants to buy out, it costs more money. The risk margin, in many ways, performs the same function. I thought that was quite a powerful analogy for him to draw.
I think the two accountants and I sent clear messages that there was a need to make some changes to Solvency II and the constraint on investments for the matching adjustment was one of the biggest issues.
The point I made, based on the research done within the profession’s bodies, was that the life community seem to be much more concerned about the constraints of Solvency II and less so about the equivalence issue; whereas the general insurance community were probably the other way around. The General Insurance Board was concerned about equivalence, but not so worried about reforming Solvency II. But, as I said to the Treasury Select Committee, that is a generalisation and there will be exceptions.
The other point that I made was the way that Solvency II is on a going concern basis. It challenges you to hold enough capital not only to survive extreme events but, through the risk margin, also to have sufficient margins thereafter to encourage recapitalisation to withstand another such event.
The profession’s written submission and my remarks to the Committee invited them to consider whether that was too much; that the cost to consumers through the cost of capital that was passed onto them was too great for a level of security that was very high indeed. Effectively, I described it as the insurance premium on insurance premiums.
I do not want to go into detail about the way the Committee operated, but I think that it was interesting that they were concerned about the role that Solvency II might have on the economy at large. They were worried, as many people of course are, about the constraints on insurers’ investments and money not being available to help the wider economy, and jobs in the wider economy.
They were worried about whether Solvency II, by its very nature, encouraged insurers all to do the same thing, whether that is because of the standard formula, on which they were definitely not keen, or because internal models were going to be similar, which might increase the risk in the system rather than reduce it. Again, I thought that that was quite a powerful argument for politicians to be putting forward.
Perhaps they were wondering what would happen to Solvency II in the future with Britain no longer having a voice at the table. My view was that it probably meant that the matching adjustment would not be high on the agenda. The Spanish use it. They have some sort of product that meets the requirements. I do not know what that product is but I know that it is not an annuity as in the UK. Few other countries use the matching adjustment in Europe. So, I did not think that was going to be high on the agenda after Brexit.
But what else they will do with Solvency II? It will be interesting to hear whether anyone has any views. When they look at something, and what they do about it, can often diverge quite widely before you arrive at the final outcome.
The other thing that was interesting was currency risk. Clearly, someone had made a point to the Treasury Select Committee about the fact that you hold capital in sterling or euros, depending on where you are, for risks that may be denominated in dollars. It is obviously a concern but it was not one that I was expecting of politicians. It seemed to be particularly interesting to the Scottish National Party member.
Mr Fulcher: Dick Rae will now talk about the Solvency II retrospective.
Mr R. A. Rae, F.I.A.: For the past year, I have been chairing a working party that has been conducting a retrospective on Solvency II with a focus on long-term guarantees. I am here to add our thoughts to the debate.
Our working party has been assessing the outcome of Solvency II against its original objectives: improved consumer protection; effective risk management; harmonisation; and financial stability. Our goals have been to look at the success of Solvency II, present our findings for discussion through a sessional paper later this year, and inform the debate for what we thought would be the European Insurance and Occupational Pensions Authority (EIOPA) review.
Brexit, of course, has added a new dimension to the debate causing the TSC to look at European insurance regulation and to consult on its suitability for the UK.
Our working party, along with the practice boards and other relevant working parties, had an input into the profession’s response to the TSC. Steven Graham, an actuary who works within the IFoA, did a superb job of bringing this together in a form on which Council could agree for our President to submit. This was followed by the oral evidence that Andrew Chamberlain provided.
As Andrew has highlighted, the consultation looks at competition, the changes we in the UK would like to see, and indeed whether to leave Solvency II. It mentions the risk margin. It asks about the costs and complexities of implementation and the transitional arrangements. In particular, it picks up the benefits of “passporting” and whether Solvency II is “a price worth paying”.
It is clear that Solvency II needs some fixes but the legislation is more than the calculation of technical provisions and capital requirements. It is about embedding risk management and good governance within the boards of insurance companies and clearly defining where their responsibilities lie. Indeed, Pillar 2 aspects, such as the Own Risk and Solvency Assessment (ORSA), are similar to the Individual Capital Assessment Standards regime we knew and are a huge improvement over Solvency I. By contrast Pillar 3, in our view, needs revisiting. It is unduly onerous and the benefits of the detailed reporting are questionable.
As a consequence, our working party feels that the UK should retain Solvency II as a framework, but take advantage of our ability to adapt it to meet our needs. Without the need to reach agreement with 27 other member states, the UK has greater scope to adapt Solvency II to address its shortcomings.
It is also worth bearing in mind that if the UK chooses to retain Solvency II then, with no voice, Solvency II could evolve in a way that does not suit the UK. As an example, there is no great attachment to the matching adjustment on mainland Europe and the risk is that it could be removed. We should aim to avoid being locked into Solvency II where we have no say and where changes could go against the UK.
Let us have a look a number of the issues and opportunities.
Passporting and equivalence through mutual recognition of the UK’s and EU’s regulatory system is a highly desirable goal. This is especially true for the international nature of business carried out by Lloyd’s of London and in the London Market. For multi-national life companies this is perhaps less important given that the current norm is to operate through local legal entities.
If we change too much, passporting and equivalence could be at risk but these agreements are made between governments. The USA enjoys equivalence with us even with significant differences between our two regimes. Even within our regime, Solvency II does not fully achieve its goal of harmonisation. German companies with long-term with profit liabilities take advantage of an extrapolation to the ultimate forward rate from the 20-year point. They do not compare to UK annuity writers that take advantage of the matching adjustment.
One of the TSC strands of questioning is the extent to which Solvency II is too prudent, principally for business with long-term guarantees.
The TSC made a good point around the fact that, because the PRA has no primary responsibility in respect of competition, allocating more money to back insurance liabilities and their capital requirements is an easy way for regulators to improve policyholder protection. New policyholders will have to pay more.
Andrew [Chamberlain] also made the point in his oral evidence that the level of prudence is a political decision. We would agree. What we can do is highlight the extent of the prudence and the impact this may have on competitiveness.
The point was made in the oral evidence that Solvency II is the most sophisticated regulatory regime in the world. Its foundation for the calculation of technical provisions and capital requirements is market consistency. Its design moved us to a framework that has demanded a true best estimate of the liabilities with a 99.5% 1-year value-at-risk capital requirement.
Omnibus II recognised that, for annuity business, it is default risk rather than spread risk that counts and this was accommodated through a higher discount rate and lower capital requirements. The liability is now linked to the assets that the insurer holds. This break from market consistency now makes it possible in certain instances to be rewarded for taking on credit risk.
At last year’s Life Conference in Edinburgh our working party held a workshop at which we asked whether aspects of Solvency II that were not market consistent should be removed. There were certainly differences in viewpoint as to how much we want Solvency II to be market consistent, although the majority strongly disagreed that the aspects that are not market consistent should be removed. Within our working party, we also do not have agreement.
It raises the question whether market consistency is a useful tool to provide a general approach or is it an end in itself. Or even whether it works at all for long-term guaranteed business.
The TSC session pointed out the flaws of a dogmatic market consistent approach but it is through market consistency that effective risk management is rewarded.
Indeed, there is a concern around the pro-cyclical nature of market consistent approaches and questions are being asked about the going concern approach of Solvency II compared to the run-off approach of Solvency I.
Pro-cyclicality and financial stability are areas where we can inform the debate. Market consistency will always be pro-cyclical. If markets fall it makes sense to hold capital against further falls. For insurers without enough spare capital this can result in their selling risky assets creating pro-cyclicality, with potentially disastrous outcomes. A good point that Andrew made in giving his evidence is that markets can over-react. They are not deep and liquid. Insurers writing long-term business do not have immediate cash flows.
In our workshop at the Life Conference there was agreement that a capital regime that provides time for insurers to formulate a measured response would be desirable. The symmetrical adjustment for equities achieves this but its effectiveness is restricted by the 10% limit on the size of the reduction.
It does, however, give time for markets to find their new level if they have over-reacted and for insurers to formulate their best response if they have not. Variations to this and applying it to other market risks, such as spread risk, is one amongst a range of solutions that could be usefully assessed.
Annuity business is probably the most important long-term guaranteed business written in the UK and, amongst a number of other issues, the risk margin has been thrown into the spotlight by the current ultra-low interest rate environment.
The concern around the risk margin in the UK is about its magnitude and its volatility. This is giving rise to a number of issues and questions. Is longevity risk really non-hedgeable? Cannot interest rate hedging be rewarded? Where did the 6% cost of capital come from? Why 6% when interest rates are so low?
For this there is a range of solutions being suggested. One proposal is to reduce the size of the risk margin by allowing longevity risk to be treated as a hedgeable risk. Certainly the rules ought to be amended so that the hedging of any interest rate volatility is rewarded. There are arguments to reduce the 6% assumed cost of capital, particularly when rates are low. There is, of course, the option to completely revisit the design of the risk margin.
As with many other aspects of Solvency II that need revisiting, there are numerous solutions, each of which warrant individual assessment before a single recommendation can be made. This is a role for more specialist working parties than our own.
Finally, there are compliance constraints and burdens. The need here is to look forward. Whilst there were lessons to be learned from the way that Solvency II was introduced and the costs associated with it, the fact is that the investment has been made and we now have a regulatory regime that is appropriate for a modern insurance industry and current technological capabilities. It also contains many features that the UK would want to retain.
There are number of burdensome and costly areas remaining that would benefit from a review. These are areas that can affect profitability, discourage new entrants and reduce the competitiveness of the industry. For example, in respect of annuity business, investment return opportunities are reduced through ineligibility of assets such as equity release mortgages and restrictions on trading bonds; the consequent need to package assets to meet eligibility rules increase costs and absorbs resources. Tight cash flow matching requirements and other eligibility rules also increase costs.
Elsewhere there are approval processes – such as for internal models and matching adjustments – that could be made more efficient. Currently they are lengthy, costly and burdensome.
Within our working party we question whether internal models are too complex and too detailed for the purposes that they serve. Furthermore, the current process to introduce changes to the internal model potentially inhibits innovation.
Finally, in terms of burdens, there are the detailed reporting aspects of Pillar 3 where the benefits of certain parts do not appear to justify the costs. The whole process of looking through to the assets within funds layered within funds requires huge amounts of processing with its associated costs, and it is not clear how the information benefits regulators or analysts.
Brexit may permit streamlining and simplifying of systems and processes that could make our industry more flexible and more competitive.
In conclusion, a combination of a more financially rational reserving and capital basis along with increased conservatism has improved protection for consumers but at the cost of higher new business premiums.
Solvency II has scored highly in the way that it rewards financially rational behaviour although, in the UK, the matching adjustment has diluted this aspect.
Through the setting of governance standards and aspects such as ORSA, Solvency II has raised the bar considerably.
We should certainly keep the Solvency II framework but take the opportunity of Brexit to focus on making it better for UK needs.
But we should not lose sight of the benefits of negotiating equivalence and passporting with the rest of the EU member states, especially for the non-life sector.
The profession has a role to play in adding its voice as to how the regime can be improved. There are, however, many views and even divergent views within the profession. The IFoA gave its written response. Andrew gave his oral response to the TSC. This meeting is our chance to add our comments to make sure that the profession captures the breadth and diversity of opinion.
The Chairman: Andy Rogan, who is senior manager at Lloyds Banking Group and is a member of the Transitional Measures Technical Provisions Working Party, is going to do the next presentation.
Mr A. I. Rogan, F.F.A.: I am going to give a brief introduction to the transitional. I will call it “the transitional” because “transitional measures on technical provisions” is a bit of a mouthful. I will cover a few hot topics, give an overview of the working party, another overview of the recalculation itself, and then conclude with a summary.
So, what is the transitional? It essentially gives firms a soft landing into the Solvency II regime. It relates to pre-2016 business, as you would expect. The transitional is essentially the difference between Solvency II and individual capital assessment (ICA) technical provisions, allowing for a restriction. That restriction, which can look at all business, essentially, looks at the Solvency II financial resources requirement, as the PRA call it, which is your current liabilities plus your technical provisions plus your capital requirements. You then compare that to your biting Solvency I equivalent of pillar 1 and pillar 2, and that is your restriction, if a restriction applies.
The transitional arises where there are differences between Solvency II and ICA: where we have things that are new or different for Solvency II, like the risk margin, contract boundaries, the difference between a matching adjustment and no matching adjustment and liquidity premiums; or things that only existed under the ICA regime such as Individual Capital Guidance (ICG) and closure expenses.
As at 1 January 2016, the transitional does actually make a considerable difference to UK firms’ average balance sheet. I would guess that is probably the difference between paying a dividend and not paying a dividend.
With it now being January, we are into the reporting season. It is likely that the range of transitionals output and disclosures from firms across the UK will increase. I am sure that there will be a wide breadth of disclosures and that certainly makes for interesting peer-to-peer comparison. It will also be interesting to see just how analysts react to the size of the transitionals specifically, especially for those companies with large annuity books and large risk margins.
A few other things have happened over Christmas. EIOPA has released guidance on disclosures of information around how firms will treat the run-off of the transitional. Transitionals have a 16-year lifetime. Regulations say that you run it off on 1 January each year. That is not quite the same as your annual disclosures that will be up to 31 December. So EIOPA have said that it is acceptable to reduce your transitional on 31 December, and also quarterly, if you so wish. If you do go down the route of running off on 1 January 2017, and annually thereafter, firms are reminded that they must show a true and fair view of their balance sheets and make sure that they show the effect of run-off in their Solvency and Financial Condition Report.
The PRA released a new consultation paper in December on the recalculation of transitionals. It is an update to the Supervisory Statement SS6/16, which was released in May last year, and that covered what is expected from firms in terms of recalculation. The working party are looking at the consultation paper and will be reporting to the IFoA about whether to respond.
There are many helpful clarifying points. It promotes a pragmatic approach to recalculation. There are some points to clarify in the text, but, overall, it is helpful on the very first read.
The PRA expects transitionals to be recalculated every 2 years. That is coming up at the end of 2017. There is potentially more political and economic uncertainty on the way and, perhaps, some more material change. That is reflective of the fact that last year we saw changes in interest rates and in risk margins. After the Brexit referendum, the PRA invited firms to apply to recalculate their transitionals, and many firms did so at the half-year.
There is now some bedding down of internal policies and methods of recalculation; firms have already gone through potentially one recalculation. That has probably been helpful so close to the implementation of Solvency II, where firms have a strong reference point to their 2015 year-end Solvency I balance sheets. How that would move forward in time, the general governance arrangements and how methods will evolve are all things to think about.
Regulations require firms to have a phasing-in plan if their pre-transitional own funds cannot cover the solvency capital requirements (SCRs). Based on the long-term guarantee report, no UK firms had a phasing-in plan on 1 January 2016. That might have changed given the interest rate volatility experienced last year.
The transitionals working party is a short-term one. We first met in July last year. We have a key focus on: the general challenges of recalculation; the options available; where you can be pragmatic and proportionate; what good practice looks like; and the emerging guidance such as the consultation paper that has just come out.
The main outputs will be a paper, which I hope will come out in the first quarter of 2017. We presented at the Life Conference and I felt that there was interest in the subject. It was so good to see members are thinking about it. Lastly, as I have already mentioned, we will be thinking about an input into an IFoA response to a recent consultation paper (CP47/16).
The working party’s areas of consideration include calculation practicalities and managing solvency in terms of how the transitional interacts with your Asset and Liability Management policy and hedging. A further consideration is good practice for communications, as well as looking to the future, in particular as the transitional runs off and becomes a less material aspect of a firm’s balance sheet. That may well support a more proportional approach to recalculation.
Moving on to the transitional recalculation itself, there are some sensible high-level questions to ask. Who is responsible for the recalculation? When do we recalculate? How do we recalculate?
We should have good governance. All firms should have a recalculation policy. A recent consultancy survey suggested not all firms have one but it is something that the PRA expects. There is a particular point about making sure that recalculations and material changes are symmetric. It is important that firms capture that in their policy so that they can demonstrate when they do apply for a change that they are being symmetric.
There is also something about making sure that you have good governance when you decide to make an application. The PRA expect the board to approve the application. The general oversight of recalculation activity monitoring needs to be an ongoing process – not just about whether you hit the triggers for a recalculation but also whether you are required to put in place a phasing-in plan.
Lastly, the audit committee will have to the recalculated transitionals once you have had approval from the PRA to recalculate. They may well get involved in approving the method or policy.
So when do we recalculate? The transitional has a 16-year run off. It runs off smoothly except for a recalculation, and those recalculations can happen in the event of a material change in risk profile, for example, a change in interest rates such as we saw in the first half of the year, or the biennial recalculation. For the former an invitation to recalculate may come from the PRA, and as well as changes in market conditions it could be because of something more structural, like a Part VII transfer or a change in the matching adjustment approval.
I have already mentioned defining material change in risk profile recalculation triggers. It is important that those are well-documented and in your policy. The PRA will certainly look at that when you make an application to recalculate. The latter is highlighted in the supervisory statement that was issued last year.
In the material change in risk profile application process, there is a lag when you apply to recalculate, from obtaining internal approval to submit an application to recalculate through to the PRA approving the application to recalculate, and all the way through to obtaining approval from the audit committee to reflect that recalculated transitional in your Solvency II disclosures.
How do you recalculate? Where do you start? There are many different issues and options to consider. The working party paper will go into these in detail.
The PRA supports a proportionate approach. There is an opportunity here for firms to fit their recalculation approach to how their business is set up and run. Some firms may be able to split their model points or other data at the drop of a hat. Some firms may find it difficult talking about pre- and post-2016 business. There is a whole raft of things to think about in terms of allocating or apportioning risk margin when you are sharing assumptions and correlations and non-hedgeable solvency capital requirements (SCRs), etc. The working party paper will go over all the things that you need to think about and also try to give some real-life examples of how you might address some of those issues.
Pressing issues at the moment are year-end 2016 disclosures and how firms will reflect transitionals and what the reactions will be. There are the 2017 biennial recalculations to follow. The response to CP47/16 is due by mid-March.
To summarise, there are some key questions to be asked both now and in the future. Who is responsible? When to recalculate? How to recalculate? How to disclose? Obviously, recalculations will be firm-specific and should be proportional, including simplifications. The methods will evolve over time and should be kept under regular review.
The Chairman: We will open up with the first question, comment or statement.
Mr D. Brooks, F.I.A.: I am from the Phoenix Group. I am also on the Solvency II Retrospective Working Party. I have a question for Andrew Chamberlain. What is your view on the difference or the consistency of the PRA’s implementation of Solvency II in the UK compared with the rest of Europe? Do you have any concrete examples where the PRA has done something that may be more prudent than the rest of Europe or the other way around?
Mr Chamberlain: It is difficult to say because most of these discussions take place behind closed doors on a one-to-one basis between firms and the PRA. Understandably, firms do not always want to reveal what has been going on in their discussions. The PRA actively discourages them from doing so. But we do have public statements made by the Aegon Group in respect of the volatility adjustment and its changing in different circumstances.
The Dutch regulator has taken a much more liberal view than the PRA. Aegon have revealed that they had to have differences in their internal models for the UK subsidiary and for the group.
It is a concrete example of where the PRA has taken a harder line than at least one European regulator. There are many anecdotal comments about the PRA taking harder lines. But, unfortunately, that is all they are. It is not possible to “evidence” that difference. But there are strong suspicions.
Mr Fulcher: It is probably worth saying here that not a single Dutch company uses the matching adjustment because the Dutch regulator has taken a much tougher line on that. I think I am right in saying that only one Dutch company uses the transitional measures, on which the regulator has taken a tougher line. Without both of these measures, the UK insurance industry would, according to EIOPA’s figures, be insolvent. So it seems to me that the inconsistency can work in both directions.
The Chairman: That point referred to Aegon. That has been our experience of dealing with different regulators and different countries. There are quite different opinions on specific areas, the one that Andrew mentioned being the most public.
Mr K. Foroughi, F.I.A.: I led our firm’s submissions to the Treasury Select Committee. There are a couple of perspectives I should like to share with you and invite comments from the speakers.
The first perspective is trying to work out what a good regulatory framework might look like.
Our suggestion was to consider a regulatory framework from three key stakeholders’ perspectives and then test to see if Solvency II is an improvement from the past and whether it can itself be improved.
The first of those three stakeholders is consumers. Solvency II has required a huge upfront cost which has clearly sunk now. So, firms have spent that money. But the suspicion is that there will continue to be a significantly higher annual cost of Solvency II. Anything that simplifies Solvency II and brings the regulatory burden back down will be much better for consumers, allowing insurers to go back to selling products and reintroduce competition. We have seen areas, particularly in the annuity market, where competition has suffered as result of Solvency II and other factors.
The second stakeholder is the government. One anecdote is that some with-profits funds have sold government bonds to buy swaps to better match the regulatory balance sheet. It is not a sensible regulatory framework where insurers are being incentivised to do that. The only winners are the investment banks, which sell the swaps. So we need a more sensible framework for the economic basis of the balance sheet for with-profits funds.
The third stakeholder is the capital providers. This can be considered from two perspectives. The first is that the insurance industry itself acts as a capital provider for the long-term market. In the early days of Solvency II we have seen more nervousness about the insurance industry acting as capital provider in the long-term asset classes, particularly at a time when the UK government is incentivising infrastructure assets. This is an unfortunate and, I hope, unintended consequence of Solvency II. I would have thought there would be a strong incentive to reverse that.
But also, in certain areas, the insurance industry relies on capital providers to help fund new business. We have seen new business costs go up. But the complexity around Solvency II is putting off many of the more standard capital providers, who do not understand what is going on with the insurance industry.
Some attempt to improve the standard formula would make it more attractive. Let us not forget the vast majority of the European companies are on the standard formula. In the UK there are many companies on internal models and others are being encouraged to be so.
The other thing that can be done on that is to reduce the annual burden of running an internal model. We have seen companies which, once the internal model is approved, have to keep doing extra work year after year to maintain it for the regulator.
The second perspective is the timescale we have to work on an improvement to Solvency II. It is now clearer that we are not targeting the first day of Brexit and that there will be some transitional arrangements for financial services that will probably include the insurance industry.
Do we want a two-phased approach, where we bring in some fairly simple adjustments early on that everyone can quickly agree with and then take more time to consider whether a more significant change to the framework as a whole is in order?
Mr Chamberlain: I will start with the last remark. That is similar to the view that the Life Board took at its meeting when it was formulating its input to the IFoA’s response. We think that to throw everything out any way hastily would be a mistake. Any changes should be well thought through, but there are some areas where early action, simplification and removal of some of the strange effects would be good.
I was taken with the comment about the capital providers not understanding Solvency II. I would go a step further and say that I am not sure anybody really understands how Solvency II works in all the different circumstances. I would venture to suggest that most experienced actuaries in previous regimes had a good idea how the balance sheet was going to respond to any given circumstance. It is somewhat more difficult today. It is so much more complicated. The behaviour, whether it is of an internal model or even the standard formula, can be sometimes counter-intuitive.
I think that it is a weakness of the system that there is less understanding than there was before. I agree with the comments that simplification is needed.
I was interested also at the suggestion that the costs are higher this year than in the past. I am not sure whether that is the fault of Solvency II or just the general regulatory creep, the extra information the regulators always seem to need – I was going to say “year on year”, but it seems more like month on month. I think that probably would have raised the costs anyway. But it is a valid point: why should those costs keep rising? Ultimately, the person who pays is the proverbial man on the Clapham omnibus. I think that it is something that we need to go back and think about. Is this really in the interests of the public?
Mr Rae: I think your comment about selling gilts and buying swaps is interesting. The traditional counterparty on swaps has been banks. If everyone moved to swaps, then you are linking the insurance industry to the banking industry. If your counterparty falls over, then that swap is there to reduce interest rate risk and suddenly you have more interest rate risk back on your balance sheet because the counterparty has gone bust.
Central clearing of course goes a long way to addressing that. But I do not know how far that goes in terms of something that is truly systemic. I think it is a valid point. There is something wrong about discounting generally within Solvency II. Otherwise, I agree with you, I think models are too complex. The 1-year value-at-risk works well for non-life companies because that is the term of their contracts but it does not work for long-term business.
Mr Fulcher: The risk-free rate in Solvency II in its early development was government bonds in the Eurozone. The UK lobbied for it to be the swap rate for the UK. We led the rest of the continent into it.
Mr J. E. Gill, F.F.A.: I have a question about the working party’s desire for the UK to take more control and make changes to Solvency II. Inevitably, over time there will be a bigger divergence between the UK and Europe. Has the working party considered how equivalence will be managed in that environment?
Mr Rae: We have not considered that. The role of our working party is to perform a retrospective. In response to the Treasury Select Committee consultation, we have produced a one-off set of opinions. Whether we take it further and how we incorporate them into our paper has yet to be decided.
Mr Brooks: I think I would have to echo what you have said. We did not go into making definitive proposals for how the UK can fix Solvency II and how we can make that happen in practice.
Mr Chamberlain: As time moves on, the EU will make changes to all sorts of things, Solvency II among them. We are going to have to decide whether we like them or not and whether we want to go along with them or not because we will not be at the table when those decisions are taken. But we may want to make changes ourselves.
The test will have to be whether or not we think that we are going in the right direction for the UK in the broadest sense.
That said, we must not lose sight of the fact that there is the IAIS – the International Association of Insurance Supervisors – which operates under the auspices of the G20 to some extent. That organisation is trying to produce global capital standards and global systems for the solvency of insurance companies. The intention is to produce some sort of level playing field and security for the system at large.
The EU is a powerful voice there, but it is not the sole voice. There are other countries that have systems not the same as Solvency II, such as Canada and Australia. The US is a country that is not quick to move to other country’s systems and is probably slower now than it was a few months ago.
So I think that it would be difficult for the working party to give an answer to that question.
Mr Fulcher: In our response to the Treasury consultation we downplayed that aspect on the grounds that it was for the insurance lawyer, perhaps, rather than the insurance actuary. In other words, what really counts as equivalence or passporting? It is not maybe where we can add value, but it is obviously quite important. I wonder whether we ought to be thinking about it more, actually.
Mr Rae: I think that the divergence will happen for long-term business, but for short-term business or non-life I do not think that there will necessarily be much divergence. If we are taking the cue that Brexit is Brexit, I see no reason why our version of Solvency II would not diverge from the EU’s. We have greater scope to make changes. I suspect that EIOPA has less opportunity to be flexible.
We could, for example, move to a gilt discount rate. That is truly risk-free. The government can print more money. Would that jeopardise equivalence? I imagine that there would be an ongoing political discussion.
Mr Chamberlain: What about periodic payment orders? Should the UK be changing Solvency II for after Brexit to accommodate a better and more realistic treatment of periodic payment order liabilities?
Mr Rae: It is intended that the Directive should pick it up because it does refer to non-life liabilities as well. But there is the fact that the indexation does not meet the more restrictive criteria around the matching adjustment.
Mr Chamberlain: The problem is that you cannot meet the matching adjustment requirements.
Mr Fulcher: The Swiss regime is deemed equivalent. It is different to Solvency II but it is similar in its broad intent. Actually, the 6% risk margin comes from the Swiss but they have various other things that are different. The Bermudans, I think I am right in saying, have a version of the matching adjustment with which we would probably be much happier. It is less binary. Even existing Solvency II equivalent regimes have flexibility. I think that there is quite a big opportunity even with maintaining equivalence.
Miss A. H. Pattni, F.I.A.: I am currently chairing the Solvency II Practical Review Working Party, which is carrying out a review from the non-life perspective. We will definitely be looking at periodic payment orders because of the limited guidance around their treatment.
A common theme is that we will be incorporating Solvency II with a few changes. But when will we have the final decision from the Treasury Select Committee? Will they recommend a list of what areas need to be improved or is this going to come from us? Also, how are the improvements to be managed? Will that be through working parties or is that something that will be considered later?
Mr Chamberlain: The Select Committees of the House of Commons set their own agendas and timeframes. They can take as long or as short a time as they see fit. Much depends on what comes out in the hearings that they have. The hearing in which I gave evidence was the first. There are several others planned.
They produce a report, which is presented to the House of Commons formally. It is effectively a report that the Government reads, and then listens to the reactions and responds to both. That will be the test of what goes on in the future.
I do not think at this stage anyone can predict the timeframe or gain any clarity as to how that would happen. The IFoA can respond now to further questions from the Treasury Select Committee or its staff, and there will be dialogue with them. It will be behind-the-scenes at this stage.
I think it is going to be an ongoing theme for the IFoA over the next 2–3 years. While there will be a transitional period, we believe that has yet to be agreed. But you clearly need to be ready for whatever happens at the end of that period. You cannot afford to sit back and wait on the basis that there is a transitional period. You have to start thinking about what you want to do and how you want to do it as soon as possible. It is inevitable that we will be forming new working parties. Whether they will be under the auspices of the Life Board and Life Research Committee or under the General Insurance Board or they will be joint working parties remains to be seen.
Dr D. J. P. Hare, F.I.A.: There are two quotations going around in my head. One of them is that Solvency II “seemed a good idea at the time”. The other quotation was from Napoleon, who said that “the one thing that man learns from history is that man does not learn from history”.
I feel that much of Solvency II has not really learnt from its beginnings. Early on in the development process came the Sharma Report, led by Paul Sharma at what was the FSA in those days. Gabriel Bernardino was a member of the working party that looked at why insurance companies failed or had near-misses. It came out with a recognition that it is nothing to do with capital; it is to do with what management does. They tell the story in that report of various imaginary near-misses but which are based upon actual events and companies that went wrong. These case studies are still well worth reading today.
A key purpose of that report was to identify the toolkit that the supervisors would need to have in order to operate effectively in the sort of risk-based regulatory regime that was being envisaged for Solvency II. Its conclusions are an important foundation for what became the 2nd pillar of the three-pillar regime. I do wish, though, that even more emphasis had been put on Pillar II in the final implementation of the regime. Instead, we seem to have spent an enormous amount of attention on Pillar I and trying to get the capital right when, in fact, most of us would say that there is no way that you can get it right because all the models are wrong – “all models are wrong, but some models are useful” is one of my favourite actuarial quotes and one which is particularly relevant in this context.
So we have ended up spending a lot of time developing models that are more sophisticated than the data with which we are able to populate them. We have not learnt from history about the importance of Pillar II because we have downplayed that by trying to get Pillar I right.
Where we have learnt from history is in the early 2000s. The feeling then was that banks understood risk and they ran themselves well and insurance companies did not.
Then during the 2000s it turned out that that was not the case. In fact, insurance companies made it through the credit crisis except those that did not hedge variable annuities particularly well. But the banks did not come through the credit crisis very well.
Interestingly, it was the consequences of that that showed one of the weaknesses of Solvency II. Paul [Fulcher] is right to highlight the history. In fact, the use of swap rates for early market-consistent calculations was only adopted by some players as a compromise – they would have like to use a higher risk-adjusted yield above gilts for certain liabilities, but felt that agreeing to the swap rate was the best industry consensus that could be achieved at the time. When the swaps rate went below the gilts rate, this was a problem. Few people would have expected interest rates to fall to where they were.
Of course, any regime that is based on discounting is going to be under pressure when you are discounting at 1% for many, many years.
What it has highlighted is that Solvency II was not as robust as people thought it would be, and that market consistency was not the answer to all the questions in the way it might have been interpreted in the mid-2000s.
Interestingly, my understanding is that the idea of the risk margin came from industry as a way of avoiding the question of what percentage solvency technical provisions should give. The answer was not to think about probability, but, rather, incorporate a margin based on a cost of capital approach. The philosophy was beautiful. The only trouble is we now have a situation where when you move from prudent reserves to best estimate plus the risk margin you end up with a higher number than if you had a prudent reserve to start with.
Unfortunately, the world became more complicated and some of the pragmatic approaches that were being pushed proved not to be pragmatic enough.
One of the aims of Solvency II was to recognise what margins were in balance sheets, and there are advantages in having that sort of level of transparency. But I wonder whether one of the solutions will be that we need to have the same level of transparency over the implementation of the Solvency II framework.
I worry when companies that had an acceptable ICA model supplementing the Solvency I regime now are told that their measurement of capital and assessment of risk is not adequate because the regulator has a model that it is using to test the model against, but that model has not had the public scrutiny that something of such a contentious nature might benefit from having.
I think the industry would benefit if there was more public debate over pragmatic judgements, recognising that we cannot assess all these risks perfectly, but we can do a good job as well-trained risk professionals. But we need to recognise that there is not going to be a perfect answer. We need to have some way of communicating the uncertainty to the politicians and regulators who set our standards. But they need to have the same level of transparency in return, particularly in terms of how much capital is required for certain risks. Otherwise, we end up having a distorted market that nobody feels they can trust.
After all, the whole point of this is so that customers can trust insurance companies. I think the level of distrust that we have now over the capital standards is not necessarily helping that, nor necessarily the value for money that some customers might have as a consequence.
That is not really a question; it is more a reflection from history. However, I would be interested to hear the responses from Andrew Chamberlain and Dick Rae.
Mr Rae: I agree about the complexity of the models. My understanding is that no Big Four company would want to audit an internal model. In fact, I do not know how comfortable the regulator feels, because as a consumer I look to the regulator to make sure that these models are correct.
In my mind, it would be nice to think that we could run the system off of simpler models. The rating agencies have fairly simple ones. The trouble is that the life insurance industry does have a habit of arbitraging these models. It seems incapable of standing back and taking a more measured view as to how to optimise, say, its asset allocation or its asset liability management.
You mentioned prudent reserving. There are two things that have gone on in the last 10 years. One is continually falling interest rates and the other is the introduction of Solvency II. Many of the problems are not to do specifically with Solvency II but to do with the interest rate environment in which we are working.
How prudent would Solvency I reserves be in the current interest rate scenario and what would the regulator’s response be to them?
Mr T. A. G. Marcuson, F.I.A.: I am speaking here as a general insurance actuary. I think that it is important to take a common-sense view.
First, it should be easy to address the things that look wrong; for example, where the long-term interest rates used do not bear any relationship to gilts or actual long-term interest rates. In addressing this, the question of whether to use gilts or swaps is an interesting one, however, having a market-based gilt rate that then seems to diverge from the market rate at long durations does not seem sensible.
Stepping back, I think it is important that we constantly challenge ourselves, given that we all work for the insurance industry (whether, for almost all of us, employed by insurance companies or by consultancies who provide professional services to them), that we do not find ourselves in a position where we are adopting a lobbying position, arguing solely against costs that the industry has to bear, or feels that it has to bear, when these are justified costs from the standpoint of wider society. We risk failing to allow properly for the externalities of weaker standards because of the manner in which these costs are borne.
By way of example, it is easy to argue against the aggregate burden on insurers of requirements to hold excessive capital, put in place better risk management or undertake more onerous reporting standards. But it can be hard to quantify incrementally the cost to society of insurer failure and the potentially very severe consequences that default and insurance market disruption can cause.
So, yes it is right to recognise that some aspects of Solvency II give rise to a cost, but we also need to recognise the wider benefits they individually and collectively contribute, even though they are sometimes difficult to quantify and attribute. We need to make sure we are taking a balanced view and thinking about the broader public interest position.
Mr Fulcher: That came up at the Treasury Select Committee meeting: the balance between consumer choice and value and security. I think, Andrew, you were taking the view that it went too far one way.
Mr Chamberlain: I do personally take the view that it has tilted too far towards security and too far away from value for money for the consumer.
The common-sense approach idea is certainly one that I would support. We need to go back to the principle. The principle is that we want the insurers to pay out claims in full at the time they are due. We want the testing that they have the resources to be able to pay out the claims when they fall due, whether that is in a week’s time or in 40 years’ time. The balance sheet, and the review of the balance sheet and the capital that is held on that balance sheet, is all about whether or not the insurers can fulfil that ultimate objective.
We have become overly hung up on more and more extreme balance sheet tests against more and more extreme balance sheet events and lost sight of what we are trying to achieve. You can end up with a balance sheet that appears to work but you will not be able to pay out claims, and a balance sheet that appears not to work but you can pay out the claims.
The problem with the swaps versus gilts issue, which is a consequence of the way Solvency II is structured, is that it artificially alters investment policy. Anything that does that is not right.
We ought to be focusing on something closer to what the new International Financial Reporting Standards (IFRS) proposal is pushing you towards, which is to look at whether there are assets capable of meeting the liability at the time, adjusting for the risk of those assets. That is the approach we actuaries have been using for years. Solvency II takes you away from that approach.
Omnibus II was called in to get you back towards that approach because they realised that going away from it did not work. Some fundamental changes will be needed. Maybe we should be looking closer at IFRS 17 for some of the solutions.
Mr Rae: You are right, there is the swaps versus gilts issue, but the business that the UK is writing today is principally annuity business.
The only area where the gilts versus swaps aspect comes in is in respect of the risk margin. Otherwise, we have a discount rate based on the assets, essentially along the lines of Solvency I. Is that a fair comment?
Mr Chamberlain: Provided that you jump through hoops to get the right assets that qualify for the matching adjustment, that is true. But again that is another aspect of the problem. There are parts of business where that is not true. Are we saying we want the only products that the life assurance industry produces to be annuities? Are we able to write sensible, longer-term policies that the consumers might need and want under this regime? I would suggest probably not.
Mr I. J. Rogers, F.F.A.: I just wanted to pick up on a couple of points that have come up through the discussion. Paul [Fulcher], you questioned whether actuaries have a role in the legal aspects of equivalence and what equivalence might mean.
And picking up on a point from Dick [Rae]’s presentation, which was about the burdensome nature of the matching adjustment and the eligibility requirements, I contend that actuaries do have a big role to play here in proposing a new regulatory regime and an interpretation of the regulatory regime which works.
By way of example, I point to equity release mortgages. Some would say that the industry has been pushed into securitising books of equity release mortgages. These contain at their heart well-understood demographic risks, which insurers are well-placed to model. To construct matching adjustment consistent default risk out of that demographic risk through securitisation is artificial and unduly burdensome.
However, let us take a look at it from, perhaps, the PRA’s perspective, which is using the legal rigour and the precision that comes from the securitisation structure to ensure that the principles of the regulation are being met.
The alternative would be to use a sort of “fixed enough” kind of approach, where you defined “fixed” as not being legally fixed but fixed subject to some sort of probabilistic test. You can see why the PRA would be sceptical.
I think actuaries, and the IFoA in providing a voice for actuaries as a whole, are well placed to contribute to this and to identify areas in the regulations, as they are currently implemented, and legal rigour is being used to ensure consistency, where there might be an opportunity for a less burdensome approach that relies more on actuarial principles of risk modelling and understanding probability but with the same degree of transparency and consistency that is ensured via a hard legal interpretation of the rules.
I welcome the working party’s work and would suggest that this is an area that would be worthy of further consideration.
The Chairman: I agree that if we do have the options through the Brexit process, we would try to loosen the bureaucratic nature of the application processes.
I remember the discussions when we were responding to the Treasury Select Committee at the Life Board on the challenges around equivalence, almost coming to: “Do you wish for equivalence or not and what do you wish to do?” The fundamental challenge is that we start off being fully equivalent and any change we make will immediately make us less equivalent. That political choice will underpin much of what we push for or desire to do. I am interested in the views of Andrew [Chamberlain] and others.
Mr Chamberlain: You have identified a clear problem. As of Brexit Day, we will be equivalent because we will be on Solvency II and then the clock starts ticking. We then are no longer under Solvency II technically because we will have Brexited, but we will have the same regime. Could we guarantee equivalence?
Of course, there is a slightly political element because the European institutions have to grant equivalence. They might or might not do so or they might take a long time to do so. That will all be part, no doubt, of the Brexit negotiations.
As soon as we then start changing it are they going to use that as a justification for saying: “You are no longer equivalent”.
Whereas, shall we just say, for argument’s sake, we have adopted word for word the Bermudan regime. This is just hypothetical. I am not suggesting that we should do it, but let us assume that we did. Would they then say: “Well, you are not equivalent but Bermuda is?” We just do not know.
Unfortunately, it will be for the European legal system to decide what equivalence means. It is not going to be what actuaries think. It is going to be what lawyers, presumably in the European Court of Justice, think.
Mr A. H. Silverman, F.I.A.: I think it is purely a coincidence that this generally works for the non-life sector, and significant players in the non-life sector would like to have passporting.
I think those two would not necessarily have gone together. Equally, the life sector it does not work for and it is not particularly interested in passporting. Again, it is a coincidence that those happen to have gone together.
In practical terms, what happens with equivalence? You might want to make some changes for the life sector. As you say, you then put equivalence at risk. So how is this going to work? Is perhaps an alternative to simply say there is no point in saying for this purpose we will adopt the Bermudan one? What if you take that argument further and say, “We will have a regulatory regime. It will simply do whatever Solvency II is with no quibble, no influence and no questions asked. But we will have another one as well, which the EU may choose completely to ignore and has, perhaps, no hope of equivalence, and just run the two”.
Mr Chamberlain: Certainly, there were some comments from the Treasury Select Committee that made me think that they were already thinking along such lines. I think it was Jacob Rees Mogg who was alluding to the possibility of such diversions. Equally, there is a precedent going back to the Solvency I regime, where for Swiss companies, I think that I am right in saying that non-life companies could passport, but life companies could not.
So there is another possible model. You will probably find that those handful of remaining composite companies would not be able to passport, but I think that is a price that could be paid. There are not very many of them.
Mr Silverman: I think that the Swiss position is that they do not passport. They just have a bespoke agreement which enables them to do a defined list of things.
Mr Chamberlain: It has the same effect. The word “passport” is not used formally. It is just a colloquialism for what goes on.
The Chairman: The discussion has highlighted that finding a common view on what we want will be as easy as finding a common view on what the country wants as a whole from Brexit. I think that we can help to influence and shape the opinions.
We are going to need to balance the views on whether or not we want passporting and equivalence. The non-life side will have different views to the life side. And then identifying the areas where we want the level of security different to the current levels will be a key part.
I think we all hope that on the Pillar 3 side we might be able to shave off a few of the challenges that we currently have. The downside, of course, is that we have 2 years to go and by then we will have developed all the systems and requirements. Any way of cutting out some of the costs on an ongoing basis will be helpful.
The final thing is to give a vote of thanks to all the speakers tonight and also to all those who participated in the discussion.
Open access
