Treasury Committee
Oral evidence: EU Insurance Regulation, HC 852
Tuesday 17 January 2017
Ordered by the House of Commons to be published on 17 January 2017.
Members present: Mr Andrew Tyrie (Chair); George Kerevan; Kit Malthouse; John Mann; Mr Jacob Rees-Mogg.
Questions 1-73
Witnesses
I: Jane Portas, Partner, Insurance Regulation, PricewaterhouseCoopers; Andrew Chamberlain, Chairman of the Life Board, Institute and Faculty of Actuaries; and Phil Smart, Partner, Head of Insurance and Investment Management, KPMG.
Examination of witnesses
Witnesses: Jane Portas, Andrew Chamberlain and Phil Smart.
Q1 Chair: Thank you very much for coming in to give evidence to us this morning, to take us much deeper into the issue of regulation on insurance. Solvency II is high on our agenda now, with Brexit. May I start with you, Jane Portas? Is the Solvency II regime suitable for the UK?
Jane Portas: Solvency II is the most advanced solvency regime in the world. It is consistent with developments in global regulation—in fact, it is influencing global regulation—and it operates on a framework consistent with broader financial services regulation. The UK market is one of the most diverse, specialist global markets. It is really important we have a regulatory regime that is commensurate with that.
Standing back, some important modifications are required to Solvency II, alongside some refinements. It is very important that the regime is capable of adapting to developments in the UK market and wider developments in global regulation, so as to remain competitive.
Q2 Chair: Do others want to add something?
Phil Smart: I would agree with Jane’s statement.
Chair: You don’t need to—it is about when you disagree.
Phil Smart: However, I think a point to add would be that some of the drawbacks of Solvency II are that they are focused very much on the European market. They don’t necessarily reflect some of the individual characteristics of the UK market.
Q3 Chair: So there are some opportunities to not being in it: what are they?
Phil Smart: I think a number of areas that you might want to look at in terms of areas of improvement are the risk margin and some of the pillar I aspects, which do not necessarily reflect some of the products that are very common in the UK market, such as annuity products; those are less common elsewhere in Europe.
Andrew Chamberlain: I would go even one step further. The principles underlying Solvency II were very similar to those that the former FSA introduced into the UK in the middle of the last decade, about 12 years ago. To that extent Solvency II is on a sound basis.
The problem is that the balance sheet underlying Solvency II has a number of unsatisfactory features that were attempted to be modified through what was called the Omnibus II directive, about two or three years after the Solvency II directive itself was passed. That, however, has produced a rather muddled method. The volatility adjustment, the matching adjustment and the transitional arrangements are all very complex and attempt to put right something that was not working in the original—and, like many of these things, they do not work either.
Q4 Chair: So you are basically saying that Jane Portas might be right that this is a very sophisticated arrangement, but we are probably better off out.
Andrew Chamberlain: I would probably not go quite as far as to say we are better off out, if we can modify, but it comes down to the vexing question of what is equivalence, to which there is no precise answer.
Q5 Chair: We are going to get on to equivalence later in the hearing. Jane Portas, what might a loss of passporting mean for business models?
Jane Portas: There are four key impacts, potentially, as a result of a loss of passporting. The first is cost, because that will require companies to reorganise their business models and potentially to establish subsidiaries where formerly they will have had passport branches. That gives rise to both operational costs and increasing capital requirements.
The second is complexity, because that introduces more matrix-type arrangements into business operating models at a time when there has actually been quite a lot of streamlining of that through Solvency II that has been achieved through passporting.
Thirdly, both those things affect competition between those companies that are operating in Europe on a subsidiary basis, as compared with operating on a branch basis—it is much more efficient to operate on a branch basis than a subsidiary basis; and, finally, it could lead to a reduction in choice for consumers if certain firms decide not to continue to operate should passporting no longer be available post-Brexit.
Q6 Chair: That is very helpful. We are going to come back to that in more detail anyway so there will be opportunities for Mr Chamberlain and Mr Smart to address it. I want just to open up one other question before I pass questioning on. We have had submissions from a number of people suggesting that the PRA should have a full competition objective comparable to the FCA for insurance. What is your view on that, Jane Portas?
Jane Portas: My view is that it is really important that the PRA is held to account for competition, but actually that the primary objective should be policy-holder protection and financial stability. I say that because it can create conflicts of interest if you have it as a primary objective. It is really important that the two—policy-holder protection and financial stability—come first.
Phil Smart: I think there is a danger that, if you don’t hold them to account—not doing it as a primary objective may not achieve that—the regulator is just encouraged to layer further prudence on top of the capital requirements. To a certain extent, that is what we have seen in Solvency II, and that has a detrimental impact on competition.
Q7 Chair: So the consumer is paying more than he or she needs in premiums, to enable the regulator to sleep more easily in his or her bed?
Phil Smart: Ultimately, that can be the effect. Indeed.
Q8 Chair: So we shouldn’t be giving Andrew Bailey such an easy night; well, he has moved on, so it will be Sam Woods?
Andrew Chamberlain: Yes, I am definitely in agreement with what you just said, Chairman. I think the fact that it is a secondary objective has led to its being left behind. I don’t think it is seen as important in the PRA.
We have to remember what this is primarily about: in the case of the insurance, it is much more about the protection of the consumer than it is about the stability of the system, which is more applicable in banking. The price that the consumer is paying now for the level of protection of Solvency II, together with PRA’s interpretations on top, is very high, particularly in some business lines such as annuities. It makes the product so expensive. The balance between security and value for money has got out of kilter.
Q9 Mr Rees-Mogg: Moving on from this to the modelling that is done—perhaps Mr Chamberlain can continue: do you think the standard model should be refined, so that there is less need for insurers to develop their own bespoke, internal models?
Andrew Chamberlain: That would certainly be a laudable thing to achieve. I think the standard formula has a number of rigidities in it that make it impossible for many organisations to operate well with it—particularly in the UK.
For example, the matching adjustment requires a separate fund with a separate calculation of a capital requirement from the rest of the business, which is one of those strange things that seems to happen with no justification. Further refinement of the standard formula to give a better model would be a good thing. However, I still think, in many cases, that it cannot adequately reflect, in any standard basis, the diversity of nature of firms that you will see, particularly in the UK market.
Q10 Mr Rees-Mogg: One of the difficulties with the standard model is when the business is not standard, and it therefore doesn’t make any sense to have the standard model. Does the whole panel agree? I noticed nodding.
Phil Smart: Indeed.
Q11 Mr Rees-Mogg: May I move on to the risk margin, which was a very common issue raised in the written responses we received? If I understand it correctly, the basic capital requirement is to ensure 99.5% confidence of surviving an event over 12 months. That is confidence to survive anything other than a one-in-200-year event. Is that correct?
Witnesses: Yes.
Q12 Mr Rees-Mogg: And the risk margin is then to be ready for something that is even worse than that?
Andrew Chamberlain: The theory behind it is that you could recapitalise after the event, because the risk margin produces a sufficient profit flow to justify the addition of new capital.
What you have is a framework that says, in effect, “After this extreme event, all insurers should be able to continue in business”, whereas previous regimes had taken the view that, after that significant event, there should be a reasonable likelihood of run-off of the liabilities. It sets quite a different standard, and therefore a much higher cost of capital, which is ultimately paid for by the consumer.
Phil Smart: For certain products, the proportion of the total solvency capital requirement, which is now taken up by the risk margin, is upwards of 45% or 50%, so it is a huge proportion of the whole capital requirement.
Andrew Chamberlain: Double the amount.
Q13 Mr Rees-Mogg: What sort of products would those be?
Phil Smart: Typically the long-term products.
Q14 Mr Rees-Mogg: Annuities and so on. What effect do you think that has on the macroeconomy?
Andrew Chamberlain: I think it has a very broad effect. I think everybody is aware that people are finding it more and more difficult to retire. It is expensive to get the income from the sums they have managed to accumulate. The implications of that are quite widespread. I am not decrying people who wish to work longer, but a lot of people are having to work longer, because they have no choice. The implications that has around the economy on the availability of jobs for young people, for the behaviours of those people who are working later, are quite widespread. So it has a significant impact.
Q15 Mr Rees-Mogg: Does it distort the assets that people invest in to meet the risk requirements?
Phil Smart: It does. If you are holding long-term liabilities, typically the sort of asset you would want to hold would be longstanding infrastructure, real estate, assets such as equity release. This actually positively discriminates against holding some of those assets.
Andrew Chamberlain: Can I differ slightly there, because I don’t think the risk margin is the best culprit there? It is the matching adjustment and the structure around it that is very detrimental.
Q16 Mr Rees-Mogg: But the key to the risk margin is the continuation of business.
To get some understanding of the cost, is it comparable with a situation in which you have a pension fund and you try to pay somebody else to take it on, as opposed to trying to meet the capital requirements yourself? The hurdle set by the pension regulator to show that you are fully funded is usually less money than you would need to get somebody to take on the pension liabilities for you.
Andrew Chamberlain: That is a good analogy.
Q17 Mr Rees-Mogg: Thank you, that is helpful. What could be done to make the risk margin more suitable? If you are already ready for a one-in-200-year event, or for anything other than a one-in-200-year event, is it really necessary?
Andrew Chamberlain: I would suggest that whether it is necessary or not is very much a question that should be answered by politicians. The level of security—
Chair: But we like to do that on the basis of high-quality advice, Mr Chamberlain. That is why we have you here today.
Andrew Chamberlain: There is a trade-off between the level of security and the price the consumer has to pay for that insurance—it is an insurance against the failure of the insurance, if you see what I mean. The price the consumer has to pay for that is something that they will not get a choice in. That choice is made by the regulatory system.
One of the things that I think has perhaps been wrong is that regulators have been allowed to make that choice. A regulator asked to make that choice, particularly with the objectives they have, is going to make the most conservative and prudent version of that. That can lead to what, in my view, are excessive costs to the consumer. But that is ultimately a political—small “p”—judgment.
Phil Smart: If your suggestion was to remove the risk margin altogether, I think a consideration that needs to go in there is the question of equivalence of the regime that may result. A removal of the risk margin altogether may be such a step that it could jeopardise the new regime being considered equivalent.
Q18 Mr Rees-Mogg: Yes, and certainly there may need to be differences for our European and non-European business, which I know some insurers are doing anyway to move some of their capital offshore to get round that issue.
Moving on to the matching adjustment, are there any changes any of you would like to suggest as to how that could be made more flexible and less constraining?
Phil Smart: On the point about asset restrictions that I referred to earlier, there are also collateral rules and ring-fencing rules that are very onerous and complex, which make it difficult to apply as well.
Jane Portas: A level of prescription is also required for meeting the matching adjustment, which makes it quite a burdensome process to work through. Perhaps trying to reduce some of that detail is important going forward.
Andrew Chamberlain: The whole thing is quite a bureaucratic structure and I would question whether you need a bureaucratic structure to achieve the end. Previous regimes, including the ICAS regime in the UK, did not require such a structure at all.
Q19 Mr Rees-Mogg: And you cannot necessarily match perfectly the liabilities that insurance companies are taking on because of their duration against the available asset classes to invest in; the weight of investable assets are shorter term than the liabilities that insurance companies are willing to take.
Andrew Chamberlain: Largely it is possible to get the duration, but the matching adjustment also requires a very high degree of certainty on the income. Mr Smart, and then you, talked about the equity release assets and so forth. Those do not have a guarantee of when the money will come out. If you have sufficient of them—using the laws of large numbers—you have an expectation that you will get a cash flow that is pretty reliable, but each asset does not have a guaranteed cash flow so it is not eligible. Those sorts of constraints are the bane of the problem.
Q20 Mr Rees-Mogg: Otherwise you just end up buying gilts—that is all you can do—and that reduces the yield on gilts. It is quite good for Government, because its borrowing is much cheaper.
Chair: Is the retirement problem even worse?
Andrew Chamberlain: The Bank of England holds a certain number of gilts anyway. They are not easy to get hold of.
Q21 Mr Rees-Mogg: So the effect on long-term interest rates is quite profound, from both regulation and what the Bank of England is doing. May I ask about the volatility adjustment? We received lots of written evidence saying that different regulators take different approaches. Do you think that the PRA should be doing what other regulators are doing, or are you happy with its current approach?
Andrew Chamberlain: Myself and my colleagues would take the view that the PRA is being unnecessarily rigid in this area. The volatility adjustment’s main purpose is to try to remove some of the exposure to the cycles; by, in the internal models, declining the ability for the company to estimate the change that would take place in that adjustment, the PRA is removing one of its positive features. It is not a well-designed tool in the first place because it is fairly crude, but taking away some of its effect is a big mistake.
Phil Smart: It is also a tool that needs to be used at times of stress as well. There is a question around whether the clumsiness of it and the way it has been constructed would actually allow the speed of reaction that would be needed.
Q22 Mr Rees-Mogg: How much change could the PRA make to it now, and how much would it not be free to make until we left the European Union?
Andrew Chamberlain: It is well documented in the public arena that the Dutch regulator has already taken a more liberal view than the PRA. As far as I am aware, no action has been taken against the Dutch regulator in regard to that, so there must be some scope.
Q23 Mr Rees-Mogg: We touched on the issue of—essentially, the answers that you are giving indicate that there are real problems with Solvency II and that it could be improved, but that the reason for sticking with it is the worry about equivalence, if we were to make radical changes to it.
Perhaps all three of you could answer briefly—before I get rude looks from the Chairman—about the balance that you think is there. In your opinion, how far can you go in reforming Solvency II to make it better for the industry while still maintaining equivalence? Do you think that there is a great deal of flexibility or is it really quite a narrow path?
Jane Portas: I suppose the first thing that I would say is let us not forget that the insurance industry has invested a huge amount of money in meeting the Solvency II requirements. There are large parts of Solvency II that are entirely appropriate for the market—subject to some reduction of prescription—so we are talking about a specific area here. As regards equivalence—
Q24 Chair: Just to clarify, are you saying that they would, therefore, have made a large part of that investment anyway, even if there had not been Solvency II?
Jane Portas: What I am saying is that there has been significant investment in the Solvency II regime, which is, in broad terms, appropriate for the industry. What we are talking about here are some specific adjustments. It is important that we bear that in mind—not to take away from the importance of this at all. From an equivalence perspective, equivalence requires regulators and regimes to deliver equivalent outcomes—it does not need to be exactly the same. There is a level of flexibility, therefore, that may be allowed through achieving equivalence. If you look at, say, the Bermudian and Swiss regimes, predominantly they are calibrated very similarly but there are some areas of difference. So it is definitely something that can be looked into. Standing back, though, the type of agreement that we might want in a post-Brexit situation may be one that is more aligned to a mutual recognition agreement that delivers similar outcomes, along the lines of what TheCityUK talked about in its report last week, which may give rise to a little bit more flexibility in the regime. I think it is very important that these issues are addressed and that we look at that closely, from both an equivalence perspective and a broader mutual recognition perspective.
Phil Smart: I would say there are elements of the Pillar I aspect of Solvency II that could be modified to be more appropriate for the UK, without actually moving so far away from the current as to jeopardise that. So modifications to the risk margin rather than removal of it would be one of those. Pillar III is another area that has very significant and complex disclosure requirements. We have seen other regimes that have gained equivalence but do not have the same level of disclosure currently required by Pillar III. That is an area that could be quite safely amended.
Andrew Chamberlain: The Institute and Faculty of Actuaries structures itself around the various practice areas: life, which I chair, and general insurance. The two parts have identified quite a difference between the importance of equivalence and the constraints of Solvency II.
Life assurance, generally speaking—and like all generalisations there are exceptions—is less reliant on equivalence and passporting. Non-life insurance is more reliant on equivalence, particularly the reinsurance markets in London, and to some extent on passporting.
There is a difference. In an ideal world, perhaps you would see us less wedded to the structures of Solvency II on the life assurance side and other long-term liabilities and more concerned about equivalence for the reinsurance market on the non-life side.
Q25 Mr Rees-Mogg: That is very important, isn’t it? During the Brexit campaign, Lloyd’s of London made a great song and dance about passporting and its position and then in October said only 6% of its business actually benefited from a passport. So, before we make a regulatory regime that is compatible with the European Union, we need to know whether it is actually important to the business we are doing, or is incidental.
Andrew Chamberlain: That’s true.
Q26 John Mann: So, can we hear the figures then? How much is some kind of bespoke equivalence system going to mean in terms of the business of the industry in totality?
Jane Portas: Two things: first, let’s go back to equivalence. We have got three types of equivalence and equivalence is aimed at creating compliance efficiency for insurance groups and for creating a level playing field. So, in particular, that is where reinsurance equivalence becomes relevant, in terms of allowing non-European reinsurers to be treated similarly to European insurers. Passporting is where we are talking about access to Europe, either through marketing and distributing products into Europe and writing them here in the UK or through establishing a permanent presence in another member state.
Clearly, after the UK leaves—after Brexit—passporting will no longer be available. A bespoke agreement could involve a form of mutual recognition that delivers equivalence but also provides some form of supervisory co-operation that would allow companies in the UK to do business in Europe, with agreements between the various supervisors on how most effectively to supervise that business. But that would not be a passporting type of arrangement.
Phil Smart: It is probably worth noting that the level of passporting from the UK into Europe is pretty much matched by movements back the other way, so this is of interest to the other 27 participants as well. Article 171 of the directive potentially gives provision for a reciprocity agreement, if it is in the interests of policyholders and maintaining the business.
Q27 John Mann: So you are saying that the other 27 countries have pretty much an equal level of business risk if no equivalence agreement of some kind is reached.
Phil Smart: Yes, there is certainly a—
Q28 John Mann: Is the consensus that it is pretty much even-handed?
Phil Smart: Yes.
Jane Portas: Outbound from the UK, there are 220 insurers that are passporting, and inbound, 726 insurers are passporting. To go back to the point that Phil Smart made, there is a framework in the Solvency II directive already that allows for third country branches, and there is article 171, which then allows the codification of a third country regime. Working with the existing provisions within the directive ought to make that a much smoother process than having to make fundamental change to the level 1 text of the directive.
Q29 John Mann: I hear that, but I am asking whether you, Ms Portas and Mr Chamberlain, agree with what I think I heard Mr Smart say, which is in essence that the value of the business coming in is pretty much the same as the value of the business going out. If you are negotiating, it is pretty important to know that.
Andrew Chamberlain: I am not sure that that is what he said.
Phil Smart: I am saying that the degree of passporting from the 27 countries in is certainly significant—it has a level of significance that compares to the passporting coming from the UK into Europe, so it is in the mutual interests of all members to pursue an agreement on this.
Andrew Chamberlain: What is more an issue is in relation to the reinsurance market. While there are significant, large reinsurers in Europe that are benefiting from the UK’s recognition—I think it is unlikely the UK would wish to withdraw from those very well-known names in the industry; the Munich Re’s and so forth of this world—the London end sees a lot more smaller reinsurers operating in the London market, not least those that operate under the umbrella of Lloyd’s. That probably is more significant to the UK than the equivalent effect to European countries. That said, of course, access to that market is of value to the insurers in Europe, and they would probably not want to lose access to the centuries of history and expertise that there is in the UK.
Q30 John Mann: Do you want to see transitional arrangements? Are you presuming that that is required?
Jane Portas: Perhaps I could answer that first. It is very important that there is a period of maintenance and stability as we go through the Brexit change, first because, depending on the outcomes of the negotiations and final arrangements, it will require firms to make some pretty significant adjustments to their business models potentially. In my experience of working with firms over many years to help them reorganise their businesses on a pan-European basis, that is a very complex process. It also involves a whole range of different directives to help facilitate that process. It takes years to make those changes. That is important and alongside that is how our own regime may need to develop in the light of the Brexit arrangements. When you are building a regulatory regime, you build it for the future market that you want to attract and regulate. It is really important to bear that in mind.
The second thing that I would say is that legal certainty is absolutely imperative, ensuring that there is continuity of cover for those policyholders at the time of Brexit and to allow certainty over fulfilment of claims. Both of those are really important as we go through this change.
Q31 John Mann: Will a quick trade deal with the United States have any impact on your industry, positively or negatively?
Andrew Chamberlain: There is an awful lot of transatlantic insurance trade already, which is not inhibited by the lack of some sort of trade agreement. I think it is unlikely to make a great difference. It is already a healthy trading relationship.
Jane Portas: One thing that is important is that the US is entering into a covered agreement with Europe relating to reinsurance. So there is potentially an opportunity for the UK to arrange something similar with the US.
Q32 John Mann: When Lord Turnbull came to this Committee, he was tremendously upbeat about the prospects for the insurance industry. He said of leaving the European Union, “It would actually help insurers if they could leave that arrangement…so…people could come to London and insure—whether its aircraft or factories or satellites—and we may find that we are better placed to offer trade.” Was he over-optimistic in telling us that last year?
Phil Smart: To the extent that Solvency II can be modified and, therefore, provide a regime that encourages competition and encourages the UK industry, that gives a remit for the UK insurance industry to operate more effectively on a global basis. The international markets are seen as very attractive areas that the London market Lloyd’s is moving into. That needs to be balanced with not giving up the rights of access to the European market as well.
Andrew Chamberlain: I don’t know what Lord Turnbull was specifically thinking but I don’t think there is strong evidence that Solvency II has inhibited in those particular fields of insurance the activities of UK insurers. What might be a difference, I suppose, is the ability of the UK to negotiate agreements with those third-party countries that it can’t do today.
Jane Portas: It is important that we remember that the London market is a world-class, global, leading, international market, writing business on a global basis. It does create an opportunity to develop further relationships with other regulators across the globe. The PRA is already in a very influential position with developing global regulation, so leveraging from that is very important to support the international opportunities of the London market.
Q33 Chair: Could you just explain what you mean by leveraging that to support opportunities in the global market?
Jane Portas: The PRA is highly respected and the Solvency II regime is very highly respected internationally. That can be very beneficial for UK firms seeking to establish business in international jurisdictions: regulators know that they are regulated under a Solvency II-style of regime by a leading regulator. That is very helpful for firms entering new markets.
Q34 Kit Malthouse: I wanted to ask about procyclicality. Does the standardisation in Solvency II and the fact that everybody involved has to take into account the same risk factors in accounting for regulatory capital mean that, when there is stress on the economy, all insurance companies across Europe will start dumping corporate bonds and equities at the same time?
Andrew Chamberlain: Inevitably, that can arise. It is most particularly a problem when you have constraints on the investment policy; the matching adjustment, as we discussed earlier in this meeting, has been one of those constraints.
Procyclicality on insurance needs to be thought about very deeply because it usually arises from balance sheet tests. The way in which regulators can or should react in times of balance sheet stress is what actually determines how quickly or otherwise companies would react. The previous UK regimes contained a great deal of flexibility in this regard and lots of loss-absorbing layers were allowed whereby if, for example, as we saw in 2008-09, corporate bond spreads went wildly wide, people were not expected to assume that that all represented the fact that the entire UK economy was about to melt down and every company you have heard of was about to default. Nobody considered that was a likelihood. It was a temporary supply-demand equation within markets.
All the Solvency II regime is based on the idea of market consistency, but market consistency as a concept depends on deep and liquid markets. In times of crisis, markets are not deep and not liquid. How you react to that change is important.
One of the criticisms that one could levy at the way Solvency II has been created is that has a cumbersome mechanism for responding to such crises and it would have to go through a process before something could happen. Previously, the FSA, the PRA or previous regulators could simply have acted overnight and taken action, as indeed happened with the FSA at the time of 9/11.
Phil Smart: That is the real concern: that in that stress scenario you have to get to EIOPA and get a view from all member states before you can respond. It is difficult to envisage how that would work in practice in a difficult position.
Q35 Kit Malthouse: So it is slow and inflexible.
Andrew Chamberlain: And unpredictable.
Q36 Kit Malthouse: And unpredictable. So does that mean it has effectively introduced greater systemic risk? If you are, a, moving slowly and, b, inflexibly or across Europe, all have to move in exactly the same way. Surely there is greater systemic risk.
Phil Smart: The market-consistent elements of Solvency II encourage behaviours which tend to move all within the same direction, so there is an increased risk there.
Andrew Chamberlain: Certainly there is an increased risk around the system; whether it constitutes a risk to the system and therefore the collapse of banks and the like is less certain. What it is likely to do is to create a situation in which some insurers dump stock at silly prices and cause themselves to fail. More likely, you will have theoretical failures because insurers will not sell the stock when the market won’t absorb them and they will be deemed to have failed. Therefore, the risk to consumers will be the disappearance of competition.
Q37 Kit Malthouse: So you have Italian banks getting into trouble and an Italian insurance company seems to be over-weighted in Italian assets and therefore has to mark everything down or dump some stock, that has an impact on stock prices generally, which means all the other insurance companies have to mark down their regulatory capital requirements. That means you might have one or two at the margin which then have to dump even more because they have to convert to cash, which starts a sort of domino effect. Is that what you are saying might happen?
Andrew Chamberlain: If you have a situation where Italian banks are failing, hypothetically speaking, you have a real situation. The danger is when people start worrying that they might fail, prices are marked right down on that basis and that triggers real actions by companies that weaken them themselves and then drive the market even lower.
Kit Malthouse: That could happen without any stress, but just the perception of stress?
Andrew Chamberlain: That is, I think, the biggest risk and the one that needs to be avoided.
Q38 Kit Malthouse: So Corriere della Sera has a huge banner headline, “Italian banks to fail”, and that could trigger this domino?
Andrew Chamberlain: Yes, that is possible.
Q39 Chair: But if you want the markets to be policed for systemic risk, do you not want them to be out there saying, “If you don’t behave yourself, if you let your balance sheet weaken, the price could be penal”?
Andrew Chamberlain: The regulator, in my view, should be looking at the real risks to policyholders. When there is a temporary market dislocation, as has happened several times in the last 20 or 30 years when there is some general perception of crisis, you do not want the failure of institutions to be triggered.
Q40 Chair: I am sorry to follow this up, but the great problem is that you do not know until after the fact whether it is a perception risk or a real one, and judging between the two is very tricky.
Andrew Chamberlain: I think when we look at the lessons—
Q41 Chair: Let me turn this into a question: do you think that the regulator brings any wisdom to the piece, in forming that judgment, that is superior to that of the markets?
Andrew Chamberlain: Yes; the regulator can bring more wisdom than the markets appear to give, because of the way that markets actually seize up in these crises and trading stops, so the prices you are being quoted are not realistic.
Looking back at the lessons of 2008-09, corporate bonds from companies right across the UK spectrum were showing incredibly wide spreads, which indicated, if you take a purist view of markets, a very high probability of all sorts of companies going bankrupt in that period.
The implications, if you believed those numbers, would be that not only would you have very few banks left through which you could have paid your policyholders anyway; they would not have been able to buy anything because the supermarkets would have all gone bust and industry would have all collapsed.
It was not a realistic view of the world to say that those spreads represented the default risks. It is that that is the critical factor in assessing whether companies should ride the storm, so to speak, or whether they should take action.
Chair: Sorry, Kit, I interrupted you.
Q42 Kit Malthouse: The issue is the standardisation across Europe: there is not the flexibility between nations around particular risk factors. It is the standardised approach that causes the problem, and the cumbersome nature of EIOPA.
Andrew Chamberlain: It is the cumbersome nature and the rigidity of the framework which forces the selection of similar assets in the first place. It is not the commonality across countries, because that is a real global phenomenon now.
Q43 Kit Malthouse: We had the Governor of the Bank of England in last week to talk about stress tests. When they do the stress tests is there any spill-over into looking at what the impact on the insurance industry will be? We know from the past that there is an interdependency, but is there any equivalent?
Andrew Chamberlain: The PRA and other authorities do gather information on the effect of particular circumstances on insurers just as much as they do on banks. I am not sure that they use the same factors though. There are different calibrations, which might itself be seen as a little strange.
Q44 Kit Malthouse: The final area I want to ask you about—I don’t know whether you can tell me—is the impact of IFRS, in particular IFRS 9, which is not going to allow you any more to manage credit losses in the way they have been managed in the past; you will effectively have to make a prediction upfront and take the loss on the chin straightaway.
That might, I guess, increase the requirement for regulatory capital, because you will effectively have to make a prediction, notwithstanding the fact that it might be priced into the price you paid for the asset, about whether there will be a further diminution in the value of the asset. Is the industry prepared for it—and does it think it is going to have a big impact?
Phil Smart: The industry is looking at that, but the bigger impact, in terms of accounting terms for the insurance industry, is the impending new accounting standard, IFRS 4 phase 2 for insurers, which is an insurance-specific accounting standard.
Currently, insurers report under a grandfathered version of IFRS, which is essentially the old GAAP with a number of modifications. A new standard is expected early this year for insurers; it will not become effective until 2020 or 2021.
The challenge they are going to have is seeing how that sits alongside Solvency II and other reporting regimes, such as embedded value reporting, and to what extent some of those will fall away and a consistent message given between IFRS 4 phase 2 and Solvency II. There a number of questions out there that remain unanswered in that respect.
Q45 Kit Malthouse: IFRS 4 has not been published yet, has it?
Phil Smart: No. There are exposure drafts out.
Q46 Kit Malthouse: Does it incorporate the same measures as IFRS 9?
Phil Smart: It will work alongside that.
Q47 Kit Malthouse: Is that going to have an impact? Managing things over time, are you smoothing the impact of these defaults? That is quite key to keeping them steady, isn’t it?
Andrew Chamberlain: To a great extent, insurers haven’t benefited in the way banks have from the previous regime—the regulatory regimes pushed it to a market consistent value for market-to-market for a long time—so I don’t think the IFRS 9 changes are significant. When IFRS 4 is—
Q48 Kit Malthouse: To give you an example: say, for instance, that you are holding a load of corporate bonds and there is a big corporate bond default. My understanding is that IFRS 9 means you would have to review the value of your corporate bonds and make a prediction of the likelihood of a default in the next 12 months.
Andrew Chamberlain: Insurers already do that.
Q49 Kit Malthouse: They price that in already?
Andrew Chamberlain: That’s already in the Solvency II regime. It was in the previous regimes as well. It is the assessment of the proportion of the spread that is likely to default that is critical, and, for instance, which the matching adjustment was designed to incorporate some of, but in a slightly clumsier way.
Q50 Kit Malthouse: Okay. So when you combine the two—with the standardisation of Solvency II and that—I don’t think you fully answered my question. I guess what I am trying to get at is whether Solvency II introduces more systemic risk? It is a sort of yes or no, I guess.
Chair: You are allowed to explain why when you give your answer.
Andrew Chamberlain: It certainly must introduce an element of more risk in the system, but I think that is for the reasons all of us gave earlier about the rigidity of the system, rather than the accounting principle and the failure of the system to have a clear mechanism to react to situations where the market is not deep and liquid any more.
Phil Smart: The counter to that is, while there may be an increment there, we have actually got additional prudence sitting elsewhere because of areas such as the risk margin and other elements.
Q51 Kit Malthouse: Right. So the domino effect is more likely, but the system is stronger to deal with it?
Andrew Chamberlain: It is in the sense of being able to pay policyholders at the end of it, because the risk margin is, if you like, extra. However, it actually increases the risk of failures—you fail because you haven’t got the risk margin.
Q52 George Kerevan: It is a reasonable presumption, given the effort everyone has put into Solvency II, not to mention the cash, that there is not going to be a rush to change it significantly in the immediate future. It therefore seems to me that everything comes down to how well the PRA regulates things. Broadly, how do you rate the PRA as a regulator of the industry?
Jane Portas: The PRA, as I have said previously, is one of the most—if not the most—highly regarded regulators in the world. It is highly respected by other regulators.
Q53 George Kerevan: For insurance?
Jane Portas: For insurance and more broadly. However, in implementing Solvency II, the PRA has taken a very robust approach. Having being a regulator myself in my career, I think it is important to understand that Solvency II is a vast regime to implement, and we have a very diverse market, as we have talked about before. So it was absolutely right for the PRA to take a considered and robust approach to implementing the regime.
Also, in the early days of implementing a regime, it is very important to ensure that precedents are thoroughly thought through, particularly given the diversity of the market. That said, there are some areas where there has perhaps been some unnecessary prescription in the longer term, and where there are opportunities to learn from that and refine the approach.
Q54 George Kerevan: Such as?
Jane Portas: Such as, there are additional requirements around approvals—so, very detailed approval processes. There has been an awful lot of work done on internal model approval, which has actually been very successful for the UK: more internal model firms than any other country. That will also now be able to be a huge learning process for the PRA in how to improve and refine as it moves forward, which I think the PRA is already considering.
Q55 George Kerevan: Are you saying that although we now have got the modelling process and it might useful in future, the process of getting there was too laborious, too complex?
Jane Portas: As I said before, we are in the early days of a vast regime. The really key point is that the industry and the regulator have been very successful in getting to this place, but should now learn the lessons from that and make appropriate refinements going forward.
Q56 George Kerevan: Could you be specific about what kind of refinements?
Jane Portas: Some of the processes that have been talked about earlier today: some of the applications around matching adjustment and volatility adjustment; we have talked previously about additional reporting requirements; the internal model process—learning form that; and considering the internal model change process going forward. There are a number of areas where there are opportunities for refinement going forward.
Q57 George Kerevan: Could I come to Mr Smart, picking up on that? What do you think of the PRA’s approach to the level of information gathering that it lays on the industry?
Phil Smart: Certainly, through the internal model approval process, the process proved quite painful in terms of the evidence required by the PRA in order to gain model approval.
I think the PRA would say that they were following the rules of the directive and Solvency II is a very much a rules-driven regime, rather than a principles regime. Therefore, in contrast to insurers in continental Europe where, as Jane says, there is a far smaller proportion of internal models, the UK industry found the whole exercise more painful.
Q58 George Kerevan: Is that because of the scale of the UK industry or was there some element of PRA being over-prescriptive?
Phil Smart: The scale and the complexity of the UK industry were certainly factors. We maybe come back to the earlier question around the competitive objectives of the PRA as well. If their primary objectives are around policyholder protection and stability, they are going to make sure that those internal models are rigorous and robust, and that probably led to the approach there.
Q59 George Kerevan: If the PRA had a slightly more nuanced approach to including competition, giving competition a higher level of support within its policy prescription, would that change things?
Phil Smart: I think you may see some changes in approach over a period of time if that was one of the—
Q60 George Kerevan: Would you like to see the regulator place more emphasis on competition?
Phil Smart: My view is there should be more of a balance. I think if the regulator’s objectives are purely around policyholder protection and stability, you can layer prudence and robustness and never stop, but actually that restricts the ability to write business; it increases prices.
Jane Portas: May I add to that? I agree that it is important that competition is “have regard to”, however it is preferable for the regulator to be held to account for competition, rather than have that in its core objectives. That is going to be important as we progress through Brexit as well.
Q61 George Kerevan: How would that work out?
Jane Portas: There are reviews in which the Treasury hold the PRA to account on competition, rather than interfering with their performance around meeting policyholder protection and financial stability, and retain it as a secondary objective but then review and hold to account.
Andrew Chamberlain: If I could just pick up on that point. The problem is that the PRA can always get out of criticism by yourselves or by the Treasury on the competition side by saying, “Ah, but primary objectives were obviously more important and therefore we weren’t able to respond in that way.”
Chair: To cover that point in more detail: we did take evidence on the competition objective in the last Parliament. I can’t remember which witness it was, but he or she pointed out that when the regulator has a duty to have regard to something, you could be confident that they will have no regard to it. Therefore it has to be elevated to the category of an objective if much notice is to be taken of it. I am not suggesting that I necessarily agree with that, but I just put that on the record.
Q62 George Kerevan: I will continue with that line: let us suppose that in a post-Brexit situation, the European authorities are more flexible in order to encourage the growth of the insurance industry. That would result in less competition in the UK, which must have some impact on the consumer and the product that the consumer is getting. Surely there is interaction between competition and value for the consumer and security for the consumer.
Jane Portas: I agree.
Q63 George Kerevan: You would still hold out for not actually giving a really strong prescriptive drive to the regulator in terms of encouraging competition.
Jane Portas: It is really important that competition wholeheartedly features in the regulatory objectives. It concerns me that conflicts of interest may arise between those objectives. Having it as a secondary objective—but with clear accountability for that—would appear to me preferable to making it a primary objective.
Q64 George Kerevan: I want to pick up on something that Mr Chamberlain said right at the beginning when he made the point that the consumer is paying a price, particularly in annuities, for the very high regulation capital requirement being placed on insurers. Do you have any numbers that would quantify what you mean by the customer paying a higher price?
Andrew Chamberlain: It is very difficult to quantify this partly because—in the light of the Competition Act—getting the information is difficult, and it is also not possible to do a direct before and after comparison, because the rate at which interest rates have moved around over the period of the introduction of Solvency II has been so great that it has made it very difficult to tell. It will be easier to tell as people start to publish numbers at the end of this year as to the amount of capital they seem to be holding, but even that will not give you an identifiable analysis to a particular class. From observing the annuity market, for instance, it is clear that the initial movement in annuity rates was probably of the order of 5% to 10% at about the time that Solvency II came in, but it is difficult to get an accurate figure and I would not want people to go away thinking that that was a scientific analysis.
Q65 George Kerevan: But it does, you think, just give a pointer?
Andrew Chamberlain: Since then it has probably risen because interest rates have fallen further, and the risk margin is highly sensitive to the current level of interest rates, so the problem will have got bigger not smaller.
Phil Smart: It is important to note that it is not just an individual annuity issue. It relates to bulk annuities as well, so it is the buy-out of pension schemes that ultimately has an impact on the corporate sponsors.
Q66 George Kerevan: To pick up a few more technical issues: in looking at the asset capital charges under the standard form, issues have been raised about the treatment of currency risk. In particular, the regime that we now have penalises the holding of own funds in a currency other than a domestic currency. I wanted to explore that, particularly given the currency volatility that we now have post the Brexit decision. Would you like to see some changes in that—in how the regulator views things?
Phil Smart: I think that change there would be beneficial. The rules currently penalise companies for holding assets in currencies that match their local liabilities, so effectively they are managing the currency risk on those liabilities, yet they are receiving a capital charge in respect of that.
Q67 George Kerevan: How would we change that?
Phil Smart: That would be a modification to the rule.
Q68 George Kerevan: That could be done just by the regulator?
Phil Smart: In a post-Brexit world, if we were amending the regime, that would be one of the amendments.
Andrew Chamberlain: The problem is that the system does not recognise that some of the capital requirement is currency-sensitive. If you have to hold own funds in your own currency and you have got dollar liabilities—to get away from Europe for a moment—the capital associated with those is dollar-related. So the natural thing would be to hold some dollar assets against that risk factor. That is not a good thing under the standard format.
Jane Portas: We would agree with that change, too.
Q69 George Kerevan: Post Brexit, the PRA is doing its own thing here. I know it may not be your areas of expertise, but to put some context on this, could you gently speculate on what you think might happen to the Solvency II regime in the EU post Brexit?
Phil Smart: The Solvency II regime will be reviewed in 2018, I believe, so that will be reconsidered in terms of the amendments that are needed. In a post-Brexit world, the PRA would not have a seat at the table with EIOPA, so we would have no influence over that. I guess the challenge for the UK is that, as the regimes diverge, is a point reached where the UK’s regime is no longer considered equivalent? That is a risk.
Andrew Chamberlain: The particular concern would be that in the UK we have a number of products that are very rarely found in other parts of Europe. The matching adjustment is being used by the UK and Spain, and Spain is using it for different reasons from the UK. So there is little likelihood that the changes we would think were necessary to the matching adjustment would ever take place in Europe post Brexit—or post now, probably.
Q70 Chair: Will you explain what the Spanish are doing with the matching adjustment, compared with us?
Andrew Chamberlain: I am not an expert in Spanish products, but they are not using them in respect of the same sort of annuity products that we are. I don’t know how they are doing it.
Jane Portas: If I could add to your point, going forward it is really important that we have a regime that is appropriate to the UK market and the future of the UK market as well. In a post-Brexit situation, we may see further opportunities to develop our market, which may require some broader adaptions to our regime and, on a proportionate basis, across perhaps an even more diverse regime, which would then make it inappropriate to be held to a Solvency II regime based on a European market.
Q71 Chair: Most witnesses—particularly high-quality experts such as yourselves—come with a clear view about what you really want to leave us with. Have you put on the record what you want to leave us with? If not, now is the opportunity.
Andrew Chamberlain: I think I have. As Mr Kerevan pointed out, the trade-off between the price the consumer has to pay for security and the level of security is, I think, the biggest question that has never been properly debated and properly resolved.
Q72 Chair: Premiums are higher than they should be, you assert.
Andrew Chamberlain: Yes.
Chair: Consumers therefore are paying very high premiums in order to ensure that Sam Woods sleeps easy in a five-star hotel; we’ve got the message.
Phil Smart: In summary, it is retaining some form of ability, probably not through passporting but through reciprocal arrangements, to operate in Europe and retaining equivalence, but at the same time driving through some UK-market-specific amendments to the regime that would make the regime more applicable to our market.
Q73 Chair: Yes. I will not explore the evidence again now, but some pretty odd features of this regime have been identified in this hearing, it seems to me. It would be helpful if you provided us with a list of those, to enable us to take that work forward in further hearings. If you think that could be helpful to us, I would like to see it; I am not demanding it.
Jane Portas: I absolutely agree that the UK regime needs to be capable of flexibility going forward. As we enter into the Brexit transition, it is very, very important that we retain stability for our market and for our policyholders and that we have a period to allow for firms to readjust to the new business models they will need to develop, depending on the final arrangements, and to ensure that policyholders are appropriately protected during that period. That is very important on the grounds of competition as well.
Chair: Good. Thank you very much indeed for giving evidence. This is the first of a good number of sessions we will be having on this subject, I expect. It is quite a difficult one in a number of ways, and not one that over the years the Treasury Committee has got so deeply involved in, but I think that will change now, particularly in the light of the regulatory changes made after 2008. Thank you very much for coming in.