Reflections on Solvency II: Part 1


Professor Karel Van Hulle was in charge of Solvency II at the European Commission from 2004 to 2013. After leaving the post in February of this year he spoke to Gideon Benari, Editor of Solvency II Wire, about what went right, what went wrong and why the industry’s strategy for handling the treatment of Long-Term Guarantees was misguided.

It feels a bit like being part of a film star’s entourage. The brief walk through the lobby of Grand Hotel Parco dei Principi, past the terrace and across the gardens to the poolside café should take no more than a minute. Not this time. I am walking with Karel Van Hulle, the retired Head of the Insurance and Pensions Unit of the European Commission and before we can make it out of the revolving door to the terrace he has already been approached by two excited ‘fans’, pleased to see him and keen to hear about life after the Commission. So am I. It’s the day before the Insurance Europe annual conference in Rome.

Reflections on Solvency II 1

Tomorrow Professor Van Hulle will take the stage as the moderator for the day. Having worked on Solvency II since 2004 his knowledge is second to none. It will be a unique privilege to have a moderator who is as knowledgeable, if not more so, about the subject than many of the panellists. For now we must pilot our way past the claques of senior insurance executives enjoying the post lunch break. We get through the door and he is immediately spotted by a group at a round table on the terrace. Lots of big hellos, hugs and handshakes. This is my true entourage moment. I stand a short distance from the goings on, close enough to be noticed but not enough to be approached. It’s an awkward place to be. A calmer sequence is repeated when we finally sit at a table in the café.

I am eager to hear his thoughts on Solvency II and the regulatory process with the benefit of some hindsight and free from the shackles of official lines and a watchful press office.

“There are very few projects that I remember in financial services where, from the outset, when we made our proposal in 2007, you had unanimous support, from industry, from finance ministries and from anybody from the outside world. So stakeholders, everybody said, this is the right way forward,” he says.

Professor Van Hulle speaks with the conviction of a retired general who knows his war still rages on in his absence. It feels like he is still very much living in the Solvency II world. That “unanimous support” began to wane as problems emerged, especially after the financial crisis showed that under the holistic balance sheet approach of Solvency II (which values both assets and liabilities on a market consistent basis) many firms would not be as well capitalised as they were under Solvency I.

“If you run into problems in the development of the project the voices of conservatism, the guys who say, after all Solvency I was not that bad because the insurance industry survived very well during the financial crisis, start to appear. The solvency margin under Solvency I shows that the insurance undertaking is doing well. That might be very different under Solvency II.”

“But of course Solvency I ignores most of the risk. These people, I would say the conservative people, who are not moving with the way the world is moving, they become stronger the longer the project stalls.” “I have often called them the devils. I have often said, in a regulatory project, what you have to do is keep away the devils in the cupboard. When a project is stalling like Solvency II now, there is a risk that these devils come out of the cupboard. And these devils will not stop. And they cry louder.”

“You can see that people who were saying in 2007 and 2009 that Solvency II was the right approach, now say the opposite because they hope that nobody will remember that they said the opposite. And they do that because unfortunately, in the insurance industry, you also have people who have a short-term view. And that is bad. Insurance is long-term. Insurance is what’s going to happen in the future. It’s not what’s going to happen during the lifetime of a CEO of a company. Yes, these conservative powers, they may indeed try to stall things, just until they resign as CEOs. They might be able to say now that they made their company great. But the day after they resign the difficulties might start to appear.”

Karel Van Hulle 1

Professor Van Hulle does not mince his words. This is a j’accuse to those in the industry whom he believes are unhappy with how Solvency II has exposed their firm’s balance sheet and are now using any technical difficulties to delay its implementation. Despite the forthright words, he is relaxed. Leaning back in his seat, wearing jeans and a check short sleeve shirt as he laments the fate of the process.

To the handful of holidaymakers enjoying the sun blazing down on this tranquil suburb of Rome we must look an odd pair: talking about “devils”, one in jeans and a short sleeve shirt, the other in a suit and tie. The scene turns all the more bizarre when pop music, the kind you hear at any beachfront café in the summer, starts booming out of what seem like unnecessarily large speakers.

Our drinks arrive. A doppio espresso each. It’s Rome. You drink coffee. Never mind the thirty-degree heat. When discussing Solvency II these days, all roads lead to the long-term guarantees debate. The problem whirs around with the peskiness of a mosquito. The size of an elephant. You can’t ignore it and if you’re not careful it will crush you. There is undoubtedly a sense of pride in Professor Van Hulle’s voice when he talks about the achievements of Solvency II. A pride, which is accompanied by a glass-half-full attitude to the process: even the long-term guarantees difficulties.

“One of the big achievements of Solvency II is that we had agreed on how to value insurance liabilities for solvency purposes. And it is the financial crisis that came in between that put into question not so much the way we designed it, that is to say to apply market consistent valuation, but it’s what discount rate you use to value these liabilities.”

The problem is that the debate is often approached from too much of an actuarial perspective, and people tend to look at market consistent valuation as a pure mathematical issue. “I have always argued that if you have a long-term liability, the way you value the liability – considering that you have a 20, 30, 40 year life span – doesn’t matter so much. I mean we need to have an agreement how to do that but I would not lose sleep if it was not what we originally conceived it to be. So I think it is important that we are a bit more pragmatic in that. And that is the challenge that we have now.”

Reflections on Solvency II 3

“Some people say, ‘ignore market risk’. That’s crazy. We should stick to a solvency balance sheet where both assets and liabilities are valued in a market consistent way. That is the basic philosophy of Solvency II.”

This is an interesting time to get his perspective on Solvency II given its odd state of limbo. The framework Directive is in force but Omnibus II, the amending directive, looms large as there is as yet no agreement on the treatment of long-term guarantee products. Everything hangs on the outcome of the trilogue negotiations in the autumn.

Meanwhile, EIOPA has issued some preparatory guidelines for consultation. The whole thing is there but not quite. What’s more, the problem has clogged up the decision making pipeline which has some pretty substantial issues that need addressing. Chief among them are third country equivalence and reporting requirements. A series of smaller yet not entirely insignificant issues are still to be addressed in the Level 2 text as well. The way he sees it the long-term guarantees bottleneck must not spoil the whole thing. “90% of Solvency II is there. Is agreed. So it’s the last bit that we are talking about.”

There’s that glint of the glass-half-fullness again. “So a company that has been preparing itself, that has set up a good governance system, a good risk management culture, is ready for Solvency II”.

He pauses. There is a sense of frustration in his voice. “It’s that last bit, the remaining 10%, that we are struggling with. That doesn’t change the whole thing.”

“Don’t get me wrong,” he hastens to add. “I am not saying that this is not important. I think it is of crucial importance. What I am trying to say is that the basic philosophy is already there.” Despite his somewhat paradoxical attitude to the long-term guarantees he is in no doubt about the importance of implementing Solvency II.

“Today, some insurance [firms] are offering products which they cannot honour because they have been badly designed. Badly designed in a sense that these products do not necessarily take account of the changing, or changed, economic environment. And Solvency I lets them go on with this without sanction. This has got to change.”

Perhaps that is why he is hesitant to accept industry’s argument that it will need more time to prepare. “Under Solvency II more companies have participated in a QIS [Quantitative Impact Study] exercise than ever happened in the banking sector. So you can’t even start arguing that companies were not prepared. Companies were prepared. Because a lot of companies have had to get themselves organised to deliver the data requested in very detailed spreadsheets. These follow-up QIS exercises were indeed practical exercises.”

When the Solvency II Level 1 Directive was adopted in 2009 the implementation date was set for 1 November 2012. During 2011 the date was pushed back to 1 January 2014 and then, when the trilogue negotiations broke down in September last year, the Commission proposed two new alternative dates: 2015 or 2016, depending on the timing of the Long-Term Guarantees Impact Assessment. Today the latter is touted as the first realistic implementation date. As for the actual state of play … take your pick from this smorgasbord of surveys:

  • 24.5%“currently ready to comply” (Moody’s, 2013).
  • 57% expect to be Solvency II-compliant by January 2014” (EY, 2012).
  • 63 % said “their ideal implementation date” was 1 January 2013 or 1 January 2014” (Barnett Waddingham, 2012).
  • “Vast majority” of participants believe they will be ready (if necessary) by 1 January 2014” (Grant Thornton, 2012).
  • 43% “confident” or “very confident” that EU insurance industry can comply by December 2012. (Deloitte, 2011).
  • 78% “feel that their programmes are on track” (KPMG, 2011).

Ongoing delays also increased resistance from industry to prepare (at least as publicly stated). Firms are taking their lead from National Supervisory Authorities (NSAs) that began to take their foot off the gas in autumn of last year. EIOPA then published the consultation on a set of preparatory, or interim, measures in an attempt to keep momentum.

Karel Van Hulle 7

Anecdotal evidence suggests that the interim measures are achieving this objective to some degree. Several NSAs have indicated that firms should adhere to the interim measures until otherwise advised and it appears that many have carried on their Solvency II work but have dropped the consultants and, often, the Solvency II label. Most of the work done is on governance and risk management. Pillar I is awaiting the trilogue outcome on Long-Term Guarantees measures, but the path of most resistance is being taken on reporting and disclosure – Pillar III.

The industry stance is clear: it is better to wait, because how can we report if we do not have the final figures?

Professor Van Hulle is unconvinced. “Well, I am a little bit hesitant to follow that line of reasoning, because one of the problems of the insurance industry is that it is an industry that is not very transparent. And there may be a lot of reasons for that. But you can no longer argue that you are representing a major economic sector which is of crucial importance for the economy, for the people, for the future of our countries and at the same time say yes but why should I be transparent about my importance.”

“It is a huge challenge. I mean put yourself in the shoes of EIOPA. They get a request from the European Central Bank: what is the systemic importance of the insurance industry? They cannot answer the question, because they have no comparable data.”

What he does accept is that where calculations are involved it would make sense to wait. “EIOPA needs to be careful not to impose disclosure requirements before there is more clarity about how to calculate the numbers. And if you talk about stages, that is an area where you could see some stages. Where we can start with disclosing some information and then increase when things become more clear. I would have no difficulty with that. But saying that you cannot be more transparent. That is an argument that I find difficult to hold.”

In the quest to unclog the Solvency II deadlock some suggest holding back on the implementation of Pillar I. “I disagree now with the people who say we have to start with Pillars II and III and then move to Pillar I at a later stage. The people who are saying that are basically saying it took us thirty years to agree on how to value insurance liabilities and we still haven’t managed to agree. That’s not a credible language.”

He pauses, takes a breath and then launches into a string of rhetorical questions. He appears to be genuinely perplexed yet certain at the same time. “Why is it that in insurance we don’t have an international accounting standard? Why don’t we have an agreed international solvency standard? Is insurance so complex? It is complex, but not that complex. So when people now say lets wait another five years, I bet you the same people – after another five years – will have the same arguments again.”

The pop music switches to something that sounds like the theme from the movie Titanic played on pan pipes. It is hideous.

Reflections on Solvency II 5

I ask him what went wrong with Solvency II. To his mind three events crippled the process: the financial crisis of 2008, changes to the Lisbon Treaty and elections to the European Parliament (both in 2009).

The crisis was a largely unforeseen event that was to be the genesis for the long-term guarantees debate. But the Eurozone crisis and the results of QIS 5 that sparked it in earnest were yet to come. “In October 2008, the Commission asked all the stakeholders whether – considering the financial crisis – the Solvency II proposal would have to be changed? And the unanimous response was: ‘no. All problems could be dealt with at Level 2’. And I still continue to believe that this would have been possible. That was the response in 2008.”

The changes made by the Lisbon Treaty had shifted the balance of power in the European Union. It made the Council and the European Parliament equal co-legislators. Now both have to agree on any legislative proposal for it to be passed. From a procedural point of view it gave EIOPA, rather than the Commission, the power to draft the Level 2 text. Previously the Parliament could only accept or reject the text. ”So the Commission is really cut out in that loop. The Commission is considered less relevant than it was under the Lamfalussy approach.”

The Lamfalussy approach, or framework, structures the legislative process into four levels: Level 1 is a principle based definition of the rules, Level 2 provides the technical details and Level 3 and 4 give guidance and ensure enforcement. The Lisbon treaty modified the process with the introduction of the European Supervisory Authorities.

For Solvency II the implication was an escalation of all the key elements of the Directive to the Level 1 text, which now required full agreement by the co-legislators. “That’s the way people have interpreted the Lisbon Treaty,” he says. “Particularly the Parliament, and that’s why you see that a lot of discussion on Omnibus II had to do with who writes the implementing measures at Level 2. And the Parliament has basically said that all major pieces of legislation should go to Level 1.”

When the new Parliament was elected in 2009 it also meant personnel changes in all three institutions, which further delayed the process – a situation which is likely to repeat itself if an agreement is not reached on Omnibus II in the trilogues this autumn.

Professor Van Hulle sees these factors as modalities, and argues that in substance the Lamfalussy process, with its principles based approach, is the right approach for designing regulation. He is therefore cautious of drawing sweeping conclusions about the process itself. “Maybe one lesson we can learn from this experience is that it is very different in Europe to get a major reform organised, if it is of European creation. In a way, the CRD IV came around relatively smoothly because it was imposed by the Basel Committee. But if it is a real European reform it is difficult, because we always tend to quarrel amongst each other. We get bogged down in the details and forget the big picture.“

These three factors combined into a Solvency II-quashing cocktail that resulted in the situation we are in today, whereby politicians are getting involved in what is an essentially technical debate.

He recalls that when the Council agreed their common approach in September 2011 long-term guarantees hardly played a role in the Level 1 text. Only the Counter-Cyclical Premium (which has now been transformed into the Volatility Balancer) featured in the text. The Matching Adjustment (then called the Matching Premium) and the Extrapolation were all left to be dealt with in the Level 2. “We would have no doubt found solutions for that because technically you can find solutions,” he adds confidently.

Karel Van Hulle 3_

But he said that the industry was afraid these measures would not be included in the Level 2 text, over which it had no effective control as it would be drafted by the Commission and it embarked on a campaign to enshrine the Long-Term Guarantees measures in the Level 1 text.

“You can tell them a hundred times that this is not right. It is obvious that it will be in the text. They didn’t trust that. So that’s why they lobbied the Parliament to have the solutions included in the Level 1 text. And then the Parliament said well if it is so important maybe it should be in Level 1. And then we ended up in a situation whereby you discuss what is possibly the most technically difficult area in Solvency II in trilogues. Basically between political people, which makes it very difficult. Because it is a technical issue. Its not the right place to discuss very technical issues.”

He believes this strategy was misguided. “The strategy should have consisted of having these matters dealt with by way of principles in the framework directive and dealing with the details in Level 2. We would have been able to do a few additional QIS exercises and we would have had a solution.”

“But people are the way they are. They have taken these options and approaches and it’s unfortunate. But that’s the way life is.”

In Part 2 Professor Van Hulle shares his views on lobbying, the Long-Term Guarantees problem and a novel approach to supervision.