The interim guidelines have added urgency to insurers' preparations for the reporting elements of Solvency II. In the first part of this InsuranceERM and Invoke roundtable, participants discuss how they are approaching the challenges
Kevin Borrett, head of risk, Unum
Lee Dobson, Solvency II programme director, LV=
Richard Hart, senior manager, EY
Anne Leslie-Bini, head of international development, Invoke
Roni Ramdin, Solvency II disclosures lead, RSA
John Scott, CRO, Zurich Global Corporate
Chaired by Christopher Cundy, editor, InsuranceERM
Cundy: Are the interim guidelines giving you an opportunity to test out your ideas and develop your processes?
Dobson: We have taken an end‑goal model and are using the interim phase as a testing, reporting and feedback loop during that cycle. The interim guidelines put some external rigour into the internal programme cycle. This is not something that one has to introduce at the beginning of 2016. It is something you need to be ready for a lot sooner than that. Waiting till the last minute is never a good thing when it comes to reporting.
Ramdin: I agree. The problem that has plagued Solvency II activity ever since I started working on it is the fact that pillar 3 has always been seen as something that is going to happen many years from now. Trying to get engagement has always been quite difficult, and all the various delays happening in Europe didn't help my cause. When Eiopa finally came out with these measures, it actually made a few people sit up and pay attention.
Borrett: That is a good point. The interim measures, in addition to being a useful calibration point within Solvency II programmes generally, have really started to shift the focus towards pillar 3 and force firms to examine whether the timing and the interdependencies of pillar 3 with pillar 1 are appropriately aligned.
France and Denmark speed ahead
Leslie‑Bini: There are a number of challenges that French insurers are finding because the ACPR [Autorité de contrôle prudentiel et de resolution] is insisting on the accelerated timeline. The first round of reporting in France is September of this year with an option of either Excel or XBRL. They have relaxed the requirement about XBRL because the timeline was just too compressed.
I think what the French did ploughing on ahead to a certain extent did accelerate the whole impetus that was given to the interim guidelines being applied across Europe. But groups that have operations across a number of countries have to cope with the French exception. How do you incorporate that into a group‑wide project without jeopardising something that is in essence meant to harmonise across Europe?
"When you do see the way things are going you cannot help but be a little bit sceptical, cynical or depressed!" Roni Ramdin, RSA
Ramdin: We do not have any operations in France but we do in Scandinavia, particularly Denmark. What the Danish regulator did from the beginning of this year is what they call the ISR Plus, which is essentially calculating a capital requirement broadly in line with Solvency II standard formula rules within 20 working days of the end of the quarter.
The local regulator is doing that off its own back, which meant that we had to bring in a standard formula tool. We had to do it just for that country, whereas obviously we would like to use a tool that could suit all of our operations across the group in a more systematic and more appropriate way.
Hart: The French regulator, Scandinavia and the Netherlands have been pushing for early reporting for quite a while. Even in the UK, I am not sure that people are going to get the full strategic solution in place with the timeline that we have. It might be a case of, 'What do we need to do to be compliant on day one, that might be a bit more sticky tape, wires, manual processes?' And then, 'What do we do longer term?' I think the French situation, if anything, is probably just a mirror to what the UK is going to go through in twelve months' time.
Cundy: I understand the regulators' approach is not to push the entire group as fast as the fastest regulator.
Hart: Yes, exactly. But the actual information itself still needs to be owned by the local CFO. They are going to be the ones having to explain the numbers.
Cundy: Is it just in pillar 3 that you are seeing these different requirements, or are there elements in the other pillars where you are looking at how to deal with different standpoints from different national supervisors?
Leslie‑Bini: In France they have already have a preparatory ORSA this year. It would be a pity to come this far and see one of the main principles of Solvency II, a level playing field with harmonised rules across the whole of Europe, not actually getting applied the way it was intended.
Ramdin: When I first started working on Solvency II I had this idealised vision of maximum harmonisation. When you do see the way things are going you cannot help but be a little bit sceptical, cynical or depressed!
Hart: I think pillars 1 and 2 are more difficult because of the diversity you see among the regulators. Firms are struggling with the ORSA: is this a 150-page document, is it a 30-page document, what does it need to contain? It is prescribed in headings but exactly what does that mean for your organisation? In pillar 3, at least there is a core which is consistent and then there seems a divergence around that – the other two pillars are more difficult still.
Pillar 3 challenges
Cundy: What have been the biggest challenges of pillar 3?
Scott: Just trying to manage the incredible number of reports that have to be produced at different legal entity and structural levels in the organisation. The worst outcome is when you are forced to produce a whole bunch of numbers which you do not even recognise because your management view might not even directly match the legal entity structure. The biggest challenge is trying to report and genuinely try to run the business on those reports but do it within a structure which does not just slavishly follow the requirements for each individual regulator.
"The biggest challenge is just trying to manage the incredible number of reports." John Scott, Zurich
Dobson: Our main challenge is time – not just to meet the compressing nature of the timetable as you get further past 2016, but for internal consumption of that information. It is our life, rather than non-life business that presents us with the greatest challenges – not least because life business data is typically spread across a number of systems which makes the task of timely and efficient integration of data more challenging.
Ramdin: You can try and break the task up into different chunks. With the quarterly numbers, the challenge is how to get the operations producing the data on an accelerated basis at the same time as all the other things that they have to do. On an annual basis, there is more time but a lot more data to produce. To complicate everything, you then have financial stability reporting which has had the effect of having some annual forms reported quarterly, at least at group level. Then you have the unknowns – like the European Central Bank's (ECB) latest proposal asking for extra data from day one of Solvency II, more than Eiopa is asking for.
Borrett: We do not have some of the additional demands, perhaps because this is one of the areas where as a medium‑size insurer we can benefit from an element of proportionality. First, in terms of governance, we very much focus on things like our data quality and data standards framework to underpin MI [management information]. That then links through to appropriate consideration of that material by the right groups. We manage on two different accounting bases, UK statutory and US GAAP, and have the interplay between those.
The third area where we have made good progress but there is clearly more to do is industrialising MI reporting processes.
Hart: The organisational implications of how you are going to report this are important. Some reports you would typically expect to come out of finance, some out of the actuarial teams, some from externals. There are significant dependencies between reports hence alignment of the data flows, process and controls across the organisation needs to be considered when designing an industrialised process.
Ramdin: We also depend on people, before they submit their data, just checking that this cell here agrees to this cell there, and that sort of thing. We have to get the training right. That is key.
Borrett: I think the importance of a do‑check‑review process in the compilation of MI cannot be underestimated. We are placing a lot of reliance on the ability to use MI to drill down in a big data world; to be able to identify themes that are emerging earlier; to help manage the business better. If the quality of the data is not right then the value of those strategic insights is greatly diminished.
"The importance of a do‑check‑review process in the compilation of MI cannot be underestimated." Kevin Borrett, Unum
Dobson: I guess the biggest challenge is the analytical review aspect too. You can obviously invest a lot of time in making sure you have the completeness checks and the data integrity checks but when you add it all up, does it look like what you expected it to be like?
Ramdin: You are absolutely right. When it comes to the balance sheet forms, it is probably easier to see which are the numbers for the Solvency II balance sheet and which are for the local balance sheet, particularly if you are using IFRS. When you look at, let's say, the A1 form, which is premiums, claims and expenses, it is not a full P&L [profit & loss] – indeed, Solvency II famously does not have a P&L. The comment I get from operations is that these numbers do not make sense. It is very hard to review and compare these to MI.
Cundy: Are you treating the interim guidelines as your dry run?
Dobson: We have not done a dry run yet. We are planning one dry run during this year and at least one more before our deadline for preparatory phase interim reporting in mid-2015. We have been through a few target operating-model considerations, and the issue of build-versus-buy. So we have some confidence that a package solution will give us an accelerated timeline, but dry runs are really important to us.
Cundy: You have done a few have you not, Roni?
Ramdin: Yes, though dry runs can mean different things. At an Ernst & Young event a few weeks ago they made the distinction between dry‑run actuals, as in using all the systems and everything that would be available, as opposed to a mock‑up, which meant having a go at populating the templates. We have just done a process of populating a few of those and we have been doing that every year.
We intend to mock‑up the remainder of the QRTs [quantitative reporting templates] again later on this year. And we hope that is going to be a system test as well. By then we will also have our QRT generator tools to XBRL tool in place as well. The next year the plan is very ambitious. We are aiming to do all our quarterly runs to time next year.
Scott: When you are getting the information from lots of different sources, the challenge is putting it all together in a format where you can then have a commentary, explain what those numbers look like and why they are different from, perhaps, the standard balance sheet or P&L.
Ramdin: Ultimately we have to be able to tell the story behind the numbers, don't we?
Leslie‑Bini: What do you anticipate being the biggest hurdles – the availability of the data or the quality of it?
Ramdin: Timeliness. We know we have the data. We will be able to churn something out. The question is, how quickly? It is going to be very, very tight. We are going to be asking people to do something for the first time at the same time as their BAU [business as usual] reporting, which in their minds will obviously have the higher priority. In fact, we are going to be tinkering with the BAU process as well in order to enable the Solvency II reporting to be done to time. So there are lots of risks, lots of opportunities for things to go wrong.
Cundy: How are other people dealing with the BAU issue?
Borrett: We are fortunate we only operate in one regulated territory. Even ahead of Solvency II, our existing models could produce data very quickly. I do feel a considerable degree of sympathy for those with local group functions consolidating across multiple territories.
Part two of this roundtable is available here