EIOPA's Implementing Technical Standards on Internal Model Approval - ready to "submit"?

EIOPA have kicked into life at the end of October like Freddy Krueger on laughing gas, releasing a swarm of consultation papers and summaries. While the releases on Colleges of Supervisors or the calculation of the risk-free interest rate will be of interest to some, my own interest was in a subset of their draft Implementing Technical Standards on Solvency II Approval Processes

I have tried to cover the distinction between ITS and the other weaponry in the hands of EIOPA and legislators here. The aggregated feedback received by EIOPA on the suite of Solvency II approvals covered by ITS is neatly summarised in this doc, with EIOPA adding dollops of proportionality into the summary, to effectively remind supervisors not to treat the applications of giants and pygmys with the same vigour. Given the audience reaction to the UK regulator's industry event a couple of weeks ago, I think 'proportionality' may need to become the PRA's middle name (though it probably works better as its first name...)

Clearly, two big moans have been common threads in the responses received, and both have been rebuffed. The concept of supervisory requests for supplementary information "stopping the clock" on one's approval window has been retained, while the idea of "no answer in 6 months equals approval" has been firmly rejected.

Internal Model approval
- "Easy, Easy"

Of more interest to me than the other approvals covered by the consultation is the Big Daddy himself, Internal Model Approvals. From a UK perspective I have covered this in multiple posts previously (here, here, here and here for a start) , and was interested to see if any more whistles and bells had been added to what is already an area causing serious fatigue within the UK's potential applicants, not to mention a few dozen corpses from the c.100 interested parties back in 2011 (slide 4 here).

I thought the following was worth highlighting;

• Given that the Delegated Acts are already loaded with granular Documentation requirements for IM applications (articles 243 & 244), the ITS insists on a few more pieces! Having cross-referenced between the DAs and the ITS, I reckon the items listed in Article 3 d, h, j, p, q and r are all new requirements, though none would be a stretch to produce.
• Expectations that your model has been in use prior to application (Art 2 (3 a ii)), and is integrated into your current system of governance (Act 2 (c)) - perhaps this is the effort which firms with capital add-ons will be compelled to do in 2016 & 2017 in advance of model applications?
• Results of your "last" Validation Process must be submitted in a Model application (Art 2 (m)) - again, suggests there will need to be a year's worth of "live" modelling in advance of making an application
• The supervisors have 30 days to assess the "completeness" of an application - this feels weirdly generous, given those 30 days count towards the overall 6-month assessment window (provided the application is complete).
• Despite the PRA's firm assertion on Oct 17th (slide 8) that conditional approval for Model applications is not available, Articles 3 (8) and 5 (4 b) both suggest otherwise, namely that proposed "adjustments" or "terms and conditions" can be accounted for in their decision. 
• As much prominence being given to Model Change process and Model Change policy as the initial Model Application process - applications can actually be rejected on the strength of the change policy (Art 5 (2)). Despite that, the articles designated for these areas (Arts 7 & 8) are relatively straightforward.

There is no real reason why these ITS for IM approval won't sail through at the Commission, so best to prepare accordingly. However, UK applicants with 2016 modelling ambitions will of course need to wait further for EIOPA's Common Application Package before they can finalise this montagne de papier.

PRA's Countdown to Solvency II Implementation Event - When I Need U(SPs)

PRA Implementation countdown
- "Feel like dancing" now?

So along with the entire UK Solvency II sticky-beak brigade, I had the unbridled pleasure of attending the PRA's 'Leo Sayer' on Friday, confidently titled Countdown to Implementation, despite going on to list a range of matters which have probably given both Internal Model and Standard Formula firms plenty of enthusiasm to try and turn the clock backwards rapidly!

Paul Fisher, Julian Adams' replacement as Executive Director of Insurance at the PRA, kicked off with a set of Solvency II vox-pops;

• Solvency II is the "main game in town"
• "The end is in sight"
• The PRA are "not gold-plating the Directive"

A nice bit of reassurance at kick-off time then - unfortunately, the clarity which was to follow from the technical specialists on a range of topics was probably not what everyone wanted to hear, given the tone of some of the audience questioning that followed! In particular (and with everything in full quotation marks below having been said by a PRA rep on the day);

• That the PRA will "have to prioritise" if everyone in the IMAP queue for 'from 2016' approval drops their applications in at the end of June, and that applicants should be "striving to submit" sooner. Suggests to me that the smaller firms in IMAP may get bumped to squeeze in the big boys.
• The reinforcement of Mark Carney's message from the other week that they will have no problem refusing permission to use models, though this was couched by Fisher in the more appropriate context of failure to meet any of the TSIMs simply "cannot be allowed".
• That, although over 90% of the UK industry was down for using Standard Formula, the PRA will be equally aggressive when challenging their preparedness as they will for IM firms.

I didn't hang around for the afternoon sessions as I had a hot date with BA, so pull whatever you can out of the Other Approvals and Regulatory Reporting slideshows. However, I would draw attention to the following from the earlier sessions:

Implementation and Policy overview

• The Old Lady of Threadneedle Street protested through multiple speakers about how they recognise that bank and insurance internal models are different, and how Insurance Supervision is now "embedded into the Bank" - I would assume to justify the credit crunch-influenced aggression now being taken by the PRA on assessing capital models (visible from Adams's speech around the time of BOE/FSA merger right through to Bailey's speech at Mansion House on Thursday night).
• An interesting slide 3 about how much more work Solvency II will generate for the PRA from go-live!
• Referred to FLAOR only once before switching to ORSA, suggesting that this disposable acronym is as much of a pain in the neck for the supervisor as it is for firms!
• Highlight what "Good" and "Bad" ORSAs have contained to-date. Clearly some firms are box-ticking, leaving an unusable report which is only skin-deep compliant as output. They were particularly scathing on Stress and Scenario Testing efforts, and implored that Reports should not be written for the PRA's benefit (though naturally must cover what the DAs and EIOPA have already set out).
• A nice piece was discussed around the expected depth of director-level knowledge of their internal models. Andrew Bulley made a useful distinction between "conceptual" and "technical" knowledge, where your dithering 80 year-old INED from the fishing industry might not be expected to understand correlation matrices, but should probably know their significance, and alternatives to them.
• For model applications to date, far too big ("encyclopedic" in cases), with too much process description, and not enough on assumptions and expert judgements.
• That it is a firm's responsibility to "ensure compliance" with the Delegated Acts, and given their lessening proximity to the legislators, the PRA flag in advance that they will not be able to give "concrete advice" to firms in future.

Standard Formula

A particularly good ground-setter, given the dearth of work published previously on Standard Formula firms and the PRA's expectations. Calendar included on the slide pack, which will be of massive use to your PM/PMO staff!

• PRA will review ALL firms ahead of 2016 for SF appropriateness - "priority" firms assessed by Q1 2015, everyone else by end of 2015.
• While SF SCR is apparently close to current ICG numbers for GENERAL insurers in the UK, it is noticeably larger in LIFE firms 
• PRA is "not promoting" SF ahead of internal models
• Vigorously directed all attendees to EIOPA's Underlying Assumptions of SF Paper - expectation that some firms won't read it, and just expect SF plus any add-ons?
• Very vocal on the "significance of the deviation" between SF SCR and one's own Risk Profile - as we know, the Delegated Acts quantify what 'significant' is in the context of capital add-ons
• An expectation that ORSAs will be used in the assessment of SF appropriateness - qualitative for sure, possibly quant elements as well?
• Range of examples of where significant divergences are being found, by Risk Category, and by Life vs General Insurer.
• A lot of emphasis on Capital Add-ons being used "only as a temporary measure", which will ultimately allow firms to PIM/IM or de-risk. They are however "patient and realistic" on how quickly that change can be done, so it sounds on the face of it like 2016 and 2017 will be targeted for Capital Add-on elimination.
• In the following session on models, a piece came up on Capital Add-ons, where the PRA confirmed that their process for handling these in future is "still developing", though they expect them to be "a lot, lot rarer" than the existing regime of ICG.

Internal Models

Calendar also provided for IM firms (slide 4), showing how tight their schedule is, and explaining why they threw the earlier curveball about all firms expecting to drop their applications in on 30th June 2015. They also touched on the following:

• Highlighting weak areas identified to-date such as over-optimistic (new?) business plans being used in capital calculations; ENIDs; omission of certain "Key Risks", and suspiciously low correlations
• That Use Test is "fundamentally important", and is an opportunity for firms to "put their money where their mouth is". They do not expect to see either end of the use spectrum i.e. no use, or blind use!
• Too much technical actuarial validation seen. Usefully suggested that validation questions may be better posed as "where might this model be inadequate", rather than "why is it OK".
• Confirmed that the PRA's SAT has now been replaced by EIOPA's CAT, which won't arrive until the back end of this year - surprisingly, no-one laughed when they said this "might create some work" for existing IMAP programmes!
• Importantly, they stressed that their powers are to Approve or Reject applications, with no "conditional" powers whatsoever. Attendees were therefore encouraged to delay applications which were thought to be unlikely to succeed, both now and in future.

Though they instinctively feel like they could have been supplied sooner, the clarifications provided in the presentations I witnesses will be hugely welcome by programme directors and PM's alike. I would venture a guess that they will be less welcome by executive committees, who may have hoped for more flexibility on the risk quantification front post-2016.

Accenture Insurance Sector research - Global Risk Management survey

Flood Risk?

A nice generic risk management benchmarking piece from the guys and girls at Accenture came out this week, and after I spent last week at the Leicester rugby game, I was happy to see another 15 "tigers", albeit this time scattered throughout the survey paper itself, presumably as a subtle metaphor for "death by tiger" risk...

It is made up of 98 C-suite respondents (nicely spread across disciplines), is Insurance sector-specific, and Global in coverage (one-third Europe, half N.America), so should be useful to any reader for trend-spotting and Board briefing.

From the document itself, I've pulled out the following;

Risk Governance

• 98% have their "risk management owner" reporting to the CEO
• 96% have a senior executive (regardless of title) as "risk management owner"
• 80% have their "risk management owner" report regularly to the Board
• 55% had a titled CRO
• A number of those stats (whilst improved since their last survey) are a poor reflection on the Global insurance industry, but perhaps reflect where corporate culture is outside of the EU/US axis
• Of the governance bodies, I was surprised to see only 60% of Life companies have an operational risk committee

Solvency II/Non EU equivalent legislation-specific

• Over 80% of Life and P&C respondents seem happy that they are preparing well for their regulatory initiatives (Solvency II or local equivalent).
• Other than Internal Model development, the main outstanding issues for Life insurers to be prepared for Solvency II/equivalent is IT architecture and Data Management/Integration. For P&C, documenting risk processes and developing a meaningful Use Test are also worrying at least half of respondents.
• Issues such as training and education, risk culture and risk governance documentation are relatively low on the priority list.
• Conversely, when asked on a 1-5 scale about specific areas of risk governance, respondents were more positive about their Data preparations than their risk governance - go figure!
• Use Test preparation remains a laggard throughout.

Generic

• Top external pressure was Legal risk, and by a good distance. Regulatory risks relatively low on the list, perhaps reflecting Europe's low weighting in the quantum surveyed.
• Risk Management seemingly well integrated with strategic deployment, but not with product development or reward.
• Poor statistics around embedding risk management into core functions.
• Two thirds of Life respondents noting that a lack of "early warning capabilities" impedes emerging risk management.
• Over half of Life companies said investment benefits ("above and beyond" continued compliance with regulations) would come from better reporting and better integration of Risk and Finance.

There is some of the softer stuff on aspirational elements of risk management thinking at the back, but if you just want to check against your peers, you can save that for a rainy day.

Pre-application for Internal Models - EIOPA's FINAL preparatory guidance for national supervisors

So the March consultation document for internal model pre-application brought a few eye-openers for those countries partaking in a less onerous application process than that favoured by the UK, with the detail in it suggesting that the UK very much had the whip hand in its drafting.

On the basis that there were still areas which even the most hardened IMAP-veteran may have winced at, it was interesting to see if anything got dropped in the lobbying stampede. On that basis, the final guidance for internal model pre-application covers the following in the preamble;

• 3.10 - That it is not in the NCAs gift to conduct pre-application preparation along the lines of provisional approval or to provide "roadmaps" to compliance (which may explain the PRA's caginess with the industry). It is purely about a firm's preparedness and suitability to submit an application
• 3.33 - On request, EIOPA have introduced a compulsion for NCAs to provide "regular feedback" to firms
• 3.35 - Confirms that not all model changes need to be reported to NCAs during pre-application, just those considered "relevant" by firms themselves

It is fair to say that the lobbying in this space has been noticeably more successful than for ORSA or System of Governance, no doubt due to the smaller sub-set of affected stakeholders having a more concentrated relevance. That said, there were still a number of rebuffs from EIOPA, particularly where the lobbying looked more like whinging about paperwork volumes! Highlights below;

Model Change Policy

3.38 - No danger of EIOPA supporting the recommendation to "fast track" model change approvals if a "major change" is required at short notice. They instead recommend "proactivity" with NCAs. Not sure what this does for the world of opportunistic acquisitions though

3.40 - Fudged the question as to whether parameter changes are considered "major", offering an answer of 'it depends' which, for me at least, leans more towards 'yes they are'.

Use Test

3.42 - Confirms that evidencing "use" is not compelling use of model outputs over and above other techniques

Assumptions and Expert Judgement

3.46 - Documentation and validation of assumption setting and expert judgements considered "crucial" in order for undertakings to counter the lack of data and subjectivity in those processes

3.47 - A guideline has also been amended to confirm that the materiality principle applies for this topic

3.48 - Only the most material assumptions will need AMSB sign-off

Methodological Consistency/PDF/Calibration

A number of changes made to clarify guidance in these areas

Profit and Loss Attribution

3.64 - No escaping the requirements to produce P&L attribution granularity at Legal Entity level, as well as by risk driver

Validation

3.68 - EIOPA do not accept that those who build models may also validate them

Documentation

3.70 - That the guidance around the documentation of the internal model should provide "...[protection] from key-person risk", which I have never seen offered as justification from the supervisory end before

Ultimately, there have been no huge concessions from the position in March, which one would think will cause a number of the more liberal EU regulators to give serious consideration to "explaining" rather than "complying" - that said, with this having been written in Union Jack ink, my British cousins should simply get their transition planning updated accordingly.

Moody's survey on Solvency II compliance preparedness - the chilly third pillar

So from what I can gather it has been a terrible week for the Girondins, with a freak hailstorm wreaking havoc in a thin strip along the vineyards of Bordeaux's Entre-deux-Mers appellation - my in-laws were seemingly spared further down the river, noting that it was merely "un peux froid".

On Ice - Solvency II programmes
and this year's white Bordeaux?

And speaking of a great deal of hard work getting aimlessly destroyed by an unpredictable European storm, Solvency II (do you see what I did there?) appears to have at least enough juice in the tank to have encouraged Moodys to survey practitioners on the preparedness of the industry to achieve compliance before the deadlines currently on everyone's lips (i.e. 2014-2015 for EIOPA Guidelines, 2016 for "go-live").

That survey is available here (short sign-up required, but worth it), with a very short summary here. The media have touched on the survey (here), but only seem to have read the summary, so I've picked through the whole shooting match to see what else was worth knowing.

The sample is small at 45 contributors, but they have all been interviewed one-on-one in Q4 2012/Q1 2013, so the responses are not too dated, particularly as many Solvency II programmes have been running on meagre rations since January of this year. Coverage of 12 EU countries is included in the 45 people, with a decent split of size and insurance type. Majority of respondents were CRO/equivalent, with a few accountants, actuaries and programme managers thrown in for good measure, and just over half are on Standard Formula.

Talking points for me were;

• That 22% have frozen Pillar 3 activity, while 11% have frozen all Solvency II activity
• Half are using Standard Formula to curb costs!
• In addition to that, 20% say that the Use Test is a barrier to using models!
• 27% are approaching Pillar 1 and Pillar 3 with a tick-box mentality (i.e. happy to use multiple manual processes/excel-based tactical solutions to deliver the balance sheet and reporting template elements), while 44% have worked exclusively on Pillar 1 at the expense of Pillar 3
• Solvency II project investment levels are "considerably more" in the UK and France compared to Germany - makes you wonder why they have such a long face!
• 67% have increased their control function staffing by 10% or more - 31% have increased by 50% or more.
• Only 7% note "capital reduction" as a perceived benefit of Solvency II, with 33% selecting improved capital planning (regardless of quantum) as a benefit.
• Only 6.7% say they are receiving "high" levels of support from their national supervisory authority.

Seemingly the stats are a hostage to the sample - I'm sure the PRA would be apoplectic if this was the position of Insurers of Britain plc, but for me the big story is the indiscriminate swelling of Risk/control functions in smaller organisations that evidently are only ticking boxes. Feels a tad disingenuous to pump the staff numbers up to demonstrate compliance, but I suppose it's not me they need to satisfy!

Lloyds directors briefing and Solvency II update - looking well

With Solvency II news and comment pretty thin on the ground this month as the guys at EIOPA count their LTGA beans, it was nice to see the trailblazers at Lloyds release their director briefing slides from last week, an event one may assume is relatively frosty after the money spent in preparation and the distinctly agitated tone on the matter from their CEO recently!

Bling - things Lloyds could have
bought with £300m

As Lloyds are already well down the road of internal model development (having almost kept to their original IMAP deadline they were able to get materials down to the FSA prior to the implementation deadline shifting), the slides are very revealing as to where the group remain lacking when attempting to meet the Tests and Standards for Internal Model approval (TSIMs). With 84% of the market 'by materiality' meeting the principles of the TSIMs, they are clearly in good nick, although by stressing 'materiality', it implies that a relatively large number of smaller members are perhaps still lagging.

The list of "common issues" found will neither surprise nor delight anyone else in IMAP, given that the same themes have been festering for a good 18 months now, and the legislative paralysis on the continent has clearly done nothing to aid the industry (in particular, the FSA letter from this time last year touches on most of these!). Specifically, they observe problems in the following areas;

• ORSA - looking far enough forward (i.e. past year 1), and using stress and scenario testing effectively
• Validation - evidencing validation work done, and following up on test failures
• Model Change - justifying the thresholds for minor/major changes, and agreeing an approach for aggregating minor changes so that they can be considered as minor/major in aggregate
• Use Test - using the model for something other than spewing out an SCR, and it would appear also that when interviewed, the effectiveness of board training and their understanding of the model is being found wanting
• Documentation - documents are either not checking off against the TSIMs, or the content is contrary to the revised controls and processes which have been developed for Solvency II

As Julian Adams made clear at the turn of the year, full compliance with TSIMs is not part of the ICAS+ agenda down at the PRA, however they will expect firms to be fully aware of where they are currently light, and what they plan to do about it. Certainly looks like the Lloyds application won't struggle in this regard, and I wish them well.

EIOPA preparatory guidelines - pre-application for Internal Models

Consultation on Pre-application for internal models guidance  (plus explanatory text)

As staggering as it is frightening, and perhaps indicative of the diversity of approaches currently on parade across the Union, the pre-application for IMs preparatory guidance is 60 pages, comprising of 72 guidelines, accompanied by a whopping 144 pages of explanatory text. This accompanies the existing 82 pages of L3 guidance on the matter released by then-CEIOPS in 2010! A sub-group of EIOPA's IRSG has been assigned to deal with the nitty gritty of this element of the preparatory guidelines.

On first read, this feels massively influenced by the UK's activities to date, and indeed anyone working in that space will recognise FSA pawprints all over the granular details contained within. This is fair I suppose - the InsuranceERM models map has the UK down for around 1/3rd of models currently in 'pre-application' across the continent.

Therefore bearing in mind the UK approach is already pretty well established, I have highlighted below areas which either diverge from what is currently being exercised on the ground, or which clarify (at least for me!) areas which were previously ripe for controversy or disproportionate/inconsistent application. Where it is common sense or continuez tout doit, I have ignored it.

Most importantly, I am reading it as a fait accompli - bearing in mind the short window of time between consultation end and period commencement, EIOPA's recent past on consultation responses (i.e. 'thanks but no thanks') and the UK's evident participation in the bulking of these guidelines, I don't see much room for lobbying swathes of this away, nor for the PRA to "explain" rather than "comply"!

GENERAL GUIDELINES

Guideline 3

• As well as nature, scale and complexity, "design, scope and qualitative aspects" of the IM should be considered when allowing for proportionality 

Guideline 4

• Any model changes pre-application look like they will be pored over by NCAs, including the associated change approval process 

MODEL CHANGES

Guideline 5 - Model Change Policy

• Policy should, as well as SCR-related changes, include changes to: system of governance (around model change); compliance with Use Test requirements; appropriateness of technical specifications and changes in Risk Profile

Guideline 6

• Approach to classifying "major" changes is expected to be objective
• Must also take into account specificities of the company (so benchmarking percentage changes against your neighbours may not be that useful) 

Guideline 7

• Aggregated change triggers must be considered, not just isolated changes
• Offsetting positives and negatives won't be acceptable to avoid a "major change" trigger!

Guideline 8

• Major/minor changes must be determined at Group and entity level

USE TEST

Guideline 9

• "No complete and detailed list of specific [model] uses" will be supplied by NCAs.

Guideline 11

• Granularity of the Risk Management System will need to match the IM in terms of categorisation
• The "structure of decision making fora" will be assessed in ensuring the IM fits to the business - extraordinary!
• Records expected to be available to show how IM outputs are designed

Guideline 12

• Assessment of training, seminars, workshops, meetings and direct interviews "should be considered" in pre-application

Guideline 13

• Will need to "ensure [IM] will be used" during pre-application, as opposed to "use it"
• Expectation that, if other tools are used in decision making, IM is improved having assessed inconsistencies against said tools

Guideline 14

• Evidence of prospective support and retrospective verification of decision making would be advisable for candidates

Guideline 15

• Must document where model is not aligned to the decision ultimately made

ASSUMPTION SETTING/EXPERT JUDGEMENT

Guideline 19

• "Materiality" in the context of assumptions will need to be both qualitatively and quantitatively assessed - should generate some healthy Risk/Actuarial function debate!

Guideline 20

• A validated and documented process for assumption setting and expert judgement will be required
• Sign-off on assumptions will need "sufficient seniority", up to and including AMSB

Guideline 21

• A formal and documented feedback should be maintained between assumption setters and users

Guideline 22

• On the transparency of assumption setting, point 1.64 here effectively asks for an Assumptions Register, as well as dictating what it expects to see in it.

Guideline 23

• Process mapping of some kind expected for the validation of assumption setting
• Independent assumption review is also expected - doesn't dictate whether this should be internal/external, but it will keep someone in clover no doubt.

METHODOLOGICAL CONSISTENCY

Guideline 26

• Methodological consistency to be validated

P&L ATTRIBUTION

Guideline 37

• Point 1.105 seems to confirm P&L attribution by risk driver is required

Guideline 39

• P&L attribution must be used at least annually in the decision making process

Guideline 40

• P&L attribution to be used in the validation process (specifically, old ones to be compared against  experience

VALIDATION

Guideline 41 - Validation Policy to contain at least

• Process, methods and tools, and their purposes
• Frequency of validation for each part of the IM, and triggers for ad-hoc validation
• Persons (not roles) responsible for each task
• Procedure to be followed where reliability of IM is questioned, and ensuing decision making process

Guideline 42

• Shies away from touching Internal Model scope when talking of validation scope - great move!

Guideline 43

• Evidence of sensitivity testing expected when determining materiality

Guideline 44

• Must document known limitations of validation process, as well as circumstances where the process falls over
• May even be asked to quantify the degree of uncertainty!

Guideline 45

• A documented escalation path would be advised

Guideline 46

• Risk Management function will be pressured, as the function with overall responsibility, to ensure all tasks are completed (if not directly performing them) - new skill set?

Guideline 47

• Evidence of how the RM function ensures that the validation process remains independent of IM design and ops should be collected/enhanced

Guideline 49

• A process will be expected to ensure the choice of validation tools used considers; complexity, nature, independence and knowledge of participants - feel this could be tricky for the smaller IMAP guys without leading to additional spend on consultants

Guideline 50

• Must be able to document the appropriateness of the validation tools used accounting for; materiality of IM part, granularity of the data being tested, purpose of the task and the expected outcome

DOCUMENTATION

Guideline 53

• Expectation of a "...clear referencing system [for IM documentation] which should be used in a documentation inventory"

Guideline 55

• An overall summary of IM shortcomings, "consolidated into a single document" will be expected
• This needs to cover at least; Risks not modelled, limitations in modelling, sources of uncertainty in results, data deficiencies, external models/data, IT limitations and governance limitations.

Guideline 56

• Potential suggestion that there should be more than one level of IM documentation to suit other audiences/uses - IM for Dummies anyone?

Guideline 57

• End-to-end User Manual expected which an Independent Knowledgeable Third Party could operate

Guideline 58

• Stress that a single document containing all model outputs (as Use Test evidence) is not required - acceptable as single docs

EXTERNAL MODELS AND DATA

Guideline 60

• Expectation that external data sets will be sense-checked against "other relevant sources"

Guideline 61

• Understanding of external models must extend to technical and operational aspects, as well as assumptions

Guideline 64

• "Material" assumptions of external models must be validated
• In point 1.163, any potential for cherry-picking features/options of external models is constrained

Accenture study on Risk Analytics - lessons for Insurance Industry

Decent piece of benchmarking from Accenture on the current usage of risk analytics as well as drivers for the future. 450 mostly c-suite level respondees from across industries (40% insurers), but the findings are targeted specifically at the Insurance industry. Interestingly, respondees leaned towards incomplete data sets, rather than a lack of data or technological capability, as the main constraint.

While it touches on a number of areas of interest for Use Test specialists, it also covers Stress and Scenario Testing (13% of Life and 21% of P&C insurers reporting that they "rarely" or "never" use stress testing in decision making), and Reporting (which suggests the main driver for reporting improvements is regulatory rather than voluntary, due to regulators "...[increasing] their focus on the quality and frequency of reporting"). Internal Modelling also gets a mention, with almost 80% of respondents saying they already use, or are planning to use, an internal model for capital adequacy requirements.

Data Governance  of course gets decent treatment here. Only 69% of insurers polled currently have a Data Policy, but 41% have a data quality department, which feels alarmingly high, particularly when drawn against Accenture's comment that "...many firms have insufficient rigor who owns data, who sets it up and who manages it". The FSA concurred with that in their preliminary Data Quality Review findings back in September, and I'm not at all convinced that a DQ department will help in this respect (i.e. BAU absolve themselves of responsibility for their data sets!)

The Accenture crew use the "Leaders" and "Laggards" analogy throughout, so benchmark away and find out which one you are!

KMPG survey on Solvency II Board Training - "Do more!"

Been slack the last couple of weeks while doing my final preparations for the Isle of Man half marathon, which I "tore up" last Sunday in, errr, 1h 36m, coming in just ahead of Brian the Snail and Albert Steptoe.

On the basis that I'm now safe from potential Olympic requirements, I am left with plenty of time to blog while my multiple friction-related injuries heal...

Starting with KPMG's latest diatribe on Solvency II Board Training, an admirable attempt to put one of the darkest sheep into the spotlight - anyone who has dragged a Policy, Solvency II briefing note, draft ORSA Report or internal model justification paper through a Board or Board sub-committee will know that getting the membership to take Solvency II seriously is not a walk in the park (particularly when you have to also explain once a quarter why the go-live date keeps moving!).

There is therefore a "comfort in numbers" which one can draw from KPMG's survey (sadly no details on sample size), with an overriding message of "do more, immediately", in particular;

• 80% of boards have received 15 hours or less of Solvency II training - I dare say in many cases this would be accumulated by tacking on Solvency II-related matters to the end of existing Board agendas over a 12-24 month period.
• 57% have covered ORSA in their training
• 30% have covered Use Test
• 44% will be embedding training objectives into their director's PDPs

Outside of those "big hitters", some interesting gaps emerge;

• 10% have not commenced training
• Half have performed 1-on-1 training with execs (perversely, more have done 1-on-1's with NEDs at 56%!). My experience would say they are equally and entirely in need of bespoke training!
• 35% have delivered training on IMAP itself, a pretty sobering number bearing in mind where we are in pre-application
• Over 20% of respondents have nothing currently planned for training around Pillar 3 reporting (22% on the QRT/SFCR side, and 26% on the RSR/ORSA supervisory report) - to have not at least briefed on these matters at this juncture is remiss, bearing in mind the consults were put out last year!
• Only a third have instigated a company-wide training programme

The section around training on the Internal Model however (p7) felt instinctively wobbly to me, with the FSA said to be "likely" to expect all directors to have a "good" understanding of the IM.

While KPMG have purloined the best bits of Article 120 (and 213 in the Level 2 draft) for the purposes of what good might look like, I would still argue that a "good" understanding is not even close to being defined from a national regulatory perspective in a manner that lends itself to effective training planning.
No doubt anyone in the IMAP space will find this useful, but you might want to check the sample size with KPMG in order to put any weight behind the conclusions if it prompts you to change programme priorities/budgets etc. Certainly an eye-opener regardless.

FSA Speech - Adams at Insurance Day Summit on Internal Model Progress (or lack of...)

Another day, another bum-kicking for the IMAP candidate firms, who have been the recipients of some public words of 'encouragement' from the FSA's Julian Adams in a speech to the Insurance Day Summit yesterday, coupled with a "Dear Firm" letter which went to all internal model applicants this week.

The same themes which have pitched up in recent presentations (here and here for example) are still for all to see - model change, model scope, validation and expert judgement/rationale all feature heavily in the speech - but perhaps the elaboration on both assumption/dependency setting and documenting and Use Test embedding are breakout features from those already tabled as being insufficient.

Useful for those outside of the UK to see exactly what is twisting the FSA's tail at the moment, while for all you UK readers, the letter is clearly a laundry list which will require immediate attention and prioritisation in your work plans. Good luck...

Central Bank of Ireland - Solvency Matters newsletter, ORSA and models

The Central Bank of Ireland have released their latest Solvency II briefing letter (thanks to Mike Claffey for tweeting it).

Important to note the following from a UK and Ireland practitioner's perspective, on the premise that the two regulatory bodies in the UK and Ireland are setting the benchmarks in many respects;

• 'No panic' message given regarding the Omnibus II 'delays' - timetable unchanged in this regard
• No references to the Commission's draft Level 2 text as a basis for undertakings to use for internal model application activity (which the FSA went out of their way to advocate at their February IMAP event). Is there a legitimate reason for this, bearing in mind how useful the document is for applicants in ensuring their documentation addresses the final requirements, even if only 'there or there abouts'?
• Model applicants should be "well advanced with the technical aspects of their model build" - this is somewhat to the FSA's take, which was that build should be pretty much complete by now (p10)
• Some nice visceral examples of Use Test evidence requirements from undertakings, particularly around evidencing Board 'understanding' of the model - I would add that these seem onerous for NEDs, but reasonable for executives, so I would be interested to see how this is applied (could you for example see your non-industry, non-executive director justifying diversification benefits to your friendly local regulator? Is that a reasonable use of project resource/Board training time?)
• Board members will be interviewed by the CBoI individually "without the support of their technical experts" - I'll get my coat then...
• On ORSA, they "recognise that individual undertakings would like to get more details on specifics of the ORSA Reporting Process", but want undertakings to focus on the contributing processes themselves - easier said than done, particularly when even EIOPA had to relent from the mantra of 'ORSA is process, not a report' when they got to public consultation stage!

I'm sure that the industry will appreciate some of the clarity in here - they may appreciate the Commission's Level 2 draft a bit more though!