Category Archives: Uncategorized

Model limitations for risk management

While model building will always be a core element of insurance, being aware of the limitations of models and techniques to manage those limitations is crucial to successful risk management.

A new article authored by Pat Renzi and Elliot Varnell for the Actuarial Post focuses on the challenges risk managers face given their responsibility for an internal or Own Risk and Solvency Assessment (ORSA) model. The article highlights three issues that risk managers need to make allowance for in managing their models: model scope and historical data; associative dependency; and the problems that arise when elegant models confound transparency and understanding.

Here is an excerpt:

…Model building is at the core of the insurance sector and we will never stop building models. But awareness of their limitations and the techniques to manage those limitations is crucial to successful risk management.

Simplifications and expert judgment are a fundamental part of building models too and should be recognized as such. However they should be applied with full transparency and in full knowledge of their limitations rather than couched in mathematical derivations where they cannot be readily challenged.

The article was published in the July 2012 issue of the Actuarial Post. Read it here.

UK Risk and Investment Conference 2012, Leeds, 25-27 June 2012

It was my pleasure to be the Co-Chairman of the Risk and Investment Conference which has just taken place between 25 and 27 June in Leeds, UK. I was the Chair for the Risk element of the conference with Margaret de Valois of Mazars the Chair of the Investment side of the conference. This blog post reviews the conference and offers some personal perspectives as the chair.

This year we managed 250 members and guests attending, which was a record and so good that we had to close registration earlier than intended. This was pleasing as it provides the Actuarial Profession Events team with some confidence that they can book a larger venue next year and continue growing the conference.

Highlights of the conference were a keynote talk from leading global financial journalist Gillian Tett of the Financial Times which I was very pleased to be able to chair. Gillian discussed the application of anthropological ideas to the cultural aspects of risk management and updated the audience on the latest geopolitical developments in financial markets. Gillian is an anthropologist by training and frequently makes use of her training in her journalism. She was joined in her session by leading anthropologist, Michael Thompson of the IIASA Research Institute and Alice Underwood of Willis Re, who have been interpreting the ideas of The Cultural Theory of Risk for practical use. Michael explained the Cultural Theory of Risk as pioneered by Mary Douglas before Alice Underwood helped put the ideas into an actuarial context and highlighted where most actuaries tend to operate – i.e., in the hierarchical solidarity. We ended the session with a short Q&A for Gillian before she had to head back to London; it was very encouraging to hear her endorsement for the work that Michael, Alice (and ex-Milliman Principal Dave Ingram) had been doing.

Continue reading

EIOPA Annual Conference – Frankfurt

I didn’t manage to get to the Life Convention this year but did manage to make it to the EIOPA conference held in Frankfurt the week before. I’ve always found it an interesting conference to attend. This year, as well as our friends in Group Consultatif, there were a number of UK Pensions representatives.

The conference followed the usual format of 3-4 panels and a couple of keynote speeches, this year by Michael Barnier, EU Commissioner for the Single Market and Gabriel Bernardino, Chairman of EIOPA. Interesting as it was to hear the views of these two gentlemen neither served as the highlight of the conference with the panels rather stealing the show. The FT did, however, pick up on Gabriel’s speech for his comments about needing to consider the risk in sovereign debt. I don’t think I was alone in thinking that these were not intended to say anything that we haven’t heard before from EIOPA. Clearly it was very good to have M. Barnier attending the event to lend support, but having been to previous CEIOPS conferences Sharon Bowles has had more in depth comment to make and for my money is a better conference speaker. What M. Barnier did say – many times – was how we all had to stick together to get through the current crisis. Undoubtedly this is at the forefront of his mind as these are dangerous times for the European Union and those who have worked hard over decades on the European Project.

What was more interesting for me were the discussions on Solvency II, Pensions and Consumer Protection each of which held the attention well.

The Solvency II debate was particularly good with Oliver Bate, Group CFO at Allianz and Karel van Hulle, Head of Insurance and Pensions Unit at the European Commission in particularly good form. H. Bate was very forthright in his criticism of Solvency II and the balance sheet volatility that Solvency II was bringing. In particular he cited that the Allianz SCR had moved by around 50% in the space of a month as yield curves shifted rapidly in the summer. This, he told the conference, made it extremely difficult to manage the firm as shareholders needed confidence over the return on capital they were getting and the management needed stability to plan the business. Both Karel van Hulle and Carlos Montalvo pushed back that the framework was necessary to protect policyholders.

The pensions debate centred on the proposals to apply a Solvency II style regime to pension schemes. A green paper had been published on this by the commission and an EIOPA consultation is currently out and being responded to by – amongst others – the actuarial profession. It was good to see a decent contingent of the UK Pensions industry there and questions asked from the floor by representatives of the BT Pensions Fund. They asked if the Commission would drop its proposals if the impact assessment showed the costs to outweigh the benefits. The answer from Karel van Hulle was that the costs and benefits were always considered but to date there had not been proposals made where the benefits didn’t outweigh the costs. I, and others, took that as a sign that the Commission was minded to proceed.

The emerging role of EIOPA in consumer protection was discussed in the final session of the day chaired by Pauline Chantillon of the ACP who has moved on to a consumer protection role in France having led the FINREQ stream of Solvency II for EIOPA. The recent Variable Annuity retail paper (which many may have missed when it came out a couple weeks ago) was referred to many times as an example of how EIOPA was rising to the challenge of protecting consumers on the conduct of business side of regulation too. One of the best points was a long list of things that a consumer would need to know in order to make optimal financial decisions to deliver the outcome they wanted. The point was well made that consumers want to buy an outcome and this is what the financial services sector should focus on giving them. It wasn’t lost on several audience members that the EIOPA conduct of business regulation on VA, if poorly drafted and executed, could yet prevent the delivery of a product line that seeks to deliver a controlled outcome for the customer.

The tone of the EIOPA conference is quite different to those in the UK and it is interesting to observe the styles of this conference in the light of discussion on culture theory with my colleague Neil Cantle and Willis Re’s Dave Ingram over recent weeks. The collegiate nature of our European friends is abundantly evident at the conference with many warm words exchanged between panellists and statements of intent to edge the European project ever-forward. There was little mention of the chaotic collapse of the Euro Zone (EZ) edging ever nearer and (if we are to believe Angela Merkel) with it the end of the European Union. The Germans I spoke to seemed certain the EZ would continue OK but echoed the German fear of hyperinflation and the quantitative easing that has been used in the US, UK and previously in Japan. Little seemed to have changed from a year earlier when I was told by a Bundesbank representative that inflation (even moderate) could never be the answer to the debt crisis. Yet a week after the conference in Frankfurt the German Bund Auction faltered and Bund yields rose above the UK suggesting the pound could become the European reserve currency of choice.

This doesn’t sound like such good news for UK insurers who could well see their discount rates pushed ever lower on a wave of gilt purchases – even if their government bond holdings appreciate in value. One thing that UK insurers with EUR exposure need to be wary of is whether they are tied into using a AAA EUR GOV bond spread. With Belgium (amongst others) downgraded by S&P on Friday pressure may be building on the French AAA rating. If it does fall below AAA the AAA EUR Govt yield curve could fall sharply as the French Bonds get excluded and the German Bunds come to dominate the AAA government curve.

The Bank of England has already told the UK banks to prepare for a dis-orderly default of the EZ in their contingency plans so it would be wise for insurers to consider how they would cope too. Fitch – at least – seems happy that the European insurance sector will not be too badly affected but its always best to have a plan in the bottom drawer just in case, as time will be in short supply and questions will come thick and fast, if and when the EZ does break up. They certainly did in the Autumn of 2008.

It’s the price we pay to live in interesting times.

Elliot Varnell

Some of the conference keynote speeches are here:

Reflections on the Life Conference 2011

As we return to our desks after another successful Life Conference, and in preparation for completing the Profession’s feedback survey, I have been collecting my thoughts on the highlights of this year’s event.

I have come to the conclusion that sessions on the state of the life assurance industry and what needs to be done to revitalise it are rarely very enlightening, because (perhaps not surprisingly) the speakers tend to say much the same things each year. The conflict between the desirability of consumption to promote economic growth in the short term and the undeniable need for greater saving to improve financial security in the longer term seems as intractable as ever.

As is often the case, the session on the economic outlook, this time presented by two economists from RBS with an emphasis on the European banking and sovereign debt crises, was for me the pick of the plenaries. Sir Richard Needham had some interesting things to say in his opening address, but I struggled to find a read-across from vacuum cleaners and milking machines to financial services.

Among the workshops I attended, two stood out. The session on Tax After Solvency II (which turned out to be a slight misnomer, as the change to basing tax on the accounts rather than the FSA returns is expected to come into force on 1 January 2013 regardless of the Solvency II start date) was full of useful information, and well presented by Matthew Little, Matthew Taylor and Andrew Rendell. Like several other sessions, it might have been even better if the time allowed had been longer, as there was little time for questions, and parts of the explanation seemed a bit rushed.

Another excellent workshop was “The sting in the tail”, delivered by John Roe of LGIM, one of this year’s hot topics and an object lesson in how to structure a presentation, as well as providing some very interesting insights into the dangers of cognitive bias when considering tail risks.

Liverpool was a revelation to many attendees I think – it seemed like a big building site when I last visited, but the transformation of the eastern part of the city centre and the dock area is impressive and the conference centre was very user-friendly.

The format of the conference still seems to work well, and the opportunity to catch up with contacts is tremendous, although I always miss many of the people I had hoped to talk to (perhaps they saw me coming!). I also need to work on my ability to conduct coherent conversations while listening to loud music – but I don’t think I am alone in that respect. Congratulations are due to the organising committee and the conference staff for putting on another very enjoyable event.

I look forward to seeing you in Brussels next year.

Nick Dumbreck

Interesting session on the future of the AFH

The session on the future of the Actuarial Function Holder (AFH) under Solvency II was a reminder we still have some work to do in mapping from existing roles to the new titles under Solvency II. 

General opinion is that the AFH role will still be required under Solvency II. But issues remain around what the AFH will do in the new environment and where does the AFH sit under Solvency II.  Is it the first, second or third line of defence?  It was the view of the speaker that the AFH sits somewhere between the first and second lines of defence (no puns about actuaries on the fence please!)

There are also subtle differences between ‘function’ and the ‘people’ actually fulfilling them. Some parts of the role could be in different lines of defence. For example, actuarial function could calculate the solvency capital requirement, this doesn’t need to be the risk management function.

But questions remain as to whether the AFH can continue to do current work – e.g. setting reinsurance strategy or investment strategy – or does the new head of the actuarial function under Solvency II need to be capable of being independent. 

There is a possible (probable?) overlap with the Risk Management function so there needs to be an awareness of any overlaps with the need for documentation of any overlaps, possible conflicts of interest, and how these are dealt with.  There is a danger that things fall between the cracks.

John, Paul, Ringo and the Economy.

It is perhaps no surprise that in the city of Liverpool, one of our most popular survey questions relates to the Beatles—more specifically, which of the Beatles songs remind you of the current economy?

Behind the scenes, we had our own challenge: our survey only provides space for 5 options. Which of the 306 Beatles songs to feature? We did our best on this admittedly subjective task, but had to leave some of these contenders out of the mix:

Fixing a Hole

Getting Better

Let it be


Crying, Waiting, Hoping

The Fool on the Hill

With a Little Help From My Friends

You Can’t Do That

So far, these are the leaders. We’re a tad surprised that Dear Prudence is not further up on the charts!

Workshop preview F6 Assets within personalised funds, does default matter?



Portfolio Bonds and SIPP’s can include individualised funds where a wide range of investments may be held. Recent years of market turmoil have forced us all to examine what is investment risk, and who actually carries this risk. In this session we will discuss the nature of default risk and counterparty risk with these particular products in mind.

As speakers we have the advantage of knowing who plans to attend our session. We expect this will be an interactive session as we will present a range of open issues and dare to suggest our own personal opinions on each one. We look forward to the debate.



Sovereign Bond spreads

I have been thinking about sovereign bonds a lot recently. Ten year Irish government bonds were yielding 9.2% on the last day of 2010 when UK yields were 3.2% and Italian were 4.9%. As at 8 November 2011 the latest yields were 8.3% for Ireland, 2.2% for the UK and 6.7% for Italy.

So we agree that the market believes not all government bonds are risk free. As actuaries we have to try to quantify how much of this spread is credit (default) risk. The Society of Actuaries in Ireland produced a useful paper in May 2011 that summarised a range of approaches to try to quantify this credit component. The full paper (and presentation and podcast!) is here:

The interesting table is included in the Appendix to the paper:

Irish government bonds, as at 31 December 2010:

Approach to estimating credit risk % of the excess spread that can be considered credit risk
Credit default swaps >100%
Market based metrics 75% – 90%*
Solvency II illiquidity premium 92%
Historical experience approach **
Bank of England’s analysis on corporate bonds >50%



* Based on data at 15 February 2011

** Depends on the analysis of historical experience.

I’m looking forward to discussion of this at the life convention in Liverpool. I expect (and hope) the International Actuarial Association attendees will bring an interesting international viewpoint to the debate.