Life & Pension Risk: Delta Lloyd was one of the first Dutch insurers to move to an economic balance sheet; does this mean Solvency II will only require a small transition?
Theo Berg, chief risk officer, Delta Lloyd: In theory, because we already run our company on an economic basis there shouldn’t be too much difference between how we operate now and in a post-Solvency II world. But in reality there will be a gap. This is because despite everyone calling Solvency II an economically-based system, it’s actually quite prescriptive. The Solvency II advice book runs at more than 300 pages, so you have to ask whether the system is rule, rather than principle, based. Companies need to ensure that thinking about risk is embedded into their systems – a fundamental approach to risk management is thinking for yourself and not just box-ticking legislative requirements. I am concerned that the current approach of Solvency II may encourage the latter approach.
LPR: Any aspects of the current format of Solvency II you disagree with?
TB: I disagree strongly with the current liquidity premium approach – instead the liquidity premium should be bigger and more widespread. And we believe the liquidity premium is different for different maturity buckets along the yield curve. But that is a technical issue. A more fundamental concern is that every single risk is already analysed for companies, and the additional experience of individual countries is limited. But the composition of financial markets is not identical in every European Union member state. Instead of the European Commission describing all the risks, companies should assess this themselves. Of course, it is possible to opt for an internal model but given the fact the regulators will want to have a benchmark against the standard formula, it is very important to align your internal approach with the standard one.
LPR: Does Delta Lloyd think an extrapolation method is the best way to establish the liquidity premium?
TB: No, we think it is important to maintain a market-based approach when valuing the balance sheet. After a lot of research we opted to use the collateralised AAA curve to estimate the risk-free curve, including a liquidity premium, to discount our liabilities. This is how Delta Lloyd values its balance sheet, and also how it reports its market consistent-embedded value figures.
LPR: Why did you opt for the collateralised AAA curve?
TB: Because these are collateralised bonds that are very illiquid and therefore they have the same profile as our liabilities; for example, you can’t just call them, and there are a lot of guarantees embedded within them to ensure that credit risk is limited. The collateralised AAA approach has also been debated by the Chief Risk Officer’s Forum as a potential proxy for the liquidity premium, and we thought that as a prudent company we could use this curve. The pre-crisis spread between the collateralised AAA curve and the swap curve was almost negligible, during the crisis this spread moved from 50 to 120 basis points. By using this curve for valuing our illiquid liabilities, we mirrored the liquidity premium effect of the fixed-income portfolio, without introducing a default risk effect in the valuation of the liabilities. As the collateralised AAA curve only consists of around 450 bonds, the use of this curve by the entire European insurance and pension industry would lead to significant demand and price changes.
LPR: ING, Aegon, Fortis plus a host of lesser-known players in the Benelux region had a horrendous time in the credit crunch. Why didn’t Delta Lloyd?
TB: The answer to this question goes back to 2004 with the appointment of my predecessor, Laurens Roodbol, who set up a risk management department and introduced risk reporting. This makes it clear that even then Delta Lloyd’s board saw risk management as an important issue. The board asked Roodbol two questions: ‘how do we do financial reporting?’ and ‘how do we integrate risk management into the heart of the business?’
Despite prevailing legislation in the Netherlands requiring companies to mark only their assets to market, Delta Lloyd took a big step by moving towards fair value accounting on both sides of the balance sheet. This gave a real understanding of the economics of the business. Instead of first trying to analyse all risk exposures perfectly, Delta Lloyd acted directly based on the directions provided from the initial risk exposure analyses. As analysis and information improved along the way, decisions could be made in more detail and more geared towards the recognised risk profiles.
The other step was integrating risk management into the company’s way of thinking so that even six years ago we made operational decisions based not just on profits, but also the level of risk contained in each product and business line. Delta Lloyd then conducted some scenario planning looking forward to the next 10 to 20 years. And it was this combination of long-tail scenario modelling with direct implementation with the fair value accounting that was the basis for operating our hedging programme.
LPR: So moving to an economic balance sheet underpinned your approach?
TB: Yes, because of the economic balance sheet Delta Lloyd saw its largest risks lay in stressed interest rates and equity scenarios – shocks to either of these parts of the portfolio were where it could lose the most value.
Based on this understanding, we designed a strategy that means we won’t be forced to sell off our assets – that is very important. Another point was, we do not want to lose too much value in volatile markets and therefore we sought out downside protection.
LPR: How did you construct your hedging approach to meet these objectives?
TB: We set up an interest rate hedging programme to protect against downward movement. Part of this strategy was to buy downside protection when markets went up – it was a deliberate counter-cyclical approach. We did this by buying out of the money swaptions. We also acquired some swaps, but mostly the strategy was to go for swaptions because the risks we are running are not linear.
LPR: Was the economic balance sheet the only factor in establishing your risk management strategy?
TB: No, the other important factor was Delta Lloyd started operating under the Individual Capital Assessment (ICA) regime that the UK’s Financial Services Authority brought in to regulate that country’s insurance sector in 2005. London-based insurer Aviva is our majority owner and when Aviva said it was going to have to produce an ICA, Delta Lloyd was given the option to introduce it locally. Rather than view this as an additional administrative burden we saw it as a good way to measure, and understand, our risks.
LPR: Did Delta Lloyd have to adapt any aspects of the ICA to meet the realities of the Dutch market?
TB: Absolutely. This is the whole point of doing an exercise like the ICA – it requires you to consider what the actual risks you are running are, and in this case Delta Lloyd was focused on the risks posed by the Dutch market. For example, longevity experience is different in the UK and the Netherlands. Another clear difference – and one that has implications with regard to what I said earlier about Solvency II as well, is real estate, which has very different volatilities in each country/region.
LPR: Did you opt for a purely derivatives-based strategy to derisk the portfolio?
TB: The strategy is not only to buy derivatives but also to lengthen the duration of the assets portfolio and ensure it is diversified. These are the two key areas Delta Lloyd focuses on when it is judging an asset liability management exercise. The second part to consider was the equity exposure. In the Netherlands, we benefit from the rule that states if you have a 5% shareholding in one company, no dividend tax or capital gains tax are payable. So it is quite interesting/rewarding from an insurance perspective to invest in these types of 5% share-holding. However, as recent experience has highlighted, the equity market are sometimes a little volatile
So the issue is how does Delta Lloyd ensure it doesn’t lose too much when equity markets fall? In 2007 and 2008 we designed a strategy that said we were prepared to lose a certain amount of money, but below that value we want to be protected. To achieve this protection, we bought some put options in 2007, and again in 2008. As part of a holistic approach to risk management, this strategy included taking into account equity exposure on both the asset side and the liability side. So for example, our unit-linked funds with guarantees have a liability side equity exposure, and this needed to be taken into account.
LPR: Did you opt to dynamically hedge your equity exposure?
TB: You can be very sophisticated and dynamically hedge the portfolio every day, but we wanted to manage the risks on a higher level and also ensure some volatility, because if you exclude all volatility the price of your hedging programme becomes enormous.
The depth of the crisis was more than we anticipated and this meant that despite the fact we were focusing on reducing our equity exposure, we lost more than we expected Our programme wasn’t complete when the crisis hit. We were going to buy options through the year in 2008 but we were unable to complete it.
LPR: What about credit risk?
TB: The last large exposure was and is credit risk due to the large fixed income block of assets. We always operated on the basis that if credit spreads increase, Delta Lloyd has the ability to hold these assets to maturity, therefore the only risk we face is that of default, which is minimal in our portfolio. Of course, this approach is fine from an economic perspective, and in our economic balance sheet the effect of credit spread was limited, however the regulatory regime in Europe is not economically/risk-based, but volume based (Solvency I).
LPR: So being regulated on a Solvency I, rather than a risk-based approach, hampered your risk management?
TB: Yes, Solvency I isn’t run on an economic basis so our company was effectively punished for running its internal risk management economically. From September 2008 we were in discussions with our regulator, the Dutch Central Bank (DNB), on what would be the right basis for valuing our liabilities because if you don’t recognise any credit spread movement in your liabilities, you are out of synch with the asset side.
LPR: Was the DNB sympathetic to your concerns?
TB: The DNB responded by saying it was bound to the existing legal framework and could not change it. So now, for example, Dutch pension funds have to report the value of their liabilities using the most liquid curve. So all these liabilities that are very illiquid are valued using the swap curve, which is the most liquid one available. That is how the DNB responded to us, and it is how they still respond to the pension funds.
LPR: How was your solvency position affected by the market conditions in 2008?
TB: We started 2008 with a large buffer, and one of the main reasons for that is that we stress-tested our buffer which limited our dividend payment to our majority shareholder Aviva, meaning that we didn’t repatriate an excessive amount of capital in the good years. While our Solvency I Insurance Group Directive (IGD) ratio currently stands at 178%, at the worst point in the crisis it went down to about 150%. This was on par with our Dutch peers. However, unlike some of the members of this group, we didn’t need state support. If it hadn’t been for the additional capital they received our solvency levels would have been considerably higher than theirs.
Also comparing our IGD ratio with others is a little difficult as our capital levels shows all effects of the top holding company. So all our subordinated debt and intra-company loans are shown transparently, giving Delta Lloyd a hard capital balance sheet position. This isn’t true of some of our competitors.