How capital shapes investment under Solvency II

In the second half of this InsuranceERM/Standard Life Investments roundtable, experts discuss the possibilities to invest in new asset classes, the shift away from gilts and the effect of transitional measures

Chris Cundy: How do you reap the benefits of diversification in your portfolios? Are you able to invest in some high risk, high‑return assets that you might not have thought possible before? 

Emily Penn: The question about diversification is how much real diversification you can recognise in your capital model. If you invest on a return-on-capital basis and you cannot properly reflect diversification in your capital, then the concept does break down somewhat. I expect that over the next two to three years, people will dedicate more resource and time to capturing those diversifications better than they currently do, to enable them to invest in asset classes that do offer diversification. 

Victoria Gibson: You are already seeing diversification from a purely asset perspective. But how much does the capital efficiency argument drive the investment strategy? Or does it become a constraint? And how can you lever that constraint to bring more stability to the balance sheet, which is supposed to be the benefit of diversification? For us, diversification has been a deliberate strategy but it has not been done with a view to optimising capital. 

Craig Turnbull: I think we have seen that logic most clearly when an insurer has felt compelled to seek more yield in today's environment; there are a range of choices, but there are some that complement the existing assets very well from a diversification perspective and that can therefore allow the efficient pursuit of yield, rather than merely investing in a way that materially increases the overall risk of the asset side of the balance sheet. 

Emily Penn: That would be things like moving away from corporate bonds into commercial mortgage lending, or corporate bonds into infrastructure.

Iain Forrester: There is a limit as to how granular your assessment of diversification can be within any capital model. But it is important to take the more economic view of diversification as well. There is value in limiting your exposure to individual counterparties, sectors, and regions even if you are unable to reflect it fully within your capital model. 

Daniel Blamont: We try to optimise assets on an economic basis, taking liabilities into account as this determines value creation. However, consideration needs to be given to capital as capital can constrain the amount of value actually distributed to shareholders and policyholders. With an internal model we have the opportunity to reduce the inconsistencies between the economic and the regulatory worlds.

"You have to prioritise which asset classes or instrument types you want to focus on because a significant amount of work is required to get them over the line"

Chris Cundy: We have seen in recent years a lot of new asset classes opening up for insurers. Are there still any flowers blooming?

Victoria Gibson: I think there is creativity. There are lots of people on any day wanting to suggest new ways that insurance companies can access something. Monetising illiquidity is still a theme, for example, and there are opportunities from the banks retreating. But we have to ask, are they going to move the dial? Or is this just going to create a lot of work, because of the issues we talked about earlier? 

That even includes some derivatives. Swaps are a classic example: you go back ten years and swaps were completely vanilla and banks were happy to price them incredibly finely. Now, the whole swap market has changed because of the clearing arrangements and capital treatment of swaps.

Emily Penn: That resource point is very pertinent to us. We are a small investment team and traditionally we would look to one of the big insurers to make that big leap first.

Iain Forrester: You have to prioritise which asset classes or instrument types you want to focus on because a significant amount of work is required to get them over the line. In the case of making a tactical investment decision very quickly, would you have to think about the areas where opportunities could come up and get comfortable with them in advance? Or would you put some mechanism in place to speed up the investment decision?   

Emily Penn: We would probably need to investigate an asset class and have a framework already set up, so that when opportunities came up we could invest in it.

Daniel Blamont: Regarding resourcing and costs, it is true that small and medium-sized insurers may not be able to diversify as much as they could, since a material investment mandate is needed to justify the costs.

Swap-spread risk

Hugo Coelho: One of the long-term structural changes is this change from gilts to swap-based investments, because of the way you discount. Is this still ongoing? 

Emily Penn: It is something that we are focused on. One of our portfolios that is heavily gilt-based is subject to the transitional arrangements, so that swap-spread risk could be nullified in transitionals, as long as they are allowed to be recalculated.

Iain Forrester: There is more to it than just the gilt-swap basis risk on the capital side. One is the return are you actually getting for that investment and at a number of terms, gilts are currently yielding significantly more than swaps. There are then the operational and cost challenges that Victoria mentioned, in terms of the swap market becoming a more expensive and a potentially more limited market to operate in. Finally, there is the potential for diversification within overall capital requirements if your model assumes a positive correlation between gilt yields falling relative to swaps and corporate credit spreads going up.  So you could see that, from a balance sheet perspective, some exposure to gilt-swap basis risk may be beneficial.

Emily Penn: We still don't know how much onus the PRA is going to put on liability‑basis risk capital charge. We have not had much feedback on that yet.

Craig Turnbull: We are currently seeing historically unusual negative swap spreads at the long end of the curve. This might increase insurers' appetite to retain gilts, earn that extra spread and bear the capital charge that might be associated with the risk that swap spreads go further negative. This negative swap spread might also create a greater incentive to apply for the MA on business where previously that decision was more finely balanced.

Iain Forrester: The insurers' decision could also depend where these assets are sitting. If these government bond assets are supporting a with-profits business where you are taking your policyholders' expectations into account, there could be a different balance of risks and rewards and you might come to a different conclusion.

Daniel Blamont: There is not just a shift from gilts to swaps, but also a shift from long-dated gilts to shorter-dated ones. Ultimately insurers should have a range of duration instruments at their disposal be it swaps, futures, gilts, gilt repo, etc to take advantage of market opportunities when they arise in a more measured way than seen currently. The current market dynamics are also an unfortunate consequence of having to report on a Solvency I basis at HY15 or even Q3 15, leaving little time to implement Solvency II changes.

"There is the potential for diversification within overall capital requirements if your model assumes a positive correlation between gilt yields falling relative to swaps and corporate credit spreads going up"

Treatment of transitionals

Hugo Coelho: You mentioned the uncertainty around transitionals and the way the use of them can delay some of these changes that you will have to do. Regarding the investment strategy, is there a big difference between legacy portfolios and the new business that you are going into?

Emily Penn: The legacy portfolios will be subject to the transitionals and if we are allowed to recalculate the transitionals, we are effectively still managing them on a Solvency I basis; the new business is not subject to that. But the volumes of new business are not significant in comparison. Over time, the decisions we make around new business will change to reflect the Solvency II regime.

Daniel Blamont: Transitionals would be most helpful for the risk margin, especially if it can be recalculated as rates move. However transitionals should not be used as a delaying tactic. Losing 1/16^th of transitionals each year can be significant and the industry should not be complacent.

Level playing field

Chris Cundy: Are internal model firms at an advantage over standard formula firms when it comes to what you can get away with investing in?

Iain Forrester: Getting an internal model or partial internal model approved is inextricably linked with the prudent person principle. If an insurer can go through the process of building up an internal capital model for an asset class, the insurer will have shown that they understand the risks of that asset class and are able to measure, monitor and manage them. 

Daniel Blamont: Internal models enable firms to reduce arbitrages that may exist under the standard formula between different asset classes or risk factors. That would make it easier for internal model firms to make a case for diversification via new asset classes. However, it is not a way to get lower capital requirements.

Emily Penn: If you want to make investments in broader asset classes, you have to get the Board happy with it and the Board will therefore expect that you have modelled and understand those risks appropriately. So I think for any insurer that wants to invest outside the very traditional space, you need an internal model…

Iain Forrester: …particularly when the standard formula has been calibrated to very broad asset classes. Hedge funds and absolute return funds can have very diverse strategies but are considered within a broad bucket, unless the insurer is able to look-through to the underlying asset positions. If you are investing in funds at the safer end of that range, there is then a rationale for internal models having lower capital charges than the standard formula.

Victoria Gibson: Theoretically, absolutely; in practice, it's not that easy. There is still a big issue with lack of data and making assumptions about strategies where there is a lack of transparency.

Matching adjustment applications

Hugo Coelho: Are there any lessons to be learned from the first round of MA applications?

Craig Turnbull: It has yet to be made public who has been successful, so I think there is still a lot to be learned. But I think it has surprised asset managers how much work has been required to craft investment mandates and servicing agreements that meet the demanding requirements of Matching Adjustment asset management.

Emily Penn: At the moment, firms are taking quite different views in certain aspects of their applications. Over the next two or three years, you will see a 'best practice' emerge.

Daniel Blamont: Most firms' aim this year will have been to get MA approval. Going forward, optimisation and relaxing of some tough constraints will be the goals, and benchmarking exercises by consultants will facilitate that process. But one would expect the PRA to ensure high standards are maintained via regular industry-wide reviews.

Read part one here.