15 April 2018
In this article, the second in a series of three, we have assessed asset optimisation and the extent to which risk can influence asset allocation. We specifically look at the Solvency II view of risk, as measured by the capital requirements and the liquidity risk on a non-life insurer's balance sheet.
Discussions concerning Solvency II asset optimisation have been vast and, quite rightly, this has been driven by the emphasis that Solvency II places on retaining enough market risk capital to withstand an adverse one-in-200-year event. A previous one-in-200-year event (not that one could even be identified) is almost certainly not what the next one will look like, meaning that modelling the distribution of the tail returns and how these returns might be correlated to other asset classes is speculative at best. Therefore, optimising a portfolio around these models does not make much sense, however having 'regard' to the capital consumption does.
In the asset management industry, portfolios are optimised by manager analysis and judgement of what is happening in the real world. They measure risk by reference to an asset's current market volatility and use correlations with other asset classes that are based on observable market movements.
The basis on which a portfolio is optimised has an impact on the asset allocation provided. Using a simplified model to that of our investment managers, we have produced the outputs below. These show the asset allocations across the efficient frontier for different levels of risk at portfolio level. They show how the 'mean variance' approach used by asset managers produces more adventurous portfolios, as opposed to a 'mean VaR' approach that might be favoured for a Solvency II capital optimised solution.
Asset allocations across the efficient frontier for given levels of risk
In practice, the asset allocation will be carried out on a more sophisticated model than that above. Our portfolio managers also do not look at the outcome from a single set of assumptions. They run their model under a range of scenarios, each with its own set of assumptions. At BMO Global Asset Management, the scenarios cover recession, stagnation, stagflation, trend, inflationary growth and goldilocks. The weighting attached to each scenario is actively reviewed and the asset allocation tilted accordingly.
This approach will resonate with insurers given its similarities to the scenario testing approach used in Own Risk and Solvency Assessments (ORSAs). From a prudence perspective, diversifying between asset classes in normal market conditions makes sense too. In stressed market conditions this will likely polarise into 'risk on' and 'risk off' asset classes, and an insurer ought to have a regard to how the portfolio might react in these scenarios too.
Solvency ratios and capital efficiency are under intense scrutiny. At BMO Global Asset Management our favoured response to this is the 'capital aware' approach. This is a constrained optimisation where the optimal economic portfolio is 'tilted' away from the least capital efficient assets to those that are more capital efficient. This makes it like other constraints within an asset management agreement whereby these do not unduly compromise a portfolio manager's ability to exercise his or her judgement.
On the non-life side in the UK, we have found that many insurers are still focused on the pure economics. As demonstrated in our last article, this could be because asset risk is not a huge feature of a UK non-life insurer's Solvency II balance sheet. Perhaps coincidentally but in Germany, where there is a greater allocation to market risk, we have experienced more interest in capital aware investing.
For an insurer, asset risk optimisation must have regard to liquidity. This is a huge constraint for insurers when it comes to the optimisation of their 'Own Funds' (the surplus assets that represent an insurer's capital base). Own Funds may suddenly be required for unexpected claims, reserve strengthening or the provision of funds to subsidiaries within an insurance group.
Insurers are feeling the pressure of providing higher returns. For some, it could be to meet policy guarantees, and for others to enhance returns on capital. An understanding of an insurer's liquidity profile is a key influence on strategy.
Insurers with constrained capital bases work hard to maximise the amount of liquidity to surrender to obtain an added return. Here, private investments in the forms of loans or mortgages are sought after.
Other insurers may be heavily constrained by liquidity, but could still be able to seek additional returns by increasing their allocation to risky assets. This is where an insurer's liquidity requirements can be both a liberator as well as a constraint.
There is greater opportunity to accommodate riskier asset strategies within the portfolio if these investments are liquid. Divesting from a risky asset is one of the easiest management actions available to an insurer to improve its solvency position, by reducing its Solvency Capital Requirement for market risk.
It is not ideal to be forced to de-risk, mainly because the timing may lead to divestment from an asset when its asset values are low. Minimising the volatility of these investments and maximising the diversification of the investment portfolio is one way to mitigate this risk. This has created a demand from both insurers and pension funds for uncorrelated asset classes. True diversification is hard to find, and there are a limited number of long short investment strategies specifically designed to meet this need. A range of multi asset investment strategies have also emerged. Some of these are low volatility targeted asset strategies. There are also absolute return strategies that seek to offer growth but with a regard to managing the downside potential. The capital treatment of these under Solvency II can typically be more complicated.
The requirement to maintain a safe liquid portfolio (as well as having regard to downside risk) can lead to a barbell strategy for Own Funds. In this strategy, the larger proportion of Own Fund assets are invested risk-free assets such a T-bills and perhaps cash on short-term deposit. This leaves the remainder to be invested in a concentration of risky assets that consume the risk budget. Their role is to generate the excess investment return across the whole portfolio. Such a strategy implicitly provides downside protection in extreme circumstances, as the expectation is that it is not possible to lose more than the total value of the risky assets.
There are other constraints that can further impact the optimal asset risk strategy. The propensity of German and French insurers that hold investments in related undertakings is one that we mentioned in the first article. In the third and final article, we will study the types of investment strategies that currently appeal to insurers in their search for returns and weigh up their pros and cons.
For professional investors only.
Past performance is not a guide to future performance. Values may fall as well as rise and investors may get back less than the full amount invested.
Views and opinions expressed by individual authors do not necessarily represent those of BMO Global Asset Management.
The information, opinions estimates or forecasts contained in this document were obtained from sources reasonably believed to be reliable and are subject to change at any time.
© 2018 BMO Global Asset Management. All rights reserved. BMO Global Asset Management is a trading name of F&C Management Limited, which is authorised and regulated by the Financial Conduct Authority. CM16075 (03/18). UK.
Dick Rae, Director, Insurance Solutions