Will Solvency II catalyse further investment strategy changes for UK non-life companies?

A recent report from JP Morgan Asset Management¹ highlighted a number of significant changes in average asset allocations in the UK non-life market since 2000, namely:

• a reduction in equities from over 20% in 2000 to 10% by 2003 and this level has been maintained since 2003
• a reduction in gilts from 35% in 2000 to 20% by 2007 with this level remaining stable since 2007
• cash holding rose from 15% in 2000 up to 30% in 2004 before gradually returning to 15% by 2013
• overseas corporate bonds have been the big winner with allocations increasing from 10% in 2000 to 25% by 2015

A lot of these can be easily explained by the equity losses caused by the tech bubble bursting, falling interest rates since the financial crisis and matching of liabilities in overseas currencies, primarily USD.  Interestingly, non-life companies have generally avoided the ‘flight to quality’ into government bonds, which has been the major driver of ever-decreasing gilt yields.

The primary focus for most insurance companies since 2002 has been avoiding large one-year investment losses and this makes sense given the high public profile of P&L results.  However, Solvency II will require publication of realistic solvency balance sheets as well as accounts and so solvency will become an additional source of comparison and commentary for analysts.  Return on capital (RoC) will become an easily identifiable measure.  Companies will look to allocate capital to risks where it maximises the return on this capital.

Will this have an impact on asset allocations as companies look to also manage the volatility and level of their Solvency II balance sheets?  As Solvency II is a risk-based capital regime, you might think this would lead to de-risking of investment portfolios and a reduction in market risk capital.  However, companies are allowed to apply diversification benefits between risks and so the optimal return on capital approach may be to increase risks that are relatively small currently, for example market risk compared to underwriting risk for a prudent company.

To investigate this, we have built a simple dummy non-life company in our SCR tool, SIIMPLIFY.  The dummy company writes 5 short-tail lines of business in both UK and US, has technical provisions of 15.4bn and total assets of 20.0bn. 

We tested the following asset mixes:

Bonds have short durations to reflect the duration of the liabilities and corporate bonds are A rated.

The resulting SCR calculations are as follows: 

As might be expected, the SCR does increase as risk is taken (and vice versa), but the effect is dampened significantly by diversification benefits. 

Whilst the SCR is important in determining the level of capital needed, RoC arguably is the more important measure. Various RoC metrics exist.  We have used Total Expected Investment Income / Post-diversification Market Risk and Counterparty Risk Capital based on best estimate and market implied investment returns for each of the asset classes.  The results are as follows:

So, based on this crude example, it appears that UK non-life insurers on average  already hold an optimal asset mix under this alternative S2 measure, and so maybe nothing will change!

In reality every company will be different; a balance will be needed between maximising investment income and RoC.  One thing that every firm should ensure is that RoC becomes part of the process for determining asset allocations.

¹Investment Trends in Non-life Insurance – Moving Ahead in Turbulent Times, August 2014