As the IAIS kicks off1 the final year of field testing before it adopts the Insurance Capital Standard (ICS) at its Annual Conference in November 2019, comparisons with the experience of Solvency II inevitably arise: the IAIS is tackling many of the same policy issues that have already been faced by Solvency II, given both regimes target “market consistency”.2
Leaving aside the question of how the ICS will operate alongside established regulatory regimes following the end of its five year monitoring period,3 the ICS that will be adopted in November could present some novel approaches to market-consistent valuation methodologies that reflect some of the lessons and experiences from Solvency II. In the context of the 2020 Solvency II review, and, potentially, the future UK prudential insurance regime, this could prove to be highly important.
This blog is part of a two-part series examining the calculation of market-consistent technical reserves in the ICS, which is one of the most important points of valuation methodology for a market-consistent regime.4
- This blog identifies some important differences in the ICS’s calculation of the risk margin adjustment as compared to Solvency II. These could lead overall, and importantly, to a less volatile and interest rate-sensitive calculation, that could be attractive to the UK in particular.
- At the upper end of the calibration ranges being tested by the IAIS, the ICS margin appears to approach the level of the Solvency II risk margin as a proportion of capital, at a level that may be expected to be stable even in a higher interest rate environment. If borne out by field testing, we expect this could well be concerning to insurers who consider the quantum of the Solvency II risk margin to be too high at present levels of interest rates.
The second blog in this series, available here, compares the calculation of long-term discount rates for the euro and sterling under the ICS and Solvency II. It discusses some factors that could, potentially, introduce significant differences in valuation and volatility between the two regimes.
Implications for firms
The 2019 field test is the last opportunity for the IAIS to test its narrowing policy choices before finalising the ICS that it will put to adoption at its Annual Conference. Consequently, it is the last year in which the ICS can be influenced through participating in field testing.
The ICS is testing a number of policy approaches that could well be instructive for Solvency II and the 2020 review currently underway. In particular, the extreme interest rate sensitivity of the risk margin has arguably been the single biggest unintended consequence of Solvency II, at least for the UK market. The ICS is testing a novel approach that, on the one hand, could demonstrate a less interest rate sensitive alternative methodology, but on the other hand could establish a calibration level that insurers may consider to be too high.
Insurers with an interest in mitigating the effects of the risk margin, as well as in the future shape of Solvency II and the ICS generally, would thus be well-advised to participate in field testing if eligible, as well as engaging in any associated consultations.
Field testing is scheduled to run until the end of July.
Valuation of insurance liabilities
Both the ICS and Solvency II value technical provisions as the sum of a best estimate and risk margin – in Solvency II terms the premium over the mean valuation that an arm’s length purchaser would require to take on the portfolio of liabilities. The methodology for calculating the risk margin is a critical driver of the overall result, and in many countries applying Solvency II, the risk margin has comprised a very significant proportion of overall claims-paying and capital resources. Data published in EIOPA’s second set of advice for the Solvency II review, for example, showed that the risk margin exceeded 20% of the capital requirement for life insurers in 24 out of 27 EEA states for which data was available as at Q3 2016, with the highest result being in the Netherlands where the risk margin comprised 76% of the SCR.
Consequently, the interest-rate sensitivity of the Solvency II risk margin methodology has been one of the most fiercely debated aspects of the regime.5
Calculation of the risk margin
For 2019 field testing, the IAIS is testing three calibrations of the risk margin (in ICS terms, the Margin over Current Estimate, or MOCE), calculated in each case using what it calls a “Percentile-MOCE” approach. The Percentile-MOCE is intended to measure the variability and uncertainty in a portfolio of insurance liabilities to a specified level of confidence; for simplicity, it is calculated in the 2019 exercise based on the life and non-life ICS risk charges assuming that the variability in the present value of future cash flows follows a normal distribution. The calibrations being tested are the 75th, 80th and 85th percentiles for life risks and the 60th, 65th and 70th percentiles for non-life risk.
In previous years, the IAIS has also tested a cost of capital approach to calculating the MOCE (the “C-MOCE”), and a “Prudence MOCE” (P-MOCE). The C-MOCE is conceptually similar to the cost of capital calculation used for Solvency II, with both fixed and variable cost of capital rates being tested. The P-MOCE is intended to capture the inherent uncertainty relating to the insurer’s future cash flows, representing the estimation of future profits for non-life risk, and a proportion of the estimated standard deviation for life risks. The IAIS is not collecting further data on these approaches in 2019 field testing, but does not indicate either that it has ruled them out.
Importantly in the context of Solvency II, any of these approaches,6 if taken at face value, should result in a less interest rate sensitive calculation than the Solvency II risk margin, and therefore a risk margin that is overall more stable over time. The Solvency II risk margin is currently calculated using a fixed 6% cost of capital rate, meaning that the margin increases when interest rates are low, and decreases as they rise; consequently, the PRA in particular has criticised its construct as inherently pro cyclical. EIOPA has recently re-calculated this rate using the same approach applied when it was originally set, and recommended it be left unchanged (EIOPA’s recalculated figures were, in fact, higher, at 6.7% to 7.8%).7 However, the European Commission is now re-examining both whether the cost of capital rate should continue to be fixed for all insurers, and the assumptions used to derive the rate.8
Overall size of the risk margin
Given current prevailing levels of interest rates, any comparison of the overall result of the ICS MOCE and the Solvency II risk margin essentially turns on the interest rate sensitivity of the risk margin.
Given this, it is striking that the ICS MOCE, at the upper end of the range of calibrations being tested this year, appears not wholly dissimilar in quantum to the Solvency II risk margin.
For 2019 field testing, the Percentile-MOCE result is approximately 26%, 33% and 40% of the aggregate life risk capital charge, and 10%, 15% and 20% of the non-life risk capital charge (in each case, post-diversification and including the effect of management actions). The P-MOCE result tested in 2018 was, for life risks, slightly over 25% of the life risk capital charge on the same basis, indicating a calibration at the lower end of the range being tested this year.9
Directly comparable figures are not publically available for Solvency II. However, EIOPA published figures for the size of the risk margin relative to the SCR across Solvency II jurisdictions for the first to third quarter of 2016 in its second set of advice for the Solvency II review. For life insurers, these ranged from 10% to 76% of the overall SCR, with a weighted average of 35%. For non-life insurers, they ranged from 6% to 31% of the overall SCR, with a weighted average of 21%.10
Comparisons with these Solvency II figures requires adjustment for changes in long term interest rates since Q3 2016 and the effect of the transitional measure on technical provisions. As these would largely be expected to increase the risk margin as a share of SCR,11 a reasonable conclusion can be drawn that the Solvency II risk margin currently represents a higher proportion of the SCR than should be the case under the ICS approach.
However, the difference is not so significant, at the higher calibrations, as to dispel the criticism that, like Solvency II, the ICS’s application of risk margin is forcing insurers to maintain a level of de facto financial resilience well in excess of the target 99.5% VaR over a one year time horizon. Under Solvency II, the burden generally falls most heavily on insurers writing long term life business, some of whom currently carry a level of risk margin that is substantially above the weighted averages calculated above.
However, in the case of the ICS Percentile-MOCE, and in contrast to Solvency II, the level of risk margin will be established in the clear knowledge of its likely quantum and impact in an era of sustained low (and still falling) interest rates.
1 The IAIS published the 2019 quantitative field testing package for the ICS on 25 June 2019.
2 As is well known to those insurers and regulators operating under Solvency II, market consistency relies on a large number of methodology and assumption choices to compensate for the lack of deep, liquid and transparent market prices for many assets and liabilities held by insurers.
4 The calculation of regulatory capital requirements is also, of course, of critical importance.
5 For example, in July 2018 evidence to the UK’s Treasury Committee, the PRA’s David Belsham stated “…a lot of these problems keep coming back to the risk margin. The capital requirement for annuities is not excessive in terms of the solvency capital. It is because the risk margin has effectively been mis‑calibrated and is too sensitive to low interest rates, which is why we are seeking to fix it.” In his keynote speech at the June 2018 Insurance Europe conference on Solvency II, European Commission Vice-President Valdis Dombrovskis noted that “We will however take a hard look at the broader design of the risk margin in the context of the review of the Directive in 2020.”
6 The exception would be if the IAIS uses a C-MOCE approach with a fixed cost of capital rate, but in our view this would be an unlikely outcome. In 2018 field testing, the IAIS tested a fixed cost of capital rate of 5%, but indicated that the rate may be revised in the future, including potentially linking the rate to “some economic variable(s)”. Fixed at 5%, a lower overall quantum of C-MOCE would be expected compared to the 6% Solvency II rate, but the calculation would be sensitive to changes in interest rates in the same way as the Solvency II methodology.
7 EIOPA’s second set of advice to the European Commission on specific items in the Solvency II Delegated Regulation, EIOPA, 28 February 2018
8 Formal request to EIOPA for technical advice on the review of the Solvency II Directive, European Commission, 11 February 2019
10 These figures represent the risk margin share of the SCR for Q3 2016 reported in EIOPA’s second set of advice for the Solvency II review, weighted according to the total SCR for each EEA state reported for FY2016 in EIOPA’s published insurance statistics, in each case excluding states for which data was not available. Note that the total SCR by EEA state is for both life and non-life insurers, so the weighting is approximate.
11 EIOPA’s 2018 report on the LTG measures indicates that technical provisions as a whole (i.e. best estimate and risk margin) are reduced by around 1.2% by the transitional measure on technical provisions. In practice, the effect of this transitional measure varies significantly between countries. The UK PRA has also indicated that, for every 100 basis point fall in interest rates, it expects a 27% increase in the overall risk margin of major UK life firms.