Jurre de Haan, Agnes Joseph, Siert Jan Vos, Jan-Willem Wijckmans compare Solvency II with the FTK in terms of the likely coverage ratio shortfall
On 15 February 2012, the European Insurance and Occupational Pensions Authority (EIOPA) presented its response to the call for advice of the European Commission regarding the revision of the IORP Directive. An important part of this call for advice1 is dedicated to the question to which extent the Solvency II framework for insurance companies could be used to create a harmonised European framework for the supervision of IORPs. The main instrument that EIOPA has proposed to achieve this latter objective is the ‘holistic balance sheet framework’. In this framework all security mechanisms within a particular pension scheme should be valued and taken into account. Although this framework theoretically offers the possibility to achieve the aims of the review, the applicability and complexity involved is still an issue of concern.
In a recent statement, Commissioner Michel Barnier of Internal Markets and Services clearly indicated that there is no intention to directly ‘copy-paste’ Solvency II rules2. This article offers support to this judgment by showing that applying Solvency II to Dutch IORPs results in a highly adverse impact for both the individual participant and the larger economy.
Comparison of FTK and Solvency II
The requirements of the current IORP directive are embedded in the financial assessment framework (FTK) in the Netherlands. The structure of the FTK is quite similar to the structure of Solvency II. Both supervisory regimes have risk-based capital requirements and market-consistent valuation of both assets and liabilities. However, there are three major differences in the detailed implementation of these principles.
The first difference is in the details regarding the valuation of the liabilities. Under FTK rules, the value of the liabilities is based on the unconditional pension benefits, while under Solvency II, conditional benefits are also part of the value of the liabilities. Since for the average Dutch IORP indexation is an important conditional benefit, Solvency II requirements will increase their best estimate value of the liabilities.
On top of this best estimate, Solvency II rules impose an additional risk margin to assure that liabilities can be transferred to a third party at any time. Finally, although both supervisory regimes use a discount curve based on swap rates, Solvency II imposes several modifications to this curve such as stabilisation of the interest rate levels for long-term maturities.
The second difference between FTK and Solvency II comes from the capital requirements. Solvency II assumes an ex-ante security level of 99.5%, which implies that the solvency capital requirement (SCR) should be sufficient to absorb unexpected shocks 199 out of 200 times on average. In contrast, the capital requirements under the FTK are based on an ex-ante security level of 97.5%. The lower security level under FTK is justified by the observation that IORPs can use several ex-post measures to mitigate risks, such as increasing contributions.
The third difference concerns the time limit for recovery in case of non-compliance with the capital requirements. Under Solvency II, measures have to be taken to ensure that the available capital satisfies the SCR within six months’ time. In extreme circumstances the supervisory authority may choose to extend the recovery period3. Under FTK, the recovery period is much longer. Dutch pension funds have 15 years for recovery until the SCR (DNB, 2012a)4 .
Solvency II rules lead to higher technical provisions and capital requirements
The effects of Solvency II regulation are illustrated by applying the quantitative requirements to a representative Dutch IORP5. The amount of assets has been tailored to reflect the total amount of assets of the entire Dutch pension sector to be able to draw generalised macro-economic conclusions. The asset portfolio of the IORP consists of 60% bonds and 40% equity and a 50% hedge of the interest rate risk. This results in a SCR equal to 17% of the liabilities under FTK rules. The initial coverage ratio of the IORP is also set to a value of 117%, to allow for an analysis of the structural effects of a change in the supervisory regime.
It is unclear how conditional and discretionary future indexation should be treated in calculating the value of pension liabilities under Solvency II. We therefore present three cases. In case A, future indexation under Solvency II is treated in exactly the same way as it is under FTK - ie, it is not part of the liabilities. In case B, future indexation is treated as a financial option, and it is included in the technical provision by calculating its value using standard option valuation techniques. In case C, future indexation is considered to be unconditional, effectively meaning that all cash flows are increased with future expected indexation up until that point.
Figure 1 displays the results of the analysis. In case A of the Solvency II rules the value of the liabilities is lower than the value of the liabilities under FTK. This is the result of several opposing effects. Under Solvency II, the value of the liabilities increases due to the risk margin (5.0%). However, the adjustments on the discount rate have a stronger downward effect of 7.2%. Note that, the impact of modifications to the discount rate is not a constant and can, for example, become much smaller, or even switch sign as the level of the swap curve becomes higher.
In cases B and C, we see that the value of the liabilities is substantially higher than under FTK rules, due to the valuation of future indexation. Since initial assets are assumed identical across cases, this pushes down the initial coverage ratios to 105% and 83%, respectively.
Capital requirements under Solvency II strongly increase across all cases. The SCR increases by over 50%, from 17% of the value of liabilities under FTK to roughly 27% of the value of liabilities. Combined with the increase in the value of the liabilities in cases B and C, the absolute size of the SCR goes up even more: the capital requirements in variant C are more than twice as high as the capital requirements under FTK.
Required capital to fund shortfall has large macro-economic impact
In figure 1, starting with FTK there is initially no funding shortfall, but a shortfall arises with the introduction of Solvency II requirements. This capital shortfall has to be restored. The funding of this shortfall, especially given the short recovery periods in Solvency II, would have large macro-economic implications. Even in case A, in which capital requirements go up by the smallest amount, the funding shortfall of €38bn would amount to more than the total annual pension contribution paid, being around €28bn in 2010 in the Netherlands6. This would have a huge negative impact on economic growth, consumption, employment and the government budget: an increase of the pension contribution rate with 1% leads to a decrease of the GDP with approximately 0.2% .
In case B, which seems to be the most likely outcome of the translation of Solvency II rules to pensions funds, the recovery contributions to be paid in a short amount of time become even more unrealistic. The €155bn shortfall equals around 25% of Dutch GDP.
It seems unlikely that increasing contributions or additional sponsor support is a feasible solution. Other measures, such as withholding indexation and reducing benefits, will be necessary. Due to the very short recovery periods out of Solvency II, these measures will have to be radical in order to achieve the required level of capital.
Strong incentive for defensive asset mix
Adjusting the asset mix is an option to mitigate the impact of the higher capital requirements under Solvency II in the short run. Figure 2 presents the results for a more defensive portfolio with only 20% invested in equity, instead of 40%. Results for case A of Solvency II are displayed in order to keep the analysis more comparable with the initial FTK results. Shifting to a more defensive investment strategy pays off in the short run under Solvency II, by decreasing capital requirements and therefore avoiding or lowering the recovery measures that otherwise would have to be taken. In the longer term, however, the drawback of this strategy is clear: expected returns on the portfolio will be lower, so expected pension benefits will decrease.
The results from figure 2 show that Solvency II attaches very high capital requirements to investing in equity. Because of short recovery periods, the capital requirements have to be paid for by current stakeholders of the IORP. This provides a strong incentive to shift to a less risky asset mix. In cases B and C the incentive to de-risk will be even stronger.
The analysis shows that de-risking can alleviate the need to immediately fund the capital shortfall. Large-scale de-risking, however, has a comparable negative impact on macro-economic growth. IORPs are an important supplier of risk-bearing capital to the European economy. Imposing Solvency II requirements on IORPs, especially given the already precarious funding situation of most funds, could lead to less capital being available for the private sector, which will have a negative impact on economic growth and employment.
Moreover, a sell-off in equities will unavoidably result in a decrease in the price of equities, exacerbating the difficulties in the financial markets. This could subsequently result in a negative feedback loop, where other institutional investors also have to sell off equity to meet liquidity or solvency requirements, with a further decrease of asset prices as a result. In the example given in this article, it can be seen that a reduction in equities from 40% to 20% for the entire Dutch pension sector, would amount to additional selling of risk-bearing assets of around €150bn. Coupled with possible similar movements for other large funded pension schemes, such as in the UK and Ireland, the impact can be quite substantial.
The statements of Commissioner Barnier that Solvency II will not be applied unchanged to IORPs because of the large impact this will have on investments and growth are very much in line with the results presented in this article. Applying the Solvency II rules on a one-for-one basis to Dutch IORPs will adversely affect economic growth through either capital injections in the fund or reduced purchasing power of participants. The required level of assets of the IORPs will increase substantially due to the higher security level and the requirement to include conditional benefits in the liabilities.
It seems unlikely that increasing contributions or sponsor support is a feasible solution to meet the shortfall. Other measures such as withholding indexation and reducing benefits will therefore be necessary. Due to the very short recovery periods prescribed under Solvency II, these measures will be radical and will have a significant impact on the purchasing power of retirees.
(1) European Commission (2011) Call for Advice from the European Insurance and Occupational Pensions Authority (EIOPA) for the review of Directive 2003/41/EC (IORP II)
(2) Barnier, M. (2012) Statement by Michel Barnier on future european rules for pension funds, February 10th 2012, http://ec.europa.eu
(3) European Commission (2009), Directive 2009/138/EC of the European parliament and of the council of 25 November 2009 on the taking-up and pursuit of the business of Insurance and Reinsurance (Solvency II)
(4) De Nederlandsche Bank (2012a), Financieel Toetsingskader (FTK) voor pensioenfondsen - Open Boek Toezicht, January 2012, www.dnb.nl
(5) Some of these results have been published earlier, see De Haan, Jurre, Agnes Joseph, Siert Vos and Jan-Willem Wijckmans (2012), ESB issue 4629, February 17th 2012
(6) De Nederlandsche Bank (2012b), http://www.statistics.dnb.nl/index.cgi?lang=nl&todo=Pen2
(7) Vuren, A.H. van (2003), Financiële consequenties pvk-regels, cpb Memorandum 56, februari 2003
Jurre de Haan is senior policy adviser at APG; Agnes Joseph is senior actuary at Syntrus Achmea Asset Management; Siert Jan Vos is policy adviser at MN and Jan-Willem Wijckmans is strategic risk manager at PGGM