After a long, drawn out process, the European legislation on prudential rules for insurance companies (Solvency II) is set for a final decision. The main text is expected to be approved early this week, following negotiations between the European Commission, European ministries of finance, and the European Parliament. This legislation will be of fundamental importance as it will dictate capital regulation for insurers.
From inertia to rush
While Solvency II is analogous to the ‘Capital Requirements Directive IV’ for bank capital, it has received less public scrutiny. This is unfortunate as parts of the proposals are still quite controversial. While the controversy slowed down the process initially, there has been a rush to address important outstanding issues in the span of only a few weeks. We fear that important aspects of the necessary conditions for effective insurance regulations have been left out in the rush to complete. For this reason, we wish to propose a short list of essential principles the adopted Solvency II rules should have to meet.
What is the issue?
The business of insurance companies is to collect premiums from the general public and companies in exchange for future promises. Insurance companies are subject to stringent regulations ensuring they hold adequate reserves so they will be able meet their obligations. These reserves are similar in nature to bank capital. Any regulatory regime needs to keep up with technical advances, and Solvency II is a major step forwards, moving to uniformed risk-based rather than volume-based solvency rules for (re) insurance companies.
What is the problem?
A major challenge for insurance companies and regulators arises from the low interest rates that followed the crisis. Under Solvency II, insurance companies are meant to use fair market valuations. As a consequence, the post-crisis present value of liabilities are higher than they would have been had the pre-crisis (higher) interest rates prevailed. This, in turn, demands higher reserves. The effect is significant given the typically long-term nature of insurance liabilities. While this may be an anomaly – stemming from a temporary, low-interest environment and thus having a short-term disproportionate impact on long-term liabilities – no certainty exists on future evolution of interest rates. Thus the low interest rates may continue to cause problems for insurance companies for some time.
What is a fair solution?
Because of the relatively higher cyclicality of asset prices and rates compared to liabilities, one might employ smoothing of the reserve levels to facilitate carrying long-term promises. For example, some of us have called for a countercyclical buffer that allows a gradual adjustment, in the expectation that the gap between assets and liabilities may be partially self-correcting, particularly with long-term liabilities. Of course, if such forbearance is allowed in difficult times, its opposite should happen in good times to avoid the problem of insurance companies becoming too thinly capitalized to absorb future shocks. Firms should build up of additional buffers when interest rates rise and risk premia fall. Earlier proposals of European Insurance and Occupational Pensions Authority endorsed this view, but only recommended weakening reserve requirements in bad times without requiring buffering in good times.
What is the proposed regulation?
Current proposals would reduce all insurer liabilities by discounting them at a higher rate than the risk free rate, and then computing the required capital buffers. This is a flawed approach, likely to lead to undercapitalization and distorted incentives for insurance companies, as shown by Koijen and Yogo (2012) for the US, where regulatory discount rates currently increase during periods of financial distress, and insurance companies were induced to sell underpriced insurance policies.
As we argued in our earlier Vox columns (2011, 2012), it is not economically logical to discount unconditional promises with a risk premium. In a more narrowly defined application for specific stable long-term liabilities, dedicated and ring-fenced insurer-specific portfolios can be tolerated as a short cut with limited risks. But this tolerance should not extend to validate insurance promises at some risk premium as a fundamental valuation procedure.
- We agree that some smoothing of reserve requirements is justifiable as market values and rates may be too variable for prudential purposes.
- We disagree with the generalized discounting approach, as it manipulates liability value rather than highlighting their excess volatility.
This may lead to reduce resilience and weakened solvency.
- The first concern is disclosure
It can make sense to grant time to insurers to adjust reserves in the present environment of ultra-low interest rates. Unlike from banks, insurers’ liabilities can be of a very long-term nature. This provides more time to recover and thus justifies a more lenient treatment; see Laeven (2012) for the full rationale.
The proposed legislation, however, is expected to define a precise formula to compute the spread to be added to the riskless discount rate. It will embed forbearance in all stages of the financial cycle by design, while requiring no extra prudential reserve building in good times.
At a minimum, the degree of temporary forbearance (the difference between required capital under fair market value and under adjusted liabilities) should be disclosed and tracked for prudential purposes.
- The second concern is prudential intervention
As the Solvency II legislation will directly set capital requirements with embedded forbearance, the national prudential authorities should be empowered to intervene, certainly if the timing of the review is set as late as four or five years. The departure from fair value implied by the discounting rule chosen needs to be monitored, to ensure it is commensurate to future circumstances and not undermining solvency.
Earlier proposals already introduced discretion in bad times, allowing prudential authorities to extend the time horizon over which required reserves may be restored. However, they did not allow in general prudential adjustments to capital buffers in favourable times, when rates rise and risk premia fall and the automatic discounting increases the reported reserves from a fair value measure. It is quite dangerous for regulators to commit in advance to forbearance in hard times, without any powers to correct rules at times when profits may be overstated in good times or understated in bad times.
- The third is timely review
The Solvency II proposal that may be approved this week will directly become law in all EU countries. All national prudential regulators, such as the Bank of England for the UK, ACPR in France and BaFin in Germany, will be bound to enforce a very novel, embedded-forbearance rule without any chance to adjust it until the proposals are reviewed.
At the critical junction of the closure of Solvency II, we identify three simple prudential principles that should be respected.
- First, since the rules will create a partially justifiable but opaque departure from fair value accounting, they must ensure disclosure of this adjustment.
- Second, the European Commission and the European Insurance and Occupational Pensions Authority should commit to a broad review within three years.
- Third, regulatory authorities should retain some discretion to intervene, if prudential concerns arise with the automatic forbearance rule.
Ayadi, Rym, Jon Danielsson, Roger J. A. Laeven, Antoon A. J. Pelsser, Enrico C. Perotti & Mario V. Wüthrich (2012), "Countercyclical regulation in Solvency II: Merits and flaws", VoxEU.org.
Danielsson, Jon, Frank C. J. M. de Jong, Roger J. A. Laeven, Christian Laux, Enrico C. Perotti & Mario V. Wüthrich (2011), "A prudential regulatory issue at the heart of Solvency II", VoxEU.org.
Koijen, Ralph S. J. & Motohiro Yogo (2012). The cost of financial frictions for life insurers, Mimeo, London Business School.
Laeven, Roger J. A. (2012). Contagion: Challenges in Risk and Insurance, Amsterdam: Vossiuspers.