Paul Sharma has over 20 years of experience as a top UK, EU and global regulator of banks and insurance companies. He was involved in the design of Solvency II since its early stages and chaired the ‘London Working Group’, which, in 2002, authored the “Sharma Report” – a pivotal document in shaping the thinking and design of Solvency II. In this exclusive article for Solvency II Wire he addresses one of the Directive’s most controversial and contested principles: the treatment of Long-Term Guarantees (LTG) under a market consistent regime. Mr Sharma argues that public disclosure is the most effective way of ensuring the LTG measures are used effectively and appropriately.
Some reliance on market consistency is essential for the sound regulation of UK and EU life insurers. Criticism of Solvency II has raised questions as to whether that reliance had gone too far, while the recent Omnibus II revisions to Solvency II have sought to answer those questions by reducing that reliance.
Historical evidence shows that valuation methods that delay the recognition of bad news expose both insurers and their policyholders to unnecessary and excessive risk. However, not all fluctuations in market consistent values of assets and liabilities are genuine and enduring news relevant to the suitability and safety of an insurer’s assets to support payment of its liabilities. What’s more, excessive reliance on market consistency introduces unwanted volatility onto the balance sheet.
To mitigate this problem, Solvency II includes a number of dampening measures. But it is important that these measures are used to deal with the excess volatility and do not result in a permanent reduction of capital requirements. The only effective way to do this is by publicly disclosing the impact of the measures so that market participants can make informed decisions about the robustness of insurers.
The case for and against market consistency
The main argument in favour of market consistent valuation is that the volatility of financial markets means that financial assets can quickly become inadequate to support the payment of policyholder liabilities.
In theory, an insurance company that has closely matched the cash flows of assets to those of liabilities need not worry about the day-to-day value of its assets or liabilities. In practice, market value fluctuations sometimes gives valuable information about the deterioration in the safety and suitability of assets that it would be imprudent to ignore. Also the promises that European insurers give to policyholders typically are of longer duration than the assets that are available to match them. Insurers rely on reinvesting maturing assets to meet policyholder liabilities, and where those liabilities included derivative-like promises may need to trade assets actively to match the liabilities.
The main argument against market consistent valuation is also the volatility of financial markets. Much of that volatility is merely random noise with no relevant information to the safety or suitability of assets. More seriously – as the global financial recession has shown – asset and interest rate market value fluctuations often reflect fluctuations in the market price for liquidity. However insurers only need to hold liquid assets to the extent to which the timing of their liabilities is uncertain.
Equitable Life, the UK insurer, was a well-known example of the under-appreciation by traditional actuarial techniques of the liability arising from long-duration derivative-like promises to policyholders. The firm promised guaranteed annuity rates that were the economic-equivalent of writing very long duration and – at the time they were written – deeply out of the money interest rate options. More broadly, life insurers often give fixed monetary promises backed by equity-assets where policyholders participate in the upside. Where the policyholders’ upside participation and downside guarantee are strong promises (which is not always the case) this situation resembles the writing of a long-duration deeply out of the money equity put option. Traditional non-market consistent actuarial techniques often struggle to adequately evaluate the time-value of such long-duration deeply out of the money options.
Conversely market consistent valuation may, during periods of stress, understate the ability of an insurer’s assets to secure the payment of its liabilities. Further, on a pure market consistent basis, the capital needs of an insurer to cover possibly temporary or irrelevant fluctuations in market value may be overstated.
Both types of error – under or over estimating the strength of assets to support policyholder liabilities – risks unnecessarily harming policyholders, destroying value for insurers and hindering insurers performing one of their economic roles (matching long term promises to policyholders with long term investments).
The crisis: uncomfortable truth or flawed solvency system?
The problem with any solvency standard (one that aims to measure a firm’s solvency by comparing the value of its assets to that of its liabilities) is that it is vulnerable to either ignoring extreme market fluctuations when it should pay regard to them or paying heed to them when it would be better ignoring them.
At the time that the key principles of Solvency II were first designed the importance of this problem was understood, but the extremity of possible market fluctuations was not.
No one contemplated the extreme market value fluctuations that were to come during the global financial crisis due to a flight to liquid assets, strong deterioration in credit conditions and central banks reducing interest rates below previous historical lows.
The QIS5 trial run of Solvency II (conducted in 2010) was made at the height of the crisis, when asset prices were extremely depressed both for equities and, more crucially, for European government debt. This showed that many insurers would either be insolvent or at high risk of insolvency if Solvency II were to be introduced at that point.
The key question is whether the solvency system (Solvency II in Europe) had failed? Does the fact that the introduction of Solvency II to its original timetable would have caused mass insurance failure mean that the system is flawed, or does it mean that it is doing its job and highlighting risk?
The answer depends, to a large extent, on whether you believe the crisis represents a temporary or permanent reduction in asset valuation (especially in respect of assets originated immediately prior to the onset of the global financial crisis) and a long-duration fall in market interest rates.
If you believe that asset prices have permanently dropped (a secular change) then you could say the solvency system is working and exposing the vulnerabilities of companies providing long-term business. If, however, you believe the crisis is a temporary aberration of the long-term average (a cyclical change) then you would argue that the solvency system is flawed.
The vulnerability of market consistency proposed in Solvency II was exposed during the recent financial crisis. It showed very clearly both why market consistency was needed, but also the difficulty of introducing it in the midst of a crisis.
Challenges of introducing Solvency II in a crisis
The problem we are now faced with is that delaying the introduction of Solvency II is simply not an option. And, whether we like it or not, we are introducing a solvency system during the tail-end of a crisis.
There is little doubt in my mind that had Solvency II been introduced five years prior to the crisis, the impact of the crisis on the life insurance industry would have been significantly lower. This is precisely what we saw in the UK, where the substance of Solvency II market consistency was largely in force from 2003 (introduced by the ICAS regime), in response to previous crises. As a result, UK insurance companies reduced their asset-liability mismatch and understood it better. And although they were put under severe strain by the crisis, they were in a much better position to cope with it than some other significant life insurance industries elsewhere in Europe.
Conversely, were we to introduce Solvency II a few years from now, then hopefully by that time the insurance companies could have recovered full health.
Introducing Solvency II during the financial crisis requires a degree of flexibility about what to do with the new information revealed by the solvency system, looking at the financial health of the existing stock of life insurance companies.
As outlined above, one’s view on the merits of Solvency II depends on how you interpret the transience, or otherwise, of the impact of the crisis. What resulted in the Omnibus II Long-Term Guarantees agreement was in effect a compromise of these two views. Given the political reality this was probably the only feasible solution to the problem.
If one holds that the solvency system was malfunctioning, and that the impact of crises is purely cyclical, then the argument follows that asset prices reliably revert to the long-term average. In that case, the holder of this view would want the solvency system to be adapted so that solvency requirements could be relaxed during the next crisis.
If, however, one believes that the solvency system was functioning and the impact of the crisis is (at least in part) secular, thus revealing a real problem, then the desired outcome is simply transitional time to cope with the situation.
On this side of the argument, one desires measures that do not affect the long-term design of Solvency II but relax its immediate vigour. They are then gradually stepped up over a number of years. For my part I believe the system is functioning and that we are unlikely to see a return to the levels experienced before the crisis.
The Omnibus II agreement does actually incorporate a very practical and good solution to addressing this problem – disclosure of the impact of the measures.
Public disclosure of the impact of the measures on the balance sheet – before and after the application of the measures – will let the management of insurance companies, the regulators and market participants take their own decision as to how believable they think the regulatory credit is.
The views expressed are the author’s own. Paul Sharma is a Managing Director, Alvarez & Marsal and co-head of the firm’s Financial Industry Advisory Services practice in London. He was formerly Deputy Head of the UK’s Prudential Regulation Authority (PRA) and an Executive Director of the Bank of England.
To subscribe to the Solvency II Wire mailing list for free click here.