Longevity Risk Management: A Matter of Life and Death?
Although living longer is generally accepted to be a good thing, it is a cruel twist of fate that longer life expectancy could cause the death of UK defined benefit (DB) pension schemes. With over 10,000 people in the UK now having reached the age of 100, the need to manage longevity risk more effectively is being placed firmly under the spotlight.
Until recently, the only way to fully de-risk a DB pension scheme was through the use of insurance (bulk annuity) policies. Although this type of solution was popular during 2008 (particularly in the form of buy-in solutions for pensioner liabilities), the position in 2009 and into 2010 has taken a very different complexion, with a particular focus on longevity swaps as a way for schemes to better manage the risk of continued improvements in member life expectancy.
Longevity risk transfer is commonplace in the reinsurance world and there is no shortage in investors’ appetite to provide the risk capital to do so. The challenge for the UK pensions industry has been to translate this into a format suitable for pension schemes. However, as the recent £3bn BMW pension liability has demonstrated, UK pension schemes now have a new weapon in the risk management armoury.
Longevity risk is different to most risks facing pension schemes in that it is ‘serially correlated’. Although equity markets can be volatile, we would not expect equity performance next year to be much affected by equity performance last year, and not at all affected by equity performance 10 years ago. With longevity, we would expect an increase in life expectancy this year to feed through to an increase in life expectancy for all future generations. Therefore a sustained increase in life expectancy on a year-by-year basis, even if the increments are small, can have a huge impact when projecting life expectancy 20 years from now, as illustrated in Figure 1.
A longevity swap enables a scheme to remove the risk that members live longer than expected. However, unlike a buy-out or buy-in, the scheme retains control of the assets and so does not remove the potential future upside from the scheme’s investment strategy, which remains a key part for most scheme financing plans.
There are a number of scenarios where this type of solution might be attractive. For schemes that want to de-risk, but cannot afford to do so, a swap will deal with the longevity risk immediately and allow the scheme to gradually remove other risks at a more affordable pace. For those schemes that have already de-risked using interest rate and inflation swaps, longevity swaps provide the missing piece in the risk management jigsaw – in effect, by combining longevity swaps with a liability driven investment strategy, a pension scheme can construct a ‘do-it-yourself’ bulk annuity.
Going forwards, any scheme considering a buy-in solution will check how this compares to a DIY solution as part of the transaction due diligence.
BMW, RSA, Babcock International and the Royal County of Berkshire have all announced longevity swap transactions, with many pension scheme trustees and sponsors looking to transact over the coming year.
Longevity swaps come in a variety of flavours, ranging from ‘bespoke’ to ‘index’ solutions. Bespoke solutions are customised to the individual pension scheme and based on the specific members in the scheme. Index solutions provide protection against general increases in population longevity, but are not tailored to the individual scheme in the same way as a bespoke solution.
Most of the activity at the moment is focused on bespoke solutions that deal with pensioner members. This is because the pricing is more competitive for large pensioner populations and because initial interest has tended to be for larger schemes looking for solutions that hedge as much risk as possible.
Index solutions lend themselves more readily to dealing with non-pensioner member risk, particularly where there is uncertainty about the timing at which members will retire and the amount of pension that will be brought into payment. For example, when members elect to take tax free cash instead of taking all their benefits as pension, the amount of longevity risk is reduced. Options such as this do not integrate very easily with a bespoke solution, which aims to match the actual pension payments as closely as possible.
Increasingly, as the market develops, we expect to see a combination of bespoke and index solutions being adopted as schemes look to deal with both pensioner and non-pensioner longevity risks.
Longevity swaps can be provided as an insurance product or as a derivative. For insurance contracts, a review of the strength of the insurer is required, similar to the type of due diligence required when considering a buy-in policy. For a derivative product, collateral arrangements are required to provide protection against counterparty risk for both the pension scheme and the provider. Standard investment documentation – similar to that used for interest rate and inflation swaps – is used, including details of how collateral is calculated and posted. At the moment, because there is not yet a liquid market in longevity risk, the collateral amount is calculated using a documented method, usually referred to as ‘mark to model’. Once the market becomes more liquid, the collateral can be calculated using a conventional mark-to-market method.
It is possible to include collateral arrangements in an insurance arrangement, just as it is becoming increasingly common for additional security to be negotiated as part of a buy-in solution. To a large extent we are seeing a ‘best-of-breed’ approach emerge, incorporating some of the standard investment collateralisation techniques into insurance products.
At the moment, sponsors and trustees typically want to understand the funding implications of entering into a longevity swap. If the swap can be entered into having little or no impact on the funding position, or the cash contributions that are being paid to the scheme, then a longevity swap can be a very attractive proposition.
This type of analysis means trustees and sponsors need to think very carefully about the life expectancy assumptions that are being used for funding purposes, how those assumptions are likely to change over the next few years and how much appetite the scheme has to continue to run longevity risk.
At the moment, providers continue to be keen to write deals – it’s difficult to know how long this level of pricing will persist, and it’s easy to be cynical about ‘buy now while stocks last’ sales pitches, but early movers may well be in a good position to negotiate favourable terms.
As with any investment transaction, careful due diligence is required to ensure that all parties are comfortable with the value for money of the product and the risks that are being eliminated or retained. One of the frequently asked questions in relation to longevity swaps is on exit terms and whether a longevity swap precludes a scheme from moving to a buy-in solution later on. It is possible to agree the terms for exiting the swap to provide some comfort on this point. The method for agreeing the exit terms tends to be negotiated on a case-by-case basis at the moment, although this is likely to become more standardised once a larger number of deals are completed.
Most of the capacity to provide longevity swaps is currently serviced by the global reinsurance market, with an estimated annual appetite of around £15-20bn from current re-insurers. While this is undoubtedly a very large number (and probably ample to satisfy the 2010 demand), it is worth bearing in mind that UK pension fund exposure to longevity risk exceeds £2 trillion – so in terms of a long-term sustainable solution, it is difficult to see that re-insurance alone will provide the universal panacea.
The alternative source of capacity is to open up a line from capital market participants. Longevity risk can be an attractive proposition for capital market investors, where it is viewed as a source of diversification to other financial assets.
The challenge in this context is to structure the transfer of risk in a format that is acceptable for investors – because while UK trustees have a desire for solutions that are highly customised to closely match the long dated and complex nature of pension benefits provided to scheme members, capital market participants generally have a strong preference for investing in products that are shorter dated, easy to understand and as liquid as possible.
Realistically, it is unlikely to envisage a situation where pension schemes hedge longevity risks directly with capital market investors. So it follows that for this market to survive, we will need to find a bridge between the pensions world and the capital markets. This is where the banks and insurers/re-insurers are likely to find their longer-term home, by providing the customised solutions that pension schemes desire, but sourcing the capacity to structure such products by intermediating the risk to the capital markets via offering simplified digestible packages of risk to investors.
Now that the market in longevity swaps has firmly established itself, the discussion changes from one of whether to remove longevity risk to whether there is any appetite to retain the risk.
What is clear is that as the UK pensions market continues to focus on understanding and managing all sources of risk, it is inevitable that the dynamics for longevity risk transfer will evolve, with both reinsurance and capital markets having a key role to play.