Company News » Special feature: Happy returns – the increasing popularity of longevity swaps

Special feature: Happy returns – the increasing popularity of longevity swaps

They're complicated, potentially costly and not without risk. So why are longevity swaps becoming an increasingly popular way to reduce final salary scheme liabilities?

For now, depressed stock market values have effectively
shelved any prospects of companies being able to offload their final salary
pension schemes through a buyout funded by the major life insurers or any of the
new players on the market. The fact that, in a buyout, schemes hand over their
assets to the provider makes it very difficult right now, since it’s not an
attractive proposition to have their assets valued and their losses crystallised
at today’s prices.

The question, then, is what this means for finance directors, most of whom
remain deeply unhappy about the cost and balance sheet volatility associated
with their final salary schemes. The increasingly fashionable answer right now
is to think about a ‘longevity swap’ as a way of taking some risk out of the
scheme.

Longevity creep adds hugely to a scheme’s liabilities since it has to fund
members’ retirement for longer – possibly much longer – than it had anticipated.
The beauty about a longevity swap is that it transfers to a swap provider the
risk that scheme members will enjoy their retirement years for what scheme
sponsors would consider an annoyingly long time.

Whether a longevity swap is worth doing depends on two things: the amount of
pain that mortality creep is likely to inflict and the cost of the swap. But
there’s a third factor: whether the trustee board is up to the task of
understanding what a longevity swap is all about.

How it works
As Jerome Melcer, a partner with Lane Clark & Peacock (LCP) explains, in a
longevity swap:
• The provider (an insurance company, bank or money market player), takes the
risk that scheme members who are currently receiving retirement benefits will
live longer than a specified average span. (No swap has yet been structured for
active or deferred members – ie, current or ex-employees.)
• In exchange, instead of paying benefits to the members, the scheme pays an
equivalent revenue stream to the swap provider.
• Ultimately, the scheme will pay out as much as it would have done had all the
scheme’s pensioners covered by the swap members (those already receiving
benefits) lived to an agreed average age.
• To the extent that they live beyond this age, the swap provider takes over
paying the benefits.

The clear benefit here is that the scheme’s liability to members is capped.
Effectively, the uncertainty over longevity is swapped out of the scheme.
Trustees can sleep more easily at night knowing they are not going to have to go
cap-in-hand to the sponsor for more funding every time the scheme actuary tells
them that longevity has gone up by another year.

A key factor to be taken into consideration is counterparty risk. Clearly, if
you are writing a (virtually) indefinite longevity swap, there is a risk that
your counterparty will not be around to back up its promises. (Anyone who thinks
this risk is small hasn’t been paying attention for the last couple of years.)

Gavin Orpin, head of trustee investment at LCP, points out that the solution
is to ensure that counterparty risk is dealt with through effective
collateralisation of the contract as time moves on. Whoever is ‘out the money’
in the swap arrangement (the bank provider, if longevity increases more than
expected; the scheme if longevity were to behave lemming-like and shorten more
than expected) must put up extra collateral – generally, a mixture of cash,
government bonds and possibly some investment grade corporate bonds.

However, this is where Orpin sees some severe difficulties ahead for the
longevity swaps market. ‘You can collateralise swap deals for inflation and
interest rate risk since there is a market that determines the price, so the
collateral can be marked-to-market on a daily basis,’ he says. Longevity is less
volatile than interest rates so the two parties involved would probably agree to
review the collateral on a monthly or quarterly basis.

But this is where the problem would arise: there is no market to determine a
value for longevity. An insurance company may use existing data and a bank would
use an internal proprietary model to calculate probable longevity. But in each
swap what is at issue is a specific scheme with its own, very specific longevity
‘challenge’. Longevity swaps can be based on a particular scheme’s mortality
experience but who determines how the collateral should move?

Take the following worst case scenario as outlined by Orpin:
• A scheme takes out a longevity swap.
• Five years later the bank defaults.
• Over the five-year period, longevity experience worsened such that the bank
owed the scheme money.
• However, the collateral the scheme received was based on the bank’s internal
model, not on what the scheme actuary specified as the longevity for the scheme.

• In such a case, the bank collateral may not cover the additional liability and
the scheme would be out of pocket.

Orpin says there is already some evidence that difficulties in resolving this
question are preventing longevity swaps moving from initial agreement to inking
the contract.

Costly option
Another critical point is the cost of a swap: it’s not cheap. Exact figures are
hard to come by and will depend on trustees’ own view of how long their scheme
members are likely to live. If trustees are more cautious (ie, their longevity
estimates are at the ‘old’ end of the range), the relative cost of the hedge
will be lower and vice versa. Orpin says that, based on pricing his firm is
seeing at present, hedging out longevity might increase the value of pension
liabilities by around 5%. So, for a pension scheme with liabilities valued at
around £500m, a hedge today might add to that around £25m.

That’s a fair amount of additional corporate cash. However, it doesn’t all
have to be funded up front and Orpin expects corporates and trustees will agree
to settle the extra cost over a long period of time. (For the first-ever
longevity swap, involving the Babcock International pension fund, trustees
agreed to funding over 20 years, according to Orpin.)

Hamish Wilson, managing director of pensions advisory HamishWilson, points
out that a much simpler approach to dealing with the longevity issue is for
employers to strike a deal with employees which basically shares the risk. ‘The
employers say: we will set aside a certain amount to deal with reasonable
increased longevity. Beyond that we will cut back on benefits to compensate for
the fact that scheme beneficiaries are living longer.

‘Pension schemes run a huge range of risks and there is no reason why all
these risks should be borne by the scheme sponsor,’ he argues. That is well and
good where the employer has some flexibility as far as the contract with the
scheme is concerned, but most sponsors of final salary schemes will not have
this kind of flexibility available to them. The pension plan will, in all
probability, have been set up before longevity was an issue, so the employer’s
covenant to meet all the liabilities will be set in stone – unless it can be
renegotiated with employees.

Gordon Fletcher, a consultant for Mercer, raises a major difficulty that
longevity swaps will have to overcome before they can become mainstream
instruments. Most such swaps, he points out, will be seen by scheme trustees and
sponsors alike as interim arrangements, a stepping stone on the way to an
eventual full scheme buyout. The problem, then, is working out what value a
buyout provider will put on a swap. At present, that is a completely unknown
factor – and with a longevity swap potentially adding £25m to a £500m scheme’s
liabilities (as in the example mentioned), this uncertainty will not make swaps
any easier to do.

Fletcher also points out that longevity swaps are only possible for a
scheme’s current pensioners. No provider, as yet, is offering swaps on deferred
members. ‘The deferred pensioners present a bigger longevity risk. Someone aged
45 has more than 20 years to go to retirement – 20 years in which anything can
happen to improve mortality. So it is great that there are providers around for
existing scheme pensioners, but we need providers to step forward for the
deferred members longevity risk as well,’ he says.

Today, banks such as JP Morgan and Credit Suisse are keen to do longevity
swaps. They pass the risk on to third-party investors and Fletcher says the
banks Mercer is talking to suggest there is a big appetite for longevity risk
deals among investors. One of the things these investors like about longevity
swaps is that they are not correlated with other asset classes or with other
kinds of risk, such as weather or property risk.

David Johnson, an actuary and consulting director at Trustee GAAPS, an
executive search firm specialising in providing independent trustees and trustee
board chairmen, argues that trustees should be extremely wary of complex new
instruments such as longevity swaps.

‘I am not saying it is not an interesting instrument for trustee boards to
have as an optional tool, but there are some very serious issues about price
right now,’ he says.

‘You are swapping a definite revenue stream for an unknown future payment.
Instead of removing uncertainty, trustees could find that a longevity swap
simply replaces one kind of uncertainty with another.’

Johnson points out that the fact there are interested providers is not, of
itself, an argument for doing deals. ‘Some very adventurous money may well come
sweeping into this space, but the value of longevity swaps to the buyout market
of the future is completely unknown right now,’ he says. ‘Often, doing nothing
is sound advice and there is a very good argument to be made for trustees not to
venture into longevity swaps until the market has matured a good deal more.’

Share
Was this article helpful?

Leave a Reply

Subscribe to get your daily business insights