Buying in bulk

A number of firms are preparing to enter the UK bulk annuities market, but the new players face asset/liability and risk management challenges. By Laurence Neville, with additional reporting by Hann Ho

Dramatic change is due to hit the UK's sleepy bulk annuities market in the coming months. Historically dominated by two firms, Legal & General (L&G) and Prudential, a handful of new players are getting ready to enter the sector, attracted by ballooning pension deficits and a perception that UK companies are more willing to offload their defined benefit schemes to annuity providers.

The potential benefits of bulk annuity purchases for deficit-laden UK companies are clear. Under FRS17, which came into force in the UK last year, companies have to value their pension liabilities using a discount rate based on AA-rated corporate bond yields and disclose this on their balance sheets. With yields on long-dated gilts falling sharply earlier this year, pension liabilities among FTSE 100 companies hit £77 billion, according to figures from Watson Wyatt. The consulting firm estimates that deficits had fallen by the beginning of March following a rebound in bond yields, but the figure remains high - around £44 billion.

At the same time, companies are under pressure from the UK Pensions Regulator to eliminate deficits within 10 years. The supervisor can force firms to make additional payments, freeze schemes and dismiss trustees if it believes firms aren't doing enough to resolve the issue, and even has the power to block merger and acquisition transactions. By offloading their pension schemes on bulk annuity purchasers, companies can achieve a relatively quick fix to their deficit problems.

To this end, new players such as UK insurance company Aviva (which owns Norwich Union), Canada Life, Pearl Group and private equity group Duke Street Partners have recently indicated that they are considering entering the bulk annuities market. A number of private initiatives are also under way. Mark Wood, the former chief executive of Prudential, is preparing to launch a company that will enter the market later this year, while Eric Viet and Amedee Levillain, formerly in the pensions group at JP Morgan, are also looking to enter the bulk annuities arena.

"Regulatory pressure and issues such as accounting rule changes mean that firms are devoting more and more time to managing their pension schemes, which is a drain on their resources and an impediment to corporate activity, such as restructuring and dividend distribution," says Levillain.

The former JP Morgan bankers' fund - which is currently unnamed - will target UK and US firms with significant pension asset and liability positions relative to their market capitalisation, and will consider alternative approaches to the traditional insurance bulk buy-out solutions, although Levillain declined to elaborate. The intention is to manage around £10 billion of liabilities by 2008.

"We believe that is a lot of scope for significantly improving the management of pension funds, in terms of both cost, as well as asset/liability management. Our aim is to create a low-cost aggregator of pension obligations thanks to economies of scale, mutualisation of risks and efficient operations," adds Levillain.

In the past, pension trustees have complained that bulk annuity purchases are too costly to consider seriously. Indeed, only a small proportion of the UK's pension schemes have so far opted for buy-outs. However, market participants claim the entry of new players will increase competition and could put pressure on margins.

But the new entrants face significant barriers. The pricing of annuities is extremely sensitive to the level of interest rates and longevity, requiring sophisticated pricing models, robust risk management capabilities and data on mortality.

In the past, underestimating increasing longevity was less of a concern for insurers because high interest rates gave them the flexibility to generate strong enough returns to offset any misjudgements. In the current low-rate environment, insurers have little room for manoeuvre and the consequences of misjudgements are greater.

"Although lower mortality experience affects the eventual benefits payable, a decision to change the longevity assumptions in calculating technical reserves has an immediate impact on both the insurer's profits and its reported net assets," says Sanjeev Shah, director at Fitch Ratings' insurance group in London.

Moreover, the challenges in assessing longevity have become greater. "Longevity is difficult to pin down because of the pace of change in medical technology," says Shah. "If a cure was discovered for cancer, for example, it could change everything." A paper presented to the Institute of Actuaries last September reported that, based on data analysis between 1999 and 2002, a 65-year old male can on average expect to live until he is 86 years and seven months - an increase of three and a half years from the previous survey, based on data between 1990 and 1994.

Insurers analyse scenarios using stochastic models to assess the risks related to changing longevity, says Shah. "Necessarily, the presumptions that underlie those models are essential to their accuracy," he says. The dominant forces in the business - L&G and Prudential - have access to vast quantities of raw data detailing mortality variables, such as type of employment and geography, to enable them to model various scenarios.

New entrants will have to rely on data published by the Continuous Mortality Investigations Bureau (CMI), which collates data from the industry. The problem for insurers relying on this information is that it is considerably less detailed than the internal data an insurance company might collect. "The information provided is often sterilised," says Shah. "Understandably, insurers like to protect their data."

Misjudging mortality can be costly. In 2002, the CMI tables published three scenarios in terms of improving mortality - with low, medium and high expectations. Most in the industry accepted the medium expectation and used it for their risk management. However, L&G - which had access to detailed data as a leading player in annuities - chose the low expectation. In July 2004, it was forced to take a £400 million hit on its profits as a result.

Mark Azzopardi, head of pensions and insurance, global risk solutions, at BNP Paribas in London, says it has become extremely difficult to hedge longevity risk. "There is no longer a reinsurance market in longevity risk and those insurers that have it on their books - L&G and Prudential - don't want any more," he says. "The dilemma for new entrants in the annuities market is essentially how much longevity risk they are willing to take, knowing that it is unhedgeable."

Instead, life insurers take the assumptions issued by the CMI, or use their own data, and back their predicted cashflows with gilts and, increasingly, corporate bonds in order to generate alpha. "That essentially means that they have a massive unhedged exposure," says Paul McGill, a member of the derivatives marketing team for the UK and Ireland at BNP Paribas in London.

There have been attempts to create a longevity derivatives market as a substitute for the hedge once provided by reinsurers. In November 2004, BNP Paribas structured a £550 million 25-year longevity bond for the European Investment Bank, with an annual payout based on an index constructed from mortality data compiled by the UK's Office of National Statistics. The payout was dependent on the proportion of survivors in a selected group - the higher the survival rate, the higher the payment - providing corporate pensions and insurance companies with a hedge on longevity risk. BNP Paribas acted as structurer, manager and book runner, while Bermuda-based Partner Re provided the reinsurance capacity.

However, the launch of the bond was delayed last year amid a lack of investor interest. Some market participants have suggested this is because many insurance firms do not yet have mandates in place to allow them to hedge longevity risk. Others have claimed that the bond's reference pool - made up only of males in England and Wales aged 65 in 2003 - was too narrow, leading to basis risk between a given insurance company's members and the reference group.

"They offer only a partial hedge given the unique characteristics of any annuity pool. The measurement of the cohort used in such transactions is critical and it's difficult to ensure this is accurate," says John Betteridge, head of the portfolio management group at Prudential.

Other attempts at providing a hedge for longevity include the Credit Suisse Longevity Index, which was launched in January and maps out historical and projected life expectancy for the US population. In addition to a composite index, which represents the total population, there are also sub-indexes for males and females, as well as for a variety of ages (50, 55, 60, 65, 70, 75 and 80). While there are currently no plans to extend the index elsewhere, the Credit Suisse index could be rolled out in the UK if it is successful in the US. Meanwhile, Afpen, the French association of pension funds, is also planning to launch a product to hedge longevity risk.

However, some in the industry privately say that longevity hedges are unnecessary and that insurers should remember that they exist to assume risk. "Of course, assets need to be bought to match payout profiles, but alpha generation - through equities or corporate bonds - is there to deal with things like longevity risk," says one market observer. "As long as assets outweigh liabilities, it's not a problem."

But investment risk is a challenge. This can be broken down into interest rate risk and credit risk. Annuities are best matched by investing in long-duration bonds of high credit quality - in other words, long-dated gilts.

However, there's a shortage of supply in the long end of the UK government bond curve, despite the launch of a 50-year gilt by the Debt Management Office last year. Strong demand for long-dated assets by pension funds and insurance companies was partly responsible for a sharp drop in long-dated gilt yields in January (Risk March, pages 28-29). It has prompted insurance companies and pension funds to consider a wider array of investments, say bankers (see box).

As such, strong asset/liability management capabilities are vital, as is a strong credit rating, notes Harish Gohil, a director in Fitch Ratings' insurance group in London. "A strong credit rating will be necessary to get access to bulk annuity business, as trustees of defined-benefit pensions funds strive to reduce counterparty credit risk for the insurers of th eir liabilities," he says.

New entrants will also need significant economies of scale to keep costs down, says Fitch - something that will be difficult to achieve in the early days, making it tricky for them to compete with the established players. Access to capital on an ongoing basis may also present a challenge for some of the new arrivals - particularly those new start-up companies.

"Prudential, like all insurers, has had to deal with mortality strengthening on our assumptions in the past and had to inject capital in order to maintain our business growth," says Betteridge at Prudential. "New entrants can raise lots of money from, for example, private equity. But what is crucial is their access to capital in the future. This is a long-term business and long-term capital is required. It is important that people look beyond what appear to be attractive margins in UK annuities."

Growing credit risk

Historically, UK insurers have not been exposed to substantial credit risk, as they have relied on gilts to provide cashflows to match their payout characteristics. However, insurers have increasingly been forced into the credit market to gain yield and, according to Guy Coughlan, global head of ALM advisory at JP Morgan in London, many have not been systematic in how they have taken on credit risk.

One example is that many annuity books have taken credit risk all along the maturity spectrum, including some very long-dated risk, says Coughlan. "It's just not an optimal strategy," he notes. "They should really balance how far down the ratings scale they go with how far along the maturity curve they go, so that most credit risk is taken systematically at shorter maturities."

Under FSA rules, insurers have to hold more capital against higher risk assets such as corporate bonds, which provides them with some protection, notes Sanjeev Shah, an analyst at Fitch Ratings. But, until recently, there has been little acceptance of the need to manage credit risk in a disciplined way. "It is unclear if any insurers have bought credit protection against their corporate bond portfolios," says Shah. "Most insurers generally don't feel it's worth it."

John Betteridge, head of the portfolio management group at Prudential, says that while the insurer is able to use credit default swaps (CDSs) to manage its credit risk, the regulatory treatment given to CDSs has still to be clarified by the UK regulator, the Financial Services Authority.

"We prefer to match our liabilities through precise stock selection and we have targets for ratings, industry diversification and other criteria," he explains. "Once the risk appetite has been defined for a piece of business, such as annuities, it's just a question of finding a way to maximise yield within that risk appetite."