PS04/16 A weighting game

insurance regulation

sept04-ps1-gif

So far the FSA’s new regulations on how UK insurance companies reserve for corporate bonds have not had a major effect on the sterling corporate bond market. But as Laurence Neville reports, when firms do implement the rules, the repercussions will be dramatic

On July 2, the UK’s financial regulator, the Financial Services Authority (FSA), at last announced finalised rules on life and non-life insurance companies’ solvency requirements. As the new regulations – snappily titled PS04/16 – were launched, bank credit strategists and insurance analysts set about frantically examining the 362-page document to identify what the implications were for insurance funds’ investments in corporate bonds. At the same time, hedge funds, bored by an uneventful first half to the year, scrambled to exploit the distortions that the new rules would inevitably create. And the money managers to whom the rules apply? They barely did anything.

After waiting all year to see how insurance companies would react to the new regulations, the market soon discovered it would have to wait the rest of the year for insurance companies to work out for themselves how they needed to react. “Little will change until the actuaries recalibrate their asset and liability allocation models,” says Corinne Cunningham, senior credit analyst at Royal Bank of Scotland. “The real-money shifts won’t happen until closer towards the end of the year.”

Denis Gould, head of UK fixed income at Axa Investment Managers, which manages funds for the insurance giant among others, concurs. “So far the only conversations we have had with the insurance company is about what data and technical information we will have to provide to them,” he says. “No doubt, there will be further discussions about the likely impact of the changes on portfolios at a future investment meeting.”

But insurers will have to make changes at some point and there is no reason why it should be difficult for them to do so by the end of the year when the new regime comes into effect. “After all,” says David Brickman, credit portfolio strategy at BNP Paribas, “they already know how to match their assets and liabilities. All that has changed is how they allocate capital to cover them.”

Allocation implications

While insurers ruminate on how their allocations will change as a result of PS04/16 (Policy Statement 2004/16), the rest of the sterling credit market has been considering the broader implications. The insurance sector is a major buyer of sterling credit. According to figures compiled in June by UBS, UK life companies invest around €280 billion in fixed income and of that €130 billion is held in with-profits funds.

Figures provided by UBS show that 67.4% of with-profits fixed-income allocations are in UK corporate bonds, indicating the impact any change on insurance companies’ asset allocation could have on the market. To put these figures into perspective, the entire UK corporate bond market is thought to be worth £350–400 billion, compared with the gilts market which amounts to £300–350 billion.

RBS’s Cunningham says that the overriding theme of any change in asset allocation by insurance companies will be the reduction of risk through a switch from equities to bonds and from corporates to supranationals and gilts. Under the new proposals, supras now carry the same risk weighting as gilts, namely immunity from the stress-test (see box below) – a change from CP195 that has been widely welcomed.

The weighting system used to assess credit risk is fairly straightforward, according to George Bory, head of European credit strategy at UBS. “It increases as you move along the yield curve and down the credit curve,” he says. Accordingly, within credit there is likely to be a move away from longer-dated corporate bonds towards shorter-dated issues, as insurers seek to preserve the yield pick-up that credit offers but avoid the high capital charges that holding long-dated paper requires.

According to Cunningham, longer money is likely to switch out of corporate bonds to gilts and supras, allowing insurers to continue to match the duration of their liabilities at a lower capital charge. Matching liabilities is considered by insurers to be secondary only to their solvency in order of importance. Both are considered significantly more important than return.

Pressure on triple-Bs

The capital charge is likely to have a significant impact on credit allocations at certain levels of the credit spectrum. As Bory notes, the capital requirements become onerous at the triple-B level. “The incremental increase in capital required climbs exponentially at this point,” he says. “It is not as black and white as saying that investors simply won’t hold triple-B.” Indeed, Howard Cunningham, portfolio manager at Newton Fund Managers, who describes non-life companies as Newton’s core business, says: “The economics of holding triple-Bs may still make sense.”

But, as UBS’s Bory points out, the tenor of any bonds held will become much more important. “We believe that the cut-off point where it makes sense to hold credit will steepen dramatically, being at around 15 years for triple-As and seven years for triple-Bs. The eight- to 12-year part of the curve looks attractive for single-A,” he says.

Brickman says that the cut-off point – or the point at which holding triple-Bs becomes prohibitively costly – is likely to be slightly further out along the yield curve. “We have observed more steepening in the curve between 20 years and 30 years than between 10 years and 20 years. The kink in the curve is between 15 years and 20 years.” He says that is likely to continue because insurance companies will have to pay more – in the form of reserved capital – for triple-Bs at 20 years than 10 years but they can still reap the benefits of credit at 15 years. But as spreads widen, the kink is likely to move towards the shorter end of the curve.

However, Lee McGinty, head of portfolio and index strategy at JPMorgan, cautions against trying to produce guidelines for how the duration curve will react to the regulations. “Every individual insurance fund will be different and there is no rule of thumb for the whole market,” he says. McGinty believes that some strategists have accorded unnecessary significance to the steepening of the sterling duration curve. The simple fact is, he says, there is a general steepening in all curves including euros and credit default swaps.

Mark Chaplin, a partner in the insurance and financial services division at Watson Wyatt, generally agrees with this. He says that although longer-dated credit will become less attractive to insurers as a result of the regulatory focus on short-term changes in price, it is not possible to say with any certainty how much the curve will steepen and at which points.

The reasons are fourfold, he says. First different companies have different strengths, some will need to minimise the capital held against credit risk while other well-capitalised funds are able to seek out larger returns. Second, each company has a different model for the second pillar of assessment: the self-assessment part. According to Chaplin, for many firms the self-assessment will often produce a greater capital requirement, making the PS04/16 credit stress-test less relevant.

Third, with-profits insurers have different policies on how to pass on profits or losses to their policyholders. And finally, the desire to match the asset and liability durations and fulfill policyholder expectations with respect to investment policy may override concerns about credit risk capital requirements. All this makes each fund so unique that predicting how they will react is unfeasible.

However, as Karl Bergqwist, head of credit research at Gartmore Investment Management, says: “Whichever way you twist it, these new rules are not going to be good for long-dated or lower-rated corporate bonds.”

Greater volatility

PS04/16 is also likely to lead to an increase in volatility, according to Cunningham at Newton Fund Managers. “The need to provide more capital to improve solvency – and therefore sell more bonds – as the market becomes weaker could create a self-perpetuating problem,” he says. “In equities the FSA has relaxed the spread tests to take account of this problem but there’s no suggestion of a similar arrangement for credit.” Insurers are required to hold 20% capital against equity investments when the equity market is rising – as determined by the FSA – however in a falling market, again determined by the FSA, this level falls to 10%. This provides a dampening effect on the market, discouraging both buying in a rising market and selling in a falling market.

The use of the square root of the spread rather than the spread itself to determine how much capital must be set aside does potentially dampen the problem but it does not solve it completely. “There is also the problem that when spreads have widened, bonds have to be stressed harder,” says Cunningham. “Therefore you might see pre-emptive selling if the market appears to be turning weaker.”

As one market participant says: “This takes the nightmare of forced selling from equities and imposes it on an already paranoid credit market.”

Newton’s Cunningham believes that PS04/16 could also result in greater volatility as a result of rating upgrades or downgrades. “The transition from one universe, investment grade, to another, high yield, may be less smooth as a result of these changes,” he says. “People will react when they think a bond is going to be downgraded, sending the spread wider. And then if it is downgraded, the change in the formula used to allocate capital to that holding will be dramatic enough to have a major effect on selling behaviour.”

However, what may counteract increased volatility stemming from PS04/16 are the new benefits of holding bonds to maturity, which could result in reduced trading by insurers. Under the new rules, a different discount rate is applied to liabilities backed by bonds that are held to maturity and those backed by the traded book, according to RBS’s Cunningham. The former permits the total yield to be the basis for the discount rate whereas the traded book is forced to use a lower, risk-free discount rate, which therefore increases the liability.

“Our understanding is that current bond portfolios of many UK life insurers would not be likely to satisfy the IAS ‘held-to-maturity’ tests the FSA plans to follow,” says Damien Régent, insurance analyst at UBS. “Insurers may be faced with a choice of either changing their asset management policies to a buy-and-hold approach, or else reducing their holdings of credit as these will be less attractive.” However, an increase in the amount of long-dated bonds held in buy-and-hold accounts can only reduce liquidity in the market.

BNP Paribas’s Brickman suggests that CDOs might be one instrument to benefit from the hold-to-maturity provision. The new regulation makes no distinction between cash-bond assets and credit derivatives. As such insurance companies can hold synthetically created credit-linked notes or collateralised debt obligations and calculate the risk capital the same way.

Whatever the outcome of insurers’ actions it is likely that the market’s initial kneejerk reaction – to sell triple-Bs and buy triple-As – was over-simplistic. As Brickman notes, there has been some reversal of those trends in recent weeks and over time there will be a clearer differentiation between which types of bonds will be winners and losers. However, it does seem certain that the days of flat sterling curves are over and that further steepening will occur as insurers act on their actuaries’ recommendations.

UK life assurance companies make up the vast majority of holdings in long-dated corporate bonds. As such there is real concern over what will happen if one life assurance firm decides to sell its long-dated corporate bonds. In this instance the question is, who would possibly be in a position to buy? Since the life companies are all governed by the same rules, it is unlikely they would want to step in and pension funds are hardly going to want to buy in what could rapidly become a sellers’ market.

Since the rules were released, liquidity in long-dated bonds has been almost non-existent with pension funds waiting to see what the life companies will do, while the life companies are running their calculations and watching what the other life companies are doing.

How does PS04/16 work?

PS04/16 updates regulations on how much capital UK insurance companies must provision for their liabilities. The regulations are derived from two previous consultation documents: CP190, a consultation paper for non-life insurers, and CP195, for life assurers. Its purpose is to rectify any shortcomings in the European Commission regulations currently in use in the UK.

Currently UK insurance companies are required simply to follow a set mathematical procedure to arrive at a figure to cover their liabilities, regardless of what assets they hold. It is widely acknowledged that this is an inadequate system as, in the case of fixed income, for example, it allows insurers to gain the higher yield offered by corporate bonds over government bonds without provisioning for the greater risk incurred in that investment.

Accordingly, the FSA has created a new system. This requires insurers to hold sufficient assets to satisfy either the current European Commission’s capital requirements or the FSA’s so-called ’realistic’ requirement - whichever is the greater. The original suggestion in CP195 for calculating the realistic requirement obliged insurers to ’shock’ the spread on each bond out to a given limit determined by its rating boundary - 120bp for double-A and 525bp for double-B - and hold additional capital commensurate with the associated fall in price. This would have encouraged investors to hold the bonds in each category with the widest spread - ie, the riskiest - since they would require the least ’shocking’ to reach the shock limit.

The new regulation adopts a new formula to shock spreads: the square root of the spread over the risk-free asset (gilts) is multiplied by a factor that depends on rating - triple-A is three and triple-B is 9.25. The price of the bond that corresponds to the current spread plus the number from the previous calculation is what the insurance company can reasonably consider the value of the asset. For bonds rated below single-B an additional value needs to be calculated: the spread widening were the market to drop by 5%.

Both CP195 and the new PS04/16 make holding longer-dated corporate bonds more expensive since they measure risk based on spread - applying a possible change in spread has a magnified effect on the price of longer-dated bonds. As a research note from BNP Paribas shows, insurance companies will be penalised by moving from 10-year paper to 30-year paper regardless of rating. For example, assuming a cost of capital of 7% for insurance companies, seven 10-year triple-B bonds would generate a net excess spread (after the cost of capital) of 79bp. But bonds over 15 years would generate only a net excess spread of 58bp over gilts.

Only users who have a paid subscription or are part of a corporate subscription are able to print or copy content.

To access these options, along with all other subscription benefits, please contact info@risk.net or view our subscription options here: http://subscriptions.risk.net/subscribe

You are currently unable to copy this content. Please contact info@risk.net to find out more.

You need to sign in to use this feature. If you don’t have a Risk.net account, please register for a trial.

Sign in
You are currently on corporate access.

To use this feature you will need an individual account. If you have one already please sign in.

Sign in.

Alternatively you can request an individual account here