The independent action group for current and ex Equitable Life policyholders, funded by contributions.

Equitable Members Action Group

Equitable Members Action Group Limited, a company limited by guarantee, number 5471535 registered in the UK

Documents: 01/05/2005 - Colin Slater's Commentary

Colin Slater's Commentary

The basic case against Ernst & Young is that they did not alert the then directors to the full cost of the Guaranteed Annuity Rate (GAR) issue. This covers the audits for the years 1997 – 2000 inclusive. This case is divided into two aspects:

  1. The provisioning issue. The Society claims that E&Y should have demanded a provision (monetary recognition of a liability) against the additional cost of the GAR on the guaranteed/contractual portion of relevant policy values.
  2. The contingent liability issue. The Society claims that E&Y should have demanded a note in the accounts quantifying the potential cost of the Hyman case. This should have covered the possibility that the court decided both that GAR applied to total policy values (including terminal bonus) and that the cost could not be ring-fenced (restricted to GAR policyholders as a group).

The argument is that if the Directors had been properly advised as to the scale of the problem they would have put the business (or parts of it) up for sale in about 1998, at which time the market was buoyant and a good price might have been achieved. Alternatively, it is argued that the directors would have reduced the bonuses and made the company more financially secure. It is stressed that this is primarily an argument that the auditors did not alert the directors.

The alternative line of argument, that the members have a claim against the auditors, is NOT being pursued. Such an approach would require at least a refinement of the leading House of Lords’ case (Caparo Industries Plc v Dickman 1990). This decided (broadly) that investors in a company do not have a claim against a company’s auditors for their investment losses. In view of the costs involved in a trip to the House of Lords, no group of Equitable Life members has taken such a case and the current Board of Directors is disinclined to support a representative group pursuing an argument along these lines.

The big problem with the ‘failure to alert the directors’ approach is that the directors already knew. This means that even if the Court decides E&Y were in the wrong, the Society will have a hard job showing that it was the auditors’ default (rather than the directors’) which caused the loss.

Even if that hurdle can be surmounted the Society faces the problem of quantification. How much bigger price would have been obtained in a hypothetical sale in 1998 than was realised in an actual sale in 2001? Would the unresolved GAR issue have made the business unattractive to a purchaser? Would a purchaser have identified the Society’s inherent financial weakness upon examining the books? If no sale was undertaken, by how much should the directors have reduced the bonuses?

Apparently, even the Society’s own expert (Mr Cryan) has advised that no outright sale or de-mutualisation would have been possible after the spring of 1998.

The Society’s re-re-re-amended case is that an earlier sale would have been on the same sort of basis as the one finally agreed with Halifax (the office buildings, equipment, sales-force & client list only), but for a better price.

Another practical difficulty facing the Society is that it cannot rely upon the evidence of the old directors (who it is also suing) to show that they would have acted differently if E&Y had not been negligent. The Society seems to hope that the old directors under pressure in court may turn upon Ernst & Young as their advisers at the time.

Of the two points at issue, the contingent liability aspect is emerging as the least weighty. No one is suggesting that the auditors should be held responsible for failing to predict the outcome of the Hyman litigation. The question seems to be whether, during the course of that litigation, the worst case scenario (GAR on full value and no ring-fencing) was so ‘remote’ that it did not have to be quantified in a note to the accounts, or merely ‘unlikely’, in which case it did. There seems to have been copious advice (upon which the auditors were entitled to rely) from eminent solicitors and leading counsel that the Society would win the Hyman case.

The ‘provisioning issue’ is much more substantial: Annuities are heavily dependent upon interest rates. Low interest rates give rise to low (expensive) annuities. The GAR policies guaranteed the policyholder an annuity upon retirement based upon a particular level of interest rates, even if at that time market rates turned out to be lower. The Society assumed the risk that if at retirement market interest rates were below that contained in the policy, then the policyholder would exercise his contractual option to require it to make up the difference.

Early policies (1957-1975) contained the guaranteed annuity option (the GAR) based upon interest rates of about 3% - 4%. In later policies (1975-1988) this was raised to about 7% - 8%. Historically interest rates peaked in about 1982 at around 15%. They fell during the 1980s and by 1993 had breached the critical 7% - 8% level. They continued downwards for the rest of the decade.

From about 1993, policyholders with the guaranteed annuity option at the higher rate were ‘in the money’. If they were old enough to enforce the option (60), then it was theoretically beneficial for them to do so.

The Society made no substantial provision (set nothing aside) to cover this risk until forced to do so by the regulator in 1998. The new board of directors now claims that some provision should have been made and the auditors were at fault for failing to demand one.

This matter is confused by the introduction of terminal bonuses in the early 1970s. The Society never intended the GAR option to apply to the terminal bonus element of total policy value at retirement, only to the contractual element. However, it never made this intention clear either in subsequently issued policies or in its literature. The Hyman case was all about GAR upon terminal bonuses. The current ‘provisioning’ case against Ernst & Young is about the cost of the GAR upon only the contractual element.

The current directors say that the additional cost of the GAR upon the contractual value of the policy was a contractual liability, for which provision should have been made. Ernst & Young accepted the old directors’ line that in almost all cases the option would not be exercised. This was because typically by the mid 1990s GAR policies contained a hefty terminal bonus content and most policyholders would have preferred to take this in full, even if this meant them buying an ‘expensive’ annuity at current low rates. The Ernst & Young defence seeks to show that because the ‘GAR cost’ would have been paid as part of the policyholder’s terminal bonus rather than as the separate cost of an additional option, no provision was required. The judge, Mr Justice Langley, described this argument as the ‘battleground’.

It is intriguing to note that both sides seem to accept that the GAR provision (assuming it to be required) would be valued using the deterministic (crude) method. The provision would be based upon the additional cost at the balance sheet date of the Society buying the annuity it had guaranteed, over the contractual value of the policy. For example, if the contractual policy value was £100,000 and the guaranteed annuity was 15% more expensive than that which could be bought on the open market, the extra cost and the provision would be based upon £15,000. The current directors’ assume deductions for the fact that policyholders almost invariably take one quarter of their fund value as tax-free cash and for the inflexible nature of the guaranteed annuity (typically flat-rate, single life and paid annually). However the underlying assumption is a crude comparison of current and guaranteed annuity rates.

This misses out the second essential factor to be taken into account in valuing an option – the time factor. Some policyholders were over retirement age (typically 60) and could take up their option at any time beneficial to them. Interest rates were falling and the guaranteed annuity might be even more advantageous to them (and more costly to the Society) if they put off their retirement. Their option must surely be worth more than that indicated simply by a comparison of current rates.

Conversely, younger policyholders could not actually enforce their option at the balance sheet date. If they were say 50, it would be 10 years before they could do so and interest and annuity rates could go up or down during the intervening period.

Over the years actuaries have evolved methods (called stochastic) to deal with the valuation of options and guarantees. These take the time factor into account. This would have allowed a provision to be made during the 1980s. The amount would then have been small because interest rates were still quite high and the provision would have been against the possibility of them falling further. However the ‘stochastic provision’ would have increased gradually as interest rates fell, thus giving an early warning of trouble ahead. This contrasts with the deterministic method, which would have been set at nil until 1993, when interest rates first fell below the critical level, but would them have increased very sharply as interest rates went down further.

It is easy to see why Ernst & Young are happy to accept the deterministic approach - the values are likely to be lower, but less easy to see why the current directors should do so.

The Old Board of Directors

The case against the Directors is that in the late 1990s they repeatedly renewed the differential terminal bonus scheme without taking legal advice and that in 1999 and 2000, when the problem was known, they failed to reduce bonuses to protect the Society’s financial position or to attempt a sale.

This rather artificial approach is because the original decision to introduce the differential bonus scheme was taken in 1993 and was out of date for a legal claim before Vanni Treves & Co took office in 2001.

The case is specifically not about the old Directors practice of repeatedly voting terminal bonuses that were not covered by assets. We now know that in very general terms this meant that the Society was paying out on average something like £200 million a year over the whole of the 1990s in excess bonuses to leavers. This was one of the main reasons for the company’s failure. Whichever way you do the sums, it was this practice that cost the company something in the order of £2,000 million and left it financially very weak.

Ironically, the GAR issue was considerably less expensive. Very few GAR policyholders were actually paid out in full (between about August 2000 and August 2001). Those that “won” in the House of Lords in the period 1993 – 2000 have been subject to a “rectification scheme” which, on anecdotal evidence, has paid out considerably less than individuals might have expected. The vast bulk of the potential GAR cost was avoided by the compromise scheme. The GARs took much less than they were apparently awarded in Court. The compromise scheme documents suggest that the GAR cost was something like £800 million.

None of the parties is keen to demonstrate ‘over-bonusing’. The Society faces a claim from the ELTA group (and perhaps others) based upon this issue. The ex-directors’ case would be greatly weakened by any admission of paying too high bonuses. Ernst & Young could find it damaging to their case to have tacitly accepted over-bonusing for an extended period, even if they were not specifically required to report upon it.

Both sides seem content to imply that it was sufficient to bring policy values and assets into line every few years, which Lord Penrose reported Chris Headdon describing as ‘smoothing across the peaks’. We know that this was wrong, since it meant that the extra cash being paid out to leavers during the long periods of deficiency was never balanced by the cash ‘saved’ during the much shorter periods of (almost) surplus. It does not appear to be in anyone’s interest to take this point.

However, bonus policy is central to the Society’s case against all defendants and it keeps rearing its ugly head. Charles Thomson had some difficulty explaining why he approved the year 2000 bonus, which was in excess of earnings, but is now claiming that the 1995-1999 bonuses (which were less than earnings) were too high.

He also had difficulty explaining why the Society was telling its policyholders in the infamous ‘Fact Sheet’ (27 October, 2004) that the asset/policy value situation was acceptable in 1999, but suing the old directors and auditors for voting bonuses that were too high in the immediately previous years.

The 1990 bonus decision is being confirmed as a really awful one. It created a big adverse gap between assets and policy values, which significantly advantaged pre-1990 policyholders and according to Bacon & Woodrow’s Peter Nowell, was still showing its effect as late as 2001.

Progress of the Case

It is becoming clear that as far as the directors are concerned, the really damaging decisions, e.g. the 1990 bonus decision and the 1993 DTBP decision without provision, are time barred and the available alternatives are very poor substitutes. As far as Ernst & Young is concerned, they may well have been wrong not to provide for the contractual portion of the GAR cost, but the wrong people are doing the suing, the directors who knew - rather than the members who didn’t.

Colin Slater
(In a personal capacity)