INTRODUCTION

Basis of evidence

  This evidence has been supplied to the House of Commons Treasury Committee further to an invitation issued by the Committee on 17 January 2001.

  The Committee's invitation was non-specific as to the aspects of the Equitable Life Assurance Society's conduct and that of other parties upon which it sought comment.

  Given free rein, we thought is would be helpful to set out our views on the background to Equitable financial philosophy, the matter of the emergence of the guaranteed annuity problem in industry-wide terms and, then, with specific focus on Equitable's treatment of the issue, with regard to its conduct, the impact or potential impact of the liabilities on the financial health of the Society and what was communicated to its policyholders at various stages of the matter. We have also commented briefly on the Equitable's deal with Halifax.

About the author

  The Committee's guide to witnesses asks that written evidence should contain a brief introduction to the persona or organisations submitting it.

  This evidence has been prepared by Mr Ned Cazalet, who is principal of Cazalet Financial Consulting.

  Cazalet Financial Consulting specialises in the provision of strategic and corporate finance advice as well as product development, training, analytical and data services to the life assurance, fund management and retail banking sectors. It has very broad and deep connections, formally interviewing at least once each year practically all heads of function of almost all UK life assurance companies (over 90 groups are covered).

  Cazalet Financial Consulting's principal, Ned Cazalet, is an adviser to HM Treasury on matters relating to the UK life assurance fund and management industries. Cazalet Financial Consulting's client base includes a wide range of life offices, reinsurers, fund managers, banks, investment banks, actuarial and consultancy practices, which breadth of contacts helps enhance our insights into the life assurance and fund management sectors. Those that have used our services in the past 3 years include the following:

ABN Amro, Allied Dunbar, American International Group, AMP, Andersen Consulting, Arkwright Consulting, Arthur Andersen, AXA Sun Life, Bacon & Woodrow, Bain & Co, Bank of Ireland, Bankers Trust, Banque Nationale de Paris, Barclays, Britannic, Brough Skerrett, Cameron McKenna, Canada Life, Centre Re, Century Life, CGU, Chartered Insurance Institute, Charterhouse Securities, Clerical Medical, Clifford Chance, Colonial, Commerzbank, Co-operative Insurance Society, Countrywide Assured, Datastream, Deutsche Bank, DLJ International, Dresdner Kleinwort Benson, Equitable Life, ERC Frankona, Ernst & Young, Fidelity, Financial Services Authority, Fox-Pitt Kelton, Freshfields, Friends Provident, FT Information, GE Capital, General Atlantic Partners, Greenhill & Co, Goldman Sachs, Grant Thornton, Halifax, Hammerson, Herbert Smith, HSBC, Hymans Robertson, Japan Research Institute of Life Insurance, Jardine Reeves Brown, KPMG, Lane Clark & Peacock, Lazards, Legal & General, Lehman Brothers, Liberty International, LIMRA, Liverpool Victoria, Lloyds TSB Group, Lovell White Durrant, Marlborough Stirling, McKinsey, Mercury Asset Management, Merrill Lynch, M&G Group, MGM Assurance, Midland, JP Morgan, Morgan Stanley, Munich Re, NFU Mutual, Nippon Life, Norwich Union, Old Mutual, Paymaster, Pearl, PricewaterhouseCoopers, Prudential, Rebus Group, J Rothschild Assurance, NM Rothschild, Royal Bank of Scotland, Royal Liver, Royal London, Royal & Sun Alliance, Schroders, Scottish Amicable, Scottish Equitable, Scottish Life, Scottish Mutual, Scottish Widows, Skipton Building Society, Slaughter & May, Societe Generale, Sun Life of Canada, Swiss Life, Swiss Re, Tillinghast Towers Perrin, UBS Warburg, Unisys, Virgin, Warburg Pincus, Wasserstein Perella, Watson Wyatt, JB Were, Wesleyan, Winterthur, Zurich Financial Services

  Mr Cazalet is widely consulted by the media as an authority on the UK life assurance, pensions and fund management sectors and, in recent months has been cited, inter alia, by the following:

the BBC (various radio and TV programmes), Bloomberg, Channel 4 TV news, the Daily Mail, the Daily Telegraph, the Economist, the Express, Financial Adviser, Financial News, the Financial Times, the Guardian, the Herald, the Independent, the Independent on Sunday, Insurance Post, Investors Chronicle, Investment Week, the London Evening Standard, the Mail on Sunday, Money Marketing, Money Management, the Observer, Pensions Age, Pensions Management, Pensions Week, Pensions World, the Press Association, Professional Pensions, Reuters, the Times, Scotland on Sunday, the Scotsman, Sky TV news, Sunday Business, the Sunday Herald, the Sunday Telegraph and the Sunday Times.

  Ned Cazalet for many years has been a regular contributor to FT Business (in particular, its titles Money Management and Pensions Management magazines) on matters relating to the life assurance, pensions and fund management sectors, and has been a guest contributor to the Daily Telegraph and Financial News.

  In recent months, Ned Cazalet has been a speaker at the Acquisitions Monthly Conference on Valuing and Pricing Mergers & Acquisitions in Financial Institutions, and has lectured at the Faculty and Institute of Actuaries' 1999 and 2000 annual Life Conventions (respectively on the techniques for analysing long term insurance businesses and on capital management within life offices). He has been a speaker on the subject of the appraisal of life office financial strength at Society of Financial Advisers' 1999 annual conference and on the same topic at a conference organised by IFA UK in 2000. He chaired the "Life Assurance 2000" conference organised by the Institute of Economic Affairs.

  Mr Cazalet also recently has appeared as a guest speaker/external adviser at non-public conferences, seminars and management/client briefings given by organisations such as CGU, CIS, Merrill Lynch Investment Management, Pearl Assurance, Prudential, J Rothschild Assurance, Royal Bank of Scotland Capital Markets, Scottish Mutual, Swiss Re, UBS Warburg and Unisys.

COMMENTARY

Preliminary comment on Equitable's communications

  We think it worth remarking at the outset that, in general, we have tended to regard Equitable Life Assurance Society as consistently having been very open with its policyholders (far more so than any other life office we can think of) in terms of communicating its strategy and its operating and financial position. In particular, in our opinion Equitable's annual reports and accounts have been models of informative clarity and easy-to-understand explanation. All of which makes the events recounted below concerning Equitable's treatment of the vexed issue of guaranteed annuity options ("GAOs") the more regrettable.

Equitable's philosophy on solvency

  At the root of Equitable's recent difficulties has been its long-standing and oft-stated philosophy of not maintaining substantial excess capital (ie funds over and above those necessary to meet its liabilities and to satisfy the solvency requirements imposed by law).

  Indeed, Equitable, implicitly at least, had tended to be critical of other life offices that maintained proportionately higher levels of excess capital. Equitable's stated position was that, unlike many of its peers, it always had sought to pay out the highest possible amounts when policies became claims (whether at maturity, death or premature surrender) and, for that reason, had not accrued the higher levels of excess capital held by most of its rivals.

  The fact that Equitable made a virtue of paying out relatively very high claims values was a strong selling point for the Society and was widely endorsed by the financial press as being a "good thing", particularly during the 1980s and early 1990s when its surrender values tended to be significantly ahead of those offered by other life offices and when its maturity values were also at or very close to the top of the league tables.

  In contrast, other life offices could be said to have "benefited" from not paying out such high claims values as Equitable, such "benefit" to some extent being represented by the ability to build up and maintain a plumper cushion of excess capital. (Think of the recent controversy surrounding the treatment of so called "orphan estate", which is another word for excess capital, accrued by AXA Equity & Law Life Assurance Society.)

  Source: Cazalet Financial Consulting.

  Assessing life office financial strength is a highly complex issue. To illustrate Equitable's position relative to a cross-selection of with profits life offices, the above table (which contains data and commentary greatly distilled from our proprietary life office rating system) shows a measure of solvency we call "WPXA" (being the amount of spare capital held by a life office relative to its in-force with profits business) together with other key pointers as to how conservatively we think the individual life office has chosen to calculate its liabilities (there is a great deal of actuarial discretion in this area), and the characteristics of its cashflow. The final score, which is out of "10", is our overall with profits financial strength rating. These ratings were assessed in November 2000 (before Equitable's announcement that it had failed to find a buyer for the Society as a whole), and it can be seen that Equitable's raw WPXA score is far lower than the generality of its peers and that, further, when account is taken of the approach to reserving, a very low rating of "2" was called for.

  To be clear, Equitable historically did not shy away from revealing its relatively thin solvency cover but, instead, promoted it as the desirable outcome of what it viewed as its virtuous approach to returning to policyholders as high as possible a proportion of the returns earned on their premiums. Given the wide reporting of this in the media, we think it likely that many persons who became policyholders during the past two decades (our memory does not extend beyond that time) would have been aware of Equitable's philosophy on not accruing and maintaining substantial excess solvency.

  The point about Equitable's relatively thin solvency cover is that it always was going to be more exposed than most other life offices to financial adversity, whether in the form of bad news on the asset side of the equation in the event of poor investment market conditions, or on the liability side in the event of unplanned liability inflation brought on by issues such as regulatory intervention (such as would be the case with pensions mis-selling or changes to liability valuation regulations) or actuarially-based factors such as the emergence of previously submerged guaranteed annuity options.

Equitable's cost base and systems

  Equitable was exceptional not just because of its solvency philosophy, but also because of its approach to cost controls and operating efficiency. On the one hand, its achievements in this area were remarkable and to be applauded but, by the same token, the development of Equitable into a very low cost and hyper-efficient life office (in terms of the acquisition of new business and the maintenance of in-force policies) meant that it was always going to be peculiarly unattractive to a prospective purchaser (even leaving aside problems with guaranteed annuities and the like) for the simple reason that it had become what might be described as a life assurance administrator so efficient and lean that there would be only very slim profits to be generated for a shareholder were the Society ever to demutualise.

  The driving down of acquisition and administrative costs achieved by Equitable was a remarkable feat of management (which has been somewhat forgotten amid all the rancour currently prevailing) and positioned the Society miles ahead of any other UK life office in efficiency terms.

  Equitable's operating superiority was driven on the new business front by a sales force that was astonishingly productive. We calculated that, in recent years, a typical Equitable sales person was generating something like up to 10 times the volume of new business achieved by run of the mill direct (non-bancassurance) sales plodder elsewhere in the industry.

  As for Equitable's wafer thin administration costs, these were enabled by the substantial volumes of new business generated year-in and year-out by the highly productive sales force, which strongly positive cashflows led to significant and increasing economies of scale.

  What is more, Equitable's focus on cost control and drive for operating superiority caused the Society to be an early mover in seeking gains from advanced technology applications. This focus paid off big time, leading Equitable to be regarded as having the best life and pensions administration systems in the UK, if not in the world. Equitable's prowess in this area has been such that it developed a thriving consultancy activity in helping develop or supply administration systems to other financial services companies in the UK (such as Marks & Spencer Financial Services, Merrill Lynch Investment Management and the Royal Liver friendly society) and overseas (the long term insurance business of the Harare-based insurer, Southampton Life, is administered remotely on Equitable Life's systems in Aylesbury).

  The table below shows the progression in recent years of Equitable's new business (split between new regular and single premiums), and shows total premium income (which additionally takes account of regular contributions continuing to be paid on the existing book of in-force business) and income and gains from investments, together with the quantum of the Society's total assets.

  Source: Equitable Life; Cazalet Financial Consulting.

  The bottom line in the table above illustrates how Equitable drove down its cost ratios during the course of the 1990s. We can put these numbers into the context of the UK life assurance industry as a whole by firstly breaking down the expense ratio total for 1998 into its underlying components that respectively represent the expense ratios relating to the acquisition of new business and the maintenance of in-force business and, secondly, by comparing those ratios to the rest of the UK life industry (we have used Equitable's 1998 data for this as, at the time of preparing this evidence, we did not have to hand comparable industry-wide data for the 1999 year end).

  Source: Equitable Life; Cazalet Financial Consulting.

  The breakdown of Equitable's expense experience set out in the table immediately above shows that Equitable was away ahead of its nearest competitors in terms of the financial efficiency of its new business and administration functions, being almost five times more efficient than the UK life industry as a whole when it came to writing new business, and nearly seven times more efficient in terms of the cost of administering in-force business.

  To reiterate, Equitable's systems and cost controls have been brilliant. The systems are very much worth acquiring by a bidder. The Equitable policy book administered on those systems, however, is "profit-lite" and unlikely (even under more favourable circumstances than currently persist) to provide a filling repast for a profit-oriented shareholder (rather like dining at Old Mother Hubbard's bistro).

  We are aware that, in recent times, officers of Equitable Life have been referred to by the media as "arrogant" and wonder whether it was the case that the Society's stupendous successes in terms of its operating efficiency caused its leaders to think that they could repeat the walking on water trick when it came to defying gravity in terms of financial strength.

The GAO problem

  The origins of the life industry's guaranteed annuity problem might best be summarised by the words of the song: "I know an old lady who swallowed a fly. I don't know why. Perhaps she'll die."

  Much has been said and written over the years on the subject of the madness of crowds and popular delusion. To the South Sea Bubble, Dutch tulip mania and the recent dotty dotcom boom and bust must surely be added the widespread granting by life offices of guaranteed annuity options on pensions contracts written in the high inflation 1970s and 1980s.

  During times when inflation was at 20 per cent and yields on government securities were seemingly perpetually to be in double digits, the giving of a promise that, come what may, the interest rate applied to your annuity once you cash in your retirement savings would be at least 7, 8 or 9 per cent (terms and conditions applying to guaranteed annuity options vary considerably from insurer to insurer) probably seemed a safe bet on the grounds that, viewed from the high 20 per cent+ uplands, it would never happen.

  Well, fifty years ago, you might have said (and truly believed) that a man would never walk on the moon, a woman would never be prime minister (or your local vicar) or that you could telephone home from on board an aeroplane flying direct from London to Tokyo.

  The "we thought it would never happen" lament now looks somewhat pathetic, especially given that the devising and monitoring of these guaranteed annuity options was and is the responsibility of that army of mathematical Mystic Megs known as the actuarial profession. From where we stand today, it is hard to conceive that, during the period when guaranteed annuity options were being handed out with indiscriminate abandon like flyers for an Ibiza night club, anyone actually worked out the precise nature of the guarantees/options they were granting. For, if they had reflected on the nature of the promises given, they would have realised that it was impossible to hedge out their components (comprising interest rate and mortality risk) and, further, that exposure to such risks was highly volatile and practically limitless. (Pensions annuity rates are calculated by life companies on the basis of the amount of the capital sum deployed to purchase the annuity and the rate of return available from government securities spread over the likely lifespan of the person taking out the annuity.)

  Let us deal with the interest rate problem first. It can be seen from the chart below that, as we have moved into a lower inflation environment, long term interest rates have fallen. The fall in long term gilt yields was so great during the 1990s as to cause guaranteed annuity options written on interest rate assumptions based on single figure gilt yields to progress from being a no-chance non-event to a horrific reality.

  The farther gilt yields fall the greater the reserving problem for life offices as policyholders' guaranteed annuity options become more and more "in the money" as the returns that such policyholders could achieve by buying their annuity income in the open market (and life offices could finance from purchases of gilts) is outstripped by the guaranteed rate incorporated in their pension contracts (which cannot be financed sufficiently from the returns attainable from gilts).

  What is more, once the guaranteed annuity iceberg began to emerge in the 1990s there was no way of hedging the problem. Leaving aside the deficiency between the gilt yields required to finance guaranteed annuities and the actual yields available in the markets, many of the pension contracts containing guaranteed annuity options were (and are) still a long way from being converted into annuities, following which, such annuities could be expected to be in-payment for perhaps 20 years or more on average.

  In other words, a 40 year old man who took out a policy incorporating guaranteed annuity terms in 1985 (when he was 25) might retire at 65 and then live and draw annuity income for a further 20 years until he dies at, say, 85. Even if, as a life office, you are prepared to take the "hit" on the gap (as it currently stands) between the gilt yield required to finance the guarantee and the actual, much lower yield currently available in the market, the problem is that our 40 year old pension saver is likely to be drawing pension annuity income up to 45 years from now, and there simply are not any 45 year gilts for you to buy!

  The second element of the GAO problem is mortality. The guaranteed annuity options granted by life offices mostly also have a mortality component (rather than just having an interest rate promise). The problem here is that longevity has increased dramatically in recent years.

  The table immediately below illustrates how, over the course of the past few decades, mortality has improved to the extent that between the commencement of a deferred annuity plan and its vesting, the holder might have seen his life expectancy increase by two or three years.

  Source: ONS.

  Such increases in life expectancy might seem innocuous but, in reality, can serve to reduce annuity payments sharply. As the next table illustrates, the improvement in mortality for a male at age 60, which took place in the 25 year span under examination is such to boost remaining life expectancy by one-fifth.

  Source: ONS; Cazalet Financial.

  The net impact on pension annuity rates is illustrated by the chart immediately below, from which it can be seen that a pension annuity compulsorily purchased by a 65 year old man in spring 1999 bought an income for life of less than 60 per cent of the income for life that could have been secured by a man of the same age in the autumn of 1990.

  Given the above, it is a no-brainer to conclude that persons contributing to personal pension contracts incorporating guaranteed pension annuity rate options would be increasingly likely to exercise such options on retirement in view of the declining value to be had from the alternative of buying an annuity in the open market.

Impact on the life industry

  The significance of all this from an actuarial reporting perspective is that, in the era of high inflation, high bond yields, lower longevity and higher open market annuity rates, it was extremely unlikely that anyone would elect to take what was then a relatively very low guaranteed rate of annuity when a much higher level of retirement income could be secured in the open market. Once this operating background started to shift adversely (from a life office standpoint), it became clear that more and more people would elect to exercise their guarantee options. In the days when the guaranteed annuity options were very unlikely to be exercised, it perhaps was not unreasonable for life offices to disregard such possibility when preparing their accounts. Once such guarantees became valuable from a policyholder viewpoint ("in the money") and were likely to be exercised, it became imperative that life offices should increase their reserves to provide for the uptake of GAOs.

  As a group, life offices began to make extra reserves for GAOs during the 1990s, tending to increase reserving as gilt yields came down and longevity improved. Such reserving was ad hoc and subject to a great deal of actuarial discretion. Between 1997 and end 1998, long gilt yields fell very sharply indeed, causing there to be a very sharp increase in the extent to which policyholders were "in the money" (and life offices were in trouble) with regard to GAOs. This market movement caused the Government Actuary's Department ("GAD") to survey life offices to determine their exposures to GAOs, which in turn led to the issue of guidance as to the basis of provisioning for GAOs. The Committee doubtless will have access to the detail of the activity of GAD in this area, but it can be summarised as meaning that, when calculating their GAO reserves, life offices were required to move to a position based on the assumption that the substantial majority (80 per cent+) of policyholders with pensions savings contracts incorporating GAOs would be likely to exercise such options.

Equitable's reaction

  When, in the early 1990s, the GAO problem began to lap at Equitable's shores, the Society (unlike the generality of its competitors) seemingly decided not to make reserves, but chose to deal with the problem in an entirely different way, with the intention of seeing to it that practically none of its customers would opt to take the annuity on the terms guaranteed in their contracts, but, instead, decide to purchase a pension annuity at the prices prevailing (whether from the Society itself or from another provider on the open market).

  Equitable sought to influence its customers to refrain from exercising the (reserving-heavy) guaranteed form of annuity by the simple expedient of manipulating its bonus rates so that a lower rate of final bonus would emerge on contracts where the policyholder had elected to take the guarantee, meaning that a smaller pension pot would be available to purchase the (higher) guaranteed form of annuity, and a higher pension pot would be available to purchase the (lower) open-market form of annuity. The bias usually would be sufficient to ensure that, compared side by side, the lower fund/higher guaranteed rate option would be seen to generate less in pension income than the higher fund/lower open-market alternative.

  Equitable's conduct would appear to have been the result of its philosophy of not holding substantial excess capital. Quite simply, the Society knew it could not afford to honour its guaranteed annuity promises and, accordingly, sought to construct its bonus rate regime so that the guarantees would be worthless (in terms of their attraction relative to the alternative of buying an annuity on open-market terms).

By GAD!

  Equitable's evasion of its annuity promises was threatened by GAD's requirement that life offices reserve on the basis that the substantial majority of their in-force pension savings contracts would result in the exercise of annuities on the guaranteed basis.

  At first blush, Equitable's payout strategy was scuppered as, notwithstanding its seeming ability to manipulate payouts so as to ensure that its policyholders would find the guaranteed form of annuity (conveniently) unattractive, GAD's insistence that Equitable should reserve on the basis that its policyholders would in fact tend to elect to exercise such guarantees meant that it would have to make extra reserves for the thick end of £1.5 billion. The problem was that Equitable's thin solvency meant that this was £1.5 billion that it really did not have spare. Crumbs!

With one bound he was free

  In the manner of a Bruce Willis or younger Harrison Ford, however, Equitable contrived to break free from GAD's tight strictures in a single, cunningly calculated leap that took it all the way across the sea to Ireland.

  The GAD-inspired accounting treatment required to be applied to Equitable's book of GAO-embedded pensions was the problem. The solution was for the Society to divest itself of this book of business (or at least the potential claims arising from it) through the magic of reinsurance. In effect, Equitable managed to park its book of GAO liabilities with a Dublin-based reinsurer (out of the reach of the UK's GAD) so that (bingo, jingo!) there simply was nothing left to reserve for in the UK.

  The reinsurer, a subsidiary of GE Capital, was no fool, however. The deal was simply a case of sophisticated regulatory arbitrage. It was Equitable's apparent belief that GAD's reserving requirement was unreasonable given that none (or, at best, hardly any) of its pension customers would want to exercise their GAOs and that, accordingly, GAD's requirement to reserve on the basis of an 80 per cent+ take up was unrealistic.

  The basis of the reinsurance was that, as pension contracts vested and (non-guaranteed basis) annuities were purchased by policyholders, such claims would be dealt with by the reinsurer, which would make recovery from Equitable on a drip-feed basis (effectively acting as a paying agent).

  Again, Equitable's motivation in resorting to pension provisioning prestidigitation would appear to have been its fundamental lack of spare capital with which to absorb the cost of making the reserves resulting from GAD's guidance.

  The required reserving for the entire UK life industry would appear to be something in the order of £10 billion+. With a couple of minor exceptions, all other life offices chose to take their GAO hits squarely on the chin, using their excess capital to meet the cost of their emergent liabilities.

HMT letter re GAOs

  In carrying out its maturity value/annuity strategy, Equitable appeared to have the explicit backing of the Insurance Directorate of HM Treasury.

  Members of the Committee doubtless will have access to the letter written, on 18 December 1998, to all life offices by Mr Martin Roberts of HM Treasury's Insurance Directorate. The key phrases of Mr Roberts' letter might be said to be the following:

"As you will know the Government Actuary's Department undertook a survey of life offices' exposure to . . . GAOs . . . earlier this year [which] indicated that the exposure to GAOs was relatively widespread . . . and had the potential to have a significant financial impact on a number of companies . . . one issue to be addressed by all companies was how the concept of policyholders' reasonable expectations (PRE) should be interpreted in the context of GAOs. The purpose of this letter is to provide some guidance to companies on the Treasury's interpretation of PRE in these circumstances.

As a starting point, we take the view that policyholders entitled to some form of annuity guarantee or option on guaranteed annuity terms could reasonably be expected to pay some premium, or charge, towards the cost of their option or guarantee.

Charging for the cost of providing a guarantee or annuity option

. . . In the case of participating [ie with profits] policies, any charge could be deemed to be met out of each premium received (or the investment return to be credited by way of bonus), and hence would impact on the assessment of bonuses, including in particular any terminal bonus that would normally be payable to the policyholders. Generally we consider that it would be appropriate for the level of the charge deemed to be payable by participating policyholders for their guarantee (or annuity option) to reflect the perceived value of that guarantee (or option) over the duration of the contract. This could be achieved in some cases through some reduction in the terminal bonus that would be payable if there were no such guarantee (or option) attached to the policy . . .

. . . it would appear possible, depending on the particular circumstances relating to the contract, that any terminal bonus added at maturity could be somewhat lower than for contracts without such options or guarantees, and that this terminal bonus could in some cases be applied at current annuity rates . . .

This is the Treasury's considered view, and is without prejudice to any decision of the courts which may effect it . . ."

Not so fast, buster

  Equitable's Nemesis came in the form of an independent financial adviser, a Mr Bayliss who, being a specialist in the retirement annuity sector, was instrumental in spreading the word that Equitable, in his opinion, had been treating its pension policyholders unfairly by juggling pension maturity values and annuity rates so as to deprive policyholders of the true value of their guaranteed annuity options.

  The agitation of Mr Bayliss and others led to the litigation that eventually saw the Society defeated in the House of Lords.

  When faced with litigation, Equitable seemed to adopt a sophisticated tone in addressing its policyholders, perhaps finding it inconceivable that the Court would ever judge other than that the Society was in the right.

  With hindsight, it might have been better if the Society had been franker with its members by being clear about why it has resorted to fast footwork on annuity rates. Let us suppose the Society had addressed its members as follows:

"Look now, this is a mutual Society and, as such, we have very limited recourse to outside capital. These guaranteed annuities are a blessed nuisance; we wrote them some time ago and now sincerely wish that we had not. At that time, all of our competitors were doing the same thing and we got caught up in the market vogue, little believing that we ever would be called upon to deliver.

As you all know, it has been a cornerstone of our philosophy not to retain relatively large amounts of surplus capital. Over the years, we have chosen, with your seeming approval, to use our resources to pay relatively high claims values on surrender and on maturity, often substantially more than our competitors. This belief in making as generous as possible a return of profits to our membership means that we do not have the resources to take the GAO hit on the chin in the manner of other life offices that have been less generous in their payouts in the past. Accordingly, something has to give. We are Equitable by name and by nature and, in arranging our maturity values and annuity rates as we now do, we simply are trying to be as fair as possible to all parties within the constraints of our modest solvency position. We can understand why some of you may feel hard done by, but assure you that we simply are endeavouring to make the best and fairest use of the Society's resources in the circumstances in which we now find ourselves. To those of you considering legal action against the Society, we would ask that you consider where the resources would come from to finance any success you may achieve in pressing your claim and urge you to refrain from such proposed action in the interest of the common good."

  Upon hearing such address, we estimate that there might well have continued to be discontent among Equitable's pension policyholders, but that this would have been tempered by a realistic understanding of the Society's true financial position and the motivation of its managers and directors in embarking upon the complained of GAO avoidance strategy.

  Instead, however, the Society's seemingly sophisticated approach and its dismissal of the GAO issue as being relatively trivial and having an impact of no more than £50 million (or, take your pick, £200 million) probably served to irritate Equitable's policyholders rather than soothe them and, when the House of Lords balloon went up, caused them to be mightily concerned to learn that a "£50 million" bagatelle was, in fact, a £1.5 billion terminal disaster.

  It is not as if Equitable's true position ahead of the commencement of litigation was not known or discernible. The following is taken from our work, LIFE 1999:

EXTRACTS FROM "LIFE 1999" (Published February 1999)

The guaranteed annuity option issue is now a matter for the Court. However, it is to be wondered whether some of those pressing to have their cake and eat it have considered where the money would come from to pay a "full" rate of TB [terminal bonus] and meet the guaranteed annuity cost, as the potential sum at issue would appear to be of such magnitude as to require terminal bonuses to be reduced in any event (there's no free lunch).

Perhaps the annuity agitators are hoping to drive Equitable to demutualise. The problem would seem to be that acquiring Equitable might be like catching a snowflake in the palm of your hand. Leaving aside the guaranteed annuity option furore, what would be the industrial logic? A bidder could hardly expect to squeeze out significant savings from cost inefficiencies (unless the acquisition was effectively a reverse takeover). Further, it is not as if Equitable is or would be capable of generating sufficient profits such as to reward a proprietor keen to show a healthy return on capital employed.

  When Equitable lost its case in the House of Lords, we were struck by the widespread policyholder astonishment at the revelation that the GAO problem should have emerged to be £1.5 billion in size. This outcome had been familiar to us for some time—we had discussed the matter of the size of the potential liability and Equitable's resorting to reinsurance with senior officials of the Society and reported on this in a document entitled "The Wolf's at the Door", which was published in October 1999 and, again, as set out below, in "LIFE 2000".

EXTRACTS FROM "LIFE 2000" (Published March 2000)

Equitable's approach to guaranteed annuity options has been that it seeks to pay the same actual money benefits whether the cash or GAO basis is selected by the policyholder, and it has stated that the cost to it of such options is unlikely to exceed £50 million. Paragraph 16 of its 1998 valuation statement contains the following advice: "If the contract guarantees minimum rates for annuity purchase the aggregate final bonus otherwise applicable is reduced when benefits are taken by the amount, if any, necessary such that the annuity secured by applying the appropriate guaranteed annuity rate after such reduction, is equal to the annuity which would be secured by applying the Society's annuity rate for an equivalent annuity in force at the time benefits are taken to the cash fund value of the benefits before that reduction, subject to a minimum value for the final bonus after such reduction of zero."

Equitable has advised that its experience is that GAOs have been exercised in about no more than 2 per cent of cases and that its experience "has been consistent with an outcome significantly lower than even the £50m figure". In the light of the foregoing experience, Equitable has determined to adopt what it regards as a belt and braces approach to the matter, and that it would be prudent to structure its reserves on the basis that 25 per cent of benefits are taken in guaranteed form and 75 per cent in cash and, accordingly, made a provision of £200m.

Notwithstanding Equitable's own experience and view as to prudential reserving, the Government Actuary promulgated reserving guidance for guaranteed annuity options to be followed by all life offices.

The Government Actuary indicated that he expected life offices to reserve on the basis that substantially all of the benefits were taken in guaranteed annuity form, irrespective of actual experience, with only relatively modest allowance for cash commutation, the exercise of open market options or the selection of different post-vesting contracts such as unit-linked and with-profits annuities or drawdown.

As a result of the Government Actuary's guidance, Equitable found it necessary to establish reserves on the assumption that c 80 per cent of the benefits were taken in guaranteed form and 20 per cent in cash. As was noted above, given Equitable's determination that the money benefits should be the same whether the guaranteed or cash option was selected, reserving in accordance with the wishes of the Government Actuary makes no pound-in-your-pocket difference to the policyholder. There is, however, a marked impact on the appearance of solvency.

The effective requirement of the Government Actuary for Equitable to assume that 80 per cent of benefits will be taken in guaranteed form has led to a significant shift across Equitable's balance sheet divide, with an additional £1.5bn reserve being shown in the 1998 returns (with consequent matching diminution on the asset side).

In order to mitigate the financial impact of having to assume that 80 per cent of benefits are taken as a result of the exercise of GAOs and to achieve a financial presentation that Equitable claims is somewhat closer to the underlying reality, the Society has entered into a reinsurance arrangement (with a Dublin-based subsidiary of ERC Frankona).

The purpose of the reinsurance effectively is to enable Equitable to assume that the composition of the relevant pension benefits taken is 25 per cent in guaranteed form and 75 per cent in cash.

The financial effect of the reinsurance is such as to enable Equitable to release part of the additional £1.5bn reserves for final bonus back to the free assets (ie back from the liability to the asset side of the balance sheet), thereby arriving at what Equitable regards as being a presentation closer to the true underlying position.

The amount of the reinsurance offset is £800m and this has been secured for a premium of £150,000. In the opinion of Equitable, such a relatively small premium indicates that there is little risk passing to the reinsurer and, therefore, that the risk of the actual take up of guaranteed annuity options beyond 25 per cent is "very remote".

Should a claim be made by Equitable pursuant to the reinsurance treaty, the reinsurer would make recovery over time from the future surplus arising within Equitable's long-term fund.

Effectively, the reinsurance enables Equitable to "park" its reserving problem (perceived by it as being unrealistic) with another institution and, if things turn out worse than expected (ie the take up of GAOs rises beyond 25 per cent), to drip feed the hit onto its balance sheet.

It is also Equitable's contention that, even after allowing for the effects of the reinsurance, there is still "effective substantial reserving" for terminal bonus. On that basis, the Society has felt it appropriate to make greater use of the future profits implicit item to give a "more realistic" position since, as future profits arise, they will fund terminal bonuses.

The contingent capital gains tax liability was estimated as being not greater than £98m as at end 1998 and a reserve of £100m was held. A mismatching provision of £600m was also held (equivalent to £1,236m if the Society had accounted on a net premium basis).

Equitable continues to be talked of as a bid candidate (such chatter inspired by the controversy surrounding the Society's treatment of guaranteed annuity options, which has seen it be transformed in the eyes of the media from the Muhammad Ali to the Mike Tyson of the industry). As we observed last year, the problem still would seem to be that acquiring Equitable might be like catching a snowflake in the palm of your hand. Leaving aside the guaranteed annuity option furore, what would be the industrial logic? A bidder could hardly expect to squeeze out significant savings from cost inefficiencies (unless the acquisition was effectively a reverse takeover). Further, it is tricky to see how Equitable's current business model would be capable of generating sufficient profits so as to reward a proprietor keen to show a healthy return on capital employed. All of this would make for a relatively slim offer from any bidder.

  In reviewing what we wrote in the second extract above (from "LIFE 2000") we are struck by the disparity between what we knew about Equitable's financial condition vis-a-vis the GAOs (ie that there was a potential £1.5bn liability that was too hot to handle and which needed to be placed in the safe, asbestos-gloved hands of an offshore reinsurer) and the gloss put on the situation in the Society's widely-circulated annual report and accounts and the statements made at its last annual general meeting (in which the £50m/£200m scenario was played up). As we remarked above, it might have been much better for all concerned for some brutal frankness to have been deployed in spelling out to its members the reality of Equitable's finances and the true, worst-case impact of a (then) potential defeat in the Courts on the matter of the GAO litigation. It seems to us that it was only after Equitable was defeated in the House of Lords that it started using the £1.5bn figure in its communications with its members who, not unsurprisingly, were confused and angry at learning of such exposure when, only a few months previously, in written and spoken communications, Equitable had been referring to a £50m/200m problem.

How big was the "hole"?

  Press coverage on Equitable, post the House of Lords defeat, relating to Equitable's financial condition and the Society's own statements on the matter have consistently referred to the £1.5bn number as if that were the size of the problem that needed to be overcome by a bidder for the Society as a whole (for which, eventually, there was none).

  In fact, the level of financial restoration that would have been required by an outright bidder (which would have taken on the long term liabilities of the Society by means of a demutualisation and transfer of such business) was probably more like £4bn.

  The reason for such a higher number is that, if we disregard the GAO issue, Equitable was running on thin solvency anyway. This already thin solvency in our estimation would have been significantly eroded by the poor investment market conditions of 2000 (in which UK equities produced negative returns and gilt investments generated only modest income) on the one hand and, on the other, liability escalation (excluding the impact of the addition reserving required for the GAOs) arising from reserving for bonus additions and other, technical issues including the impact of new, more stringent actuarial valuation regulations in May 2000.

  The asset and liability scenarios would have acted like a pincer (for all with profits life offices, not just Equitable) in the 2000 calendar year, causing significant erosion of solvency across the UK life industry. Equitable's skimpy excess capital meant that (excluding GAOs) it was going to be more adversely affected by these circumstances than pretty well any other insurer.

  To return to the specific matter of the GAO liabilities, it is crucial to note that, although the figure of "£1.5bn" has been bandied about, the true total, being based on ever-changing mortality and interest rate factors, is highly volatile and, if longevity were to continue to improve and/or if long term interest rates were to decline further, the actual amount would be set to increase—possibly substantially. For that reason, any acquirer of the long term business of Equitable (which there was none) would have been well advised to take a very conservative view of the potential ultimate value of the Society's GAO liabilities—this could have been done by reserving on the basis that long term interest rates would turn out over time to be lower than they are at present and by ensuring that the mortality bases used incorporated an allowance for significant improvement in annuitant life expectancy. On the basis of such justifiably cautious and prudent approach to reserving, it would not be surprising if a potential bidder looked at Equitable and factored in a GAO liability figure of £2bn or perhaps even £2.5bn, making the Society an even more eye-watering nettle to grasp. (Incidentally, if in such circumstances a buyer did make reserves of, say, £2.5bn and the actual outcome was £1.5bn, it would be expected that the demutualisation would have been constructed on the basis that the £1bn of over-provisioning be released gradually for the benefit of the pre-demutualisation with profits policyholders rather than constitute a windfall profit for the purchaser.)

Why no outright buyer?

  The value of a life office has three components: the amount of its capital base; its embedded value (ie the present value of the future profits predicted to arise over time from its in-force business); and the goodwill element (which reflects the value of the brand name and its distribution capability).

  As for its capital base, Equitable practically had none once the impact of the poor markets, general liability inflation and the impact of the GAOs had been factored together. Indeed, it might be argued that, using very conservative assumptions (actuaries have some degree of discretion over how conservative or otherwise they choose to be in putting a present day value on future liabilities), Equitable's capital position was negative as at end 2000.

  Equitable's embedded value would have been very small relative to the size of its in-force book of business. As we noted above, this outcome would have resulted from the fact that it ran its business on a very low cost base that left very little meat on the bone as far as profit to be shared with a prospective proprietor was concerned. The embedded value of Equitable's in-force book was probably not very much more than £1bn.

  As for the goodwill element, how long is a piece of string? Equitable's image had already been tarnished but, before the announcement that it had failed to find a bidder, probably not terminally so. In conducting a valuation on a behalf of a prospective purchaser, we would tend to have made some allowance for the value of the distribution channel (in the context of Equitable's prior to 8 December 2000 reputation) and have made some calculation of the value to an acquirer of Equitable's systems (which had had a very good reputation among insurance industry back office anoraks).

  Adding all this together, we are talking about a fundamental value of Equitable (pre 8 December 2000) of perhaps £2bn at best. This had to be set against the estimated £4bn of required financial repair, however. Even allowing for the fact that part of the financial repair might have been effected by means of debt finance and financial reinsurance arrangements (ie by non-equity financing), it always seemed impossible to breach the very wide chasm between Equitable's intrinsic value and the required rescue finance.

A LITTLE XTRA FROM THE HALIFAX

What's "not"

  Contrary to what has been reported in some newspapers and newswires, Equitable Life Assurance Society most certainly has not been "bought" by Halifax plc. The arrangements entered into with Halifax are in respect of certain of the operations of the Society and do not involve the demutualisation of Equitable and/or the transfer of its long term business to any person. Accordingly, Equitable will remain a mutual life office owned by its members.

The deal(s)

  The Halifax deal is effectively in three parts, and may be summarised as follows:

An initial certain £500m

  Of this amount, £200m is by way of consideration for Equitable's administration systems (which thereafter will be exploited by Halifax for its own benefit, subject to the provision of administration services to Equitable on a not-for-profit basis), a contract to manage (on a fully commercial basis) Equitable's c £30bn of funds (formerly managed in-house by Equitable), and the Equitable Life sales force (which will transfer to Halifax and be deployed to sell Halifax products to the Equitable customer base).

  The remaining £300m does not, in fact, represent "new" money introduced into Equitable, but arises from the financing by Halifax of part of Equitable's book of business. By way of background, about 75 per cent of Equitable's in-force business is with profits, with the remaining 25 per cent being made up of a £7bn tranche of unit-linked and other non-participating business (such as term assurance and annuities-in-payment). This £7bn book of non-participating business is a source of profit for Equitable's with profits policyholders and is estimated (in present value terms) to be likely to generate profits of £300m spread over many years in the future. Halifax, via a reinsurance arrangement, will advance Equitable the £300m up front and, over time, recoup this amount (together with appropriate margins for the provision of the financing) by collecting in the profits from this tranche of business as they gradually emerge.

A payment of up to £250m

  As mentioned above, Halifax has acquired the Equitable sales force (believed to be c 350 in number), which will be re-branded as "Halifax Equitable" and be deployed to sell Halifax products to the Equitable customer base. Halifax appears to have taken a pretty cautious approach to this element of the deal in that, aside from the seemingly nominal consideration for the sales force forming part of the £200m payment for Equitable's operating assets described above, Halifax will only put its hand in its pocket if the re-branded sales force comes up trumps in terms of new business production. The amount eventually paid by Halifax for the sales force is capped at £250m and could, in an ultra-worst case scenario, be zero.

  Although the Equitable sales force was the hottest act in town when it came to selling bucket loads of life and pensions policies badged with the (then) untarnished Equitable name, it remains to be seen what sort of reception (a handshake and a cup of tea, or a jab in the posterior with a sharp stick, perhaps) the typical Equitable customer will give to Johnny Salesman when he comes a-knocking at the door with a smile and a cheery "It's me again!".

A £250m peace dividend

  It clearly would be in the interest of the Society as a whole for it to resolve the guaranteed annuity option problem by having policyholders with pensions in deferment relinquish their options. We are some way away from seeing detailed proposals on this matter, but it does appear that the GAO policyholders will be asked to accept an up-front sum by way of an addition to the value of their policies in return for giving up the guarantee option.

  Will this find favour with the policyholders? Any such proposal is likely to require the consent of the majority of both those policyholders not having policies with GAOs and, separately, those with policies incorporating GAOs. As those policyholders without GAOs have nothing to lose and something to gain, we should expect that they would be overwhelmingly in favour.

  However, it is not at all clear that the GAO policyholders will vote for such a scheme, as it appears that the likely proposal will see the value of the GAO group's policies enhanced by 20 per cent as a result of the up-front capital boost but that, on the basis of current market conditions, the subsequent loss of the GAO would cause a concomitant 28 per cent fall in annuity income in retirement. In short, the GAO policyholders (who only a few months ago were being encouraged to rejoice in the splendid victory achieved on their behalf in the House of Lords) are being asked to take a haircut.

  A further problem in persuading GAO policyholders to vote for the deal would appear likely to arise if, as suggested, the GAO policyholders' vote is subject to weighting according to size of policy. Given that, in aggregate, we would expect those policyholders nearest retirement to have the largest accumulated pension funds and that the losses of value from taking the haircut would be more immediate and certain for such persons having greater voting weight, it would seem a "hard ask" to expect there to be a rush to sacrifice soon-to-be-enjoyed retirement benefits for the good of the Society as a whole.

  If Equitable did succeed in persuading its policyholders to vote for what it calls the "compromise" (a reasonable-sounding euphemism, much less challenging than "haircut", "sacrifice" or "potential loss" do you not think?), this would not strengthen its financial situation considerably, notwithstanding the £250m kicker from Halifax, as the Society would incur substantial expenditure in buying out the GAOs. The real benefit would not be the modest improvement in Equitable's solvency position, but the fact that the fundamentally unstable GAO problem would be eliminated. (Remember that GAOs have interest rate and mortality components, making for a very tricky-to-manage volatile liability position as: (a) mortality experience changes, usually negatively; and (b) long term interest rates fluctuate, each slight fall causing the liability to increase. The combination of these factors means that the £1.5bn GAO liability could quite easily turn into £2bn or more over a short period of time or, were long term interest rates to rise appreciably, decline to £1bn or less.)

15 February 2001