2. Contextualisation

Somewhere around 1886 in the Eastern Cape, South Africa, counsel for the plaintiffs in Clair v Port Elizabeth Harbour Board and Kennedy v Port Elizabeth Harbour Board Footnote 1 argued that the evidence of actuaries afforded a good basis as to the measure of damages, citing a number of authorities to support this ascertainment. The learned judge appears to have agreed, as he based his award for damages to a large extent on annuity values derived from the ‘Carlisle’ and ‘Northampton’ tables, these tables being well known to nineteenth century actuaries.

Over time South African courts have increasingly relied on actuarial evidence (Howroyd & Howroyd, Reference Howroyd and Howroyd1958:65; Milburn-Pyle & van der Linde, Reference Milburn-Pyle and van der Linde1974; Koch, Reference Koch1983). In the experience of the author, South African actuaries are commonly called by the courts to give expert testimony in cases involving the assessment of quantum of damages in lieu of future loss of earnings. This sometimes happens in even the most simple of cases.

In contrast, authors have commented on the historical reluctance of the English courts to rely on actuarial evidence. Refer, for example, to Street (Reference Street1962); Prevett (Reference Prevett1968); Owen & Shier (Reference Owen and Shier1986) and Burrows (Reference Burrows1987).

In 1984 an Inter-Professional Working Party consisting of actuaries and lawyers released a series of actuarial tables known as the Ogden Tables. These tables are published by the UK Government Actuary's Department and rely on actuarial principles for their construction.Footnote 2 Increasingly these tables gained general acceptance by the English courts. The Evidence Act 1995 made it clear that the Ogden Tables may be admitted into evidence. The widespread use of the Ogden Tables is an indication of the English legal fraternity's acceptance of the authority of the actuarial principles underlying the tables, though not necessarily of the need of actuarial expert testimony.

Hodgkinson & James (Reference Hodgkinson and James2007:467–470) trace the history of actuarial evidence in the English courts over the past few decades. They report of a long war waged in the 1970s and 1980s to convince the courts to allow actuarial evidence. Prevett (Reference Prevett1968:295) remarks that until then the help of an actuary was sought in exceptional cases only. Some detail of various battles of this war can be gleaned from Prevett (Reference Prevett1968) and Prevett (Reference Prevett1972).

It would appear that the war is not over yet. Daykin (Reference Daykin2009) suggests that the current English approach is not sensible in damage cases where the compensation is likely to be large, and proper actuarial advice should be sought under such circumstances.

It is hoped that this paper will add a few arrows to the quivers of English actuaries who wish to offer their professional opinions in the English courtrooms for the benefit of the public at large.

3. The question of a surprising divergence

The English courts are still reluctant to admit the testimony of actuaries, as reported by Hodgkinson & James (Reference Hodgkinson and James2007:467–8). South African courts however, readily admit and frequently insist on actuarial testimony.

Erasmus (Reference Erasmus1975) has analysed the history of the South African law of damages, and concluded that the quantum of damages is established primarily by application of rules and concepts found in English law.

South African law of evidence is firmly rooted in its English parent. In fact, the Civil Proceedings Evidence Act 25 of 1965 has set South African common law of evidence as the English law of evidence as it stood on 30 May 1961, the day before South Africa became a republic.

Thus, the divergence in the approaches of the South African and English courts towards actuarial expert testimony is somewhat surprising. This, then, is the research problem.

Contrary to actuarial intuition, the problem is not solved by looking at it from an actuarial perspective. The answer lies primarily within the field of law, particularly the law of damages and the law of evidence, and the consequences of the operation and interaction of these branches of law, in circumstances where actuaries and other professionals maintain that their opinions are of an expert nature (as understood in legal parlance) for the benefit of the courts.

The author being a South African practitioner, the reader should appreciate that the literature considered will be biased towards South African law. However, the law of England and the law of South Africa are near identical as far as the procedural aspects of law in context of this research are concerned. This should therefore not invalidate the analysis.

4. Outline of paper

Sections 5 to 8 sketch the environment within which this research is conducted, for the benefit of an actuarial audience. Section 9 states South African rules of assessment of damages. Sections 10 to 12 outline the procedures followed by English and South African courts in the process to assess damages. Sections 13 to 15 examine the English law for the establishment of damages. It then contrasts the English approach with the South African approach, in particular relating to the practice of admitting actuarial expert testimony. A summary of the author's findings appears in Section 16. The author's conclusions are stated in Section 17.

5. Terminology

A review of the literature shows that scholars of law use a variety of terms when referring to the same concept. Thus, for the purpose of an actuarial audience, some elements of terminology need to be introduced first in order to limit confusion.

6. Scope of consideration

In the interest of academic rigour, and taking into consideration the fragmented nature of the common law operating in conjunction with the rigidity of a formal logic set out below, the environment of the analysis has to be carefully delineated. If this is not done, then the analysis, summary and conclusions may be invalidated.

In what follows, it will be assumed that the damage and damages under consideration will satisfy the following limiting criteria:

• – the damage relates only to the period following the date of assessment, i.e. it is prospective, (i.e. we are ignoring any losses between the date of commission of the wrong and the date of assessment, on the presumption that such losses have been agreed upon and are not part of the dispute);

• – the plaintiff had no promotional prospects over the balance of his career;

• – there is only one plaintiff and only one defendant;

• – the plaintiff has zero residual earnings capacity;

• – it has already been established that the defendant is liable;

• – it has already been established that there is no contributory negligence on the side of the plaintiff;

• – the matter has reached such a stage that evidence in open court is now being considered, in order to establish the quantum of damages;

• – it has already been established that the damages will be settled in the form of a lump sum, not an annuityFootnote 5; and

• – likely compensation is so large that the matter is heard in the High Court.

7. The doctrine of precedent

It should be kept in mind that both England and South Africa are common law jurisdictions. They do not have comprehensive purpose-made statutory laws that govern the assessment of quantum of damages (although a sprinkling of statutory law has been added over the years). Instead one has to turn to judge-made rules of law as contained in the (either reported or unreported) judgments of the higher courts. This approach is commonly referred to as the ‘doctrine of precedent’. This doctrine dictates that there is ‘a general duty of judges to follow the legal rulings in previous judicial decisions’ (Hahlo & Kahn, Reference Hahlo and Kahn1973:214). Thus if in a particular case the material facts are the same as those that existed at the hearing of a previous case then the previous ruling must be followed. A hierarchy of courts exists and the higher up the hierarchy the stronger the authority of such court's pronouncements. In particular, a court of appeal carries more authority than the court of first instance, and may override any rules made by the court of first instance.

In what follows, these rules are loosely referred to as ‘general’ rules of law.

This system is commonly found in the anglophone jurisdictions, such as the United States of America, Australia, Scotland and parts of Canada (besides the two jurisdictions under considerationFootnote 6). Continental Europe generally uses a substantially different system.

As a result of the operation of a common law system, the quantum rules will differ from one jurisdiction to the next.

It would appear from a study of the writings of legal scholars that the extent of generally recognised South African rules of law for the assessment of damages is limited. Koch (Reference Koch1993:39) expresses surprise at this lack of detailed legal rules, given that there have been so many reported court judgments that address this issue. In fact, South African courts seem to shy away from laying down hard-and-fast rules.

On the other hand, the English courts do not share this reluctance.

It is suggested that this difference in approach lies at the root of the difference in treatment of actuarial expert testimony by South African and English courts.

8. Formal systems: law and logic

Part of the backdrop of this paper concerns the structure of a formal (or logic, or mathematical) system. In terms of such a system, there are definitions and a list of well-formed statements called axioms that are accepted without proof. Building on the axioms one can create valid (but not necessarily true) statements called theorems, obtained by applying logical deduction.

Extending this to the study of law, the formalist view of law holds that law can be treated as a science with its own set of fundamental rules resembling a formal system's axioms and theorems. There are some doubts among legal scholars and practitioners as to the validity of the formalist view, and it is by no means universally accepted. It is commonly argued that law as a discipline does not operate within a vacuum and some real-life practicality needs to be injected. However, for the purpose of this paper it is assumed that the formalist view of law is valid. The quest is then to establish a set of axioms the courts in South Africa and England assume and use without proof, when establishing the quantum of damages.

Due to the fragmented nature of the common law system, it is to be expected that the axioms in the two jurisdictions are unlikely to be the same. This is indeed the case.

9. Contextualised axiomatic quantum rules in South African law

Regarding the quantification of damages within the narrow environment sketched in Section 6, it is suggested that South African common law contains only two rules that can be considered to be fundamental in an axiomatic sense.

10. Establishing the quantum of damages

The physical court procedure used to establish the quantum of damages rests heavily on two branches of law, namely the law of evidence and procedural law. There are two major international systems upon which the law of evidence and procedural law can rest (Schmidt & Rademeyer Reference Schmidt and Rademeyer2000:12). The Anglo-American approach, commonly used in anglophone jurisdictions, is adversarial. The Continental approach, commonly used in continental Europe, is inquisitorial. The major difference in approach lies in firstly, the involvement of the judicial officer such as the judge or magistrate, and secondly the admissibility rules. In terms of the Anglo-American approach, the presentation of evidence to the trier of fact is tightly regulated, and only ‘admissible’ evidence is allowed to be heard. The judicial officer plays a passive role, and decides what the facts are after listening to each side presenting its case. In nature the format of the proceedings is similar to a boxing match with the referee not getting involved in the fight. The Continental approach allows the judicial officer to be actively involved in the process, asking most of the questions and becoming involved in the dispute. Also, evidence is seldom excluded and there is much less focus on the ‘admissibility’ or otherwise of evidence.

English and South African law of evidence and procedural law follow the Anglo-American approach.

11. Procedural law

Procedural law sets out the procedure that the court has to follow when it goes about its business of (amongst others) establishing a fact such as the quantum of damages.

In the case under consideration, the plaintiff and defendant are not in agreement as to the quantum of damages. The trier of fact now considers evidence including testimony, in order to make an assessment of the damages. An outline of the procedure is given in appendix A.

12. Law of evidence

The law of evidence regulates the manner in which facts are proven in court. Schmidt & Rademeyer (Reference Schmidt and Rademeyer2003) provides a thorough treatment of the law of evidence.

A fact is established after consideration of evidence. ‘Evidence’ has a particular meaning at law. Carstens (Reference Carstens2001:50) defines evidence as ‘the testimony of witnesses and the production of documents and other exhibits, which can be used for the purpose of proof in legal proceedings.’ A ‘witness’ is ‘a person who gives formal or sworn evidence at a hearing’.

‘Testimony’ in a legal sense is a subset of evidence. It is a solemn statement or declaration of fact. It may be oral or written, but under oath in either case. A false oath is perjury, which constitutes a criminal offence.

When giving testimony in court, witnesses may only testify about things that they have perceived with their senses, e.g. what they have seen. Witnesses are not allowed to offer their opinions. An opinion as such is deemed ‘inadmissible evidence’ and may not be presented to the court, subject to certain exceptions.

Meintjes-Van der Walt (Reference Meintjes-Van der Walt2001) analysed one important exception to this rule – that of expert opinion testimony. Such testimony is admissible provided it is sufficiently relevant. It is relevant if, according to Meintjes-Van der Walt (Reference Meintjes-Van der Walt2001:238), ‘the expert by reason of special knowledge or skill, is better qualified to draw an inference than the trier of fact’, the trier of fact being the court. However, relevance by itself is not sufficient for the testimony to be admissible. The opinion of the expert must give appreciable assistance to the trier of fact, implying that the evidence must pass a cost-benefits test. It is for the court to rule on the admissibility or inadmissibility of the testimony. When a matter falls within the common knowledge of the trier of fact then the opinion of an expert witness as such is superfluous and should not be admitted as evidence. However, from the analysis of Meintjes-Van der Walt (op. cit. at 244) it appears as if South African courts are quite lenient in allowing expert opinion testimony. The practice seems to be to readily admit testimony that purports to be of an expert nature, and then to leave it to the trier of fact to determine the level of reliance to place on the testimony, only after the testimony has been considered. Thus the hurdle for admissibility seems to be low and the hurdle for reliance is higher. When the hurdle for reliance has not been cleared, the trier of fact will attach little value to the testimony.

13. Contextualised axiomatic quantum rules in English law

Koch (Reference Koch1993) compares the laws of damages assessment in South African and English law. It would appear that the English lawmakers have been more enthusiastic in devising general rules of law (in the sense of axioms) to assist in the assessment of damages.

Given this enthusiasm, it is to be expected that the rules will differ according to the type of damage sustained. Hence it becomes necessary to delineate the environment carefully, as in Section 6.

14. Actuarial evidence in England

15. A possible weakness in the English axiomatic rules?

Within the narrow environment set out in Section 6, and taking into account the rules in Section 13, it is clear from logical deduction that, in the English courts, a ‘theorem of actuarial testimony’ exists, which holds that ‘actuarial testimony is [generally] inadmissible’Footnote 17.

On the other hand it is suggested that most actuaries would be of the opinion that the above theorem is not necessarily true, and that many instances exist where actuarial testimony can lend appreciable assistance to the courts.

One way in which it can be shown that the theorem is not necessarily true, is to show that one of the rules in Section 13 is not necessarily true.

In the view of the author, the one rule that is most doubtful is Rule 2: ‘Use the multiplier-multiplicand method’.

Actuaries should recognise this rule as the simple actuarial principle of the valuation of an annuity. Ciecka (Reference Ciecka2008) records that these principles have been established by the work of the 17^th century Dutch mathematicians Jan (Johan) de Witt, Jan Hudde and Christiaan Huygens.

Since the time of the three mathematicians, major scientific advances regarding economic value have been made. Rubinstein (Reference Rubinstein2006) provides a concise chronicle of some of these advances.

It is submitted that economic value is the modern tool to assist in determining the sum of money as envisaged by Lord Blackburn in section 13 above. The axiomatic use of a primitive actuarial tool such as the multiplier-multiplicand method that disregards some major developments in the theory of economic value over more than three centuries is surprising.

Within Section 6's narrow environment, English and South African courts would probably arrive at very similar answers for the loss, and so the English courts’ exclusion of actuarial testimony under those circumstances seems reasonable.

However, based on the author's personal experience of South African legal process, and a reading of English law, even a small broadening of the environment can lead to a divergence in the answers.

A particularly simple example that might add clarity involves the relaxation of the ‘no promotional prospects’ assumption in Section 6.

The assumption is now made that the plaintiff is a young graduate at the start of his career. Amongst labour economists, as well as actuaries with some training in pension fund mathematics, it is common cause that his promotional prospects have a significant impact on the expected present value of his work-life earnings.

An assumption typically made by South African non-actuarial experts in the field of remuneration systems is that a new entrant to the job market can expect real promotional earnings growth of 5% per annum, until reaching a peak at the age of 40–45.

Consider a scenario of (say) an individual aged 21 exactly with retirement age of 65 exactly. Using typical South African demographic and economic assumptions, the impact of the promotions can easily be quantified. In this case, relaxation of the ‘no promotional prospects’ assumption increases the value of work-life earnings by approximately 82%.Footnote 18

In this scenario, the English courts have to turn to the Ogden Tables. To evaluate the promotional aspects, the court has to turn to paragraph 23 of the Explanatory notes. This paragraph contains a brief but highly technical discussion on how to adjust the tables for promotional prospects. In this case, the trier of fact has to apply paragraph 23 and dissect the plaintiff's future lifetime earnings into at least 25 sections (more if a pension is involved). These sections run from age 21 to 22, age 22 to 23, and so on up to age 44 to 45 and then finally age 45 to 65. The trier of fact then has to perform a calculation for each section in the manner set out in paragraph 23, and then has to add the results for all the sections to arrive at a global amount.

For this exercise a spreadsheet and some actuarial ability would be helpful. It is submitted that it is doubtful that ‘the most stupid circuit judge’ together with ‘the two most stupid advocates’Footnote 19 (in Sir Michael Ogden's words) is likely to arrive at the correct answer.

Alternatively to paragraph 23 of the Explanatory notes, there is scope to allow for ‘contingencies’ which leaves a back door open for an implicit allowance for promotional prospects, by means of an adjustment by the trier of fact to the multiplier or multiplicand. This is quite subjective though and not subject to any scientific scrutiny.

South African courts, on the other hand, will allow explicitly for the possibility of real promotional earnings growth, on the strength of the testimony of experts in the field of employee remuneration systems. In practice, the court will turn to an actuary to assist the court in determining the likely numerical effect on the work life earnings, of real promotional earnings growth as identified and enumerated by other non-actuarial experts. South African courts are allowed this leeway, as they are simply following South African rules 1 and 2.

There is an additional benefit of the South African approach, which is also available to parties to a dispute in the English courts. The presentation of expert actuarial testimony allows counsel for the counterparty to cross-examine the witness and thus firmly establish the validity of the opinion. It also (in all probability) allows the counterparty to appoint a different actuary to present testimony in the matter. Thus, a peer review mechanism is introduced and the court has added comfort regarding the opinions of the respective experts.

English courts are also simply following their own 4 legal rules. Unfortunately, the outcomes may be significantly divergent for most but the simplest cases. It would appear that in this particular case the aims as stated by Lord Blackwell will not have been met.

This outcome is corroborated in part by an empirical study of Lewis et al. (Reference Lewis, McNabb, Robinson and Wass2003) that compares awards for damages in Britain and the USA. They found strong evidence that British claimants are not fully compensated, in cases where the claimant has promotional prospects.

Matters can become much more complex, and South African triers of fact are prone in such cases to turn to the assistance of testimony of actuaries and other experts. Complications include:

• – progressive taxation systems in conjunction with promotional increases producing cross-correlation effects;

• – A variety of possible life-time earnings paths both before and after the loss-causing incident, each with a particular probability associated with its realisation;

• – periods of unemployment when the plaintiff has to undergo medical procedures as a result of the sequelae of the loss-causing incident (a medical specialist would normally be able to estimate the probability of a particular procedure, the age at which the procedure is likely to occur and the likely time off sick due to the procedure);

• – sick leave policy of employer;

• – periods of unemployment due to family responsibility;

• – different tax regimes, depending on the source of earnings;

• – taxation of fringe benefits;

• – different retirement ages before and after the loss-causing incident, and hence different pension levels;

• – different tax regimes for pension benefits and earnings before retirement;

• – part of pension taken as a lump sum;

• – pension lump sum falling under a different tax regime as a regular pension; and

• – non-integer ages and durations.

In South Africa, the trier of fact would usually consider actuarial expert testimony on the financial impact of these factors. It is not clear what the approach in England is, when dealing with these complications.

16. Summary

The legal procedural systems of South Africa and England are (as far as influence on the outcome of the deliberations) essentially the same, when applied to the assessment of future pecuniary damages.

However, and very surprisingly, South African courts embrace actuarial expert testimony whereas English courts do not, except in some narrowly defined circumstances.

This divergence arises out of a difference in the substantive law of damages as interpreted by the higher courts in the two jurisdictions.

This is also surprising, as South African quantum of damages is established primarily by consideration of rules and concepts found in English law.

In terms of substantive law, the South African courts’ view is that no (or little) hard-and-fast rules can be made regarding the assessment of damages. Each case must be judged on its own merit. Within the adversarial system of adjudication, South African courts are in doubt as to which of the litigants to believe and are then more likely to turn to expert advice on matters in which the courts do not possess the necessary expertise.

However, the English approach has been to lay down simple rules of assessment that the courts can apply without deference to external expertise.

The doctrine of precedence enforces general compliance with these views.

English substantive law views loss as a simple multiplication exercise, with some judgment thrown in. The trier of fact is considered to be well equipped to perform this exercise.

South African substantive law views it as a solution to a complex stochastic processes problem, with some judgment thrown in. Once entering on more complex terrain such as stochastic processes, the courts are bound to call in the advice of experts, even more so in adversarial systems such as ours.

For as long as the body of English substantive law views loss of earnings as a simple multiplication exercise, with a sprinkling of judgment added, it is the opinion of the author that it is not likely that the English courts will frequently turn to expert actuarial or similar testimonyFootnote 20.

It is submitted that the question of assessment of future pecuniary damages is essentially a stochastic processes question. In a simple case this may be reduced to a discrete process. A simple and easy-to-use method such as the multiplier-multiplicand method will then produce an acceptable answer at low cost. In a more complex case, it becomes difficult (or impossible) to reduce the stochastic process to a simple discrete process. Assistance from actuaries and other experts is called for. Acceptance of this position is likely to enhance the relevance of actuarial and other expert testimony, as has been seen in South Africa.

Such a move is likely to enhance the status of science in the courtrooms, in general. A vast number of professionals and scientists in various fields are well equipped to assist the courts in their fact-finding missions, if only they were permitted to do so. A reading of relevant South African case law shows that the courts rely extensively on expert testimony. From own experience, South African lawyers rely extensively on actuarial expert opinion in pre-trial negotiations. This assists in speedily narrowing down the points of disagreement within an adversarial legal system such as used both in England and South Africa.

Contrary to what might be thought, this system saves time and money as it gives the parties to the dispute a fairly accurate indication of what a likely reward of damages will be, while still treating the parties to the dispute fairly. Thus the risk of lengthy, costly trials to assess future pecuniary damages can be reduced.

It has been said that the English approach saves cost, and provides simplicity, certainty and consistency across cases.

There is a substantial risk, however, of an unfair award of damages when such a simplified approach is followed in a complex case involving large losses. In the opinion of the author, this risk outweighs the benefits of the other considerations, under such circumstances.

18. Disclaimer

This paper, although it covers material relating to the law, has been written by an actuary and not a lawyer. Its sole purpose is to stimulate debate within the actuarial profession. It does not purport to constitute legal advice. The author does not accept any responsibility for any reliance placed on it by third parties. Any reader who wishes to obtain legal advice or advice relating to matters where legal input may be of benefit, should seek such advice from a suitable qualified legal expert. It might be noted however that the author has given expert actuarial testimony regarding loss of earnings, both written and oral, on a large number of occasions.