SUBMISSIONS OF COUNSEL ASSISTING
THE SPECIAL COMMISSION OF INQUIRY
INTO THE MEDICAL RESEARCH
Allens Allen Allen & Hemsley, then Allens Arthur Robinson
Amaba Amaba Pty Ltd, known as Jsekarb Pty Ltd
Amaca Amaca Pty Ltd, to February 2001 known as James Hardie & Coy
Coy James Hardie & Coy Pty Ltd to February 2001
Current Data James Hardie’s asbestos claims data for the period 1 April 2000 to
31 December 2000
DOCI Deed of Covenant and Indemnity dated 15 February 2001 between
JHIL, Coy and Jsekarb (Ex 1, vol. 6, tab 60)
DOCIA Deed of Covenant, Indemnity and Access dated 31 March 2003
between JHINV and ABN60 (Ex 122, PGM2, vol 1, tab 49)
February Report The Trowbridge letter report dated 13 February 2001
JHA James Hardie Australia Pty Ltd
JHFC James Hardie Fibre Cement Pty Limited
JHIL James Hardie Industries Limited to October 2001, thereafter ABN
JHINV James Hardie Industries NV (formerly RCI Netherland Holdings
JHNV James Hardie NV
JHR James Hardie Research Pty Ltd
Jsekarb Jsekarb Pty Ltd to February 2001, thereafter Amaba Pty Ltd
KPMG KPMG Actuaries Pty Ltd
MRCF Medical Research & Compensation Foundation
Trowbridge Trowbridge Consulting Ltd, later Trowbridge Deloitte Ltd.
A: THE CURRENT POSITION OF THE MRCF
Actuarial firms Trowbridge and KPMG have given different estimates of the present and future
liabilities of Amaca and Amaba as at 30 June 2003. Those estimates are differentiated by the following
i) number of future reported claims;
ii) claim costs;
iii) nil settlement rate;
iv) superimposed inflation.
The last is the major source of variance ($356.5m out of $483.6m – see Wilkinson, Ex 252, p
a) In relation to superimposed inflation, the evidence suggests that an appropriate rate may be
either 4%, 6% reducing to 2%, or 2%. Which is appropriate? What impact would the choice of
rate have on the estimate of liabilities (for example, adoption of 4% would increase Trowbridge’s
estimate from $1,089.8 million to $1,866.1 million (Ex 3 vol 3 ps 613, 653); adoption of 6%
reducing to 2% would increase KPMG’s estimate from $1,573.4 million to $1,958.1 million (Ex
252 p 109))?
b) What estimates should the Commission accept?
c) What amount, or amounts, should be found to be the “future asbestos related liabilities” of
Amaca and Amaba, for the purposes of Term of Reference 1? How is that amount to be arrived
Some evidence suggests that an analysis directed at establishing a level of funding for Amaca and
Amaba that might have a reasonable likelihood of satisfying all legitimate claims would proceed in
different fashion from the analyses undertaken by Trowbridge and KPMG.
(a) In particular, such an analysis might have to make conservative allowance for the following
i) number of future reported claims
ii) claim costs
iii) cashflows and discount rate
iv) superimposed inflation
v) excluded claims (see below)
vi) propensity to sue.
Is this correct?
(b) If so, could these matters be catered for by including in the models already proposed by
Trowbridge and KPMG:
i) more conservative allowances (where assumption exists);
ii) sensitivity analyses;
iii) a buffer or contingency factor; and/or
iv) an explicit estimate of the probability of the assessed net present value of claims?
1. The Commission has received in evidence two actuarial estimates of the liabilities of
Amaca and Amaba for asbestos related diseases as at 30 June 2003. Although
referrable to a time now a year past, they are the most current reports available.
2. Those reports are by Messrs Minty and Atkins of Trowbridge Deloitte (Ex 3, v 3, t
10) and Mr Wilkinson of KPMG Actuaries Pty Ltd (Ex 252). The Commission also
has the benefit of a report by Mr Whitehead of Taylor Fry (Ex 251), which explains
the nature of the actuarial task in this context, and provides a commentary on the
3. Issues 1 and 4 raise a series of questions concerning the actuarial assessments, all
directed to the ultimate issue: what is a reasonable assessment of the liabilities facing
Amaca and Amaba, or, to be more precise, what is the best assessment of the sum
which would need to be available now to achieve reasonable confidence that all
legitimate future claims will be satisfied. The latter formulation is more precise
because the evidence indicated that the purpose for which an actuarial report is
required will affect the nature of the estimates made, and the methods adopted to
Estimating the liabilities as at 30 June 2003
4. The basic elements of the Trowbridge and KPMG estimates are as follows:
Future Meso reports 4,149 4374
Average non-nil cost $280,000 301,750 (includes a “large
claim” allowance of
$41,750: Ex 252, p 105)
Nil claims (Meso) 20% 17.5%
Expected Peak (Meso) 2011/12 2011
Discount Rate 5% 4.49-5.56%
Superimposed Inflation Nil 2%
NPV 1089.8m 1573.4m
NPV with SI sensitivity 1,866.1 (4%) 1958.1 (6% reducing to
5. Mr Wilkinson has analysed the impact of the major points of difference in Table E.5:
6. The major point of difference is the allowance of SI. Ultimately, the evidence of
Mr Minty was that it was appropriate to make such an allowance if one was acting on
instructions that made such an allowance permissible. His evidence was that as at
30 June 2003 past experience would suggest that an allowance of 4% might be
appropriate for the future, and he did not disagree with the view of Mr Whitehead in
that respect (T 3313.31 – 3314.10; 3316.50 – 3317.31; Whitehead at T 3210.42 –
7. Application of a 4% allowance for SI would increase Trowbridge’s assessment of the
NPV of claims above that of KPMG.
8. Mr Wilkinson’s 2% SI figure was not one he defended with conviction, as against the
“sensitivity” he proposed of SI running at 6% for 5 years, decreasing linearly to 2%
over 5 years, then remaining at 2% in the long term (Ex 252, p 109; T 3396.1 –
3397.33). Two factors support a view that the latter is a more appropriate assumption
than the 2% adopted by Mr Wilkinson as his primary position. The first is that, as Mr
Wilkinson acknowledged, SI was likely to be higher than 2% earlier in the period,
and having regard to the effect of discounting, it would be more accurate to reflect
that in the SI assumption. Secondly, as Mr Whitehead said (T 3211.25 - .35) one of
the matters which contributes to SI is improving medical care for those affected by
mesothelioma in particular. Such improvements tend to increase damages (see also
Attrill at T 1207.30 - .56). In his report Mr Wilkinson said (Ex 252, para 1.23):
9. Increased use of Alimta seems likely to occur, as the Food and Drug Administration
(US) announced on 5 February 2004 that it had approved Alimta for use in
combination with Cisplatin for the treatment of patients with malignant pleural
mesothelioma.1 Material published with the announcement indicated that in clinical
trials patients treated with Alimta and Cisplatin had a median survival of 12.1 months
compared to 9.3 months for patients treated with Cisplatin alone.2
10. The other points of distinction between the actuaries are of less significance.
However as regards claim numbers, there are some factors which Mr Wilkinson’s
exposure model did not take into account that would suggest an earlier peak then he
forecast (see T 3386.55 – 3388.18; 3389.5 - .51). That would reduce his claim
See http://www.fda.gov/bbs/topics/NEWS/2004/NEW01018.html, to form part Ex 313
See http://www.fda.gov/cder/drug/infopage/alimta/alimtaQA.htm to form part of Ex 313
numbers. As regards claim costs, this is a matter that may relate indirectly to the SI
assumption – a higher claim cost assumption makes lower SI more acceptable.
11. Taking all these matters into account, it would be appropriate for the Commission to
adopt a figure of $1.9 billion as a starting point for estimating the liabilities of Amaca
and Amaba. It is necessary to describe that figure as a starting point because, as the
evidence emerged, it became clear that the techniques used and assumptions
employed in the actuarial reports from which that number was derived were
techniques and assumptions designed for an inquiry different from that which is of
present concern. In particular, none of the reports was directed at assessing the sum
that would need to be provided on a once for all time basis in order to have a
reasonable degree of confidence that all legitimate claims would be met (“defining an
12. In the first place, that each of the models was deterministic as opposed to stochastic
meant that there was no allowance for probability in the assessed NPV of claims.
Each derived a central estimate or median NPV, which is the outcome in the centre of
the distribution of possible outcomes, such that 50% of the outcomes are above it, and
50% are below. To establish a fund by reference to a central estimate is to establish a
fund that has only a 50/50 chance of meeting all its liabilities. (See Minty at
T 821.10-17: Wilkinson at T 3383.28 - .47; Marshall at T 915.29 - .38).
13. A reasonable degree of confidence that all claims would be met implies a higher
probability, and therefore the selection of a higher NPV (cf Marshall at T 918.43 –
919.37 who refers to probabilities of the order of 90%). Such an exercise was not
attempted by the actuaries, but there can be no doubt it would significantly increase
their assessments. For such an exercise Mr Wilkinson would have adopted more
conservative assumptions, including as to costs and SI (T 3397.35 – 3398.27).
14. In addition it would be necessary to make allowance for areas of potential liability
which were excluded from the analysis. These include various potential liabilities
explicitly excluded from consideration by the actuaries. These are stated in the 2003
Trowbridge report as follows:
“a lowering in the standard of causation in a number of lung cancer
claims, reversing the generally accepted condition that evidence of
asbestos is required before lung cancer will be attributable to asbestos
claims arising from mental anguish associated with asbestos exposure
cross-claims by Amaca…
environmental, land remediation or clean-up claims
general emergence of new sources of claims not currently represented in
the Amaca database
claims arising outside of Australia
costs incurred due to the charge-back from NSW Dust Diseases Board
(DDB) of amounts paid by it before the settlement of a claim at common
15. Claims for exemplary damages may also be taken to be excluded.
16. An indication of the magnitude of the risks associated with some of these potential
liabilities (in particular, remediation, DDB reimbursements, exemplary damages and
US claims) appears from James Hardie’s own records (see Ex 57, v 1, pp 14 – 16; the
Forrest QC and Watson opinion on exemplary damages, Ex 61, v 4, pp 294-5 and 305
– 319; and Attrill’s evidence at T 1165.40 –1166.1; 1217.47 – 1220.5). It is
necessary to have regard both to the likelihood of the risk materialising and the
magnitude of its impact if it does. Some of the documents in which these matters are
discussed are subject to non-publication orders. For present purposes, it is sufficient
to note that the risks appear to be significant, and that in some cases (remediation in
particular) the potential liabilities were very large.
17. In addition to the excluded risks referred to in section 8 of the Trowbridge 2003
Report, there is evidence that two other areas for potential increases in claims were
effectively omitted from consideration by the actuaries.
18. The first of these is what are sometimes called “third wave” claims. They were
explained by Mr Whitehead at T 3211.47:
“<47> …the first
<48> wave of claims are generally linked to the mining and
<49> milling of asbestos and the production of …
<50> …asbestos products, so those would
<51> be people working in mines or the mills or working in the
<52> plants where asbestos was used to produce, for example,
<53> asbestos sheets. The second wave of claims would arise
<54> from the people who make use of those products. In the
<55> case of asbestos it was largely people in the construction
<56> industry for example and ship building and the workers
<57> were exposed and others were exposed to the asbestos in
<58> using and installing the asbestos containing product. We
< 1> then have the situation where that asbestos remains in the
< 2> environment until it is ultimately removed and there is
< 3> on-going exposure, sometimes call secondary exposure,
< 4> which could potentially lead to further claims in the
< 5> future. That exposure is on-going and is based on the fact
< 6> that, as I say, there’s a significant amount of asbestos
< 7> in the Australian building environment which would lead to
< 8> people being exposed to asbestos and having mesothelioma
< 9> cases and those are in the third wave claims.”
19. Mr Minty accepted that Trowbridge’s model did not fully allow for such claims, and
in particular, did not make full allowance for exposure to asbestos being an ongoing
matter (T 3320.17 – 3321.50). Mr Wilkinson agreed that the exposure data that
underlay his model made no allowance for third wave claims (T 3389.53 - .57).
20. It is difficult to assess the possible impact of such claims. On the one hand, it may be
accepted that the removal of asbestos which is now in situ is likely to involve much
lower levels of exposure than the original mining, milling, manufacturing and
installation of asbestos. This view is reflected in some American literature (see B
Price, “Analysis of Current Trends in United States Mesothelioma Incidence”, (1997)
145 American Journal of Epidemiology 211 at 217). On the other hand, while
standards of practice and risk awareness in relation to occupational asbestos exposure
may be relatively high, it is doubtful if that is so for the large number of “do-it-
yourself” home renovators and repairers in the Australian community. Despite an
invitation to do so (T 2598.33-47) James Hardie, which had been responsible for 70%
of Australian Asbestos consumption, (Ex 251, p 3-21) adduced no evidence that it
had ever conducted or been responsible for a public education campaign designed to
educate the public about the risks of the huge volumes of James Hardie’s asbestos
products still in homes and workplaces in Australia.
21. The second additional area of risk which seems to have been excluded from these
analyses is that of further increases in propensity to claim. Increases in propensity to
claim are likely to be the main explanation for the fact that mesothelioma claims
against James Hardie have increased at a much higher rate than the incidence of
mesothelioma in the community (see Whitehead at T 3208.51 – 3209.2; Minty at
T 3326.47 - 55; Marshall at T 3444.35-40, Ex 262 at p 7).
22. The fact that mesothelioma claims against Amaca are continuing to increase (rather
than having reached a distinct peak or plateau), and the fact that, while James Hardie
accounts for 70% of Australian asbestos consumption, it has claims made against it in
a much smaller proportion of the total number of cases, together suggest there
remains potential for the rate of claims against Amaca further to increase (Whitehead
at T 3209.33 – 3210.14; Minty at T 3328.3 – 3329.20). Mr Wilkinson’s impression
was that propensity to claim was not likely further to increase, but he acknowledged
that if one were defining an Adequate Fund it would be necessary to do more research
into the question (T 3401.53 – 3402.33).
23. As mentioned above, the evidence supports a conclusion that the purpose for which
an actuarial report is obtained has significant bearing on its content (see Ex 251, pp 3
– 31 to 3 – 35; Minty at T 3318.7 – 3320.15; Marshall at 913.6 – 915.3; Wilkinson at
T 3397.49 – 3398.2). In particular, the evidence indicates that if one were engaged in
defining an Adequate Fund, it would be necessary or appropriate:
to make allowance for the risks which were identified as “excluded” in
section 8 of the 2003 Trowbridge Report;
to make allowance for third wave claims, to the extent that the existing model
did not make adequate allowance for them;
to make allowance for the possibility of further increases in the rate of
propensity to claim against Amaca and Amaba;
to adopt a conservative discount rate (ie, a risk-free rate such as the
Commonwealth bond rate)
possibly, to provide for a “buffer” or prudential margin.
See: Minty at T 3319.54 – 3320.15; 3321.35 - .50; 3329.11 - .34; 821.5 – 823.57.
Wilkinson at T 3397.48 – 3398.27; 3399.26 – 3400.49; 3402.20 - .33.
Marshall at T 913.6 – 915.6; 3444.35 – 3445.44.
24. To some extent, perhaps a large extent, these considerations would be allowed for by
adoption of some of Mr Wilkinson’s sensitivities. His table of sensitivities is as
follows (Ex 252, p 109):
25. On Mr Wilkinson’s evidence, it seems that if defining an Adequate Fund he would
have applied, cumulatively, the sensitivities for SI, claims size and claim numbers
(“notification curve”) producing an estimate of total liabilities of approximately
26. This estimate would at least make some allowance for those areas of risk not
encompassed by the existing assessments. It would probably not meet Mr Marshall’s
standard of what was necessary for a high degree of certainty that all claims would be
met (see T 918.43 – 919.37).
27. On the evidence before the Commission it should be found that as at 30 June 2003 the
sum necessary to provide reasonable confidence that all James Hardie’s asbestos
victims who were able to establish claims against Amaca or Amaba would have their
claims satisfied in a sum not less than $2,240 million.
To what extent is the escalation in Trowbridge’s assessment of liability between the February Report and
the present attributable to:
(a) any change in claims administration or policy on the assumption of control by MRCF;
(b) “judicial inflation” (ie, an increase in the size of awards for equivalent cases);
(c) unanticipated increases in costs;
(d) unanticipated increases in numbers of claims;
(e) other changed circumstances?
28. One of the remarkable features of the circumstances examined by the Commission
has been the dramatic increase in the assessed exposure to asbestos claims of the
James Hardie subsidiaries from 1996 to 2003. The best estimate NPV in the
Trowbridge reports over that period are as follows:
October 1996 $230m
September 1998 $254m
June 2000 $294m
February 2001 $322m
August 2001 $574m
October 2002 $752m
September 2003 $1,089m
29. For various reasons (in particular, the impact of discounting) these numbers are not
directly comparable. Table 4.7 in Mr Whitehead’s report gives an analysis on a like-
Table 4.7: Comparison of Calculated Discounted Values of Projects
Payments after 31 March 2003 as at 31 March 2003, Discounted at 5% pa
Calculated Discounted Value of Projected Payments – Mesothelioma only
Valued at 31 March 2003 at 5% pa
Oct – Sep – 98 Jun –00 Feb – 01 Aug – 01 Oct – 02 Sep – 03
% pa % pa % pa % pa % pa % pa
Discount Rate 5% 5% 5% 5% 5% 5% 5%
$ mns $ mns $ mns $ mns $ mns $ mns $ mns
Calculated Discounted Values at 31 March 2003
Settlement Payments 153.2 150.7 211.0 266.4 525.5 688.8 877.1
Legal Expenses 51.1 71.5 52.7 65.6 67.4 73.1 93.1
Total 204.2 222.2 263.7 333.0 592.9 761.9 970.2
% change from previous 9% 19% 26% 78% 29% 27%
Cum % Change since Oct 1996 Report 9% 29% 63% 190% 273% 375%
Cumul % Change since Feb 2001 Letter 78% 129% 191%
Note: The projected cash flows for later reports are for year ending 30 June rather than 31 March. For the
purposes of this comparison, we have treated all projected cash flows as being for years ending 31
March. We do not expect this approach to create any material distortions in the results presented.
(Ex 251, para 4.6.13)
30. Mr Wilkinson’s report gives a break down of the increase in his estimate of the
liabilities as between February 2001 and June 2003:
(Ex 252, p 107)
31. It will be observed that the major components of the change are increases in the
expected number of claims and the average costs. As for claim numbers, the shift
appears to be due to the higher levels of claims being experienced by Amaca and
Amaba by June 2003:
(Ex 252, p 98).
32. It will be observed that claim numbers for all classes of liability have increased
substantially since 1999/2000, and that the increases are consistent and substantial for
mesothelioma claims, which are the major component of the liability.
33. As for claims costs, the figures are as follows:
(Ex 252, pp 99-100).
34. Further evidence about the increase in assessed future claims can be found in the
statistical analyses of claims filed with the Dust Disease Tribunal (“the DDT”)
undertaken by Turner Freeman solicitors that are annexed to the Statement of
Mr Jorg Probst dated 8 July 2004 (Ex 311). The tables attached to Mr Probst’s
Statement show the following:-
(i) Since 1999/2000 there has been a substantial increase in claims filed with the
DDT against non-James Hardie defendant, so the increase in claims is not
unique to James Hardie subsidiaries (see Annexure B – Table 3);
(ii) James Hardie subsidiaries have always been the primary defendants in a
significantly higher proportion of claims filed with the DDT than other
defendants (see Annexure F - Table 7); and
(iii) Since 2000, there has been an increase in the percentage of claims in which
the James Hardie subsidiaries are the primary defendant (see Annexure F -
35. On the available evidence, the increase in assessed future claims has been driven
largely by the increasing claim numbers. The most likely explanation of this
phenomenon, in the absence of evidence of a corresponding increase in the incidence
of mesothelioma in the wider community, is an increase in propensity to claim.
Evidence was given by Mr Gardiman and Mr Gordon as to the advertising conducted
by their firms over the period since the creation of the Foundation (Ex 230 and 233).
While the evidence suggests the possibility that marketing activities, in particular in
South Australia and for a brief period in Queensland may have contributed to an
increase in propensity to sue, it does not indicate that this is likely to be the major
reason for the increase in claims. There is some evidence of a natural tendency for
James Hardie to feature increasingly as a defendant in asbestos claims. According to
James Hardie was the pre-dominant organisation with regard to asbestos
manufacture and supply in Australia. As such, as one conducts more and more
asbestos litigation, one becomes increasingly familiar with the ubiquitous nature
of Hardies product. This emerges from the review of historical documents and
discovery; as well as the taking of instructions from plaintiffs and workmates
alike. It also occasionally emerges from advice from other defendants, in the
context of requests to join Hardies as an extra defendant. It has emerged for
example, in litigation against the Victorian Housing Commission that Hardies
was the exclusive supplier of AC product for the construction of Housing
Commission homes; likewise in Victorian schools; likewise that the vast majority
of asbestos product going through the Port of Melbourne was bound for James
Hardie. The acquisition of such knowledge inclines one to include Hardies when
a new similar case arises.
(Ex 233, pp 3-4)
36. Both witnesses deny any change in policy as regards suing James Hardie subsidiaries
as a result of the establishment of the Foundation.
37. The evidence does not offer a definitive explanation for the observed increase in
claim costs. However Mr Gordon’s evidence does provide some assistance:
I believe the range of damages claimed on behalf of asbestos victims has
increased in the period inquired after. I would offer the following general
i) For a mesothelioma victim with no claim for economic loss, the range of
damages usually achieved has moved in my experience from a median
figure of about $200,000 to a median figure of about $250,000 in Victoria
and New South Wales; and in Western Australia, from about $150,000 to
ii) Claims for damages for voluntary care have played a part in this increase.
Greater damages are now claimed under the principles enunciated in
Griffiths v Kirkmeyer; and claims are now occasionally made under the
principles enunciated in the NSW Court of Appeal decision in Sullivan v
iii) In some cases, the firm will seek an allowance in damages for the prospect
of a verdict for punitive damages. In my opinion, consideration of the
prospect of a punitive damages award is required more commonly in
Amaca litigation today than was the case five years ago.
There has also been some increase in the values ascribed for general damages for
pain and suffering and loss of enjoyment of life.
(Ex 233, pp4-5).
38. Mr Attrill’s report in Appendix 3 to the “Asbestos Liabilities Management Plan to
YEM 03” (Ex 7, MRCF1, v1, p0103) also offers some assistance. In this report he
offers as one of the reasons for the increase in claim costs for the YEM 01 the fact
that there had been an increase in end user claims and that end user claimants tended
to be younger thus increasing the number of larger claims.
39. Taken together, the evidence suggests that the increase in claims and claim costs has
been a natural progression, caused by forces and circumstances operating at the time
of the establishment of the Foundation, and not caused or accentuated by its
ISSUE 3 - Insurance
Does the MRCF have insurance recovery rights or rights of reimbursement that are not allowed for in
the 2003 Trowbridge and KPMG reports? If so, what is their value, and what impact would they have on
current estimates of the ability of Amaca and Amaba to satisfy their present and future liabilities? In
(a) On what basis did Trowbridge assess the anticipated insurance recoveries for Amaca
and Amaba? Why has the anticipated amount of insurance recoveries increased
between Trowbridges’ February 2001 report and the 2003 report?
(b) As regards policies issued by insurers in the HIH group, what is their number, face
value and scope?
(c) To what extent may there be rights of recovery pursuant to the HIH Claim Support
Scheme in respect of those policies (see Ex 220-223)?
(d) To what extent does s.562A of the Corporations Act make available claims against re-
(e) Insofar as the policies are claims made policies in respect of previous periods of
insurance, may claims be made now in reliance on s.54 of the Insurance Contracts
(f) What were the results of the recent inquiries on behalf on MRCF in the London
40. It would be premature to address this issue before receipt of Mr Wilkinson’s report on
insurance matters (see T 3414.44 – 3415.30) and additional information anticipated
from the MRCF.
ISSUES 5 – 29 - OVERVIEW
41. Mr McGregor gave evidence that the JHIL Board’s decision on 15 February 2001
was to fund the estimated liabilities of Coy & Jsekarb (T 1471.51 - .57; T 1454.14 -
.19). This could mean that the Board decided to fund the Trowbridge best estimate
NPV of the claims. Or, it could mean that the Board made an independent
assessment of the amount necessary to ensure Trowbridge’s estimated claims would
be satisfied. The latter seems to be the correct interpretation, both because JHIL did
not attempt to do the former (Trowbridge’s assessed NPV of all claims on a best
estimate basis was more than $320 million, whereas the assets available to the MRCF
were not estimated at more than $295 million), and the JHIL Board seems to have
attributed significance to the JHIL cash flow model, which was immaterial on the
42. [There is another possibility– that the JHIL Board thought it was doing both. This
arises because it seems the Board was, in February 2001, led to believe that the NPV
of all future claims was $284m, when this was in fact only the NPV of the first 20
years’ claims – see McGregor at T1577.55-1578.18 and Shafron at T1605.35-.47;
T1641.21-25. This possibility may treated as the same as the second for present
43. The second approach entails that critical to the decision to establish the Foundation
was JHIL’s independent assessment of the amount necessary to fund the anticipated
claims. When the evidence is examined a difficulty emerges: JHIL’s independent
assessment was so superficial and so flawed that it suggests the possibility that it was
not a serious or genuine exercise so far as those who conducted it were concerned.
44. A focus of the next series of issues is how JHIL came to form a view that the MRCF
was “fully funded”; whether there was there a reasonable basis for such a view; and
whether any such view was honestly held?
ISSUE 5: Omission of Current Data from the February Report
The February Report was, at the request of JHIL, prepared without reference to the Current Data. Some
evidence suggests that Messrs Minty and Marshall wished to have, and asked for, the Current Data for
the purposes of preparing the February Report. Other evidence suggests that Mr Minty or Mr Marshall
said that the Current Data was not required, or would not make any difference to their report.
(a) What evidence as to the circumstances resulting in the non-inclusion of the Current Data in the
February Report should be accepted?
(b) Would the Current Data have affected the estimates in the February Report, and if so, to what
extent? In this regard, there is evidence that:
v) Trowbridge, with knowledge of the Current Data, would have increased its estimated
20 year NPV from $286 million to $373 million and the total (50 year) NPV from $322
million to $437 million (in each case, with no allowance for super-imposed inflation)
(see Ex 50, p 338; see however Wilkinson, Ex 252, pages 77-84);
vi) KPMG, with knowledge of the Current Data, would have increased its estimated NPV
from $694.2m to $1044.5m (in each case, with super-imposed inflation at 2%) (Ex 252,
(c) To what extent, if at all, was JHIL influenced to ask Trowbridge to prepare the February
Report by reference to the data as at 31 March 2000 by:-
i) a concern that this reference to the Current Data would unduly delay completion of the
ii) a belief that data after March 2000, or perhaps June 2000, was not “clean”?
What evidence should be accepted
45. One of the early issues of concern for the new directors of the MRCF was that the
asbestos litigation costs for YEM 2001 were much higher than had been assumed in
the February Report and the cashflow model. The former projected full year net
costs on an undiscounted “Best Estimate” basis as follows:
YEM 2001 $21,566,258
YEM 2002 $22,838,310
(Ex 7, MRCF1, tab 16, p 0307. For the cash flow, see Ex 7, MRCF2, tab 19, p 0350).
46. The actual result for YEM 2001 was reported by Mr Attrill on 11 April 2000. It was
about $31,7 million (after adding back a receipt from QBE, which was not allowed
for in the February Report and was treated separately in the cash flow model - Ex 7,
MRCF1, tab 26).
47. According to Mr Cooper, this information prompted him to do an analysis which
suggested that if that rate of costs continued the Fund of the MRCF would be
exhausted in 10 or 11 years, even at an earning rate between 8.7 and 11.7% (Ex 7,
MRCF1, t 33; Cooper, Ex 5, paras 126-134).
48. The discrepancy between the estimates of Mr Cooper and those derived from the cash
flow model and the February Report is explained by the fact that the February Report
was prepared by reference to James Hardie Asbestos claims data to 31 March 2000,
and its experience in YEM 2001 was significantly different. The significance of the
absence of this data appears starkly from the KPMG actuarial report referred to in
issue 5(b)(ii) – with it there was an increase in the NPV of claims from $694m to
49. How it came about that the February Report was prepared without reference to the
Current Date is a question of some significance, on which the evidence is
inconsistent. The primary conflict is between the evidence of Mr Shafron on the one
hand and Mr Minty and Mr Marshall on the other as to an exchange between them on
19 January 2001. The occasion was a meeting, also attended by Mr Attrill and Mr
Morley, at which those present discussed the significance of the recent Watson and
Hurst paper and the basis on which Trowbridge might do a new report updating its
assessment for James Hardie in light of Watson and Hurst. Mr Shafron’s evidence
was that in the course of the meeting the following exchange occurred:
“Me: We have a current report from you which uses March 2000
claims data. Will you need to see more recent claims data
to do this work?
Mr Minty and
Mr Marshall: I don't think so. When you have 10 or more years of data,
you've got a lot of data points to draw your trend lines. It
would be unlikely that an additional short period of data
would make much difference.
(Ex 17, para 140).
50. Both Mr Minty and Mr Marshall reject this version of events (see Minty, Ex 51,
para 18; T 806.5-.19; Marshall, Ex 54 para 10 (substituting “140” for “138”)).
Mr Minty’s evidence was as follows:
“In order for us to update the March 2000 position, we’re going to need
James Hardie’s up-to-date claims information including the number of
claims reported and settled together with settlement amounts. We would
also like to get any other information you have in relation to emerging
trends nationally for ARD claims, including new legal precedents. We
have the national experience from the November presentation, but we
need to see how your data relates to that in order to calibrate the national
model to reflect James Hardie’s own claims.”
In response, either Mr Shafron or Mr Attrill (I cannot now recall which of them)
said words to the following effect:
“We’re unable to get updated claims data to you in the time available.”
Mr Shafron said words to the following effect:
“Anyway, we don’t think there’s anything in the data that would affect the
results because nothing significantly different from what you projected has
occurred during the period. We want you to proceed on the basis of the
claims data you had from us and the assumptions you used in the March
2000 report, just updated with your current views on likely national
(Ex 50, para 31).
51. Mr Attrill’s evidence of the meeting on the 19th did not shed any light on this
question. His recollection was largely confined to his note of the meeting, which was
not specific in this respect (Ex 57, vol 4, pp 970-971). However he gave other
evidence that tends to support Mr Minty’s account. His evidence was that on 16
January 2001 he had a conversation with Mr Shafron as follows:
PJS: “Well Wayne, as you know from yesterday’s meeting the Trustees will require
an updated Trowbridge report.”
WJA: “Okay, do you need a fresh data dump for Trowbridge?”
PJS: “No that is not necessary. We will run with the data from the March 2000
report, except that we will ask Trowbridge to update the future claim numbers
having regard to their presentation in November. We are going to need the
report pretty quickly. Can you call David Minty and see how long it’s likely
(Ex 57, para 105).
52. He said that he particularly recalled the request, as it was unusual (T1009 .6-.17). In
oral evidence he said that Mr Shafron explained that he was concerned about delay
53. The conversation is significant because it sits ill with Mr Shafron’s account of the
conversation on the 19th – if Mr Shafron had on the 16th decided he wanted the report
done by reference to March 2000 data it is difficult to see why he would raise the
issue afresh on the 19th.
54. There is reason to accept Mr Attrill’s evidence of his conversation with Mr Shafron
on the 16th. While Mr Shafron did not recall the conversation, he did not deny it
(T 1705.42-1707.30). More importantly it is corroborated by Mr Attrill’s note of a
conversation with Mr Minty on the 16th January 2001 in which he records this
“Can run with data we have”.
(Ex 57, v 4, p 956).
55. Mr Attrill’s evidence was that this was likely to have been Mr Minty’s response to Mr
Attrill saying words to this effect:
“We need a report done quickly. Peter Shafron has asked whether you could do it
on the following basis – run with the March 2000 data, [and] update with the
work done by Watson and Hurst?”
(T 1133.53 – 1134.2).
56. The tension with Mr Attrill’s evidence also affects Mr Morley’s evidence of the
meeting on the 19th. He recalls this exchange:
Mr Shafron: Could you update your June 2000 report to take account of Watson
and Hurst. Do we need the latest claims data to do that?
Mr Minty: I’ve got ten years of data. A couple of quarters won’t make that much
Mr Shafron: The data is pretty much in line with what you have looked at before.
Could we have a revised report as soon as possible covering 20 years?
(Ex 121, para 203).
57. It will be observed that this evidence provides some support for both the other
versions of the conversation. However, it permits a confident acceptance of the
evidence of Mr Minty that on the 19th he was told by Mr Shafron that:
“we don’t think there’s anything in the [current] data that would affect the results
because nothing significantly different from what you projected has occurred
during the period.”
(Ex 50, para 31; see also the evidence of Mr Shafron at T 1709.45 - .50; 1713.27 -
58. Such a statement is to be expected as an explanation by Mr Shafron as to why
Trowbridge should feel comfortable proceeding without the current data. Such an
explanation is likely to be called for only by a perception that Trowbridge would
otherwise prefer to do the report by reference to Current Data.
59. In addition Mr Minty’s evidence is supported by evidence of his statements both
before and after January 19, 2001. First, there is evidence that, as at December 2000,
Mr Minty regarded it as preferable to have the next 6-9 months data for the purposes
of updating James Hardie’s assessment to account for the new insights deriving from
Watson and Hurst’s paper. In an email to Mr Marshall on 4 December 2000 he
recounted a conversation that morning with Mr Attrill:
“He asked if we could put the revised projected numbers into their valuation
model and let him know the impact. I said I would not like to do that without also
reviewing the proportion of claims that we ultimately expect them to incur as a
proportion of total Australian numbers since the result would also be sensitive to
that. I said it would be a largely proportional increase if only the underlying
model numbers changed and nothing else.”
(Minty, Ex 50, Tab 11).
60. Mr Minty’s record of this conversation is corroborated by Mr Attrill’s email to
Mr Shafron of 4 December 2000, which records Mr Minty saying that he “would
want to look at the proportion of claims JHC actually receives (updated from
31 March 2000) as compared to the new claim projections” (Ex 57, v 4, p 801).
61. Secondly, when Mr Minty and Mr Marshall attended a meeting of the Board of
Amaca on 6 August 2001 they informed the Board that a request for the most current
claims data (as at January 2001) was rejected by James Hardie. (Ex 7, MRCF 2,
62. In addition, there is an inherent plausibility in Mr Minty’s evidence that “as a normal
part of our work of this nature we like to have information which is … as up-to-date
as possible.” (T. 813.20 - .23). Mr Marshall confirmed this (T 871.10-.19).
Moreover, the particular task Trowbridge was asked to undertake made a request for
current data material. Part of the exercise was to “calibrate” James Hardie’s
experience to the chosen model (Marshall T 882.7 - .22). For that purpose a current
picture of the James Hardie’s experience was necessary, in order to “fit” it to the
picture Watson and Hurst derived from the national experience (Marshall T 871.20 -
.26). Indeed, it might affect the choice of an appropriate curve – Berry medium, or
Berry high (see Ex 57, v 4, p 971; and Ex 3, v3, p 470). Mr Shafron understood this
63. On the other hand, some matters may be pointed to in support of Mr Shafron’s
version. First, in cross-examination by counsel for James Hardie, Mr Minty said
(T 796.5 – 13):
< 5> Q. And I suggest to you that in January and February
< 6> 2001, you didn't ask for this sort of information because
< 7> you believed that you had sufficient information to
< 8> establish trends to which you could apply the revised
< 9> Berry medium and Berry high curves--
<10> A. We had--
<12> Q. --do you agree with that?
<13> A. I agree with that statement.
64. However, this exchange suffers the difficulty that the question was double-barrelled,
and the answer can be read as a response only to the issue of his belief as to the
sufficiency of his information for the purpose of applying the new curves. Mr
Minty’s evidence remained that he had asked for Current Data (see T 813.12 - .23).
65. Secondly, according to Mr McGregor, at the JHIL Board meeting on 15
February 2001 there was discussion about the higher claims figures disclosed by the
1st February 2001 asbestos report (Ex 2, v 3, pp 555-558) and management
(Messrs Macdonald, Shafron and Morley) were asked whether this experience had
any impact on the Trowbridge numbers based on data to March 2000. He gave this
evidence (at T 1442):
<51> And we also were aware by our own experience in the
<52> December quarter, or the quarters up to and including the
<53> December quarter, and we asked what was the effect of
<54> those increased figures, and the Trowbridge answer, as
<55> conveyed to us, was that they said that one quarter did
<56> not interrupt a long-term trend.
(See also McGregor, Ex 80, para 29).
66. This might be seen as some corroboration for Mr Shafron’s account. The evidence
has some difficulties however. The tenor of it is that Trowbridge had been asked for
its view of the implications of the increase in claim numbers and payments. Though
this is what one might have expected to have occurred, on Mr Shafron’s evidence it
had not. Ultimately, it seemed that Mr McGregor’s evidence on this topic was not
particularly reliable. At T 1445.53 - .57 and again at T 1446.26 - .30 he clearly
conveyed that Trowbridge had been asked about the significance of the increase, but
when asked squarely to confirm this, he retreated (T 1446.56 – 1447.14).
67. Moreover, Mr McGregor’s evidence highlights the implausibility of Mr Minty saying
something to the effect that a short period of data would not affect a long-term trend.
In the case of James Hardie, Mr Minty had noted a recent levelling of mesothelioma
claims (Ex 57, v 4, p971). This, in its terms, is not a long-term trend. If there was a
long-term trend, it was one of increasing claim numbers (see table B.3 in the 2000
Trowbridge Report, Ex 2, v 4, p 900). In truth the situation for James Hardie in
late 2000 seemed poised in the balance – new data might tend to confirm that the old
trend of increasing claims had stopped, and claims were indeed levelling; or it might
suggest that the recent levelling was merely a temporary respite and the underlying
trend remained one of increasing numbers of mesothelioma claims. These
considerations suggest the inherent likelihood of Trowbridge expressing a preference
for the availability of current data.
68. Another possibility is that Mr Minty did say something to Mr Shafron to the effect
that six months or so of data would not make much difference to the result but that he
said this in a conversation in late November or early December. This possibility is
suggested by the wording of Mr Shafron’s memo to Mr Macdonald of
9 November 2001 (Ex 85, p 3; put to Mr Shafron in cross-examination at T 1716).
The difficulty however is that there is no evidence to support such a finding save the
language of this letter written nearly a year after the event.
69. A further possibility which should be noted is that Mr Shafron was confusing the
relevant conversation with something Mr Attrill had reported to him on
18 April 2000. By email of that day Mr Attrill reported to Mr Shafron the results of a
discussion with Mr Minty as to why his preliminary estimate of the liabilities was so
much higher ($300 – 350 million) than in 1998 ($254 million). Mr Attrill wrote:
“I expected the figure would go up (given the wharf claims and JHC’s increasing
settlement share), but I must says not by this much. I asked David Minty whether,
given the tight timeframe to produce the figure, he deliberately built in a safety
margin, and he said he did.
On the assumptions, he said they have pretty much left the legal costs as they are.
I pointed out that legal costs are down 15% in YEM00 (total YEM00 litigation
costs up 8.5% over the previous years), but he said that a few years’ figures
probably wouldn’t affect the overall result.” (emphasis added).
(Ex 57, v 1, p 106).
70. The notion that a “few years [good litigation costs] figures” probably would not affect
the overall result is sensible enough. However a rejection of the significance of all
current claims information is another matter, particularly where the critical question is
whether mesothelioma claims have in fact levelled, and the current data suggests not.
What was the potential impact of the Current Data?
71. The evidence cited in the issue 5 indicates that the actuaries are agreed that bringing
into account the Current Data would have had a significant impact on the assessment
of James Hardie’s asbestos liabilities in February 2001. Mr Wilkinson (Ex 252,
pp 77-84) cavils with some of the adjustments Trowbridge attributed to the
significance of the Current Data in their letter report of 26 September 2001 (Ex 3, v 3,
t 6). These criticisms are not that the adjustments were inappropriate, but that they
should have been made in any event. Mr Wilkinson seems to accept that an increase
in the total NPV of future claims of $138 million was warranted on Trowbridge’s
figures (Ex 252, para 7.2.3), which implies an increase in the NPV of all claims,
discounted at 7%, from $322.6 million to $460.6 million. In addition, his own
percentage increase in the assessment attributable to allowance for the current data is
not materially different from Trowbridge’s (50% as opposed to 52%. See Ex 252,
p 86, table 7.7).
Why was the Current Data Withheld?
72. Even on his own version of events, it is clear that Mr Shafron preferred that
Trowbridge proceed without reference to the current data. What is less clear is why
he had this preference. His explanation in cross-examination was that he believed,
based on his experience, that provision of Current Data would delay the report
(T 1419.34 - .57; T 1715.49 - .52). This evidence is not inherently implausible. It
may derive some support from Mr Marshall’s evidence that if detailed Current Data
had been provided it might not have been possible for Trowbridge to do their report
in the time required (T 878.25-.40).
73. In addition, Mr Shafron’s evidence of a concern about delay may be supported by the
results of his approach to Trowbridge 10 months before. On about 10 March 2000
Trowbridge was retained by Allens to up date its 1998 report (Ex 57, v 1, p 39, 41).
Later that day Mr Shafron rang Mr Attrill and asked him, it seems, to arrange an oral
briefing for Mr Minty by Mr Attrill and Mr Geoff Watson [presumably
Mr Watson SC], and to ask Mr Minty to have a “reworked number” by 31 March
2000, or at least one week before a Board meeting to occur on 15-16 April 2000 (Ex
57, v 1, p 39; see also Minty, T 793; Attrill T 926-7; Shafron at T 1753-4).
74. Mr Attrill also recalls putting this timetable to Mr Minty, and the effect of Mr Minty’s
response was that it could be done (Ex 57, v 1, p 39). On 16 and 23 March, Mr Attrill
sent Trowbridge disks containing the claims database and other relevant material (Ex
57, v 1, pp 44-91). On 4 April Mr Minty and Mr Attrill discussed timings for a
briefing, and settled on Thursday, 6 April (Ex 57, v 1, p 92).
75. The briefing occurred (Ex 57, v 1, pp 97-99). Further data was provided on 12 April
(Ex 57, v 1, p 100). On Friday, 14 April Trowbridge provided a “first draft”
conclusion that the discounted NPV had increased from $254 million to between
$300 and $350 million (Ex 57, v 1, pp 101-2). Having referred to the main reasons
for the increase, Mr Minty went on:
“Karl Marshall and I still need to recheck all of our figures and review some
aspects of the data so please treat these numbers as indicative only, as we
discussed the other night. The work that Howarths and your litigation team have
undertaken to clean up the individual claim database has made some significant
changes to the nature of the data items with which we have been supplied,
although the database should be much more reliable for further work once we
have come to grips with these.”
76. Ultimately, a draft of the management summary of the report was faxed to Mr Attrill
by Trowbridge on 9 May 2000, assessing the NPV of all claims at $293 million (Ex
57, v 1, pp 121 – 130).
77. However the picture is made less clear by other evidence. Mr Shafron’s letter to Mr
Macdonald of 9 November 2001 (responding to Sir Llew Edward’s letter of
24 September 2001) does not assert that the Current Data was not provided because
of a concern about delay, although it does refer to that circumstance (Ex 85, p 3).
Rather, under the heading “James Hardie Claims not provided”, Mr Shafron said:
“The claims data that Trowbridge produced and the numbers in our modelling
were March 2000 numbers. The report which Trowbridge made available to the
Foundation director, makes clear on its face that the underlying data was March
2000 data. They were the most up-to-date “clean” numbers that we had at that
(Ex 85, p 3).
78. This account is a little odd. If Current Data was not provided because “clean”
numbers were not available, it is not to be expected that Mr Shafron would have
raised the subject with Mr Minty except to say that there were no more recent “clean”
numbers. Moreover, none of the other contemporary evidence refers to the
“cleanliness” of numbers as a matter of relevance (cf Shafron at T1755.36-1756.11).
Mr Shafron’s explanation of what he meant by “clean numbers” (T 1714-1715; T
1755-7) was unpersuasive, and not easily reconciled with the quality of data in fact
available to James Hardie (as revealed by Ex 57, v 4, p 878-911). This suggests that
the reference to this consideration in Ex 85 may have been an ex post facto
rationalisation, perhaps prompted by a recollection of the reference to “cleaning up”
the data in Mr Minty’s email of 14 April 2000. Concern about delay resulting from
provision of Current Data to Trowbridge may also have been merely an excuse.
79. That non-provision of the Current Data was not motivated by genuine concerns about
delay is also suggested by the circumstance that Mr Shafron made no inquiry as to
whether Trowbridge could make use of the Current DatA in the time available (see
T1755.5-.35). If it be assumed that he had a genuine interest in obtaining the best
possible estimate from Trowbridge in the time available (cf T1720.35-.44) this would
have been the natural thing to do, particularly in circumstances where, as will appear,
Mr Shafron must have been conscious that recent data might cast doubt on
Trowbridge’s perception that James Hardie’s mesothelioma experience had levelled
80. If Mr Shafron’s explanations for non-provision of the Current Data are not plausible,
the available inferences are that it was not provided because he perceived that it might
be unhelpful (in the sense of tending to produce a higher assessment by Trowbridge),
or, that he had a concern about delay but was indifferent as to the accuracy of the
ISSUE 6: The Evidence as to the Incoming Directors’ Understanding
There is evidence that the incoming MRCF directors did not appreciate that the February Report and
the cash flow model produced by Mr Harman of JHIL did not employ the Current Data for the “best
estimate” and “high” iterations of the model? If this is correct, should they on or before 15 February
2001 have appreciated this was the case, having regard to:-
i) the language and content of the February Report, including the language of the 15 February
version of the report;
ii) the terms of the cash flow models they saw;
iii) what the incoming MRCF directors were told by Trowbridge, JHIL representatives and their
iv) documents seen by them; and
v) any other circumstances?
81. For the reasons that follow, it is submitted that the question raised by issue 6 should
be answered in the negative.
82. First, it seems clear that while there was some reference to use of March 2000 data in
the presentation to the incoming directors on 13 February 2001, the specific
references were only to the use of that data as regards the “Current” model described
by Trowbridge. This may be seen in Mr Cooper’s copy of the February Report
discussed at that meeting– he has noted “March 2000” opposite a bar for the Current
model (Ex 7, MRCF 1, p 0308). Similarly, on his copy of the 9 February 2001
“Current projections” cashflow, Mr Gill wrote “March 2000” (Ex 29, p 70). He
wrote no equivalent notes on the cashflows for the other two scenarios. However
they are described in such a fashion as to suggest that they were based on the latest
“assumes High Scenario (updated 8 February) projections”
“assumes Best Estimate (updated 8 February) scenario projections”.
(Ex 29, pp 71-2).
83. These descriptions were apt to mislead.
84. Secondly, it seems that there was no clear statement to the directors that all the
scenarios were based on 10 ½ month old data. This is consistent with Mr Minty’s
evidence (Minty, Ex 50, para 44) and Mr Marshall’s evidence (Marshall, Ex 54,
para.23). Indeed, Mr Minty left the meeting concerned that the point might not have
been clear, and proposed an amendment to the February Report to clarify the position
(Marshall, Ex 54, para 25).
85. Thirdly, the February Report, as distributed and discussed on 13 February, suggested,
if anything, that it was based on up-to-date data. In his statement (Ex 5, para 82)
Mr Cooper says:
“When I read the Draft February Report I noted the opening paragraphs which
“We refer to your letter dated 30 January 2001 on the above subject. You
have asked us to revisit the claim number assumptions that we adopted for
our draft advice on the future costs of asbestos-related disease claims in
view of recent work that Trowbridge Consulting have carried out to
estimate the impact of such claims on the insurance industry.” (emphasis
I was not shown a copy of the 30 January letter from Allen Allen & Hemsley to
Trowbridge and I did not ask to see a copy of that letter. Nor had I seen the
previous draft advice referred to, though I had asked Mr Attrill during our 11
January meeting if I could be provided with previous Trowbridge reports.
Nevertheless, I placed emphasis on the words I have underlined in forming the
view that Trowbridge had relied upon the most update information available in
preparing the report that was provided to us during this meeting. Neither the
report nor Mr Minty’s briefing gave me reason to believe otherwise.”
86. However, even the ultimate version of the February Report was less than perfectly
clear in this respect. The changes appear “marked up” in Ex 50, t 21 as follows:
“We refer to your letter dated 30 January 2001 on the above subject. You have
asked us to revisit the claim number assumptions that we adopted for our draft
advice on the future cost of asbestos-related disease claims as at 31 March 2000
in view of recent work that Trowbridge Consulting have carried out to estimate
the impact of such claims on the insurance industry in Australia.”
87. It is unlikely any of the incoming directors would have noticed the change of
wording. The revised letter was given to them simply as part of the “Red Folder”, a
large volume of completion documents (see Jollie, Ex 36, paras 82-83, Cooper, Ex 5,
paras 91 and 109 and Gill, Ex 29, para 13). The changes were not marked up and no
one drew attention to the alteration. Even if one of the directors had been astute
enough to notice the change in the version sent on 14 February, they would have been
unlikely to attribute any significance to the change of language, which left in place a
proposition that claim number assumptions as at 31 March 2000 had been revisited in
view of the recent work. What was called for, and did not appear, was a clear
statement that the report was prepared without reference to James Hardie’s claims
data for the period 1 April 2000 to date.
88. Fourthly, consistently with the above, Mr Attrill’s notes of the meeting on 13
February record the following:
MG [Gill]: “Current Model” – March 2000?
DM [Minty]: Yes.
Mr Shafron (at T 1627.18 - .47) was prepared to accept the accuracy of this note in
preference to para 183 of his statement (Ex 17).
89. Fifthly, Mr Minty accepted a proposition put to him by Mr Meagher SC that at that
meeting Mr Minty said that the numbers which he was presenting as his best estimate
were based on the experience and the best information that he had available (T 805.5
- .9) – a statement which called for explicit qualification if the available information
was incomplete or inadequate.
90. Finally, it is perhaps noteworthy that in Mr Shafron’s memo to Mr Macdonald dated
9 November 2001 (Ex 85) he did not suggest that the incoming directors had been
told that all the scenarios were based on March 2000 data. He did say that the
directors were given, by means of Mr Attrill’s presentation on 15 January,
information as to claims up to 31 December 2000. That, however, is not to the point.
The directors may have appreciated that the 2000 data showed a deterioration in
claims. What they were not to expect was that Trowbridge had not taken this into
account in the February Report.
ISSUE 7: Excluded Claims
The February Report does not appear to have:-
(a) allowed for US litigation costs and settlements, remediation claims, exemplary and punitive
damages, Dust Diseases Board reimbursement claims, and the risk of change in the medical or
(b) made full allowance for so-called “third wave” or “downstream” claims;
(c) provided for further increases in propensity to sue.
To the extent that such matters were not allowed for:-
(d) Having regard to the actual purpose for which the February Report was intended by JHIL to be
used, and was in fact used, should allowance have been made for such matters?
(e) Would allowance for such matters have made any difference of significance to an assessment in
February 2001 of the liabilities of Coy and Jsekarb or the likely longevity of the MRCF?
91. Like the 2000 Trowbridge report, the February Report made no allowance for types
of claim for which there was no established claims history. Some of these exclusions
were made explicit in the presentation on 13 February (though not in the report itself).
Dust Diseases Board reimbursement
Overseas claims arising from product export (including US claims)
Any changes in the legal or medical environment.
See Ex 57, v 4, p 1058 (Attrill); Ex 29, p 68 (Gill), and compare Ex 50, p 29 and Ex
57, v 2, p 307 (2 June Draft of the 2000 Trowbridge report).
92. Like the 2003 reports, the February Report made no, or no adequate allowance for
third wave claims, and no allowance for potential increases in propensity to claim.
93. As will appear from evidence to be mentioned in respect of subsequent issues, it
should be found that the Board of JHIL was relying on the claim forecasts produced
by Trowbridge in order to assess whether there was Adequate Funding of future
claims (in the sense described in relation to issues 1 and 4). The same applies to the
outgoing and incoming directors and Coy and Jsekarb, subject to the qualification that
the interest of the latter in the “life of the fund” may have been satisfied by
confidence in 20 or so years of life before insolvency.
94. The evidence dealt with in relation to issues 1 and 4 is applicable to the February
Report, so far as the need for allowance for excluded risks is concerned. That is,
given the purpose of assessing the ability of the Fund to pay all claims, these matters
should have been allowed for.
95. The effect of allowing for them cannot be considered separately from the question of
the overall effect of assessing the liabilities on an Adequate Funding basis.
According to Mr Marshall, funding to a high degree of certainty would be likely to
increase the assessment by 300 – 400% (T 918.50 – 919.36). By reference to the
February Report this would imply a range of $966 - $1,288m.
96. Perhaps a better guide, although still rough, is to scale up the 2000 Trowbridge
sensitivity table for the NPV increase between March 2000 and February 2001 ($294
m to $322 m; a 10% increase):
Base Case: $322 m
Add: High Claim Numbers $126 m
Add: High Avg. Claim Size $ 43 m
Add: High Claim Inflation $130 m
Add: Low Discount Rate $ 51 m
(Compare Trowbridge at Ex 2, v 4, p 892 and Wilkinson at T 3397-8, 3399 – 3400).
The upper end of Mr Wilkinson’s reasonable range for the 2001 estimate is $825 m
97. Needless to say, such assessments would have made a very substantial difference to
an assessment in February 2001 of the likely longevity of the MRCF. Mr Harman’s
cashflow model established that the Foundation would not be able to satisfy all claims
unless the total liabilities were no greater than those assessed by Trowbridge, AND
the fund’s earnings on liquid assets averaged 11.7% p.a. (with no significant adverse
volatility in the early years).
98. Mr Harman had assumed, wrongly, that the Trowbridge Report was comprehensive
as to classes of potential risks (T 1297.43-47) and that the “best estimate” represented
a “substantial likelihood” not merely a 50/50 chance (T 1298.38 - .51).
ISSUE 8: James Hardie’s Mesothelioma Experience
Evidence has been given that at the meeting on 19 January 2001, Mr Minty said words to the effect that
James Hardies’ mesothelioma numbers had levelled off recently.
(a) Was this an accurate statement of James Hardie’s mesothelioma claims experience as at
(b) If not:-
i) Were the executives of JHIL who were present on 19 January 2001 aware of this?
ii) Ought they have been so aware?
iii) Did they at any time communicate an accurate statement of the current position to
iv) If not, why not?
v) Was the omission deliberate?
99. The 2000 Trowbridge report (Ex 2, v 4, t 14) recorded past and projected future
mesothelioma claims and events for James Hardie is set out on the following page
(table B.3, p 900).
100. The first column (Number of Events) of Table B.3 shows an upward trend,
continuing in YEM 2000. The second column (Number of Claims) shows an
upward trend to YEM 1998, then a plateau (the YEM 2000 claims number was
recognised as possibly too low because of a time lag between events and claims -
see T 3322.29-3324.6). The fourth column projects claims (as opposed to events)
continuing at the 1999 – 2000 level until YEM 2006. Then there is a decline, as
appears from figure 4.1 on p 865.
101. Table B.3 and figure 4.1 reveal Trowbridge’s assessment as at June 2000 that the
number of mesothelioma claims against James Hardie had peaked. This perception
was referred to by Mr Minty at the meeting with Messrs Shafron, Morley and Attrill
on 19 January 2001. Mr Attrill’s note records him, in the context of a discussion of
the impact of Watson and Hurst on previous estimates, saying that:
J[ames] H[ardie] meso[thelioma] numbers levelled off recently – so Berry
medium may be the best line for JH. JH may have had different exposure to the
rest of the community.
102. Two things must have been obvious to those present as a result of this statement.
The first is that Mr Minty’s understanding was that James Hardie’s mesothelioma
claims were no longer increasing. The second was that this fact was of particular
significance in forming a view as to James Hardie’s future liability, in light of
Watson & Hurst, because it might impact on the choice of model or curve
appropriate to James Hardie.
103. In the circumstances, it must also have been obvious that if the Current Data
suggested that James Hardie’s mesothelioma claims had not yet peaked, then to
permit Trowbridge to proceed to update their report without knowledge of the
Current Data would be to permit them to proceed on a false premise.
Did James Hardie’s Current Data suggest that its mesothelioma claims had not
104. The evidence only permits an affirmative answer to this question. The actuarial
response to the current data is unequivocal (see the discussion in relation to
Were the executives of James Hardie who were present on 19 January aware of
this? Ought they have been so aware?
105. It is convenient to consider these issues together.
106. Periodically Mr Attrill’s department had to participate in an operating performance
review (“OPR”). This involved the presentation by Mr Attrill of a written report to
a sub-committee of the General Management Team (“GMT” – including Messrs
Macdonald, Shafron, Baxter, Ashe and Morley) comparing the results for the period
with the objectives outlined in the then current plan. The Asbestos Liabilities
Management Plan to YEM 03, as at 24 May 2000, is Ex 57, v 1, pp 1 – 38.
107. On or about 29 November 2000 Mr Attrill obtained from an employee in Mr
Morley’s section data that would enable him to estimate claims costs to the end of
YEM 2001 (see Ex 61, v 5, t 12; T 1162-3). His estimate took into account
malignancy claims notified as at that date which were likely to resolve in the fourth
quarter (see Ex 61, v 5, p 42), and a $2 million allowance for new claims. The total
(leaving aside proceeds of the QBE settlement) was $30.2 million, representing a
50% increase on the previous year (see Ex 75, PJS 1, v 7, p 2605). The estimate
was accurate – the reported total on 11 April 2001 was $31.69million (see Ex 7,
MRCF 1, v 2, p 0379).
108. On 13 December 2000, Mr Attrill’s written report was discussed in an OPR
telephone conference. His note of the meeting has Mr Morley (apparently)
outlining the numbers:
2001 $24.1m (crediting $6.1m from QBE)”.
This adopted Mr Attrill’s estimate of 29 November – that 2001 was expected to
be a $30 million year as contrasted with less than $20 million experienced in
109. The report itself (Ex 57, v 4, pp 803 – 844) records that:
Cost of settlements and damages awards was 25% higher than forecast and
56% higher then the same period in YEM 99.
This was due to a 58% increase in the number of claims settled.
New claim numbers (126) were up 70% on the previous period, and 116 of
these were product or public liability claims.
The greater majority of claims settled were mesothelioma claims.
An increasing proportion of new claims were for mesothelioma contracted by
end users (59 of 64 for HY 01, as opposed to 36 out of 41 in HY 00).
110. Mr Attrill says in the report:
“The increase in claims may be a temporary phenomenon. However, if it
continues it may necessitate a re-rating upwards of James Hardie’s long-
term expected claim numbers and liabilities because: (a) the number of
people who used or were exposed to James Hardie’s AC [Asbestos
Cement] products is very large; (b) lighter exposure to asbestos appears
to be sufficient to cause mesothelioma; (c) lighter exposure is known to
increase disease latency ... which in turn may limit insurance recoveries
… and may extend the peak in claims received.”
(Ex 57, v 4, pp 808-9).
111. Appendix three of the OPR report is headed “Increase in forecast asbestos-related
costs for YEM 01.” A copy, taken from the version at Ex 7, MRCF 1, p 0103, is as
follows (the manuscript markings appear to be those of Mr Shafron noting the prior
year figures mentioned by Mr Morley in the meeting on 13 December: see also
T 108.29 – 110.8):