Exhibit 99.1

James Hardie media release James Hardie Industries NV Media enquiries please contact: Julie Sheather on Tel: 61 2 8274 5206 Mob: 0409 514 643 or Steve Ashe on Tel: 61 2 8274 5246 Mob: 0408 164 011

7 June 2004

James Hardie Releases New Actuarial Review

James Hardie today provided to the Special Commission of Inquiry into the Establishment of the Medical Research and Compensation Foundation (MRCF), and the ASX, an independent actuarial review of future asbestos liabilities of the former James Hardie subsidiaries Amaca and Amaba, now held by the MRCF.

The review was prepared by KPMG Actuaries (KPMG) for James Hardie and ABN 60 Pty Limited (formerly James Hardie Industries Limited) and assesses actuarial advice provided by Trowbridge Consulting between 1996 and 2003.

The report was commissioned to assist the Commissioner to determine the situation facing the MRCF and the former James Hardie subsidiary companies and to provide KPMG’s view of historical and current projections with which to address many of the issues being considered by him.

The figures produced by KPMG highlight “an unforeseeable upward trend” in claims numbers and average claimant costs in recent years.

Based on information that would have been available to Trowbridge at the time it prepared its actuarial report for ABN 60 Pty Limited in February 2001 (and upon which the funding for the MRCF was based), KPMG believes their best estimate would have been in the order of $694m. The equivalent Trowbridge figure was $323m.

The estimate produced by KPMG for future liabilities as at June 2003 is $1,573m which compares to the Trowbridge estimate of $1,090m. The KPMG figures include some $432m in legal costs. Among a number of differences, KPMG has included an additional $356m for superimposed inflation* that was not included by Trowbridge.

James Hardie’s CEO Mr Peter Macdonald said the different figures were extremely concerning to the directors of the company who believed that the funding set aside at the time the Foundation was created would meet the most likely estimate of future anticipated claims.

Directors are considering the implications of the KPMG report.

End.

* Superimposed inflation is inflation above the underlying rate of inflation and is sometimes called judicial inflation. In this sense, it reflects that the rate of increase in court awards can increase at a higher rate than underlying inflation.

Media/Analysts Enquiries:

Julie Sheather – Vice President, Public Affairs
Telephone:  61 2 8274 5206
Mobile:  0409 514 643
Email:  [email protected]

Or

Steve Ashe – Vice President, Investor Relations
Telephone:  61 2 8274 5246
Mobile:  0408 164 011
Email:  [email protected]

Facsimile:  61 2 8274 5218

www.jameshardie.com

Disclaimer

2

	James Hardie Actuarial Expert Witness Report
KPMG Actuaries Pty Ltd	4 June 2004

EXECUTIVE SUMMARY

Scope and Purpose

I have been asked by Allens Arthur Robinson to provide an expert witness submission to the Special Commission of Inquiry into the Establishment of the Medical Research and Compensation Foundation (“MRCF”).

I have been asked to consider the appropriateness of the advice given by Trowbridge Consulting and the nature of the actuarial assumptions they made. I have also been asked to provide an actuarial reassessment of the asbestos-related liabilities as at 31 March 2000, 13 February 2001 and 30 June 2003.

My assessment considerations have predominantly focussed on the period 2000 to 2003. I have only considered the valuations at 1996 and 1998 for the purpose of performing comparisons with the valuation at 31 March 2000.

Background

Over the years 1996-2001, James Hardie Industries Ltd (“James Hardie”) received advice from Trowbridge Consulting, the actuarial subsidiary of Deloitte Touche Tohmatsu (“Trowbridge”), in relation to the level of asbestos liabilities to which it was potentially exposed.

The MRCF subsequently received actuarial advice from Trowbridge in the period 2001 to 2003.

Overview and Key Observations on 2000 Valuation and 2001 Update

It is my opinion that in three key respects the work of Trowbridge for the 31 March 2000 draft valuation and the 13 February 2001 update advice falls outside of the bounds of what would be regarded as reasonable actuarial advice.

The three key areas are:

§		The peak and duration of the future number of claims notifications;
§		The rate of superimposed inflation on claim amounts assumed; and
§		The rate of future nil settlements assumed.

I do not agree with the Trowbridge view in 2000/2001 that the peak in the number of reported mesothelioma claims had already been reached by the year 2000.

	James Hardie Actuarial Expert Witness Report
KPMG Actuaries Pty Ltd	4 June 2004

Information available in the market at the time, and epidemiological views, suggested that this was not so at a national level nor would it have been reasonable in respect of James Hardie, taking into account the nature and incidence of James Hardie’s exposures. The nature of exposures within Australia and for James Hardie, as apparent in 2000/2001, indicate to me that the peak in reported claims would not be likely to be reached until at least 2010. In these circumstances, I would consider that an assumed peak of at least 2005 would be at the lower end of a reasonable range.

Trowbridge formed a view in 2000/2001 that there was no need to allow for future superimposed inflation. It is my assessment that this was at the lower bound of the range of assumptions made by Australian actuaries for this factor at that time.

Public Liability, Workers Compensation and CTP portfolios had all exhibited significant levels of superimposed inflation during the period from the mid 1990s. General market practice and assumptions were allowing for some rate of superimposed inflation (although not necessarily at the same level as those historically exhibited).

Within the provisioning of asbestos-related liabilities, I was, and my clients were, making some allowance for future superimposed inflation at that time.

I do not agree with the Trowbridge view in 2000/2001 that the rate of nil settlements for mesothelioma claims would be 25% in future. In my opinion, their own analyses did not support this view but instead supported a rate of between 15% and 20%.

As a consequence of these three key aspects, it is my view that the assessment of the liabilities by Trowbridge set out in the draft 31 March 2000 report and the 13 February 2001 update letter did not reflect a reasonable actuarial assessment.

Overview and Key Observations on 2003 Valuation

As requested, I have performed an independent valuation of the MRCF asbestos liabilities as at 30 June 2003. This assessment is materially different to that made by Trowbridge as at the same date. The key differences between our respective assessments are:

§		Trowbridge, in 2003, continued to assume that future superimposed inflation would remain zero. Given the continuing trend of actual emerging superimposed inflation and my understanding of general actuarial practice in 2003, it is my assessment that this is outside the range of reasonable actuarial assumptions at that time.
§		Trowbridge, in 2003, were adopting assumptions in the areas of the number of future claims (in terms of the peak and duration of the reporting pattern), the average costs of claims and the proportion of nil settlements that are, overall, optimistic relative to my assessment.

	James Hardie Actuarial Expert Witness Report
KPMG Actuaries Pty Ltd	4 June 2004

Summary of Liability Assessments

The table below provides a summary of the provisions recommended by Trowbridge to James Hardie, or the MRCF, and those assessed by me.

Table E.1: Comparison of reserves

*This is the full cashflow. The 20-year cashflow was $286.5m. The figures above make no allowance for the present value of the QBE settlement of $3.1m p.a. for 15 years from 2000.

It should be noted that the provision at June 2003 could be affected by the following factors:

§		The emerging experience in relation to reported claims and average costs in the period since June 2003;
§		Changes in the yields available on Commonwealth Government bonds which would affect the discount rates selected; and
§		Impact of any exceptional judicial or epidemiological developments in the period since June 2003.

Extent of Deterioration of Asbestos-Related Liability Provisions of James Hardie

It is apparent from the table above that the assessments of the liabilities by Trowbridge resulted in significant revisions upwards.

It is apparent from the above table that my own assessments of liabilities have also shown a significant trend upwards. That said, the extent of my increase is (in proportionate terms) less than that resulting from Trowbridge’s assessments.

	James Hardie Actuarial Expert Witness Report
KPMG Actuaries Pty Ltd	4 June 2004

The emerging experience in the numbers of claims in which James Hardie (or Amaca / Amaba) have been co-joined as a defendant have shown an unforeseeable upward trend. This upward trend has also, in my experience of those markets, been reflected in the UK and US where similar unpredictable trends have emerged, in spite of the best attempts of actuaries to predict the development of notifications. This is also true for the epidemiological studies which, whilst the peak period of incidence has not changed substantially, had not foreseen the absolute level of emergence of claims.

Furthermore, the average claimant costs (including claimants’ legal costs) have shown an unforeseeable upward trend. This is in part due to inflationary pressures but is also likely in part to be a consequence of an increase in the percentage of the total claim borne by James Hardie through co-sharing, which may be a consequence of what is often termed ‘deep pocket syndrome’.

Part of the deterioration has also resulted from the change in the discount rate to be applied to cashflows. At March 2000 and February 2001, a discount rate of 7% per annum was selected, whilst in June 2003 a rate of broadly 5% per annum was selected. The impact of this change is that if I had re-valued the June 2003 liabilities at 7% per annum the liability assessment would have been $1,285.2m, i.e. the change in the discount rate has contributed $288.2m.

Based on my assessment at February 2001 of a liability of $694.2m, the liability at June 2003 would have increased to $752.8m at June 2003. The increase of $58.6m relates to the fact that the unwinding of the discount, at around $45m p.a., exceeds the annual payments in the early years so that the provision would be expected to increase.

Table E.2 identifies the sources of the change in the assessment of liabilities from $694.2m to $1573.4m.

Table E.2: Analysis of change in KPMG liability assessment:
February 2001 to June 2003

	James Hardie Actuarial Expert Witness Report
KPMG Actuaries Pty Ltd	4 June 2004

Within my assessments I have assumed that insurance recoveries will be made at a rate of recovery consistent with the gross claim payments. I do, however, note that in my experience within the insurance and reinsurance markets, there appears to have been little activity with regards to recoveries on the Amaca / Amaba insurance contracts in the last few years. As such, if this deferral of recoveries continues it may cause additional cashflow strains to be borne by the fund, potentially unnecessarily so.

Survival Ratios

One often-used benchmark of the adequacy of an asbestos liability assessment is the “survival ratio”.

The survival ratio is a ratio of the discounted liability amount to the annual rate of payments being made, where the annual payments rate can be assessed based on the most recent year or based on the average of the last three years (say). This ratio gives a crude measure of how long the fund would last at the current, or recent, rate of disbursements (ignoring interest effects).

The ratio provides a useful benchmark for comparing different insurance companies’ liability assessments, and also for a particular company over time to gauge the extent to which the liability assessment remains consistent with the emerging experience.

It should, however, be noted that this is only a high-level indicator, and that between companies and time periods, comparisons need to take into account differences in:

§		Timing of exposure;
§		Maturity of claims;
§		Payment/settlement patterns; and
§		Trends in other elements and percentage sharing of claims.

The graph below shows how the survival ratio, based on the actuarial assessment of the liabilities as recommended by Trowbridge, has changed over time. The figures in the chart are net of insurance which James Hardie had placed to cover various elements of the portfolio. The ratios are shown based on 1-year payments and 3-year average payments.

The considerable difference that has emerged between the 1 year and 3 year average ratios in the last three years reflects the significant increase in the rate of payments over this period. Gross payments have increased by 135% from $22.8m in 2000 to $53.6m in 2003 — the current payment rate is well above the last 3-year average.

	James Hardie Actuarial Expert Witness Report
KPMG Actuaries Pty Ltd	4 June 2004

Figure E.1: Trends in implied survival ratios for James Hardie
using Trowbridge recommendations

30.00 25.00 20.00 15.00 10.00 5.00 0.00 Jan-96 Jan-97 Jan-98 Jan-99 Jan-00 Jan-01 Jan-02 Jan-03 Jan-04 Survival Ratio, based on average of 1 year Survival Ratio, based on average of 3 years

*The figure for February 2001 is based on the all-years cashflow provision of $322.6m rather than the figure that was adopted (being a 20-year cashflow of $286.5m). The survival ratio based on the curtated cashflows would be 9.9 and 13.1 respectively.

In my experience in the Australian market, survival ratios of between 20 and 35 are the norm and anything outside that range would require consideration of the nature of the exposures. Even then, where the company has a survival ratio in that range, it does not necessarily imply the liabilities are adequate.

As stated already, it is also important not just to consider the absolute level of the survival ratio but the trend in the survival ratio. It can be seen that there was a systematic reduction in the survival ratio for the portfolio up to February 2001, followed by an immediate and significant increase back towards a level I would regard as more typical.

Furthermore, consideration should also be given to the maturity of the payments. Companies with relatively immature historic payments might require higher survival ratios than those for whom payments have been relatively more mature, although this in part may be a result of the nature and timing of exposures.

KPMG’s Assessment at 31 March 2000

KPMG’s central estimate of the liabilities at 31 March 2000 is $539.7m. This is based on cashflows being discounted at a rate consistent with ED88. If instead liabilities were discounted at a rate more akin to risk-free rates, the assessment of the liabilities would instead be $571.2m.

The table below shows an analysis of variation between the central estimate I have assessed and that assessed by Trowbridge.

	James Hardie Actuarial Expert Witness Report
KPMG Actuaries Pty Ltd	4 June 2004

Table E.3: Analysis of variation of liabilities at March 2000

KPMG’s Assessment at 13 February 2001

KPMG’s central estimate of the liabilities at 13 February 2001 is $694.2m. This is based on cashflows being discounted at a rate consistent with ED88, being approximately 7% per annum.

It should be noted that if a rate more akin to a risk-free rate were used that the liabilities would be assessed as $804.9m, based on a risk-free rate (averaged across durations) of approximately 5.9% per annum.

The table below shows an analysis of variation between KPMG’s central estimate and that assessed by Trowbridge.

	James Hardie Actuarial Expert Witness Report
KPMG Actuaries Pty Ltd	4 June 2004

Table E.4: Analysis of variation of liabilities at February 2001

KPMG’s Assessment at 30 June 2003

KPMG’s central estimate of the liabilities at 30 June 2003 is $1,573.4m. This is based on cashflows discounted using a yield curve for Commonwealth Bonds at 30 June 2003.

Trowbridge’s assessment of the liabilities was $1089.8m, this being based on a uniform discount rate of 5% per annum. KPMG’s assessment of liabilities at a discount rate of 5% per annum would be restated to $1,627.1m.

In considering this assessment, and the change in assessments set out above, it must be stressed, that as has been exhibited both by James Hardie’s and MRCF’s past development, and that of the Australian and worldwide market more generally, the assessment of asbestos liabilities is subject to considerable uncertainty and actuarial judgement at any given point in time. This is in part a consequence of the extremely long-tailed nature of the liabilities and also a reflection of the extent to which the liabilities are significantly influenced by factors such as legal developments and epidemiological advancements.

As such, it should be noted that the assessment of $1,573.4m above, whilst a best estimate, could be subject to further as-yet-unforeseeable development.

The table below shows an analysis of variation between KPMG’s central estimate and that assessed by Trowbridge.

	James Hardie Actuarial Expert Witness Report
KPMG Actuaries Pty Ltd	4 June 2004

Table E.5: Analysis of variation of liabilities at June 2003

Market Overview

As I have already noted, it can clearly be seen that both within Trowbridge’s valuation and in my assessments that the liabilities have shown continual increases in reserves, albeit to different degrees.

It should be recognised that provisioning of asbestos-related liabilities has proven problematic for actuaries and insurance companies across the world for the last five years or more.

The problem of the regular need for further increases in the levels of provisions set aside has not been restricted to Australia, and certainly not to James Hardie or the MRCF. Both the United States and Equitas have experienced some significant deterioration in their asbestos-related liabilities in the last five years.

In the United States, the net reserves and net ultimate losses have shown a significant upward trend between the end of 2000 and the end of 2002. As can be seen in the chart below, the net reserves in the United States almost doubled from $10.5bn to $20.9bn in the 2-year period. The net ultimate cost of claims increased by more than 40% in the same period from $32.1bn to $45.5bn.

It is also noted that the trend of further increases in reserves continued in 2003, with the following companies announcing increases in provisions amounting to more than $6bn:

§		Travelers adding $2.6bn (14/1/03),
§		Hartford adding $2.6bn (12/5/03),
§		Allstate adding $0.5bn (Q2 and Q3),
§		ACE adding $0.3bn (27/1/03), and
§		Liberty Mutual adding $0.3bn (Q3).

	James Hardie Actuarial Expert Witness Report
KPMG Actuaries Pty Ltd	4 June 2004

Figure E.2: Net Asbestos Liabilities – US Insurance Industry

50.0 45.0 40.0 35.0 Billion 30.0 25.0 US$ 20.0 15.0 10.0 5.0 - - 1997 1998 1999 2000 2001 2002 Net Incurred in the Year Net Paid to Date Net Reserve Net Ultimate Loss

It should be noted that “unimpaired claimants” have had a material effect on these figures in recent years.

The survival ratio for US asbestos-related liabilities is 14.0 and 14.3 based on 1-year average and 3-year average payments respectively.

Equitas, the reinsurance vehicle set up in 1996 to carry Lloyd’s pre-1993 liabilities, has also been affected by the worldwide adverse developments in asbestos-related liabilities. In 2000 it increased provisions by £1.5bn, in 2001 it increased provisions by £1.7bn, and in 2003 it increased provisions by £0.4bn.

At the end of the 2002 financial year, Equitas’ survival ratio for asbestos liabilities (being the ratio of reserves to annual average payments) was 23.6.

	James Hardie Actuarial Expert Witness Report
KPMG Actuaries Pty Ltd	4 June 2004

CONTENTS

	James Hardie Actuarial Expert Witness Report
KPMG Actuaries Pty Ltd	4 June 2004

Tables

	James Hardie Actuarial Expert Witness Report
KPMG Actuaries Pty Ltd	4 June 2004

Figures

Appendices

	James Hardie — Actuarial Expert Witness Report
KPMG Actuaries Pty Ltd	4 June 2004

1.1		Background of the interaction of James Hardie Industries and the Medical Research and Compensation Foundation with Trowbridge
		Over the years 1996-2001, James Hardie Industries Ltd (“James Hardie”) received advice from Trowbridge Consulting, the actuarial subsidiary of Deloitte Touche Tohmatsu, (“Trowbridge”) in relation to the level of asbestos liabilities it was potentially exposed.
		The Medical Research and Compensation Foundation (“MRCF”) has subsequently received actuarial advice from Trowbridge in the period 2001 to 2003.
1.1.1.		James Hardie: 1996-2001
		Over the period 1996-2001, Trowbridge provided four actuarial assessments of the costs of future asbestos-related disease liabilities of James Hardie to its Board of Directors (the “Board”).
		In October 1996, the first valuation report was provided to James Hardie based on data to 31 March 1996. The actuarially assessed liability was $230.2m. This report was signed by Mr David Minty and Mr Geoff Atkins, both Fellows of the Institute of Actuaries of Australia (“IAAust”).
		In September 1998, a second report was provided which assessed the potential liability based on data to 31 March 1998. The actuarially assessed liability was $253.8m. This report was also signed by Mr Minty and Mr Atkins.
		In June 2000, James Hardie received advice from Trowbridge of the level of liabilities required based on data to March 2000. The actuarially assessed liability was $294.7m.
		The authors were Mr David Minty and Mr Karl Marshall, a Fellow of the Faculty of Actuaries. The report was not finalised or signed and it remains in draft form to date.

	James Hardie — Actuarial Expert Witness Report
KPMG Actuaries Pty Ltd	4 June 2004

		In November 2000, two actuaries employed by Trowbridge presented an update to the market on the development of asbestos liabilities within Australia. This was presented to the Institute of Actuaries of Australia’s 8th Accident Compensation Seminar and the authors were Mr Bruce Watson and Mr Mark Hurst. The report indicated a continuing deterioration in asbestos-related disease claims, especially with respect to the numbers of claim notifications, which far exceeded any modelling previously published in Australia.
		In February 2001, James Hardie received updated advice from Trowbridge. James Hardie requested this advice take into account the developments in the market since the time of their previous advice (June 2000). This analysis was presented in the form of a brief update letter dated 13 February 2001. It showed cashflows for 10, 15 and 20 years. The report showed a discounted liability of $286.5m for 20 years of cashflows (based on a 7% discount rate). The calculations appear to have an effective date as at 31 March 2000. The authors of this letter were Mr Minty and Mr Marshall.
1.1.2.		The Medical Research and Compensation Foundation (MRCF): 2001-2003
		The MRCF was set up on 16 February 2001. The MRCF encompassed the liabilities of the two companies James Hardie & Company Pty Limited (renamed Amaca Pty Ltd) and Jsekarb Pty Limited (renamed Amaba Pty Ltd).
		In August 2001, Trowbridge issued its first valuation report to the MRCF, based on information up to and including 30 June 2001. The authors were Mr Minty and Mr Marshall. The report indicated a central estimate liability of $574.3m. This prima facie appears to indicate that a considerable increase in liabilities had been identified within the 6 months since the 13 February 2001 letter had been presented to the Board of James Hardie.
		Two further valuation reports have since been issued, based on data to 30 June 2002 and 30 June 2003 respectively. Each of these reports has shown further considerable increase in liabilities relative to prior valuations: to $751.9m in June 2002 and to $1089.8m in June 2003. The authors of these reports were Mr Minty and Mr Atkins.
		A chronology of the various Trowbridge presentations to the market and valuation reports is attached in Appendix A.
1.2		Terms of reference
1.2.1.		Who commissioned this expert report
		This report has been commissioned by Allens Arthur Robinson, formerly Allen, Allen & Hemsley (“Allens”), acting as lawyers to James Hardie.

	James Hardie — Actuarial Expert Witness Report
KPMG Actuaries Pty Ltd	4 June 2004

		This report has been prepared as an expert witness submission to the Special Commission of Inquiry into the Establishment of the MRCF.
1.2.2.		Scope of engagement
		I have been requested to consider the nature of the actuarial advice provided by Trowbridge to James Hardie, and subsequently to the MRCF.
		I have been asked to consider the appropriateness of the advice given by Trowbridge, and the nature of the assumptions they made. I have also been asked to consider what a reasonable actuary might have concluded and further to advise on my views (“KPMG’s assessment” or “my assessment”) of the extent of liabilities that should have been determined by Trowbridge, on the basis of a reasonable actuary’s view at each of the key dates: being 31 March 2000, 13 February 2001 and 30 June 2003.
		I have, in the case of the March 2000 and February 2001 valuations, made no use of hindsight, and considered only the data, information and views that were potentially available to the Trowbridge actuaries at the times they provided their advice. I have also considered the impact of extra data on my liability assessment.
		Such sources of information available to Trowbridge informing their views could include epidemiological studies, actuarial reserving bases used in the UK and US (which could have been available from Trowbridge’s associated partner firms), and their own knowledge and research. In my opinion, Trowbridge’s advice, as highlighted in their published papers, should have considered the nature of the techniques and methodologies commonly used to assess asbestos liabilities elsewhere in the world.
		In respect of the 2003 valuation, I have taken account of hindsight information. I note the recent indication that the NSW Government is intending to over-turn a previous legal decision in respect of Orica vs. CGU. This valuation makes no allowance for the impact Orica may have had.
1.2.3.		Areas of work not in scope
		This report does not cover any issues relating to the establishment of the MRCF more widely.
		It is understood that certain cashflows estimated by Trowbridge were applied within a broader “investment model”. This report does not cover that modelling work.

	James Hardie — Actuarial Expert Witness Report
KPMG Actuaries Pty Ltd	4 June 2004

		There have been discussions about potential legal reforms to help balance asbestos and other general liabilities, including steps to create a Proportionate Liability structure or other tort reforms specifically relating to asbestos. It is noted that these developments, which may act to distribute costs differently amongst the joined parties in proportion to their relative contributions. No explicit allowance for this within any of the analysis or assessments within this report.
		No allowance has explicitly been made for any impact of the increased usage of Alimta (Pemetrexed) or other drugs increasingly being used to treat asbestos-related cases. Such treatments may increase the lifetime of malignant pleural mesothelioma sufferers and thereby increase the cost of settlements.
1.2.4.		Accounting standards
		It is noted that Trowbridge’s valuation reports were prepared with due regard for the relevant accounting standards prevailing at the time. Specifically, I note that the relevant draft standard for the March 2000 valuation and the February 2001 update was ED88.
		It is noted that ED88 specifically referred to “discontinuing liabilities at the market yield as at the reporting date on high quality corporate bonds”.
		It is also of note that there was to be no consideration of any margins for risk and uncertainties required by ED88.
		Taking the above into account, my assessment of the liabilities in 2000 and 2001 is a central estimate discounted at a yield commensurate with that achievable on high quality corporate bonds rather than the risk-free rate indicated within actuarial professional standards. My assessment will also make no allowance for a prudential margin.
		The relevant accounting standard for the assessment of liabilities at 30 June 2003 was AASB 1044 (which replaced ED88). This accounting standard specifically refers to “the pre-tax rate that reflects current market assessments of the time value of money and the risks specific to the liability”.
		I have interpreted this as implying the use of a risk-free rate of discount and have used such a rate in my assessment of the liabilities at 30 June 2003.

	James Hardie — Actuarial Expert Witness Report
KPMG Actuaries Pty Ltd	4 June 2004

1.3		KPMG Actuaries and the Curriculum Vitae of Richard C Wilkinson
1.3.1.		KPMG Actuaries
		KPMG Actuaries Pty Ltd (“KPMG Actuaries”) is an actuarial consultancy providing specialist actuarial and financial services advice to the insurance and financial services industries in Australia and South-East Asia.
		KPMG Actuaries, in its current form, was established in 1996. It was first established as Edwards Consulting in 1984 as a life insurance consultancy.
		KPMG Actuaries now consists of a General Insurance practice, a Life & Financial Services practice and a Health Insurance practice.
		KPMG Actuaries’ General Insurance Practice provides valuation and audit services to a diverse client base across the industry and for a wide variety of portfolios. It also provides advice in respect of liquidations, mergers & acquisitions, as well as corporate restructurings.
		KPMG Actuaries is a self-owned practice, owned by the Directors of KPMG Actuaries. It is not a subsidiary of KPMG, but is an affiliate member of the worldwide KPMG network.
1.3.2.		Curriculum Vitae of Richard C Wilkinson
		I, Richard Wilkinson, am a Fellow of the Institute of Actuaries, and have been since 1974. I am also a Fellow of the Institute of Actuaries of Australia.
		I am the General Insurance Practice Leader, and a Director, of KPMG Actuaries in Sydney, Australia.
		I have occupied my current role of Practice Leader of KPMG Actuaries’ General Insurance practice since April 2002. Prior to undertaking this role, I was the Head of KPMG LLP’s international insurance actuarial practice between 1995 and 2002 and was based in London.
		I have worked within the Australian insurance industry since the 1980s, when I first became involved in the liquidation of NEM. NEM had asbestos exposure through its workers compensation portfolio, together with a moderate amount of general and product liability risks. Since that time I have developed extensive first-hand experience of the Australian insurance industry. Through this experience, I have developed extensive knowledge in provisioning for asbestos-related liabilities within the Australian insurance industry both on the direct and reinsurance side.

	James Hardie — Actuarial Expert Witness Report
KPMG Actuaries Pty Ltd	4 June 2004

		Since 1998 I have advised the KPMG Insurance Audit practice in Sydney. Some of their clients have material exposure to asbestos liabilities, within both general and products liability and workers compensation portfolios.
		Additionally, since March 2001, I have been working for the Joint Liquidators of HIH as the Chief Actuarial Advisor to the liquidators. I provided evidence to the HIH Royal Commission in that role.
		In my previous roles, especially with KPMG London, I had developed an extensive exposure to asbestos reserving and methodologies used in the UK, European, US and Australian markets, especially via a number of my clients within:

	§		Lloyd’s of London,
	§		The London Market,
	§		The captive insurance market,
	§		Insolvent companies, and
	§		A large manufacturer which produced an extensive number of products which contained asbestos, being Turner and Newell.

		Many of the exposures and claims I have reviewed emanated from the US. On the reinsurance side from Australia, I was involved with the audit of a company called General Accident Group who had significant exposures to the Australian market. I was also involved when they commuted their considerable exposure with CSR.
		In respect of the insolvent companies for whom I continue to provide advice, their combined liability to asbestos is over US$1.25bn, with the main element coming from the US. I also continue to advise one direct company in the UK that wrote liability business and has a reasonably large exposure to UK asbestos. The emergence of claims for that company is following a similar pattern to that currently being experienced in Australia.
		I was involved in the formation of Equitas, the reinsurance vehicle set up to ring-fence all pre-1993 Lloyd’s liabilities. Equitas was eventually created in 1996. I acted both as the actuarial adviser to the “exceptional cases team” and also advised and reviewed the level of liabilities allocated to the managing agent who had the largest single exposure to asbestos in Lloyd’s for 1992 and prior, this being the Merrit Group.

	James Hardie — Actuarial Expert Witness Report
KPMG Actuaries Pty Ltd	4 June 2004

1.3.3.		Assistance and peer review
		I, Richard Wilkinson, am acting as the expert witness in this matter. I have been assisted considerably by Mr Jefferson Gibbs and Mr Neil Donlevy. Mr Greg Martin has provided a peer review of the analysis undertaken and of this expert witness report.
		Jefferson Gibbs is a Director of KPMG Actuaries and is a Fellow of the Institute of Actuaries and an Accredited Member of the Institute of Actuaries of Australia. Jefferson has extensive experience in asbestos reserving of UK and US liabilities, having worked for KPMG’s general insurance practice in London. Jefferson chaired the Institute of Actuaries’ Employers Liability Working Party in 2000. The Working Party’s paper includes an overview of the analysis of and methodologies used in the evaluation of direct asbestos liabilities. He has developed considerable insight into the issues facing the Australian industry in relation to asbestos-related diseases through work performed with, and for, both KPMG Actuaries and KPMG clients.
		Neil Donlevy is a Senior Manager of KPMG Actuaries and is a Fellow of the Institute of Actuaries and an Accredited Member of the Institute of Actuaries of Australia. Neil has experience valuing US and UK asbestos (and other pollution and health hazard) liabilities underwritten in the London Market by an insurance company for whom he worked. He has developed considerable insight into the issues facing the Australian industry in relation to asbestos-related diseases through work performed with, and for, both KPMG Actuaries and KPMG clients.
		Greg Martin is the Managing Director of KPMG Actuaries. Greg is a Fellow of the Institute of Actuaries of Australia and is the Senior Actuary (as defined under the Code of Conduct) of KPMG Actuaries.
1.4		Professional standards and compliance
		This report is not a valuation report; it is an expert witness report. As such, it is my opinion that the report is not required to comply with IAAust general insurance professional standard PS 300. Nonetheless, I have been cognisant of PS 300 and its implications when preparing the report.
		I also note the obligations I have in complying with the IAAust’s Code of Conduct. In particular, paragraph 9 of the Code of Conduct states “members must realise that there is room for honest differences of opinion on many matters. Members must avoid unjustifiable or improper criticism or malicious injury to the reputation of another member.”

	James Hardie — Actuarial Expert Witness Report
KPMG Actuaries Pty Ltd	4 June 2004

		I am aware of the need to act with integrity in my relationship with other members.
1.5		Data and information available to KPMG Actuaries
		The following information has been provided by Mr Michael Ball, of Allens, and his team:

	§		Trowbridge’s presentation to the 8th Accident Compensation Seminar dated 29 November 2000, written by Bruce Watson and Mark Hurst;
	§		Trowbridge Reports (“Review of Potential Exposure to Asbestos-Related Claims”) to James Hardie, as at March 2000 and 13 February 2001;
	§		Trowbridge Reports (“Review of Potential Exposure to Asbestos-Related Claims”) to MRCF, as at 30 June 2001, 30 June 2002 and 30 June 2003;
	§		Letter of Engagement from Allens to Trowbridge dated 30 January 2001;
	§		Letter from Trowbridge to Amaca dated 29 August 2001 (“Change in Assessment of Potential Asbestos-Related Liabilities”);
	§		Letter from Trowbridge to Amaca dated 29 August 2001 (“Use of Emerging Data in Assessment of Potential Asbestos-Related Liabilities”);
	§		Letter from Trowbridge to Amaca dated 26 September 2001 (“Application of updated Data to Assessment of Potential Asbestos-Related Liabilities”);
	§		Letter from Trowbridge to Amaca dated 20 March 2002 (“Assessment of Potential Asbestos-Related Liabilities”);
	§		Letter from Roy Williams (of Allens) to Wayne Attrill (“Project Green”), dated 23 June 2000;
	§		The Commission of Inquiry’s Issues Paper, issued 25 March 2004;
	§		The Commission of Inquiry’s Revised Issues Paper;
	§		The Expert Statement submitted by Professor B.K. Armstrong;
	§		The Witness Statement submitted by Andrew William Smith (a paralegal of Turner Freeman) to the Inquiry dated 2 April 2004;
	§		A letter written by Steve Ashe, of James Hardie, dated 8 August 2000;
	§		CDs containing various spreadsheets used by Trowbridge in their valuation work between 1998 and 2003;

	James Hardie — Actuarial Expert Witness Report
KPMG Actuaries Pty Ltd	4 June 2004

	§		12 files submitted by Trowbridge to the Commission (including the witness statements of David Minty and Karl Marshall);
	§		Datasets used by Trowbridge in relation to the valuation of MRCF liabilities as at 30 June 2003.

1.6		Reliance and limitations
		I am limited in the extent of advice I am providing. The advice is limited by the availability of the data provided by Allens and the Commission. Amongst the considerable amount of information supplied, access has been given to Trowbridge’s working papers, but completeness of these has not been verified.
		It would be usual for an actuary to seek access to claims handlers and assessors, together with legal advisors and litigators, who were advising on the individual asbestos claims. This would facilitate a better understanding of the key issues driving initial claims estimates and outcomes specific to James Hardie.
		The direct interviews would allow the actuary to understand the way in which the entity recorded its case liabilities and whether these fully reflected the legal advice given. This also helps to establish how practices may have changed over time and how this is reflected in the incurred cost development of those claims in aggregate. The valuations I have performed in order to prepare this report have not had the benefit of this input and have therefore been more ‘desk-based’ valuations than desirable.
		It is emphasised that the projection of the ultimate cost of asbestos liabilities is affected by extremely complex factors and in the large part is based on the projected epidemiological factors which emerge from exposures many years previously, and for which detailed information is not available, and therefore for which the claims exposures to both employees and users of asbestos products are subject to considerable uncertainty. The cost per claim is subject to considerable variation with every case, and is highly sensitive to many factors, including emerging medical and legal developments, which I cannot predict at this stage.
		Medical developments could include the emergence of drugs as successful treatments, either to cure plaintiffs or extend their life.

	James Hardie — Actuarial Expert Witness Report
KPMG Actuaries Pty Ltd	4 June 2004

		Legal developments could include the admission of new sources of claims or identification of increased levels of secondary exposures. They may also include changes to legislation regarding the level of benefits payable to plaintiffs, which could materially increase or decrease the adequacy of the liabilities assessed.
		I have had to make assumptions over the nature of the legal and epidemiological environment prevailing in the future in formulating my assessment of the liabilities at 30 June 2003. I have only allowed for known legal and medical advancements to date, and assume a continuation of these going forward. It is my opinion that this is the approach that a reasonable actuary would make in the circumstances.
		I do, however, caution that as a consequence, the variation of the environment from this assumption could result in the liabilities assessed exceeding or falling short of the ultimate cost of claims.
1.7		Distribution and use
		The purpose of this report is limited by its terms of reference, discussed in section 1.2, and should not be used for any other purpose.
		This report is provided to Allens and it is to be made available to all parties involved in the Special Commission of Inquiry. KPMG Actuaries and I provide our consent for this report to be made available to all such parties.
		The report should only be distributed in its entirety (including appendices) and judgments about the conclusions and comments drawn in this report should only be made after considering the report in its entirety.
		KPMG Actuaries and I will not be liable for the financial consequences of any third party, including those third parties involved in the Special Commission of Inquiry, relying upon information or conclusions contained within this report.

	James Hardie — Actuarial Expert Witness Report
KPMG Actuaries Pty Ltd	4 June 2004

2.1		Overview of relevant standards
		Australian actuaries, as professionals, are required to ensure that their work and advice meet certain standards. These standards seek to ensure that a user of advice receives at least the deemed minimum quality of the work, analysis and advice, and that the advice is based on certain norms (unless clearly stated otherwise).
		There are professional guidance notes and a Code of Conduct in addition to the professional standards.
		The two standards most relevant at the time of the work undertaken by Trowbridge for James Hardie were:

	§		Professional Standard 300 “Actuarial Reports and Advice on Outstanding Claims in General Insurance”, (effective date July 1994) of the Institute of Actuaries of Australia; and
	§		Code of Conduct (effective date April 1998) of the Institute of Actuaries of Australia.

		A copy of each of these is attached in Appendices C and D respectively.
		I provide below an outline to each of these standards and the requirements each bestows upon the advising actuaries.
2.2		Professional Standard 300
2.2.1.		Applicability of PS 300
		PS 300 applies to “actuaries preparing estimates of the liabilities for outstanding claims of general insurance companies, reinsurers, self insurers and providing advice on the liabilities to be set aside to meet those liabilities in the balance sheet...”
		It is my opinion that, in terms of PS 300, James Hardie was a self-insurer. This is based on the fact that James Hardie knowingly self-retained risk exposures below specific insurance covers. James Hardie purchased insurance (which has the characteristics of reinsurance) to manage its retained risk, with the cover in some cases being quite limited. It is also noted that the workers compensation portfolio was almost totally insured with Allianz, via its acquired subsidiary MMI.

	James Hardie — Actuarial Expert Witness Report
KPMG Actuaries Pty Ltd	4 June 2004

		It is my view that the liabilities valued by Trowbridge have every characteristic of the liabilities found in a general insurance company, and as such any actuarial reporting and work performed by Trowbridge was “governed” by PS 300. The relevant versions of the standard were effective as at July 1994 and April 2002.
		Some key elements of PS 300 are summarised below, sourced from the July 1994 standard.
2.2.2.		Procedures
		The actuary should perform the following when advising on outstanding claims liabilities:

	§		Seek written terms of reference and purpose of the advice;
	§		Collect necessary data;
	§		Select a valuation model and assumptions;
	§		Reconcile results with the previous investigation;
	§		Analyse variability and sensitivity;
	§		Formulate conclusions; and
	§		Present a written report.

		The actuary should also consider collecting and analysing additional data.
2.2.3.		Data
		The actuary should ensure that he/she:

	§		Understands the procedures for administering and accounting for the insurer’s claims;
	§		Is conversant with the policy terms/liability exposures, legislated benefits and legal environment, and any reinsurance arrangements;
	§		Is familiar with legal and social trends affecting the liabilities; and
	§		Has sufficient and appropriate data, which is complete and consistent. Reasonable steps must be taken to verify overall consistency of data with financial records.

	James Hardie — Actuarial Expert Witness Report
KPMG Actuaries Pty Ltd	4 June 2004

2.2.4.		Liabilities
		The actuary should subdivide the data into appropriate groupings, having due regard to the credibility of the data once sub-divided, and should then select the most suitable model for each group.
		The actuary needs to select appropriate assumptions and parameters which will have due regard to past experience and make allowance for future trends. The actuary should not spread a change in basis over more than one valuation.
		The actuary should make appropriate allowance for future inflation and also for the rate of discount, which should be set having due regard for the risk-free rate and the assets held by the insurer. Where the discount rate differs from the risk free rate, the actuary should explain the reasoning for adopting a different rate.
		It is the responsibility of the actuary to ensure the calculations are carried out correctly.
		The actuary should analyse the change in results since the last valuation and understand the sources of movements. The actuary should also compare and investigate how the results have changed (especially if significant).
2.2.5.		Uncertainty
		The actuary has a responsibility to explain the nature and consequence of the uncertainty of his estimate and also provide sufficient information to portray the range of reasonable uncertainty.
		This will most often be done by use of sensitivity and scenario analyses.
		The actuary should enunciate clearly the nature of the key components of this uncertainty. This is often illustrated by probabilistic distribution tables to show the range and breadth of possible outcomes.
2.2.6.		Reporting
		The report should state its purpose and/or the terms of reference.
		The report also needs to state if it complies with PS 300 and if not, why not and in what respects it does not comply.
		The report should explicitly cover the nature and accuracy of the data used. It should identify the valuation model and key assumptions and what has changed since the last report.
		It should identify the results, the uncertainty and a comparison of actual experience against that expected at the last valuation.

	James Hardie — Actuarial Expert Witness Report
KPMG Actuaries Pty Ltd	4 June 2004

		The actuary should discuss data verification, limitations on the extent and quality of data and any reliance placed on the auditor for checking.
		The report should state the assumptions and how they have been derived.
		If relevant, any qualifications need to be stated.
2.3		Code of Conduct
2.3.1.		Applicability of the Code of Conduct
		The Code of Conduct covers some general points about the conduct of actuaries, about actuarial advice and about the contents of reports.
		The Code of Conduct is an overarching standard.
		A summary of the relevant components of the Code of Conduct follows.
2.3.2.		Actuarial advice
		The advice given by the actuary should be unbiased.
		Where an area of work is well established, an actuary should not accept an assignment unless they have the relevant practical experience or are to be guided by an actuary of such relevant experience.
2.3.3.		Third party transmission
		If the actuary has reason to believe that their advice will be transmitted in whole or part to a third party, the actuary must take all reasonable steps to ensure that any significant implications of the advice are stated, and that the advice is not presented in a way likely to give a misleading impression.
2.3.4.		Contents of reports
		When providing a report, the actuary should state whether the report complies with the relevant Professional Standard, Guidance Note or Mandatory Guidance Note published by the IAAust (in this case PS 300). If it departs from the standard the report must state this and provide justification for the departure.
2.3.5.		Data
		The actuary’s report should describe or clearly identify the data used and the methods and assumptions used. In particular, the actuary should draw attention to the consequences and implications of using these.
		The actuary should ensure that the advice given is based on sufficient and reliable data and on adequate and appropriate assumptions.

	James Hardie — Actuarial Expert Witness Report
KPMG Actuaries Pty Ltd	4 June 2004

		If the actuary gives advice which does not follow the above requirement then the actuary must qualify their report.
2.4		Definition of a “reasonable actuary”
		In my opinion a reasonable actuary is an actuary with sufficient relevant experience in the area upon which advice is being provided. In the case of asbestos liabilities he or she would have substantial, but not exhaustive, awareness of the various epidemiological and legal developments taking place within Australia affecting the emergence of asbestos related claims, and who has at least broad awareness of the developments and trends in asbestos claims in UK, Europe and US, given the worldwide nature of the problem.
		A reasonable actuary is an actuary who will accord with all material aspects of Professional Standards, Guidance Notes and Codes of Conduct and will offer appropriate advice, recognising the requirements and circumstances of the user of their advice.
		As described in those standards, a reasonable actuary will present unbiased results. A reasonable actuary should complete sufficient checks on data reasonableness to be satisfied that it was sufficient for the actuary’s purposes.
		A reasonable actuary should provide clear communication of the key results, assumptions, sensitivities and issues involved in their advice.
		A reasonable actuary should apply an analysis and methodology consistent with the Actuarial Control Cycle. This is discussed further below.
2.5		Review of experience to date (“actuarial control cycle”)
		One of the key aspects of the work of a reasonable actuary is the use of what the actuarial profession currently calls “the Actuarial Control Cycle”.
		The Actuarial Control Cycle is a concept that was documented at Macquarie University by an education committee from the IAAust in 1994-95. It now forms an integral part of the professional examination of the IAAust and is a compulsory part of actuarial qualification. The concept was well established in the Australian actuarial profession before 2000.
		The actuarial control cycle framework can be described pictorially as follows:

	James Hardie — Actuarial Expert Witness Report
KPMG Actuaries Pty Ltd	4 June 2004

Figure 2.1: The actuarial control cycle

Global Economic and Commercial Environment Define Problem Actuarial Control Cycle Monitor Experience Develop Solution Professionalism

		The principle of the control cycle is that when an actuary has a complex issue to address or advise upon, the actuary first identifies what the key aspects of the issue are. This helps to identify the scope of the project and also to identify the approach that may need to be used.
		The actuary would then construct a model (or solution) and specify the parameters and assumptions to feed into the model and make forecasts of the future experience.
		The actuary should then, at successive valuations, monitor how the actual experience emerges. The actuary should also consider how appropriate the previous assumptions and parameters were and from this, the actuary should refine (or maintain) the model, the parameters and the assumptions taking into account the emerging experience and external factors. This process allows the actuary’s work to evolve and also for their advice to react to developing information.
		The actuary should, in all of their work, consider the environment and external developments to that environment (be it legal, medical, social or economic).
		Part of the actuarial control cycle is that the actuary should monitor and identify the causes of changes in the valuation result and attribute the change to each of the significant items that have changed (e.g. changes in future assumptions, emerging experience etc).
		The actuary should also, when revisiting a model, consider not just whether the parameters need to change but also whether the model itself should change in light of emerging experience and external developments.

	James Hardie — Actuarial Expert Witness Report
KPMG Actuaries Pty Ltd	4 June 2004

3.1		Overview
		This section provides some context for the overall Trowbridge approach and the key assumptions. In a number of cases I refer to studies that were not available to Trowbridge at the time of the various reviews. I have also separately documented information which has become available more recently.
3.2		Establishing the peak and duration of future notifications
3.2.1.		Epidemiological studies
		A bibliography of various epidemiological studies referenced is attached in Appendix E. Appendix B provides a chronology of these studies and how their availability relates to the timing of the various Trowbridge advices. The chronology also shows the relevant and material court awards and legislative changes during the period.
1.		Australia
		Key studies in Australia are Berry (1991), Berry et al (2004), Leigh et al (2002) and (2004).
		Berry et al (2004) indicates that experience in Wittenoom1 appears to be better than previously forecast, being at the lower end of the Berry (1991) models.
		Leigh et al (2002) and (2004) indicate the incidence of mesothelioma has increased four-fold in 20 years. They indicate that Australia has the highest reported incidence rates in the world. The report indicates that the incidence of mesothelioma in Australia is as high as liver cancer and the mortality rates are as high as those for kidney cancer amongst males and uterine cancer amongst females.

	James Hardie — Actuarial Expert Witness Report
KPMG Actuaries Pty Ltd	4 June 2004

		A number of these and other reports disclose the relationship between exposure and reporting, i.e. the distribution of the latency periods for mesothelioma.
2.		UK and Europe
		Key studies in the UK and Europe are Peto (1995), Peto (1999), Health & Safety Executive (HSE) (Hodgson & Darnton) (2003), Peto (2004) and Pelucchi, Decarli et al (2004).
		Peto’s 1995 analysis of UK mesothelioma indicated a peak for UK claims by reporting year of 2,700-3,300 claims per year in around 2020. The HSE 2003 report indicated that improved modelling suggested a peak of between 1,950-2,450 in 2013; that is earlier, and lower. Peto (2004) indicated an improvement relative to previous estimates with a peak of around 2,400 in 2015-2020. Peto (1999) addressed European exposures and confirmed the peaks in Europe to broadly accord with UK findings. Peaks of between 2013-2020 were estimated across a variety of countries.
		Pelucchi & Decarli (2004) revisited the Peto (1999) analysis and they also indicated an improvement in Europe relative to previous estimates, consistent with the indication of moderate improvement indicated by both the HSE and Peto research in the UK.
3.2.2.		Nature of worldwide exposures and deaths
		Figure 3.1 shows the nature of the exposures for the UK and Australia in terms of net asbestos consumption (Production and Imports less Exports) by year.

Figure 3.1: Analysis of exposure and deaths from
mesothelioma: UK and Australia

3,500 3,000 2,500 2,000 1,500 1,000 500 0 Aus : Peak = Wide views Nos in 2000 = 348 Nos in 2001 = 685 UK Peak in 2013 at 2200 deaths Nos in 2000 = 1650 Nos in 2001 = 1850 Australian Peak Exposure 4-6 years later than UK 0 10 50 5 95 00 050 5 045 50 19 019 051 9 1 9 1519201925193019 319 4019 4519 5019 551 9 6019651970197519 819 819 901 9 2 0 2 0 2 0 10201520 220 220 320 3520 402 0 2 0 UK Exposure Aus Exposure (standardised to UK at 1967) UK Deaths (HSE) Aus Diagnoses (standardised to UK at 1981)

	James Hardie — Actuarial Expert Witness Report
KPMG Actuaries Pty Ltd	4 June 2004

Sources:     Treasure T, Walter D, Swift S, Peto J (2004); Radical Surgery for Mesothelioma (BMJ)

			Australian Number of Deaths: Australian Mesothelioma Register: 2003
			UK Number of Deaths: HSE Statistics (www.hse.co.uk)

		Australian Exposure Statistics: Bureau of Mineral Resources; RJ Hughes 1978; Leigh et al 2002, Virta 2003 (Extrapolation by KPMG from 10-year intervals), US Geological Survey (produced annually)
		The number of mesothelioma deaths have been standardised to the UK at the 1981 level (the Australian volume of deaths was 25% of the UK at that point). It can be seen that the Australian exposure is some 4-6 years later in its peak than the UK and that it peaked in the early 1970’s. The UK exposure peaked in around 1966/67.
		The current view in the UK is that the peak of future deaths will arise in, or around, 2015 (HSE Central Estimate is 2013; Peto 2015-2020).
		Given this, one might infer that for the Australian Industry as a whole, the peak may also lie somewhere in the 2nd or 3rd decade of the 21st century. Some epidemiologists in Australia have expressed a view that the peak will now occur in 2023 based on their modelling, whilst other Australian sources indicate that the peak might be as late as 2030.
		The number of mesothelioma deaths in the UK is expected to peak at between 1,950 and 2,450 per year (HSE Central Estimate 2,200) with a total number of future deaths in the period 2004-2050 of around 55,000 (compared with 22,000 already incurred in the period to 2003).
		In Australia, the number of diagnoses or notifications to 2003 is approximately 8,000 (source: Malignant Mesothelioma in Australia, Australian Mesothelioma Register Report 2003). This can be broadly equated to deaths given the extremely short period between diagnosis and death for mesothelioma. Those affected by malignant mesothelioma usually will die within a couple years of the onset of the disease. Many malignant mesothelioma patients are not diagnosed in a timely manner so the time of diagnosis and death are often very close together. Receiving no treatment for malignant mesothelioma is associated with a typical survival rate of between four months and a year.

	James Hardie — Actuarial Expert Witness Report
KPMG Actuaries Pty Ltd	4 June 2004

		In the US, some market commentaries are now indicating that the peak will arrive in the US in the period 2010-2020. However, it should be recognised that the US experience has, in part, been distorted by the ‘unimpaired claimant’ effect.
3.2.3.		The insurance industry
		It is noted that within the insurance industry there exists a degree of variation in the assumed year at which the peak of notifications will arise.
		In 2000, a considerable number of insurers assumed that the peak of notifications would arise in the period 2002-4.
		It should be noted that the existence of such an assumption does not necessarily imply this as being a reasonable assumption for the James Hardie assessment, and nor does it provide independent support to Trowbridge’s assumed peak of notifications for James Hardie’s liabilities.
		This is because:

	§		The peak assumed within the insurance industry was significantly influenced by Trowbridge’s published presentations;
	§		The peak relates to insurance exposures, not those of a direct asbestos manufacturer;
	§		James Hardie’s exposure was more readily measurable than that for many insurers where exposure from their insured clients is often not available or difficult to quantify. As such, this extra information ought to have been factored in;
	§		James Hardie’s main element of exposure is to public and product liability. For workers compensation claims, the exposure will in the main be related to when the company no longer produces or manufactures asbestos-related products. For product and public liability, the exposure remains a considerable time after the last date of production of products containing exposure, reflecting the “working life” of those products, so that the exposure therefore extends.

	James Hardie — Actuarial Expert Witness Report
KPMG Actuaries Pty Ltd	4 June 2004

3.3		Superimposed inflation
3.3.1.		Analysis of inflation within liability portfolios
		Market experience, at the time of 2000, suggested that total inflation had been running at up to 8% per annum above base wage inflation for a number of years for a range of liability-related classes. Trowbridge would have been aware of this impact on such classes of liability; including workers compensation, public liability and CTP. All of these portfolios were exhibiting growing trends in superimposed inflation. Actuaries were making appropriate allowance for this exhibition of superimposed inflation both in pricing and reserving for these portfolios.
		For public liability claims, superimposed inflation had generally exhibited between 6% and 8% per annum on average over the period 1995-1999.
		For workers compensation, superimposed inflation had generally exhibited between 4% and 8% per annum during the latter part of the 1990s.
		For CTP in NSW and QLD, superimposed inflation in the latter part of the 1990s was exhibited between 2% and 6% per annum over the period.
		In relation to asbestos-related disease liabilities coming through the insurance industry, superimposed inflation of between 5% and 10% was exhibited during the 1990s, with public liability claims generally experiencing higher rates of superimposed inflation.
3.3.2.		Market treatment of allowance for superimposed inflation
		Market practice in respect of allowing for superimposed inflation within valuation models, especially in relation to asbestos-related liabilities, has generally been to assume an average rate to apply over all future years. This was particularly the case in 2000 and 2001.
		Market practice has also been to aggregate together a number of contributing factors to the overall rate of superimposed inflation, being:

	§		The rate of pure (‘judicial’) superimposed inflation,
	§		The impact of medical or other developments, and
	§		The effect of an ageing population of claimants on the rate of inflation of overall damages, a component of which relates to loss of earnings.

It is common practice to assume that the third item has a negative effect on superimposed inflation and that this is offset against the other factors.

	James Hardie — Actuarial Expert Witness Report
KPMG Actuaries Pty Ltd	4 June 2004

		The general approach, and it is the approach that I have taken within this report, is to implicitly offset these various factors in arriving at the average rate of superimposed inflation to apply over all years.
		Therefore, whilst superimposed inflation may have been high in recent times, there are some factors which might potentially act to dampen this in the future, being the effect of the ageing population of claimants.
		As claimants age, the component of the claim which relates to loss of earnings or pension diminishes, reflecting the shorter future working life or life expectancy of a healthy individual. As such, this diminishing component of the claim acts to partially suppress the overall inflationary nature of claims.
		It should be noted that this assumption of an ageing population providing a negative effect on inflation would be influenced by changes to retirement age and also life expectancy more generally.
		It should also be recognised that the existence or potential existence of secondary exposures, to individuals who were not directly exposed through mining or manufacture of asbestos products could also lead to a younger population being exposed.
		Such secondary exposure could therefore have the opposite impact to an ageing portfolio and exacerbate the apparent rate of superimposed inflation in the future.
		Whilst it appears that recent levels of superimposed inflation has been of the order of 6% per annum or more for a number of years, I have not assumed that this would represent the long-term average position, and that the long-term average would be somewhat lower. My assessment is that a rate of 2% per annum averaged over all future years is consistent with the impact of the various factors.

	James Hardie — Actuarial Expert Witness Report
KPMG Actuaries Pty Ltd	4 June 2004

Figure 3.2: Analysis of age of claimant by report year

y=0.5816x + 59.486 Claimant’s age at report year report year 100 90 80 70 60 50 40 30 20 earlier 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 0% 25% 75% 90% 100% mean linear (mean)

		The chart above shows the minimum (0%) and maximum (100%) ages of claimant for each report year. The age of the 25th, 75th and 90th percentile observations are also shown.
		The mean age of claimants is shown, together with a trendline of how the mean has progressed over time. The chart shows that the average of age of a claimant has been increasing by approximately 0.60 years with each passing year.
		This analysis has provided me with an insight to the extent to which the effect of ageing has affected historic average claims costs and also provides me with a more robust basis for making adequate allowance for the effect of ageing upon future rates of superimposed inflation.
3.4		Approach used to evaluate asbestos-related disease liabilities
		The approach below describes what, in my opinion, a “reasonable actuary” could pursue by way of methodology and approach in evaluating the liabilities for asbestos-related diseases.
		I recognise that the extent to which this can be followed is dependent on the information available from the client or the market, or which might be able to be made available from a variety of sources.

	§		Identify exposures by year (source, State, type of asbestos product, nature of occupation);
	§		Distribution of latency period by disease type;

	James Hardie — Actuarial Expert Witness Report
KPMG Actuaries Pty Ltd	4 June 2004

	§		Identify “to date” position;
	§		Conclude “Ultimate” position by year;
	§		Inference of disease incidence per unit exposure;
	§		Allow for legal/environmental factors, such as:

	§		Increasing propensity to sue;
	§		“Unimpaired claimants”;
	§		“Downstream” claims;
	§		New sources of claim; and
	§		Change in mix of exposure by occupation etc.

	§		Project future notifications;
	§		Project future settlements (this will be done for each disease type modelled);
	§		Analyse historic average settlement costs (@ 100% share);
	§		Inflate to current money terms;
	§		Analyse James Hardie share historically;
	§		Trended view of James Hardie share prospectively;
	§		Considerations of legal developments on average costs (@ 100% level and on James Hardie share);
	§		Inflate to future settlement year using Average Weekly Earnings (AWE) plus future superimposed inflation allowance;
	§		Loading for legal cost elements similarly;
	§		Measure insurance recoveries consistently with gross claims experience; and
	§		Review insurance contracts in terms of:

	§		Years on risk;
	§		Policy limits;
	§		Policy terms;
	§		Disputes of coverage; and
	§		Recoverability/bad debt liabilities.

	James Hardie — Actuarial Expert Witness Report
KPMG Actuaries Pty Ltd	4 June 2004

4.1		Methodology adopted by Trowbridge
4.1.1.		Overview
		Based on a review of the working papers and the review of the reports produced by Trowbridge in their assessment of the asbestos-related liabilities of James Hardie and then the MRCF, it appears that Trowbridge’s methodology, in general terms, has not changed since they began advising James Hardie in 1996. It is generally consistent with the existing practice adopted by Australian actuaries when considering asbestos provisioning.
		It entails a projection of future numbers of claims to be reported multiplied by an individual average cost to settle those claims (subdivided by type of asbestos disease claim), plus an adjustment on the value of claims already reported but not settled. There are two key components to such analysis:

	§		The ultimate costs associated with claims that have been reported but are open or could reopen (discussed further in Section 4.1.2 below).
	§		The ultimate costs associated with claims that have been incurred but have not yet been reported (“IBNR”). This is discussed further in Sections 4.1.3 and 4.1.4 below.

4.1.2.		Outstanding costs of reported claims
		Trowbridge note that individual estimates (“case estimates”) were available for the open claims on a “low”, “best estimate” and “high” basis.
		Trowbridge effectively added the known reported open claims to its numbers and averages model for IBNR claims discussed below.
		It appears that Trowbridge’s approach effectively used case estimates to provide a rule-of-thumb only. In doing this a great deal of useful information that may be imparted by the case estimates of these open claims falls outside of the information considered.
		It is, and was, common practice to make more use of the paid amounts and case estimates on the open claims. This is discussed further in Section 4.3.2 below.

	James Hardie — Actuarial Expert Witness Report
KPMG Actuaries Pty Ltd	4 June 2004

4.1.3.		Numbers of claims
		The methodology used by Trowbridge is to first estimate the number of future claims to be reported in each year by class of liability, i.e. public or general liability and workers compensation+ (see Section 4.1.7) and by disease type (see Section 4.1.8).
		This is done through the application of a “Berry Curve” or similar curve.*
		From the assumed number of future claims reports and from the current open claims, the total number of future settlements is estimated by year. A settlement pattern is applied to map claim numbers from year of reporting to year of settlement. Generally the settlement pattern is different for new reports and currently open claims and is derived by reference to past settlement patterns which Trowbridge have analysed.
		Trowbridge then estimate the proportion of claims settled which will be settled for nil claim costs (‘Nil claims’), again based on an analysis of the historic rate of nil settlements.
4.1.4.		Average claims costs and inflation
		Costs are modelled separately for claim settlements and legal components. The average claim settlement cost specified is for a non-nil claim settled and is in current dollar terms. It is derived separately by disease type and class of liability, and is derived from examination of the historic settlement costs by year. Based on a review of the working papers and the actuarial reports, it does not appear that these historic costs were inflated to current money terms.

*Professor Berry is an epidemiologist who has conducted a considerable number of studies into the development of mesothelioma in Australia. One of his most notable studies, particularly for actuaries, is his 1991 research paper “Prediction of mesothelioma, lung cancer, asbestosis in former Wittenoom asbestos workers”.

The Berry curve is derived from consideration and studies into the exposure and incidence of deaths of miners at Wittenoom, Western Australia. The key components to any curve used are (a) the peak in the notification curve, and (b) the duration of future claims notified from the valuation date. It is also important, when developing and utilizing such a curve, to take into account the detailed exposure information underlying the liability relative to that of Wittenoom mine.

	James Hardie — Actuarial Expert Witness Report
KPMG Actuaries Pty Ltd	4 June 2004

		The legal costs of nil and non-nil settlements are set separately and are also established by reference to the analysis of settled legal cost data for each of these groups.
		The average claim costs are then projected to the date of settlement by inflating from the valuation year to the settlement year using an assumed future average claims inflation rate (Trowbridge set this equal to its estimate of future wage inflation).
		There does not appear to have been any separation or analysis of claims between “small” and “large” in the setting of the average cost assumption.
		No allowance for superimposed inflation (i.e. inflation above the base measure of wage inflation) is made by Trowbridge.
4.1.5.		Insurance recoveries
		The majority of James Hardie’s insurance contracts, being those upto 1986, were written on an exposure-year basis. That is, recoveries are made against claims which have been notified to the company by first allocating the claims back to the relevant years of exposure from which the claims arose and spreading the cost over those (potentially multiple) exposure years accordingly (usually based on a proportional allocation). From 1986, the contracts were generally written on a ‘claims made’ basis. That is, recoveries are made against claims which have been notified to the company in the year.
		The insurance recoveries, as estimated by Trowbridge, are modelled by the following method:

	§		Calculate the future cashflows on general liability claims, including an allowance for legal costs (net of an allowance for unallocated loss adjustment expenses (“ULAE”);
	§		Assume that the percentage split of cover by insurer is based on an historic assessment of the estimated recoveries to date;
	§		Assume that 1.5% of the claims costs relate to Jsekarb;
	§		Identify the implied contributions by year of each insurer;
	§		Quantify the expected cash recoveries from each insurer recognising the limits of exposure of the insurance contract, separately for JH&Coy and Jsekarb, and the level of cover estimated to have already been used;
	§		The QBE commutation was assumed to imply no coverage remained.

	James Hardie — Actuarial Expert Witness Report
KPMG Actuaries Pty Ltd	4 June 2004

4.1.6.		Cashflows and discounting
		From the number of future settlements and the average cost of settlement, by settlement year, the claim and legal costs by class of liability and disease type are projected for each future financial year on a cash basis. The insurance recoveries are netted off the gross cashflows to derive the net cashflows, and any additional costs relating to specific liabilities (such as wharf claims) are added. The cashflows are discounted at a uniform discount rate.
		The discount rate in 2000 and 2001 was set relative to anticipated “market yields... on high quality corporate bonds”.
4.1.7.		Class of liability
		The data was segmented by class of liability into general (public and product) liability and workers compensation claims. The claims relating to Waterside (Wharf) Workers were (where applicable) separately analysed.
4.1.8.		Disease types
		For general liability, the data was segmented by disease type into:

	§		Mesothelioma / non-mesothelioma (used for pre June 2001 reports);
	§		Mesothelioma / lung cancer / asbestosis / Asbestos-Related Pleural Disease (ARPD) and others (used for June 2001 and onwards).

		For workers compensation, the analysis by disease type was generally not made and all claims were aggregated together.
4.1.9.		Data gathering and checking
		Data was combined by Trowbridge from two sources maintained by James Hardie, namely:

	§		An “individual claim register”;
	§		An “accounting database” by matching claims identifiers.

		Trowbridge notes that “this matching process was not straightforward”.
		Trowbridge reconciled claim payments between the Individual Claim Register and the accounting database. Trowbridge note in their reports that other “checks, for reasonableness and internal consistency” were made, but further detail is not given. Trowbridge made the following observations:

§

“We found a number of anomalies in the data” – 31 March 2000 valuation report, and

	James Hardie — Actuarial Expert Witness Report
KPMG Actuaries Pty Ltd	4 June 2004

§

“Some of the individual claim information was missing or incomplete” – 30 June 2003 valuation report.

4.1.10.		Excluded items
		There are a number of areas which Trowbridge “have not quantified due either to their speculative nature or lack of specific data”.
		The three areas of excluded items (as per the 31 March 2000 report) were:

	§		Claims originating outside Australia and New Zealand;
	§		Claims due to non-disabling conditions (e.g. stress);
	§		Environmental, property or land remediation.

Additionally, it should be noted that a number of other items are implicitly excluded from consideration. These include:

	§		Any claims which do not have a report date assigned to them, and
	§		Any current or prospective legal developments which cannot be readily quantified owing the lack of specific data or being speculative in nature.

4.2		Key assumptions
		Appendix F shows the progression of the key assumptions made by Trowbridge over time, between 1996 and 2003.
4.2.1.		Key assumptions
		The key assumptions required for the Trowbridge model are:

	§		Number of future reported clams;
	§		Pattern of settlements of future reported claims and of currently open claims;
	§		Average settlement cost and average legal cost;
	§		Rate of nil settlements;
	§		Future inflation;

	§		Base
	§		Superimposed

	§		The discount rate; and
	§		Insurance recoveries.

	James Hardie — Actuarial Expert Witness Report
KPMG Actuaries Pty Ltd	4 June 2004

4.2.2.		Key considerations
		This section addresses the types of questions, thoughts and considerations Trowbridge ought to have addressed when setting their assumptions. Section 4.3 will address the extent to which this was done.
		Number of future reported claims

	§		When is the peak?
	§		How long will future notifications continue for?
	§		What is the nature of James Hardie’s exposure?
	§		How does this differ to US, UK exposures and what are UK and US views on peaks and durations?
	§		Does data move (e.g. through re-designation of claim types from non-mesothelioma to mesothelioma)?
	§		What are the sources of the reported claims by occupation, by exposure period, by age of claimants? Is there any visible or predictable trend in this?
	§		What is the distribution of the latency period?
	§		What is the impact of the filing of multiple claims per claimant against more than one James Hardie entity?
	§		Is there any evidence of increased co-joining of James Hardie (or the MRCF) owing to a perceived ‘deep pocket’?

Pattern of settlement

	§		What has been the past settlement pattern?
	§		Any views on litigation or other developments which may alter this?
	§		What is the best estimate prediction for the open claims?
	§		Changes in settlement strategy by James Hardie (or MRCF)

Average cost of settlement (claims and legal)

	§		Recent and historic average settlements (inflated to current money terms)
	§		What are the trends in the “100% share” costs?
	§		How is James Hardie’s share changing (is there any increase results from a perceived ‘deep pocket’)

	James Hardie — Actuarial Expert Witness Report
KPMG Actuaries Pty Ltd	4 June 2004

	§		Landmark legal cases potentially impacting settlements, together with a quantification
	§		Impact of re-designations (i.e. from non-mesothelioma to mesothelioma)
	§		Any moving data historically? (i.e. changes to the historic number of claims reported)
	§		Any medical developments e.g. earlier identification, extending life?
	§		How are case estimates tracking between report date and settlement?
	§		Is there any notable distortion from ‘large’ claims? Are any open claims expected to settle for very large amounts?
	§		What is expected in respect of open claims on a best estimate basis?

Nil settlement rate

	§		Recent and historic average rates of nil settlements
	§		What is the pattern of nil and non-nil settlements?
	§		What is expected in respect of open claims on a best estimate basis?
	§		Can we project ultimate settled at nil and non-nil by a report year cohort? Is such a protection showing any trends?

Future inflation (superimposed)

	§		Australian market views on historic superimposed inflation
	§		James Hardie’s own historic experience of superimposed inflation
	§		Segmenting claims data e.g. by ordering historic settlements
	§		Measuring observed inflation
	§		Age profile of claimants by year
	§		Landmark legal cases which may cause judicial inflation or deflation pressures
	§		Exposure and claims mix by state
	§		Have case estimates tracked well to settlement? Are case estimates reasonably estimating inflation to settlement date?
	§		Real inflation rate relative to selected discount and observations of this in recent periods.

	James Hardie — Actuarial Expert Witness Report
KPMG Actuaries Pty Ltd	4 June 2004

4.3		Reasonableness of key assumptions and approach undertaken by Trowbridge
4.3.1.		Overview
		With the benefit of hindsight that comes from later data and valuations, it is clear that the assumptions and approach of earlier valuations have led to assessments of liabilities that have ultimately proven significantly too low. The comments below are set out to represent my views without the benefit of such hindsight.
4.3.2.		Review of outstanding costs of reported claims
		As at 31 March 2000, case estimates on a best estimate basis were $25m, approximately 9% of the total assessed liability of $295m.
		The approach used by Trowbridge in respect of known claims is not typical of the approaches seen elsewhere in that it makes limited use of the information available for open claims. The standard approach used is for a projection based on a “report year” grouping or “cohort”.
		The projections would typically use standard chain ladder methods applied to:

	§		Cumulative paid claim amounts;
	§		Cumulative paid plus case estimates (‘Incurred’) claim amounts;
	§		Cumulative numbers settled at nil;
	§		Cumulative numbers settled at cost; and
	§		Reported numbers (as recording of reports can be subject to processing delays).

These projections would allow the actuary to:

	§		Identify any past processing delays or changes to processing (in doing so raise the potential for similar distortions going forward);
	§		Examine the current and ultimate expected split of nil to non-nil claims, improving on the current practice of only relying on settled claims for this, and helping to identify more quickly any emerging trends in this split;
	§		Examine the settlement pattern (by claim number and amount), in turn adding to the data available in setting this assumption for cashflow modelling;
	§		Explicitly model the emerging cashflows from open claims through the paid claim projection, again adding information to this modelling;

	James Hardie — Actuarial Expert Witness Report
KPMG Actuaries Pty Ltd	4 June 2004

	§		Examine the total outstanding costs for open claims; and
	§		Examine the implied settlement cost of open and/or recently notified claims to ascertain the extent to which settlement costs may be increasing or decreasing. This provides an early detection of any adverse trends emerging in recent claims experience, allowing the actuary to respond as appropriate.

		This approach would provide a check, which is to a degree independent, of the result of the method used. In turn understanding any differences would allow the actuary to further refine the assumptions used in the more material IBNR component.
		Setting out such information would also aid in data checking and the application of the “Actuarial Control Cycle” (discussed in Section 2.5). Specifically:

	§		The change in cumulative paid amounts should be reconcilable to the company cashflows.
	§		The triangle of reported numbers would show the changing data in historic periods, allowing a check for data completeness and consistency with prior data sets. This comment also applies to examination of paid and incurred claims triangles.

4.3.3.		Numbers of claims
		Trowbridge have used a Berry Curve, calibrated to the recent levels of reported claim numbers. There is no explicit discussion or visible consideration of whether the curve is appropriate, or the extent to which the curve needs explicit adjustment. As far as can be ascertained, no consideration has been made of the nature of the actual exposures of James Hardie, such as tonnage of asbestos used or volume of products produced or sold.
		The Berry Curve is based on a population of miners employed by CSR at Wittenoom mining Crocidolite (“blue asbestos”), which is associated with asbestosis and mesothelioma. The Wittenoom mine closed in 1967. The workers in the mines had very heavy exposure to asbestos dust and the levels of claim will likely have been heavier, more severe and expected onset of asbestos-related diseases would have been quicker than those exposed via James Hardie. The longer-term fall-off of claims would also likely have been earlier owing to the heavier associated mortality.

	James Hardie — Actuarial Expert Witness Report
KPMG Actuaries Pty Ltd	4 June 2004

		James Hardie ceased using Crocidolite asbestos in 1968 (source: Asbestos: The Relentless Peril, Adrian Leonard (2003), page 113) but continued producing insulation products until 1974 using Amosite (“brown asbestos”), and brake linings and asbestos cement were produced until 1987 using Chrysotile (“white asbestos”) (source: Trowbridge March 2000 Report, page 13).
		It is not apparent whether any consideration or analysis has been performed of the mix of claims by source or occupation to identify the extent to which the patterns in numbers might be changing as a result of any changes in the nature and source of claims being notified.
		In my view, the approach used by Trowbridge is less sophisticated than models potentially available to Trowbridge. I note that exposure modelling is not always used; it is especially difficult for some companies and insurers where product liability exposure information is not available. However, for companies with direct asbestos exposure, such as CSR and James Hardie, it is my opinion that such information is both important and more readily available and has been taken account of in my assessment of liabilities.
4.3.4.		Average claims costs and inflation
		Trowbridge split the claims into legal elements and the claims/damages elements. This is a reasonable approach.
		The average cost is derived by considering past average settlements. This is a reasonable approach.
		The average costs are analysed historically without inflating the historic costs into current money terms. In my opinion, the standard approach used by actuaries is to convert historic costs into current money terms firstly, in order to more fully understand and extrapolate any trends in average costs. This can then be used as the basis of assessment of the average costs. Such information also provides an improved framework for better gauging the existence of superimposed inflation within past claims experience.
		Trowbridge do not appear, from the papers made available, to have examined the distribution of settled claim by size, in particular to examine the potential for the pattern of large claims to distort the average costs.

	James Hardie — Actuarial Expert Witness Report
KPMG Actuaries Pty Ltd	4 June 2004

		From the papers provided it appears that Trowbridge have not separated out small and large claims (e.g. claims in excess of $1m in current money terms). Typically the actuary would segment claims in this way and then analyse the trends in small and large claims (and their relative frequencies and trends therein) in forming an overall average cost per claim. This evaluation is also critical where insurance is on an excess of loss basis.
		As regards claim inflation the basic approach is standard but my view is that the combined rate selected was not reasonable, as I have noted in Section 3.3. The rate of future superimposed inflation is a key assumption and I would expect:

	§		Commentary on wider market observations;
	§		Cognisance of, if not comparison to, other long tail liability classes such as workers compensation generally and public liability generally;
	§		Consistency with Trowbridge’s own presentations to the market place;
	§		Examination of “actual to expected” experience relating to visible recent inflation;
	§		Discussion of the combined impact of the real discount rate, i.e. discount less overall inflation in the context of the very long term projection;
	§		Commentary relating to the views of the relevant claims handling persons;
	§		A view on overall settlements (as opposed to the more limited observations that come from examining the James Hardie share only which may be distorted by changes in this share over time);
	§		Discussion of any material or potential changes to medical treatments, diagnostic approaches and interventions; and
	§		Discussion of any material, or potential, changes to the legal environment.

		In all cases such considerations significantly challenge the use of a zero superimposed inflation assumption. The Trowbridge assumption of a zero rate of superimposed inflation does not appear reasonable in light of the above matters.
		As I have noted in Section 3.3, there do exist potential compensating factors such as the impact of an ageing population of claimants who generate lower loss of earnings-related settlements thereby depressing the overall rate of future superimposed inflation.

	James Hardie — Actuarial Expert Witness Report
KPMG Actuaries Pty Ltd	4 June 2004

		As I have noted previously, the general practice was to offset this against the other factors giving rise to superimposed inflation within one aggregate figure as opposed to separate consideration of each of these factors. I have taken this general approach in my assessment.
4.3.5.		Insurance recoveries
		The various reports provide limited information relating to the insurance programme. I would ordinarily expect a summary of the insurance programme to be provided. To the extent that cover is on an “excess of loss basis” (i.e. covers individual claim amounts above a specific excess point or deductible) I would expect greater clarity on whether individual claim costs would trigger recoveries.
		I would also seek confirmation that the excess points are not subject to indexation (the adjustment of the excess relative to inflation) or if indexation did exist a discussion of how this has been modelled.
		There has also been no tracking of emerging experience or actual versus expected analysis of insurance recoveries. No allowance has been made within the estimate of the recoveries for the potential bad debt, or non-recoverability, of these recoveries. Although I note this practice of allowing for non-recoverability has become more standard only in recent years.
		No allowance has been made for the potential of disputes between Amaca and Amaba and their insurers. This is a relatively standard practice. Such disputes might relate to the elements of the claim which are covered by the insurance contract, the basis of allocation of exposure and also disputes over disclosure (which are often present in asbestos-related contracts in my experience). In essence, full recoverability has been assumed.
4.3.6.		Cashflows and discounting
		The approach taken in respect of cashflow modelling and discounting of liabilities is appropriate as a method.
4.3.7.		Classes of liability
		In general the approach of considering workers compensation claims separate to the liability claims seems reasonable. I would normally expect to see consideration of product liability claims separate from public and other liability claims.
4.3.8.		Disease types
		In general the approach of considering mesothelioma separately from other disease types seems reasonable and is common.

	James Hardie — Actuarial Expert Witness Report
KPMG Actuaries Pty Ltd	4 June 2004

		It would appear that James Hardie had been able to resist claims coming from people who had a fear of contracting asbestos-related diseases as a result of exposure to any products either through manufacturing or handling of products thereafter, such claims, when made, are often referred to as ‘unimpaired claims’. Any such claims would require separate identification if they were notified and accepted as claims.
4.3.9.		Data gathering and checking
		One of the responsibilities of the actuary is to perform checks on the data being used to content themselves as to the veracity of the data on which the valuation is dependent.
		In my experience, the actuary would normally refer to the types of validations performed and any issues that arose from this validation process (and explain the extent of the significance of any issues that arose).
		There is an absence of reference in the reports to certain checks which ought to be completed, which I feel are on the critical path to the valuation, specifically:

	§		Cross checking against the data provided for prior reviews; and
	§		Reconciliations against the general ledger entries (or equivalent) relating to asbestos costs.

There is an absence of any reference in the reports to material observations relating to the data, specifically:

	§		Identification of the movement in historic data (claim numbers and settlement amounts) within settled claims cohorts;
	§		Identification of moving claim numbers within claims grouped on a reported basis; and
	§		Identification of claims ignored (e.g. those with no report date).

		It would be usual for such checks to be completed and an understanding of the causes of this to be demonstrated, as well as any potential impact to the assumptions.
		Overall, in my opinion, the steps taken with respect to data do not appear, based on the information I have available, to be reasonable.
		Further support on this is given in the sections discussing specific reports, particularly in Section 8.

	James Hardie — Actuarial Expert Witness Report
KPMG Actuaries Pty Ltd	4 June 2004

4.3.10.		Excluded items
		In respect of site clean up costs and ‘unimpaired claims’, I have concluded that the treatment of excluding potential liabilities associated with these items was reasonable at the time on the basis that:

	§		It is my understanding that, to date, unimpaired claims have generally not been admitted in Australia; and
	§		Site clean up costs were not considered material to James Hardie, although in this respect greater disclosure of any claims paid to date would have given greater support to Trowbridge’s treatment.

		US asbestos claims were known to be problematic and yet the report offers no comment as to the potential exposure. It is noted that the scope identified in 1996 excluded US claims on the grounds they would be speculative. This was reasonable when only 1 such claim was noted to have existed.
		However, by 2000, there had been 11 claims filed. It is my opinion that such claims had then become somewhat less speculative and that Trowbridge could have done more in respect of the assessment of the potential for US exposures, in particular:

	§		To the extent that there were a number of known reported potential claims (11 as at 31 March 2000) the report could have made this clear. This would illustrate that a best estimate was above nil (especially given the legal costs of defending such potential claims);
	§		Furthermore, Trowbridge could have illustrated potential ‘US Scenarios’, e.g. by extrapolating using the Australian or a US numbers curve as benchmarks.

		I would ordinarily expect that where the actuary has excluded a sub-set of liabilities from their deliberations that they would highlight the need for the client to form their own view and make such allowance as the client felt appropriate.
		It should be emphasised to date that there is no evidence that James Hardie have incurred any large class action type claims unlike many other companies who produced asbestos-related products which found their way to the United States.

	James Hardie — Actuarial Expert Witness Report
KPMG Actuaries Pty Ltd	4 June 2004

5.1		Background
		The March 2000 Trowbridge report (in draft) was issued on 16 June 2000 and documented an actuarial assessment of the future asbestos-related disease liabilities of James Hardie taking into account data as at 31 March 2000.
		The authors of the report were Mr Karl Marshall and Mr David Minty and the report estimated a liability of $294.7m. This liability was the present value of all future cashflows discounted at 7% per annum.
		The liability made allowance of $10m for Australian wharf claims, whilst no allowance was made for the cashflow benefit of the QBE settlement. No allowance was made for US exposures in spite of 11 claims having been lodged. Trowbridge state that they “have not made any estimate of the potential exposure as there is little information available on which to base any assessment and the areas are therefore considered too speculative for an actuarial assessment”.
		The report notes other potential exposures such as environmental and site clean-up costs and claims relating to stress resulting from potential exposure to asbestos. No liability was estimated or recommended for any of these exposures.
		The report states that data quality was good and has improved. Trowbridge state the results of their valuation are not materially affected by any discrepancies. Trowbridge also state that data reconciliation has been performed and the results are reasonable.
		The report addresses sensitivity testing on the following factors:

	§		The numbers of claims;
	§		The peak and duration of future notifications;
	§		The average cost of claims;
	§		Superimposed inflation;
	§		Discount rates.

	James Hardie — Actuarial Expert Witness Report
KPMG Actuaries Pty Ltd	4 June 2004

5.2		Key observations on Trowbridge’s valuation basis
		In this section the valuation basis used by Trowbridge for March 2000 is reviewed. For each assumption, I have considered the emerging experience in the intra-valuation period (1998 to 2000), the change in the basis that Trowbridge have made relative to the 1998 valuation, and I discuss the reasonableness of that change and the assumption, together with matters I feel ought to have been addressed.
5.2.1.		Numbers of claims
		Analysis of emerging experience
		Mesothelioma numbers for the last three years (1998-2000) had risen considerably relative to numbers estimated at the 1998 valuation (see table B.3 of Trowbridge’s report and Table 5.1 below). The table below shows the reported mesothelioma claims at each valuation, and highlights where the historic data has “moved”. It should be noted that no comment was made by Trowbridge regarding this movement.

Table 5.1: Reported mesothelioma general liability claims
at the 1998 and 2000 valuations

		*Projected to ultimate for the additional days of reports to 31 March 1998/2000. Actual 86 at 1998 and 90 at 2000, both years projecting further 4 claims +Estimates at 1998 valuation.
		The reported number of claims for the 1997/98-year of report had risen from an assessed ultimate of 90 to 101. I have inferred that this was in part related to processing delays. It is also noticeable that earlier reporting years also show some movement.
		Generally, there ought not be an increase in the numbers of claims reported in a previous year other than potentially to allow for processing delays.

	James Hardie — Actuarial Expert Witness Report
KPMG Actuaries Pty Ltd	4 June 2004

		The number of claims reported in 1999 is 10 higher than was forecast at the 1998 valuation whilst the claims for the 2000-year of report appear to be broadly in line with the forecast at the 1998 valuation. However, it should be recognised that there clearly exists potential for further development of the actual experience for 2000 (given the strong development of the 1998-year of notifications since the previous valuation).
		Non-mesothelioma claims numbers for the three years (1998-2000) appeared to be considerably below the numbers estimated at the previous valuation (see table C.3 of Trowbridge’s report and Table 5.2 below). The historic data also “moved” and the table below highlights this movement. It is noted that no comment has been made by Trowbridge regarding the movement.