The Australian Statistician v Leighton Contractors Pty Ltd
[2008] WASCA 34
•26 FEBRUARY 2008
THE AUSTRALIAN STATISTICIAN -v- LEIGHTON CONTRACTORS PTY LTD [2008] WASCA 34
| (2008) 36 WAR 83 | |||
| SUPREME COURT OF WESTERN AUSTRALIA | Citation No: | [2008] WASCA 34 | |
| THE COURT OF APPEAL (WA) | |||
| Case No: | CACV:137/2007 | 1 NOVEMBER 2007 | |
| Coram: | STEYTLER P McLURE JA NEWNES AJA | 25/02/08 | |
| 19 | Judgment Part: | 1 of 1 | |
| Result: | Appeal allowed | ||
| A | |||
| PDF Version |
| Parties: | THE AUSTRALIAN STATISTICIAN LEIGHTON CONTRACTORS PTY LTD PUBLIC TRANSPORT AUTHORITY OF WESTERN AUSTRALIA |
Catchwords: | Civil practice and procedure Production of documents Claim of public interest immunity Scope of the doctrine Whether confined to governmental functions Relevance of confidentiality orders |
Legislation: | Australian Bureau of Statistics Act 1975 (Cth), s 5, s 6 Census and Statistics Act 1905 (Cth), s 10, s 11, s 12, s 13, s 14, s 19 Census and Statistics Amendment Act 1981 (No 2) (Cth) Public Transport Authority Act 2003 (WA), s 6, s 12 |
Case References: | Air Canada v Secretary of State for Trade (No 2) [1983] 2 AC 394 Alfred Crompton Amusement Machines Ltd v Customs and Excise Commissioners (No 2) [1974] AC 405 Alister v The Queen (1984) 154 CLR 404 Commonwealth v Northern Land Council (1993) 176 CLR 604 Jackson v Wells (1985) 5 FCR 296 Jacobsen v Rogers (1995) 182 CLR 572 Leighton Contractors Pty Ltd v Public Transport Authority of Western Australia [2007] WASC 196 R v Young (1999) 46 NSWLR 681 Relationships Australia v Pasternak (1996) 20 Fam LR 604 Rogers v Home Secretary [1973] AC 388 Royal Women's Hospital v Medical Practitioners Board of Victoria (2006) 15 VR 22 Sankey v Whitlam (1978) 142 CLR 1 |
JURISDICTION : SUPREME COURT OF WESTERN AUSTRALIA TITLE OF COURT : THE COURT OF APPEAL (WA) CITATION : THE AUSTRALIAN STATISTICIAN -v- LEIGHTON CONTRACTORS PTY LTD [2008] WASCA 34 CORAM : STEYTLER P
- McLURE JA
NEWNES AJA
- Appellant
AND
LEIGHTON CONTRACTORS PTY LTD
First Respondent
PUBLIC TRANSPORT AUTHORITY OF WESTERN AUSTRALIA
Second Respondent
(Page 2)
ON APPEAL FROM:
Jurisdiction : SUPREME COURT OF WESTERN AUSTRALIA
Coram : LE MIERE J
Citation : LEIGHTON CONTRACTORS PTY LTD -v- PUBLIC TRANSPORT AUTHORITY OF WESTERN AUSTRALIA [No 5] [2007] WASC 233
File No : CIV 1570 of 2006
Catchwords:
Civil practice and procedure - Production of documents - Claim of public interest immunity - Scope of the doctrine - Whether confined to governmental functions - Relevance of confidentiality orders
Legislation:
Australian Bureau of Statistics Act 1975 (Cth), s 5, s 6
Census and Statistics Act 1905 (Cth), s 10, s 11, s 12, s 13, s 14, s 19
Census and Statistics Amendment Act 1981 (No 2) (Cth)
Public Transport Authority Act 2003 (WA), s 6, s 12
Result:
Appeal allowed
Category: A
Representation:
Counsel:
Appellant : Mr T Howe QC & Mr A J Power
First Respondent : Mr C G Colvin SC & Mr B D Luscombe
Second Respondent : Ms M J Paterson
Solicitors:
Appellant : Australian Government Solicitor
First Respondent : Mallesons Stephen Jaques
Second Respondent : State Solicitor for Western Australia
Case(s) referred to in judgment(s):
Air Canada v Secretary of State for Trade (No 2) [1983] 2 AC 394
Alfred Crompton Amusement Machines Ltd v Customs and Excise Commissioners (No 2) [1974] AC 405
Alister v The Queen (1984) 154 CLR 404
Commonwealth v Northern Land Council (1993) 176 CLR 604
Jackson v Wells (1985) 5 FCR 296
Jacobsen v Rogers (1995) 182 CLR 572
Leighton Contractors Pty Ltd v Public Transport Authority of Western Australia [2007] WASC 196
R v Young (1999) 46 NSWLR 681
Relationships Australia v Pasternak (1996) 20 Fam LR 604
Rogers v Home Secretary [1973] AC 388
Royal Women's Hospital v Medical Practitioners Board of Victoria (2006) 15 VR 22
Sankey v Whitlam (1978) 142 CLR 1
(Page 4)
1 JUDGMENT OF THE COURT: These are our reasons for giving leave to appeal and allowing the appeal from the decision of Le Miere J rejecting the appellant's claim of public interest immunity in respect of certain information in documents which came into existence for the purpose of constructing a producer price index known as Table 48.
Facts
2 Leighton Contractors Pty Ltd (first respondent) is the contractor for construction of the Perth component of the railway extension between Perth and Mandurah. The contract is contained in a project deed between the first respondent and the second respondent, the Public Transport Authority of Western Australia (PTA), entered into on 14 February 2004.
3 The project deed, so far as relevant for present purposes, provides for adjustments to the progress payments to be made by PTA to the first respondent by a rise and fall mechanism to be calculated in accordance with annexure G of the project deed. The rise and fall calculation depends upon the application of two indexes, one of which, known as the Table 42 index, was produced by the Australian Bureau of Statistics (ABS)) and entitled 'Price index of materials used in building other than house building, excluding electrical materials and mechanical services'. Annexure G to the project deed provides that if an index is discontinued there must be substituted 'the nearest index consistent with the intention of this annexure G so as to give effect to it'.
4 The ABS discontinued the Table 42 index on 30 June 2004. The PTA asked the ABS to reinstate the Table 42 index. The ABS declined to do so but early in 2005 agreed to create a new index, called 'Price index of materials used in building other than house building' and known as the Table 48 index (Table 48), which would provide an index from June 2004. Table 48 could not, however, be compiled using information already in the possession of the ABS. In order to produce Table 48 it was necessary to collect information from 19 respondents from the Table 42 index and to obtain further information from, or to 'enrol', 11 new respondents. Information therefore had to be collected from a total of 30 respondents in Perth and the information had to date back to June 2004. Interviews were conducted by ABS officers in Perth in June 2005. In the course of each interview, the respondent concerned was assured that the information it provided would remain confidential.
5 Information in two types of documents was the subject of the appellant's claim for public interest immunity. That information (the redacted information) was contained in confidential exhibits produced to
(Page 5)
- and inspected by the court. The documents were reports of interviews conducted by ABS officers with survey respondents (interview reports) and ABS working documents collating the raw data supplied by the survey respondents as recorded in the interview reports.
6 The redacted information in the documents the subject of the claim (Documents) related to nine special collection items which, together with 34 other items, was the source material for the construction of Table 48. Special collection data specific to Table 48 was collected retrospectively for the period June 2004 to September 2005. Table 48 was first published in October 2005.
Findings of primary judge
7 The primary judge referred to the well-known passage of Gibbs ACJ in Sankey v Whitlam (1978) 142 CLR 1, 38 for the general rule that the court will not order the production of a document, although relevant and otherwise admissible, if it would be injurious to the public interest to disclose it, but it is the duty of the court to determine whether the public interest which requires that the document not be produced outweighs the public interest that a court of justice in performing its functions should not be denied access to relevant evidence.
8 His Honour referred to authorities for the proposition that a public interest which gives rise to public interest immunity requires a dimension that is governmental in character. He went on to say in respect of the documents sought by the first respondent:
The contact report forms do not contain information that relates to decision making at the highest levels of government or is governmental in character. The information contained within the forms was collected solely for the purpose of constructing the Table 48 index. The Table 48 index was designed, constructed and produced by the ABS at the request of and for the purposes of the [PTA] and Main Roads. Neither the information in the contact report forms nor Table 48 are used by the ABS to construct other indexes. Table 48 is not used by the Australian government or any of its agencies. The information in the contact report forms or in Table 48 does not affect the proper functioning of the Australian government [50].
9 His Honour noted that the information had been disclosed to the ABS in confidence, but voluntarily, and that the ABS encouraged the voluntary provision of information as it has been found that data collected in that way was more reliable than data collected using the coercive powers of the ABS.
(Page 6)
10 His Honour concluded:
The disclosure of the contact report forms with the names and addresses of the respondents and other identifying material redacted to the counsel and solicitors for the parties, will not harm the functioning of government or otherwise harm the public interest.
The respondents provided the information in the contact report forms to the ABS interviewers. They were assured that the information would be kept confidential. Indeed, the Act requires that the information be kept confidential. However, the evidence before the court does not establish that if the court orders suitably redacted copies of the contact report forms to be disclosed to counsel and solicitors for the parties that will cause respondents to ABS surveys to less willingly co-operate with the ABS in providing information in the future.
There may be a public interest in securing the voluntary supply of information to statutory agencies even where there is a power under the statute which enables the agency to acquire the information compulsorily. However, that is not sufficient to give rise to a public interest immunity. In this case the disclosure of the limited information required by [Leighton Contractors], which excludes the name and address of the respondents, does not give rise to public interest immunity. The limited disclosure of the interview reports to counsel and solicitors for the parties is not likely to lead to a reduction in the quality of the data collected by the ABS, increase the costs of the ABS or threaten the quality of the statistics produced by the ABS [54] - [56].
11 The primary judge concluded that limited disclosure would not harm the public interest but if (contrary to his finding) limited disclosure was injurious to the public interest, the risk of harm was slight and was outweighed by the public interest in the court having access to relevant evidence.
12 The primary judge reached similar conclusions in respect of the second category of documents, the working documents.
13 The primary judge ordered that the two categories of documents should be produced, redacted to remove the names and addresses of respondents, their employees and other representatives, that one copy of each document should be produced to each of the first respondent and the PTA for inspection only by counsel and solicitors, that no copies of the documents were to be made, and that at the conclusion of the trial the documents were to be returned to the ABS.
14 It is against those orders that this appeal is brought.
(Page 7)
Grounds of Appeal
15 The grounds of appeal are in substance as follows:
(1) The primary judge erred in finding that the documents in issue do not contain information that relates to or affects government of decision-making or is governmental in character;
(2) The primary judge erred in considering whether harm to the public interest would actually arise as a matter of probability, rather than whether harm to the public interest could arise as a matter of real possibility;
(3) If the primary judge did conclude that there was no real possibility or risk of harm to the public interest, that finding was contrary to the evidence and in so concluding the primary judge failed to take into account the principle that, in assessing a public interest immunity claim, a court should give significant weight and respect to the views of deponents of affidavits filed in support of the claim;
(4) The primary judge erred in approaching the balancing exercise on the basis of a finding that the risk of harm to the public interest from disclosure to the parties' legal representatives was 'slight';
(5) The primary judge erred in approaching the balancing exercise without assessing the strength or weight of the public interest in favour of disclosure.
Ground 1 - The scope of public interest immunity
16 The first issue in the appeal was whether public interest immunity was capable of applying. The primary judge held that the documents the subject of the claim did not attract public interest immunity because the information contained therein was not governmental in character.
The statutory and factual context
17 The Australian Bureau of Statistics (ABS) was involved in the preparation of Table 48. The ABS is established by the Australian Bureau of Statistics Act 1975 (Cth) (ABS Act)(s 5(1)). The ABS Act also provides for an Australian Statistician (Statistician) (s 5(2)). The ABS consists of the Statistician and the staff of the ABS (s 5(3)). The Statistician controls the operations of the ABS and has such other
(Page 8)
- functions powers and duties imposed by or under any Act (s 5(4)). The functions of the ABS are set out in s 6 of the ABS Act. They include:
(a) to constitute the central statistical authority for the Australian Government and, by arrangements with the Governments of the States, provide statistical services for those Governments;
(b) to collect, compile, analyse and disseminate statistics and related information.
19 The Statistician is obliged to compile and analyse statistical information collected, and publish and disseminate the results of any compilation and analysis (s 12(1)). However, the Statistician is prohibited from publishing or disseminating information in a manner that is likely to enable the identification of a particular person or organisation (s 12(2)).
20 Section 13 of the Statistics Act relates to the release of information. It provides:
(1) Notwithstanding anything in this Act (other than this section), the Minister may, by legislative instrument, make determinations providing for and in relation to the disclosure, with the approval in writing of the Statistician, of information included in a specified class of information furnished in pursuance of this Act.
(2) Without limiting the generality of subsection (1), determinations may make provision:
(a) as to the persons to whom the information may be disclosed;
(b) as to the persons, being the persons from whom the information has been obtained, whose consent is required for the disclosure of the information; and
(c) specifying terms and conditions subject to which the information may be disclosed, including, but without limiting the generality of the foregoing, terms and conditions as to the requiring of a person to whom the information is, or is to be, disclosed to give an undertaking, in writing with respect to the disclosure of
- the information by that person, including an undertaking not to disclose any of the information to any person.
- (3) Information of a personal or domestic nature relating to a person shall not be disclosed in accordance with a determination in a manner that is likely to enable the identification of that person.
21 Section 19 is the secrecy provision. It provides:
(1) A person commits an offence if:
(a) the person is, or has been, the Statistician or an officer; and
(b) the person, either directly or indirectly, divulges or communicates to another person (other than the person from whom the information was obtained) any information given under this Act.
23 The reasons for the confidentiality provisions in the Statistics Act are explained in the second reading speech relating to the Census and Statistics Amendment Act 1981 (No 2) (Cth) (Amendment Act) which amended the Statistics Act. It states:
Preservation of the confidentiality of information provided to the Australian Bureau of Statistics is regarded by the Government as of the utmost importance. Moreover, it is central to the relationship of confidence and trust that exists between the [ABS] and those businesses and households which are asked to provide information for statistical purposes.
24 The second reading speech also notes that in making decisions under s 13, the Statistician will continue to be guided 'by an overriding concern to maintain the confidence and trust of respondents to statistical collections'.
25 For the purposes of these proceedings, no relevant distinction was drawn between the Statistician and the ABS. The Statistician/ABS publishes economic statistics including the consumer price index and producer price indexes. Users of economic statistics include Australian government agencies and departments such as the Reserve Bank of Australia and Treasury. Economic statistics are also used by the Treasury departments of the States and Territories and by non-government bodies such as financial institutions, business, academia and the general public.
(Page 10)
- Price indexes are freely available on the internet. Producer price indexes are used as short-term indicators of inflationary trends, for economic monitoring, as deflators in parts of the national accounts and for indexation in contracts (in the public and private sector).
26 The Statistician collects data by way of census or survey. A census directly measures all respondents in a population. A survey takes a sample of the population and the data collected is used to infer the characteristics of the population. Price indexes are compiled using surveys and are determined by the price of a basket of goods bought and sold. Data collected from respondents for producer price indexes includes commercial-in-confidence information such as pricing policy and total sales and purchases.
27 Table 48 is a producer price index and like all ABS price indexes was freely available for general use. However, as previously noted, it was prepared at the request of the PTA for use in its contract with the first respondent for the construction of the Perth component of the railway between Perth and Mandurah. The PTA is an agent of, and has the status, immunity and privileges as the State of Western Australia (Public Transport Authority Act 2003 (WA), s 6). Its main function is to provide and operate safe and reliable public passenger transport services (s 12).
28 There was no evidence that Table 48 was actually used by the Australian Government or its agencies. However, there is evidence that it was used by the Victorian Government.
29 The parties conducted the hearing below and the appeal on the basis that the Statistician and the ABS were acting intra vires in constructing and publishing Table 48.
Public interest immunity - Legal principles
30 The general law is that a court will not order the production of information or a document, although relevant and otherwise admissible, if it would be injurious to the public interest to disclose it: Sankey v Whitlam (38).
31 An objection may be made to the production of information or a document because it would be against the public interest to disclose its contents or because it belongs to a class of documents which in the public interest ought not to be produced: Sankey v Whitlam (39); Commonwealth v Northern Land Council (1993) 176 CLR 604, 616.
(Page 11)
- The parties accepted that public interest immunity cannot be waived: Air Canada v Secretary of State for Trade (No 2) [1983] 2 AC 394, 436.
32 Public interest immunity is not absolute. It has to be weighed in the balance with the competing public interest in the due administration of justice: Sankey v Whitlam (38 - 39).
33 A view has been expressed that public interest immunity only applies in relation to the conduct of governmental functions: R v Young (1999) 46 NSWLR 681, 693 (Spigelman CJ); Royal Women's Hospital v Medical Practitioners Board of Victoria (2006) 15 VR 22. In the latter case the Medical Practitioners Board of Victoria seized a patient's medical records kept at the Royal Women's Hospital pursuant to a search warrant. The hospital's claim that the documents were protected from production on the ground of public interest immunity was rejected at first instance and on appeal. Warren CJ held that public interest immunity did not apply because the documents in question were not governmental in character. Maxwell P held that the documents did not attract public interest immunity because the information in the documents was wholly unrelated to decision-making 'at the highest levels of government'. Charles JA held that public interest immunity was restricted to what must be kept secret for the protection of government at the highest levels and in sensitive areas of executive responsibility. He concluded that the holding of medical records by a hospital was not a governmental function and therefore did not attract immunity.
34 Maxwell P's opinion that the information must be related to decision-making 'at the highest levels of government' is inconsistent with authority, even in relation to class claims: Jacobsen v Rogers (1995) 182 CLR 572, 589 - 590. Our preliminary view is that the statement of Charles JA is consistent with the case law and is a correct general statement of the scope of the immunity. However, it is unnecessary to determine whether information must be connected with the performance of a government function and if so, the nature and degree of the necessary connection. On any view, there is a sufficient connection on the facts of this case as discussed below.
Analysis
35 The appellant objected to the production of the redacted information in the Documents on the basis of both its class and contents. In determining the connection between the information or documents the subject of a claim for public interest immunity and any function of government, particular attention must be given to identifying the relevant
(Page 12)
- public interest sought to be protected. The case law demonstrates that the public interest can be formulated at differing levels of generality (see Jacobsen v Rogers (589 - 590)). At its most specific, the appellant relied on the public interest in the confidentiality of the information supplied by survey respondents; or more precisely, the public interest in the protection of the identity of the respondents to whom the confidential trade information related. The uncontradicted evidence was that such confidentiality was necessary to ensure the voluntary cooperation of survey respondents, the reliability (accuracy and completeness) of the information provided and its provision in a timely manner.
36 There was uncontradicted evidence that disclosure of the redacted information in the Documents would directly identify the respondents and the confidential information they provided in relation to the nine items which formed part of the data for the construction of Table 48. That is the contents claim. The class basis for the claim is concern within the respondent community generally about ensuring the confidentiality of their information. In his affidavit of 25 September 2007 a Deputy Australian Statistician Mr Harper stated:
(22) Disclosure to the legal representatives, even on a confidential basis, carries with it a real risk of engendering significant concerns within the respondent community about the 'safety' of their sensitive trade information. Such concerns would not be limited to Table 48 respondents.
(23) A very large percentage of respondents require clear and unqualified reassurance that their information will be absolutely protected. The assurance of confidentiality given by the ABS to respondents is not a self-imposed precaution - it is a vital aspect of the ABS' ability to collect information and discharge the statutory functions given to it. The release of the redacted information even to legal representatives on a confidential basis would have a long-term and significant impact on the ABS' ability to give the level of assurance to businesses and householders which they require.
(24) … [T]he impact of not being able to give unqualified assurances of confidentiality would not be limited to Producer Price Indexes but would have a serious effect on all surveys undertaken by the ABS - including the Consumer Price Index … which [is one] of the most important macro-economic measures available to Government to manage the economy at a national level … [T]he importance of giving, and respecting, unqualified assurances of confidentiality is internationally recognised.
(Page 13)
37 As is evident from [50] of the reasons of the primary judge, he focussed on Table 48 as if it was a separate and distinct exercise of power that could and should be considered in isolation.
38 The primary judge erred in concluding that information in the interview reports was collected solely for the purpose of constructing Table 48. The uncontradicted evidence was that the interview reports recorded, in some instances, information given by respondents for the purpose of constructing another producer price index, Table 17. However, nothing turns on this error.
39 The approach of focussing on Table 48 is erroneous for a number of reasons. For the purposes of determining whether public interest immunity applies, there is no material distinction between Table 48 and other ABS economic statistics. The Statistician and the ABS provide statistical services for the Australian and (by arrangement) State governments. Table 48, like other price indexes, was such a statistical service for government. The fact that price indexes are generally available and known to be used for non-government purposes does not affect the position. The PTA, an agent of the State, requested the preparation of a replacement producer price index as part of the exercise of its main statutory function of providing public transport infrastructure which is a governmental purpose. However, we do not rely on this for the government function nexus. The effect of focussing solely on Table 48 is to ignore the class aspect of the claim for public interest immunity and the associated public interest in issue.
40 Before identifying that public interest, it is necessary to mention the general approach taken by the primary judge to the issue of whether public interest immunity applied to the information in the Documents. The primary judge had previously upheld a claim for public interest immunity for these and other documents: Leighton Contractors Pty Ltd v Public Transport Authority of Western Australia [2007] WASC 196. In the subsequent application in which the appellant's claim for public interest immunity was rejected (the subject of this appeal), the dispute between the parties had significantly narrowed to the question whether the redacted information in the Documents, which was capable of identifying the respondents who supplied it, should be the subject of limited disclosure, being to counsel and solicitors for the parties to the litigation.
41 In determining whether public interest immunity applied to information that would identify an individual survey respondent as being the source of the commercial-in-confidence information provided, the
(Page 14)
- primary judge confined his consideration by reference to the proposed limited disclosure to counsel and solicitors of the parties. The appellant contended that public interest immunity could not be avoided by the simple expedient of confining disclosure to the parties' lawyers. We agree that the initial question of whether documents or information attract public interest immunity is to be determined on the basis of unrestricted production or disclosure for the purposes of the litigation. Public interest immunity protects complete, not partial, secrecy. This is implicit in the rule that ordinarily there is no disclosure of the documents or information to the legal representatives of the parties even for the purpose of determining the objection to production on the ground of public interest immunity: Commonwealth v Northern Land Council (619); Alister v The Queen (1984) 154 CLR 404, 469; Jackson v Wells (1985) 5 FCR 296, 307 - 308; Relationships Australia v Pasternak (1996) 20 Fam LR 604, 614 - 616. However, in our view the effectiveness or appropriateness of disclosure limitations of the type imposed in this case may be relevant factors to consider at the balancing stage of the process.
42 Returning to the public interest associated with the class claim, the evidence established that unrestricted disclosure by the Statistician or ABS of the redacted information for the purposes of the litigation would, or had significant potential to, seriously damage the proper workings of the Statistician and ABS in the collection of reliable raw data in a timely fashion for all price indexes (not just Table 48). Such an outcome would impair the efficient or proper functioning of the Australian Government in the management of, inter alia, the economy. The evidence relating to the need for maintaining confidentiality is consistent with the statutory framework in which the Statistician and ABS operate. The importance of confidentiality in establishing and maintaining the trust and confidence of the pool of actual and potential respondents is the basis for the onerous confidentiality requirements in the Statistics Act. On any view of the proper scope of public interest immunity, the appellant's class claim has a sufficient connection with governmental function. We would uphold ground of appeal 1.
Grounds 2 and 3 - Potential for harm
43 The appellant contended (ground 2), relying on [54], [55], [56], [58], [59] and [69] of the reasons, that the primary judge concluded that public interest immunity was not available unless harm to the public interest would actually arise (as a matter of probability). The appellant contended that immunity was available when harm to the public interest could arise (as a matter of real possibility).
(Page 15)
44 The appellant contended in the alternative (ground 3) that if the primary judge's reasons should be construed as finding that there was no, or no real, possibility or risk of harm to the public interest, the findings involved errors of fact or law.
45 What is not in doubt is that the primary judge determined the question of the applicability of public interest immunity by reference to the proposed limited disclosure to counsel and solicitors for the parties. On that basis he held that public interest immunity was not available and thus avoided the balancing exercise. We have already concluded that he erred in taking this approach.
46 We accept the appellant's submissions that a claim of public interest immunity is properly invoked so as to require the court to undertake the balancing exercise if disclosure of the information in question gives rise to a real risk of harm to the identified public interest relied on, which in this case was the risk that disclosure of the confidential information would impair the responsiveness (in terms of reliability and timeliness) of survey respondents and thus impair the proper workings of government agencies. The incurring of the identified risk is itself injurious to the public interest. That is consistent with the authorities, including Sankey v Whitlam (39); Rogers v Home Secretary [1973] AC 388, 410 - 411; Alfred Crompton Amusement Machines Ltd v Customs and Excise Commissioners (No 2) [1974] AC 405, 434.
47 However, we are not persuaded the primary judge made the error. Even if he did, it would not have affected the result. On our reading of the primary judge's reasons he found that, with orders limiting disclosure of the information to counsel and solicitors of the parties, there was no, or no real, risk of harm to the public interest or alternatively, the risk was slight. As previously noted, the option of limiting disclosure in that way is relevant when weighing and balancing the competing public interests.
48 However, it is convenient to deal at this stage with the appellant's challenge to the finding as to the level of risk associated with limited disclosure. The primary judge ordered that the Documents be produced for inspection by counsel and solicitors acting for each of the parties, which Documents could be redacted (edited) to remove only the names and addresses of the respondents, employees of the respondents and other representatives of the respondents. The evidence established that other information in the Documents was capable of identifying the respondent who provided the confidential information. In particular, the evidence was that the detailed product descriptions in the Documents would enable
(Page 16)
- identification of respondents, especially by building industry insiders. It is also necessary to bear in mind the particular context in which the order was made. The Documents contained commercial-in-confidence information supplied by Perth producers of building products and one of the parties to the litigation, the first respondent, is a large consumer of products of that type in this State.
49 A panoply of orders can be made in the litigation process to protect relevant and admissible confidential information. In formulating appropriate orders, it needs to be borne in mind that the confidential information will or may need to be communicated not only to lawyers for the parties but witnesses, including expert witnesses, and will or may be referred to during the trial and in the reasons for judgment. The usual confidentiality orders would limit disclosure to specified persons or classes of persons (lawyers, experts etc) and impose confidentiality obligations on the recipients of the information. Steps would also need to be taken to suppress confidential information adduced in evidence at trial and in reasons for judgment. The intention of those orders is to limit the scope of disclosure and seek to limit the wider (and uncontrolled) dissemination of the information. That intention can be frustrated at a number of levels.
50 There is the risk of inadvertent disclosure by counsel or solicitors to their client and its witnesses, including expert witnesses. This risk was recognised in Jackson v Wells (307 - 308). There is also the risk of inadvertent communication during the course of the trial of confidential information which unknown to the lawyers is capable of identifying the respondents. This risk is significant in this case. The persons with the information (counsel, solicitors and the court) may not have sufficient knowledge to accurately assess whether particular information is capable of identifying a respondent.
51 There is the potential for other difficulties. Counsel and solicitors may from knowledge gained in preparing for trial or from their own investigations be able to identify the respondents who provided the confidential information. There is nothing in the orders made by the primary judge which would prevent the lawyers from subpoenaing the respondents to adduce any relevant evidence, oral or documentary.
52 Confining production of information to counsel and solicitors (even with associated orders which were not made in this case) is not a complete or necessarily satisfactory protection of the public interest in the confidentiality of the information. Further, we are not here concerned
(Page 17)
- solely with the public interest in preserving confidentiality. There is the further public interest in maintaining the trust and confidence of respondents and the proper and efficient workings of the Statistician and the ABS.
53 Mr Harper gave evidence (set out earlier) that disclosure of the information in question to legal representatives entailed a real risk of harm because of perceptions of survey respondents. Mr Harper's evidence is consistent with the second reading speech for the Amendment Act which emphasises that preservation of confidentiality is central to securing and maintaining the trust and confidence of survey respondents. There is nothing in the circumstances of this case to undermine Mr Harper's assessment of the risk. To the contrary, the fact that confidential information identifying the respondents would be disclosed to the legal representatives of a large consumer of products of the type produced by the respondents for use in the course of a trial is likely to reinforce the basis for concern in the respondent community. Although limited disclosure may in appropriate circumstances reduce the risk of harm to an extent that materially influences the balancing exercise, this is not such a case.
54 We are satisfied that there is no evidentiary foundation for the primary judge's finding that there was no, or only a slight, risk of harm to the public interest. The evidence established that there was a real risk of significant harm to the proper workings of the Statistician and the ABS. We would dismiss ground 2 but uphold ground 3. However, that is not the end of the matter; it remains necessary to weigh the competing public interests.
Grounds 4 and 5 - Balancing exercise
55 We have concluded the primary judge erred in finding there was no or only a slight risk of harm in disclosing the contents of the Documents to counsel and solicitors for the parties. We therefore uphold ground 4.
56 Having concluded the primary judge made material errors of law and fact in his decision to reject the appellant's claim that the redacted information in the Documents attracted public interest immunity, it is open to this court to itself determine the appellant's claim for public interest immunity, having the necessary materials to do so. In these circumstances it is unnecessary to deal with ground 5 which relates to the primary judge's approach to the balancing exercise.
(Page 18)
57 The importance of the redacted information to the conduct of the litigation is a relevant factor. The appellant accepted that the redacted information was relevant to the issues in the litigation. However, it is not centrally relevant to the pleaded issues. A number of claims depend upon the proper construction of the contract. Those claims are unaffected by the redacted information. The only claims to which the information may relate are contained in par 16(bb) and (e) of the Third Further Amended Substituted Statement of Claim which are in terms:
(bb) there were errors in the design, construction and methodology used to create [Table 48];
Particulars
Particulars of the errors are provided in an email from Matthew Berger, Director of Producer Price Indexes of the Prices Branch of the ABS, to Geoffrey Neideck, Assistant Statistician of the Prices Branch of the ABS, dated 23 May 2007;
Further particulars may be provided after further discovery is given by the ABS.
…
(e) it was not calculated using data collected at or about the period to which that data related, but was instead calculated using data collected well after that period.
Particulars
The data collected for 'special collection' items for the period from the September Quarter of 2004 to the September Quarter of 2005 was collected in the second half of 2005.
58 There is no evidence that the errors referred to in Mr Berger's email involve or relate to raw data collection. Further, the first respondent does not rely on raw data errors to draw an inference of systemic or methodological error. To the contrary, the first respondent relies on methodological deficiencies to inferentially cast doubt on the reliability of the data. However, we accept that the Documents may (or may not), by reference to the presence (or absence) of information, assist in challenging the evidence of ABS witnesses as to the actual methodology implemented.
59 The first respondent relies on a number of factors which it contends weigh in the balance in its favour. They are first, the information was volunteered rather than provided under direction; secondly, there was no evidence that the particular respondents to the nine special collection
(Page 19)
- items for Table 48 were given an assurance that the information would be absolutely confidential; thirdly, as the information was provided voluntarily without an express assurance of absolute confidentiality, the respondents in relation to the nine items would have an expectation the information may be disclosed pursuant to compulsory court procedures; fourthly, the information sought by the first respondent was historical information; and finally that Table 48 was, to the appellant's knowledge, provided for the purpose of the contract between PTA and the first respondent not for general government purposes.
60 It is not significant that the survey respondents did not have to be directed to provide the requested information. The uncontradicted evidence was that the use of compulsory powers detrimentally impacted on the reliability and timeliness of survey responses. Further, the evidence was that the respondents who provided raw data on the nine items were assured that the information would be kept confidential. We do not accept that the respondents would reasonably understand or expect that the confidential information would be provided under compulsory court procedures particularly to a large consumer of products of the type they produced in the same geographic market. In any event, it is the perception of the respondent community as a whole that is relevant. Further, all information collected for producer price indexes is historical to one degree or another. We were not persuaded that pricing and other confidential trade information from June 2004 would be of no current relevance or interest to product consumers or trade competitors. Finally, as discussed earlier, there is nothing about the background to the construction or purpose of Table 48 that justifies considering it in isolation.
61 As there was a real risk of significant harm to the public interest even with limited disclosure to counsel and solicitors of the parties we concluded that, having regard to all relevant factors including the degree of relevance of the redacted information to the issues in the litigation, the balance favoured the appellant's claim for public interest immunity.
28
13
4