Report 17 - Access to Information Review Task Force

THE NATURE AND STRUCTURE OF EXEMPTING PROVISIONS AND THE USE OF THE CONCEPT OF A PUBLIC INTEREST OVERRIDE

PART IV - ISSUES AND DISCUSSION

Prior to embarking on a discussion of whether or not there are changes to the Access to Information Act which could be made in order to create a legislative regime which is more generous in providing access to government information, it is important to note a number of extremely important features which the legislative approaches we have reviewed have in common.

Firstly, there is a clear recognition in all cases that, while there is a public interest in the disclosure of information held by the government, there are equally strong public interest reasons for protecting certain types of information from disclosure. Broadly speaking, information which does not belong to the government but, rather, has been received from a third party, either voluntarily or under compulsion, but in circumstances where the relationship with the third party is to be fostered, must be protected. It is important to protect not only the third party information but also the relationship with the third party. Accordingly, the legislative approaches are quite unanimous in recognizing that individuals and third party commercial entities have a proprietary interest in their personal information or commercial information and that it is in the interest of the government, meaning the public, to protect not only the relationship with the third party individual or commercial entity such that the state is able to continue to collect information, but also to protect that proprietary interest which, in the case of commercial entities, may have a substantial commercial value. In the case of personal information, the legislation which we have examined recognizes not only the proprietary interest which individuals have in their personal information but also attempts to recognize, on a broader scale, the general proposition that there should be protection of personal information and that the right of the individual to control, or at least limit, the dissemination of his or her personal information is now recognized as a human right.

In addition, the legislation consistently recognizes that governments do receive confidential information from other governments, whether they be domestic governments, such as municipalities or provinces, or allied foreign governments. Again, the access legislation which we have reviewed consistently recognizes that the relationship which allows for the candid exchange of information must be fostered. It also recognizes that there will be a number of circumstances where the information that is received from a third party government is in fact the proprietary information of that third party government.

There is also a consistent recognition that the government has a specific interest in matters such as law enforcement, international affairs and defence, government financial decision making and recognition that a certain amount of confidentiality is required to foster the candid exchange of ideas and information among the public service and between the public service and the executive in order to facilitate the government policy and decision making process. These are also public interests to be protected.

It is perhaps this latter area, that is the protection of information which relates to the public service and legislative policy and decision making processes, that is the most contentious. Few would disagree that information which, if released, would compromise a country's ability to combat organized crime ought to be protected. However, there are widely differing views about the amount of decision making information which can be released without compromising the decision making process itself. It is really with respect to this latter type of information that a critical analysis of various approaches to access to information suggests that there may be models which are more conducive to fostering a culture of openness and sharing of government decision making information than that currently expressed in the Access to Information Act.

Accordingly, the question that needs to be posed is whether the legislation as currently drafted, including in particular the structure of the exempting provisions, is conducive to the release of as much information as possible, or are there other approaches or structures which might work better?

Two approaches which are markedly different from that seen in the Access to Information Act and in the provincial Freedom of Information and Protection of Privacy legislation which we have reviewed are those which have been adopted by New Zealand and by the United Kingdom. The basis for withholding information from disclosure is couched in terms of "good reason for withholding it" (New Zealand) or the similar concept in the UK legislation of requiring, in most cases, that the "public interest in maintaining the exemption outweighs the public interest in disclosing the information." In both cases, the government is specifically put to the task of weighing the competing public interests - disclosure of the information vs. exemption of the information.

Both models still rely on mandatory exemptions for some information, particularly third party information such as personal information and third party commercial information. However, the preponderance of information, particularly information which relates to the public service and its decision-making process and its advice to the executive, is subject to the public interest balancing test. While, as we have pointed out above, the Access to Information Act appears to require a similar weighing of the public interest in disclosure as against the public interest in invoking an exemption through the use of the concept of discretionary exemptions, it would appear that the decision of a head of an institution is not usually predicated on the clear and direct balancing of these two interests which the New Zealand and UK models require.

The UK legislation is new, and there is no experience with its operation to draw on. However, there appears to have been a conscious decision to ensure that the public interest test is a distinct and identifiable stage in the process of deciding whether information should be released or whether an exemption should be claimed for all or part of it. This step, as expressed in the Freedom of Information Act, 2000, is clearer than the more general public interest override which appeared in the Code of Practice, which governed the release of government information before the enactment of this legislation. ⁽⁴¹⁾

New Zealand, on the other hand, has had almost twenty years of experience with its legislation.

By all accounts, the New Zealand model appears to be working well. In a paper presented at the conference "FOI and the Right to Know" in August of 1999, Sir Brian Elwood, Chief Ombudsman of New Zealand, had the following to say:

"It is realistic to conclude that New Zealand has now achieved across the total public sector as open a process as governance as any yet devised, which has established a right to request access to official information whilst preserving the capacity for a government to operate effectively. The right balance between apparently irreconcilable objectives, openness and secrecy has been found, and shown to work." ⁽⁴²⁾

Prof. Rick Snell appears to agree. In contrasting the success of the New Zealand model with the Australian model, he notes that the New Zealand model has been successful in being able to interact to a greater degree than the Australian with three dimensions of the government, the legal, the bureaucratic, and the political. The Australian model, on the other hand, was, he argues, conceived of, designed for, and largely operates within the legal framework. The political and bureaucratic elements of a successful freedom of information regime appear to consistently be subordinated to the strictures of the legal elements. ⁽⁴³⁾

To carry this analysis into the Canadian experience suggests that the Canadian model, being more akin to the Australian model, has probably also been less successful in engaging the bureaucratic and political elements of government than has the New Zealand model. The issue for the Task Force is to determine whether the New Zealand experience owes its success to the structure of its legislative model or whether the success is more attributable to a differing climate of government, a more accepting philosophy of open government and the positive leadership displayed by the Office of the Ombudsman and the Attorney General, who has taken on the task of ensuring enforcement of the Ombudsman decisions. ⁽⁴⁴⁾

Is the Canadian Legislation Sufficiently Clear that Access is the Rule?

The federal Access to Information Act follows the pattern of most of the similar legislation which is reviewed in this report. There is a clear statement in the introduction to the legislation that access is to be the norm and exemptions the exception. The Federal Court of Canada has recognized this principle and applied it in its review of decisions to exempt information. Nevertheless, there remains a perception that the Act is not being applied in as liberal a fashion as it might be and that information is being unnecessarily withheld. However, it is not completely clear that this perception is necessarily true. There are few reliable statistics which allow us to evaluate the operation of the legislation. The Office of the Information Commissioner records statistics regarding complaints, but complaints only represent a subset of requests. The data kept by departments is at the gross level only. We know how many requests were received and how many resulted in some or all of the information which was requested being released. These statistics, for 1999-2000 ⁽⁴⁵⁾ are actually quite encouraging. During that period, in 40.5% of requests, the requestor received all of the information which he or she had requested. In another 33.7% of the cases the requestor received some of the information requested. In 18.6% of the cases the request was not processed, either because there was insufficient information provided by the requestor, no records existed or the request was abandoned. In 2.3% of the cases the request was treated informally, and one expects to the satisfaction of the requestor.

The requested records were exempted in their totality in only 2.8% of the cases.

The most commonly applied exemption was Section 19 - personal information - which was used in 28% of exemptions. The second most common exemption was Section 20 - third party information - which was used in 24% of exemptions. The next most common exemption, however, was Section 21 - operations of government. ⁽⁴⁶⁾All other exemptions were applied less than 10% of the time.

But what do these statistics mean? The fact that requestors got all or part of the information they requested in 74.3% of the cases looks like a fairly positive statistic, particularly when 18% are effectively excluded from assessment of exemptions on the basis that they were not processed. Indeed, of the requests that were processed, over 90% resulted in some or all of the information which was requested being released.

This ratio is somewhat better than it was in 1995-1996 when only about 66% of the requests resulted in the requestor receiving some or all of the information requested. ⁽⁴⁷⁾ However, even then, only 3.1% of requests were met with all requested information being exempted.

However, the academics who have studied freedom of information legislation, have not been particularly complimentary to the Canadian experience. Alasdair Roberts of the School of Policy Studies, Queen's University has written,

Frequent users of the ATIA say that neglect and adversarialism are becoming more serious problems within the federal government, but there has been little analysis done to assess these concerns. This paper uses data contained in annual reports prepared by federal departments and agencies and finds evidence that neglect and adversarialism is becoming a more serious problem. On average, federal institutions take longer to respond to requests than they did five years ago. Requests are now less likely to result in full disclosure of information. Statutory provisions that authorize the withholding of information by departments are also used more often. The frequency with which complaints against federal institutions are upheld by the Information Commissioner has increased substantially. ⁽⁴⁸⁾

Professor Rick Snell, in a paper entitled Administrative Compliance and Freedom of Information in Three Jurisdictions: Australia, Canada and New Zealand, writes,

The level and type of non-compliance varies within and between the three jurisdictions. On a sliding scale of concern Canada would occupy the highest level of concern. Australia would occupy the next slot although displaying an increasing drift towards a general state of non-compliance. New Zealand would occupy the zone of least concern due to a number of factors contributing to compliance covered in the final section of this paper. These factors include the legislative architecture, history and nature of the freedom of information constituency in New Zealand. ⁽⁴⁹⁾

In his latest Annual Report ⁽⁵⁰⁾, the Information Commissioner identified 3 problems with the implementation of the legislation. Delay - many departments are unable, on a consistent basis, to respect the timelines for dealings with requests. Excessive Secrecy - the Act is all too often administered as a secrecy statute. The attitude is "if in doubt, keep it secret" and there has been little or no effort to put information into the public domain on a proactive basis. The Commissioner listed 5 causes for the persistent problems he has identified:

- inadequate resources;

- absence of targeted educational programs;

- poor procedures and practices (including the matter of poor information management);

- inadequate delegation to, and classification of, Access Co-ordinators; and

- slowness of Ministers/Deputy Ministers and senior managers to change the culture of secrecy by force of leadership.

He then goes on to identify 5 ways in which each of these problems can be addressed.

In addition, the Information Commissioner makes his own recommendations for reform of the legislation in his most recent Annual Report. These recommendations will be considered below. The question, however, is whether reform of the legislation is necessary or whether reform will enhance the legislation and address the criticisms that have been levelled against the legislation.

Accordingly, while there is considerable criticism of the Canadian legislation, there is little, if any, statistical data which would support the criticisms. The percentage of requests which are denied in whole or in part is not necessarily increasing nor are complaints to the Information Commissioner. However, the percentage of those complaints which are found to be unsubstantiated remains relatively stable and, perhaps more importantly, the percentage of complaints which the Commissioner manages to resolve continues to be very high. In his 2000-2001 Report, the Commissioner tells us that of 265 complaints based on refusal of access which he found to be substantiated, his office managed to resolve 263. All of the available statistical data must be considered very carefully. The fact that information was denied as a result of a request is not as important as knowing why it was denied and whether or not the exemption was applied in a fair and considered manner. Similarly, complaint statistics are difficult to interpret. We do not know how many requestors may have made multiple complaints. Some requestors file complaints automatically when they are denied information. Others may be unhappy with the response to their request, but have given up on the process and do not bother to file a complaint.

There is no comparative data which would allow a comparison of responses between countries such as Canada and New Zealand. Nor has anyone established benchmarks to evaluate compliance. Accordingly, it is difficult to evaluate the effectiveness of the current legislation and it is difficult to answer the questions that we have to ask ourselves. What would be effective legislation? Some information will always be exempted and should be exempted. How does one measure the effectiveness of legislation? Indeed, is there any mechanism which would provide an accurate comparative and benchmarked criteria to evaluate and compare the effectiveness of freedom of information legislation across jurisdictions?

However, the Commissioner's criticisms are legitimate. In particular, this writer agrees with the assessment that there has been a culture of secrecy and that many departments have not used the discretionary exemptions to facilitate the release of information. There has also been a lack of direction and leadership from the Government, particularly Treasury Board. There are few guidelines which attempt to set out the criteria and the factors to consider when assessing the public interest in disclosure of information as against the public interest in exempting it which is represented by the exempting provision. There has been little attempt made to develop methods of analysing information to determine whether, and to what extent, there is actually likely to be injury of the type contemplated by the exemption if the information in question were to be released. In this writer's view, the failure to take a systematic, proactive approach to the evaluation of information holdings, with a view to applying exemptions as sparingly as possible, is a failure which should be shared by the Government and the Information Commissioner. The relationship between the Commissioner and the Government is not a positive one at the moment and the current climate is not one which is conducive to promoting a culture of access as opposed to a culture of secrecy.

Accordingly, it is not clear that modifications or amendments to the legislation will be able to address the problems. There are, however, a number of changes which have been suggested by other commentators and which are worthy of consideration in light of the differing approaches that have been taken in other jurisdictions such as New Zealand and The United Kingdom. In particular, this paper will now examine whether there are changes that could be made to the use of class exemptions or if more exemptions could be expressed in terms of an injury. We will also look at the use of mandatory exempting provisions and whether there could be a broader provision for the release of otherwise exempt information in the public interest. Finally, this paper will consider the recommendations for reform outlined by the Information Commissioner in his latest Annual Report.

Are Class Exemptions Necessary?

In 1987 Open and Shut ⁽⁵¹⁾ the Standing Committee on Justice and Solicitor General recommended that, "each exemption contained in the Access to Information Act and Privacy Act be redrafted so as to contain an injury test and to be discretionary in nature". In its response, The Steps Ahead ⁽⁵²⁾ the government rejected the approach of making all exemptions injury-based, particularly "significant injury". It responded as follows:

The concept of significant injury also increases uncertainty about what information is or is not exempt. It is often a complex task to identify the reasonable likelihood of an injury. The significant injury concept would require even further refinement - an assessment of the degree of the injury. The resulting uncertainty could increase the risk that sensitive information which should be protected would be disclosed. It could cause an increase in the number of complaints to the Commissioners and applications to the court. Finally, the application of exemptions based on a significant injury test would be more time consuming and could result in processing delays. Results such as these would be counterproductive to the efficient and effective operation of the legislation, a goal to which the government is committed.

In his latest report, the Commissioner agrees that class exemptions are unnecessary except in two cases. ⁽⁵³⁾ He would agree with keeping section 19 - personal information and section 13 - confidences of other governments as class exemptions. He does not, however, agree that the test needs to incorporate the concept of significant injury.

Can Some Class Exemptions be Replaced with Injury or Harm Exemptions?

The Information Commissioner concedes that a class exemption is the appropriate approach for personal information protected by Section 19. He states that, making it [the personal privacy exemption] discretionary and subject to an injury test would radically alter the current balance between the Access to Information Act and the Privacy Act. Indeed, as can be seen from Chart 3 ⁽⁵⁴⁾ where provincial legislation addresses the concept of undue or unjustified invasion of privacy as part of the exempting provision, it is in the context of legislation which addresses both freedom of information and protection of personal information.

Australia addresses the exemption for personal information in terms of "unreasonable disclosure of personal information" about any person (including a deceased person) - Section 41. Similarly, in the United States the exemption under the Freedom of Information Act is expressed in terms of whether disclosure would constitute a "clearly unwarranted invasion of privacy" - Section (b) (6); in Ireland, the exemption is a class-based one - Section 28; in New Zealand the exemption is a class-based one - paragraph 9(2)(a). In the United Kingdom, separate legislation applies.

The writer agrees that the current structure of the Canadian legislation, with separate regimes for access and privacy should also continue to treat the exemption for personal information as a class exemption.

The second circumstance in which the Commissioner is prepared to consider keeping the class exemption is in case of information received in confidence from another government - Section 13. However, while he recognized the importance of respecting the confidences of other governments, he referenced the practice in the provinces of British Columbia, Alberta and Ontario which treat this information as an exemption which combines elements of class and injury by using the term "could reasonably be expected to reveal a confidence", and concluded that the exemption could be phrased as both an injury-based one and a discretionary one.

The writer does not agree with the Information Commissioner's categorisation of these exemptions in the three provinces referenced as being injury-based. They are class-based exemptions. The fact that a confidence of another government could or could reasonably be expected to be revealed is sufficient to invoke the exemption. It need not be demonstrated that revealing the confidence would result in any particular consequence or injury. Paragraph 33 (1) (b) of the Australian legislation expresses this exemption as a class-based one. Similarly, the other countries which we have reviewed exempt such information without requiring that any injury as a result of disclosure be demonstrated: Ireland - Section 24 (2); New Zealand - Section 6 (b); United Kingdom - Section 27 (2); and United States - Section (b) (1) (A) (by Executive Order). In fact, in all of these cases, the fact that the information was received in confidence, presumably with an expectation that it would be maintained in confidence, really seems to be the exemption. Indeed, injury to Canada's ability to have full and frank discussions and exchanges of information with foreign states and institutions and with the provinces and their institutions seems to be an inevitable consequence to the release of such information if the providing entity does not consent. There is also the consideration that the information belongs to the entity which provides it. ⁽⁵⁵⁾ In this writer's view, it would be difficult to express this particular exemption in any terms other than as a class exemption. That is not to say, however, as discussed below, that the exemption needs to be a mandatory one.

The other class-based exemptions in the Access to Information Act are discussed below. In most cases, the equivalent exemptions in other jurisdictions, particularly in the provinces, also tend to be expressed as class-based exemptions. ⁽⁵⁶⁾ A notable exception is policing information which is treated by the provinces as an injury-based exemption.

Subsection 16 (1), Paragraphs (a) and (b)

The head of a government institution may refuse to disclose any record requested under this Act that contains

(a) information obtained or prepared by any government institution, or part of any government institution, that is an investigative body specified in the regulations in the course of lawful investigations pertaining to

(i) the detection, prevention or suppression of crime,

(ii) the enforcement of any law of Canada or a province, or

(iii) activities suspected of constituting threats to the security of Canada within the meaning of the Canadian Security Intelligence Service Act,

if the record came into existence less than twenty years prior to the request;

(b) information relating to investigative techniques or plans for specific lawful investigations;

By and large, the provincial legislation which we examined treats policing information as an injury-based exemption. ⁽⁵⁷⁾ There would appear to be little basis for treating all information which falls into this category as automatically subject to exemption without the need to demonstrate any injury. This is not the approach at the international level where the exemption for policing information is usually based on the demonstration of some kind of injury or prejudice. While arguably the fact that the exemption is a discretionary one means that the exemption will not necessarily be invoked, reference to an injury is the more common approach and it is not at all clear that this approach could not be used in the Access to Information Act. As they have become more familiar with the legislation and as they have gained a level of comfort in their ability to identify information which can be released without any prejudice, government institutions, particularly the Royal Canadian Mounted Police, have tended to exercise the discretion to release more liberally. The time has probably come to make all aspects of the exemption for policing and law enforcement information predicated on the demonstration of injury or prejudice. Both the Information Commissioner and the Court have been willing to defer to the expertise of the government in assessing injury from the release of information relating to international affairs and defence as described in section 15 and there is no reason to think that this approach would not be carried over into an assessment of injury under paragraphs 16 (1) (a) and (b). The Information Commissioner has recommended that section 16 be based solely on the demonstration of injury. ⁽⁵⁸⁾ The writer agrees with that recommendation.

Subsection 16 (3)

The head of a government institution shall refuse to disclose any record requested under this Act that contains information that was obtained or prepared by the Royal Canadian Mounted Police while performing policing services for a province or municipality pursuant to an arrangement made under section 20 of the Royal Canadian Mounted Police Act, where the Government of Canada has, on the request of the province or municipality, agreed not to disclose such information.

This is a unique exemption which does not have any equivalent in either the provinces or other jurisdictions which have been canvassed. Neither is there any data which we are aware of to assist in determining the prevalence of the use of the exemption. One assumes that the equivalent information should be available, unless an exemption applies, under the applicable provincial statute on the basis that it is under the control of the Attorney General or the Solicitor General of the province in question. The exemption is no doubt phrased as a mandatory class exemption on the basis that the information in question really does not belong to the government of Canada, but rather to the province in question and there is a contractual obligation to keep it confidential. Neither the Standing Committee in its 1987 review or the Information Commissioner, in his most recent report, has commented specifically on this particular provision. Given the unique context of this exemption it is the writer's recommendation that it remain as a class exemption.

Paragraph 18 (a)

The head of a government institution may refuse to disclose any record requested under this Act that contains

(a) trade secrets or financial, commercial, scientific or technical information that belongs to the Government of Canada or a government institution and has substantial value or is reasonably likely to have substantial value;

Only this paragraph in Section 18 is expressed as a class exemption rather than as an injury-based exemption requiring that some harm be shown to a governmental financial interest. The Standing Committee commented on aspects of this paragraph in its report. ⁽⁵⁹⁾ It recommended, in its comments on Section 20, that the term "trade secret" be defined. It also made a recommendation that a specific provision for the release of product or environmental testing to mirror the provision in Section 20(2) be included in Section 18. In his 2000-2001 Annual Report the Information Commissioner has made a similar recommendation. ⁽⁶⁰⁾ He also recommended the addition of the word "monetary" to the concept of "substantial value". He does not elaborate on why he has made this recommendation.

The provinces, with the exception of Québec, tend to treat all exemptions relating to government financial interests as injury-based. Ontario also recognizes a class test in its equivalent to paragraph 18(a). The Ontario provision uses the term "monetary value or potential monetary value".

In this writer's view, a properly defined trade secret is something that is inherently valuable and it is an integral part of the identification of a trade secret that it is something that is kept confidential and which would result in some negative financial consequence if released. The difference between the federal statute and the equivalent provisions in the provincial statutes is really more a matter of what words have been chosen for the definition. If a trade secret, by its very nature, has a value which is dependent on the fact that it is secret, the very act of releasing it will diminish its value and cause some injury to the body that no longer has the advantage of the secret. That being the case, even though expressed as a class exemption, the exemption is really one that can be considered as an injury-based one. This is particularly so with the addition in this provision that the trade secret must have substantial value.

The changes recommended by the Commissioner to define "trade secret" and to add the word "monetary" would clarify the exemption. Otherwise no change is recommended.

Subsection 20 (1), Paragraphs (a) and (b)

Subject to this section, the head of a government institution shall refuse to disclose any record requested under this Act that contains

(a) trade secrets of a third party;

(b) financial, commercial, scientific or technical information that is confidential information supplied to a government institution by a third party and is treated consistently in a confidential manner by the third party;

Other than the comments discussed above regarding the addition of a more precise definition of "trade secret" the Standing Committee did not comment on the exempting provisions of Section 20. The Information Commissioner describes Section 20 as "one of the most used, abused and litigated exemptions". ⁽⁶¹⁾ He recommends the abolition of paragraph (a) trade secrets, and paragraph (b) relating to confidential information, on the basis that there is adequate protection in paragraph (c) to protect both types of information.

No doubt this observation is correct. However, it is not clear that modifying the exemption would have any impact on the practice of using the Access to Information Act as a tool for attempting to gain information about business competitors. Nor is it at all clear that an amendment would reduce the number of times that section 20 is invoked as an exemption, the time consuming efforts that the consultations required by sections 27 and 28 involve, or the number of times that third parties try to block or delay the release of information by making an application to the Federal Court.

The Federal Court has been very careful about when it will allow paragraph 20 (1) (b) to be used. It has developed a strict test which follows the test laid down by the courts in an action for breach of confidence, and has insisted on the objective confidentiality of the information. ⁽⁶²⁾ Merely marking information as "confidential" is not enough. Similarly, the subsection is not applicable just because the government and a third party decide that the information will be treated as confidential. In the writer's view, the Federal Court has been very careful to allow this exemption to be applied only in limited circumstances. It is the nature of the third party interests which are involved which leads to both the use and any abuse of this exemption. No change is recommended.

Section 21 (1)

The head of a government institution may refuse to disclose any record requested under this Act that contains

(a) advice or recommendations developed by or for a government institution or a minister of the Crown,

(b) an account of consultations or deliberations involving officers or employees of a government institution, a minister of the Crown or the staff of a minister of the Crown,

(c) positions or plans developed for the purpose of negotiations carried on or to be carried on by or on behalf of the Government of Canada and considerations relating thereto, or

(d) plans relating to the management of personnel or the administration of a government institution that have not yet been put into operation,

if the record came into existence less than twenty years prior to the request.

This section is the one that has caused much criticism of the Act. As we have seen, it is the section that is invoked most often to exempt information after the exemptions for personal information and third party information. The Standing Committee was of the view that the language of this section is far too broad, and made two recommendations:

That section 21 be amended not only to contain an injury test but also to clarify that it applies solely to policy advice and minutes at the political level of decision making, not factual information used in the routine decision-making process of government. The exemption should be available only to records that came into existence less than ten years prior to a request. ⁽⁶³⁾

The Information Commissioner, in his 2000-2001 Report, wants this section to be amended to address, as the legislation in Ontario and British Columbia does, types of information which are not covered by the exemption. Examples would be factual information and material, public opinion polls, statistical surveys, economic forecasts, environmental impact statements and reports of internal task forces. He also recommends that there be an attempt to define the term "advice". He is complimentary of the attempt contained in the Treasury Board policy manual. In addition, he recommends that the exemption should be clearly limited to communications to and from public servants, ministerial staff and ministers and that it should be subject to a public interest override. Finally he recommends that paragraph (1) (d) be amended to specify that plans which have been rejected do not qualify for exemption. ⁽⁶⁴⁾

The Commissioner does not explain how this section would be re-cast as an injury-based exemption. His recommendation is directed at making sure that the exemption is sufficiently constrained that the abuses which have concerned him will no longer be available. All of the provinces which we reviewed for this report, treat the exemption for advice and recommendations as a class-based exemption.

The writer tends to agree with the recommendations of the Commissioner but would leave the exemption as a discretionary, class-based one. The changes he proposes, coupled with a more defined approach to the application of discretionary exemptions, as discussed below, should go a long way toward addressing the current problems which have been identified with this exemption.

Section 23

The head of a government institution may refuse to disclose any record requested under this Act that contains information that is subject to solicitor-client privilege.

The Standing Committee recommended that Section 23 be amended and that the exemption should only be applicable where litigation or negotiations are underway or are reasonably foreseeable. ⁽⁶⁵⁾ The Committee was of the view that the exemption should only be available, "if the record sought would genuinely impair the confidential relationship existing between the lawyer and his or her governmental client".

The Commissioner recommends two amendments to Section 23. First, he is of the view that the exemption should be subject to an injury test. Secondly, he wants the legislation clarified to make it certain that application of the severance provisions in Section 25 will not result in a loss of privilege for the information which is exempted. ⁽⁶⁶⁾ He also recommended that this section be subject to a public interest override.

In all of the other jurisdictions that we canvassed for the purposes of this report, the exemption for information which is subject to solicitor client privilege is expressed as a class exemption and there is no requirement that an injury be demonstrated. The issue of whether or not any particular opinion should be released is addressed by the fact that this exemption is usually stated as a discretionary one.

In the writer's view, this exemption should continue as a class exemption. Again, perceived abuse or over-use of the exemption should be addressed as part of the process of clarifying and explaining the use of discretionary exemptions.

Section 24

The head of a government institution shall refuse to disclose any record requested under this Act that contains information the disclosure of which is restricted by or pursuant to any provision set out in Schedule II.

This section then goes on to provide that there was to be a review by the legislative committee established to review the legislation which was to report on whether and to what extent the provisions in the schedule are necessary. The schedule lists a number of statutory provisions which independently govern whether or not information which is under the control of the government can be released e.g. Income Tax information which is strictly controlled by provisions of the Income Tax Act. While the disclosure provisions of a number of statutes are included in the schedule, there are a number of provisions which one might expect to see included which have been excluded. It is not always clear what the rationale for including one provision and not another is. Clearly some mechanism is required to ensure that the disclosure requirements of the Access to Information Act do not undermine legitimate confidentiality requirements in other legislation. Section 24 is as good a mechanism as any. It should, however, be reviewed on a yearly basis to ensure its continuing consistency and relevance.

Cabinet confidences

The treatment of Cabinet confidence information by the Canadian federal government is unique and the approach in the Access to Information Act of excluding such information from the coverage of the Act is consistent with this overall treatment. When the Access to Information and Privacy Acts were enacted in the early 1980's, they were part of a package of legislation which included amendments to the Canada Evidence Act. ⁽⁶⁷⁾ These amendments set out a procedure for claiming general public interest privilege in documents and information when those documents and information were the subject of a subpoena - often in civil actions or in criminal proceedings. When the public interest privilege being claimed relates to matters involving international relations, national defence or security, the objection to disclosure of the information must be reviewed by the Federal Court of Canada no matter where the objection was brought. ⁽⁶⁸⁾ When the objection is made on the basis that the information or documents constitute confidences of the Queen's Privy Council - Cabinet confidences - there is no review of the certificate by which the objection has been filed. ⁽⁶⁹⁾ A certificate is filed by the Clerk of the Privy Council certifying that the documents constitute confidences and there is no ability on the part of the courts to look behind the certificate. The court may only comment on and review the form of the certificate. ⁽⁷⁰⁾ Section 39 displaces the common law approach to the review of a claim for Cabinet confidence privilege which is discussed by the Supreme Court of Canada in Carey v. Ontario. ⁽⁷¹⁾ The common law approach is to allow the courts to review the claim for Cabinet confidence privilege and determine, on a case by case basis, whether the public interest in protecting the information from disclosure is outweighed, in the circumstances of the particular case, by the public interest in disclosure of the information.

This restrictive approach is carried over into the access and privacy legislation. By virtue of Section 69, the Access to Information Act simply does not apply to Cabinet confidence records. This means, of course, that the Information Commissioner, like the courts, has been barred from even examining information for which the Cabinet confidence privilege has been claimed in order to satisfy himself that the claim has been properly made and that the information in the records does indeed meet the criteria of Cabinet confidence.

In most of the jurisdictions which we have surveyed, Cabinet confidence information is covered by the access legislation but is subject to a mandatory class exemption. The definition of Cabinet confidence tends to vary, particularly in terms of how long the absolute bar against release will prevail. However, the absolute bar, including the inability of either a commissioner or the court to review and validate the claim for Cabinet confidence exclusion from disclosure, is a unique Canadian federal legislative feature.

The Federal Court of Canada has recently indicated that it may become more aggressive in policing the use of the Cabinet confidence privilege claim, particularly in the context of the Act. In a decision released in April of 2001, the Trial Division discussed the question of Discussion Papers which are covered by Paragraph 69(1)(b). Discussion Papers are documents which are prepared for the purpose of presenting background explanation, analyses of problems or policy options to Council for consideration by Queen's Privy Council in making decisions. By virtue of Subsection 69(3), Discussion Papers lose their absolute immunity from disclosure a) once they become 20 years old; or b) if the decisions to which they relate have been made public, or if not made public, if four years have passed since the decisions were made. In this case, The Information Commissioner v. Minister of Environment and Ethyl Canada Inc ⁽⁷²⁾ the Information Commissioner argued that his investigation of the refusal to release the records in question and the claim for Cabinet confidence privilege had led him to conclude that the former content of Discussion Papers had been moved to other Cabinet confidences, primarily into the Analysis section of the Memorandum to Cabinet document. Accordingly, he was of the view that the refusal to disclose was not justified as it was based not on the content of the information but on the form of the record in which it was disclosed.

The Federal Court, Trial Division agreed that the privilege should be viewed in terms of the content of the information, not in terms of the format of the document in which it appears. Cabinet cannot organize its system of cabinet papers in such a way as to circumvent the intent of Parliament that Discussion Papers should be released in certain circumstances simply by eliminating the document known as a Discussion Paper and putting the information in some other form of document. ⁽⁷³⁾

The question is, however, whether the absolute exclusion of Cabinet confidences from the operation of the Act is really necessary. No one doubts that there is a very significant public interest involved and that Cabinet secrecy is to be preserved in most cases. However, other jurisdictions manage to get by quite nicely on the basis of the common law approach with review of a claim for Cabinet confidence privilege reviewable by the courts.

In Open and Shut ⁽⁷⁴⁾ the Standing Committee made a number of recommendations regarding the issue of Cabinet confidence which included narrowing the definition and the development of a procedure whereby such a claim could be properly reviewed by the Federal Court. The writer is of the view that these recommendations provided a sound basis for reform of this area and that the federal government has failed to articulate any sound reason for continuing the absolute exclusion of claims for Cabinet confidence privilege from Court scrutiny or for continuing to exclude records containing Cabinet confidence information from the operation of the Access to Information Act. ⁽⁷⁵⁾

Can Some Mandatory Exemptions be Replaced with Discretionary Exemptions?

As we have seen, the Access to Information Act only has five mandatory exemptions. In four of those cases the information is seen as belonging to another party and is not treated as federal government information. In the fifth case, Section 24 relating to information which is exempted by another statute, the theory would appear to be that access to the information is more properly dealt with in the context of some other statute, usually one which contains a separate and comprehensive regime for the collection and use of that information.

In the case of Section 13 - information received in confidence from another government, Alberta, British Columbia and Ontario have discretionary exemptions. The exemption in Manitoba is mandatory and Québec has exemptions which combine discretionary and mandatory elements.

Subsection 16 (3) - Provincial Policing information, does not have any equivalent in the provincial legislation.

Section 19 - Personal Information, is a mandatory exemption in all of the provinces.

Section 20 - Third party commercial information, is a mandatory exemption in all of the provinces.

Section 24 - referencing other statutory provisions does not have an equivalent in the other provinces. Québec has a somewhat similar provision in Section 29.1 dealing with decisions where disclosure is prohibited. That exemption is a mandatory one.

Where the legislation in other national jurisdictions has equivalent provisions, they are usually mandatory in nature for this kind of information, particularly personal information and third-party commercial information. New Zealand treats personal information and third-party information as information which can be withheld, but it does not include this type of information under the category of information for which there are "conclusive reasons for withholding" under section 6. Information received in confidence from other governments is addressed under section 6.

Generally speaking, the Canadian approach to the use of mandatory exempting provisions is consistent with the approach that has been taken in the equivalent provisions of provincial legislation and in the equivalent provisions of the legislation from other countries which we have examined.

In the writer's view, there is no reason why these exemptions should not continue to be mandatory. In particular, section 13 addresses information received from foreign governments, sometimes in circumstances where Canada is dependent on the confidence that foreign governments have in Canada's ability and willingness to safeguard this information. On balance there appears to be too much potential for loss of this trust and goodwill if the exemption were to be changed to a discretionary one.

Clearly there will be statutory regimes, such as that contained in the Income Tax Act, which will, and should, have their own comprehensive provisions regarding access to information and government obligations to safeguard information. We do not see any particular advantage as between the federal system of listing such statutory provisions in an exempting section of the Access to Information Act as opposed to providing that the access legislation prevails unless the other statute says otherwise.

In the writer's view, what is important is that a systematic and consistent approach be taken to ensure that information held under other statutory regimes is only exempted from the access legislation when absolutely necessary, that such exemptions be carefully circumscribed and that the interplay between the access legislation and other statutory regimes be reviewed on a regular basis.

Is the Discretion Being Properly Applied to Facilitate Access?

Theoretically, the fact that the majority of the exemptions in the legislation are discretionary ones, allows for the balancing of the public interest in disclosure as against the public interest in exempting information. The discretion would be exercised in favour of invoking the exemption only where it is necessary to do so in order to protect a public interest which would be harmed by the disclosure of the information. This balancing exercise is available whether the exemption is expressed in terms of an injury or in terms of whether or not the information in question is within the class described by the exempting provision. Indeed, the Treasury Board Policy on Access to Information ⁽⁷⁶⁾makes it clear that a discretionary exemption leaves the government institution with an option to disclose the information where, "no injury will result from the disclosure or where it is of the opinion that the interest in disclosing the information outweighs any injury which could result from disclosure". The policy does not use the term "public interest" just "interest". The term "public interest" appears to be reserved for the public interest which needs to be protected by the exempting provisions.

While the legislation has a mechanism for the balancing of interests, and while the policy does address the balancing, there are few guidelines to assist access co-ordinators in exercising the discretion and trying to achieve the appropriate balance. Hence the perception, probably an accurate one, that when there is any doubt about what to do the doubt is resolved in favour of invoking the exemption. For instance, the Guidelines in the Treasury Board Manual ⁽⁷⁷⁾ dealing with Section 23 - Solicitor Client Privilege, address situations in which the exemption should be used. The Guidelines do not do much to address those situations in which it would not be necessary to invoke the exemption because there would not be any injury resulting. By contrast, the Guidelines which address the application of the exemption in Section 21 - advice and recommendations, are more comprehensive and do direct the access co-ordinator to ask questions about the information, the governmental interest which is being protected and the impact that disclosure of the information might have.

Generally speaking, in this writer's view it is a fair comment that the discretionary exemptions in the Access to Information Act are not being applied in a way that facilitates access to information. Rather, the current practice is, by and large, to invoke the exemption if it is available.

As indicated above, the main reason why the decision whether or not to invoke a discretionary exemption is not more explicitly focussed on the question of whether the public interest in disclosure of the information outweighs the public interest in exempting the information from disclosure is a lack of direction and guidance on this issue. In the writer's view both the Treasury Board and the Information Commissioner could do more to develop practice directions and guidelines to assist Government Institutions in determining when to apply discretionary exemptions. For instance, the exemption in section 23 for solicitor client privilege is a good example of one where more guidance is needed. The current Treasury Board Guideline states as follows:

Section 23 of the Access to Information Act provides that a government institution may refuse to disclose any record that contains information that is subject to solicitor-client privilege.

This discretionary exemption ensures that communications between a government institution and its solicitors are protected to the same extent as is legal advice in the private sector. The privilege also extends to materials prepared by or for the solicitor expressly for the purpose of providing advice or presenting a case in court.

The decision to invoke solicitor-client privilege is the responsibility of the client. However, government institutions should consult their legal advisor prior to invoking solicitor-client privilege in order to determine if the information is in fact privileged and also before disclosing such information in order to ascertain if the disclosure could injure the government's legal procedures or positions. Institutions should be aware that waiver for part of the section 23 information may result in waiver for all of the related information that is subject to the privilege. This exemption should be used when the disclosure of information could:

(a) circumvent the normal process of discovery in cases presently before the courts; or

(b) prejudice the government's legal position in present or future litigation or negotiations; or

(c) impede the ability of government institutions to communicate fully and frankly with their legal advisors.

In Solosky v. the Queen, [1980] 1 S.C.R. 821, the Supreme Court of Canada identified three criteria required for the privilege to exist:

1. the communication must be between a solicitor and client;

2. it must entail the seeking or giving of legal advice; and

3. it must be intended to be confidential by the parties.

In addition, the Solosky decision confirmed that any consultation for legal advice, whether litigation is involved (or foreseen) or not, is protected by the privilege.

The Guideline should, however, address the following issues to help access coordinators understand the exemption so that they can make informed decisions and recommendations about whether or not to apply the exemption in the circumstances of a particular case:

• A description of solicitor client privilege, its origins and the public interest which the privilege addresses;
• Factors which are unique to the concept of solicitor client privilege in the public sector;
• Factors to be considered in assessing the harm that would result in the case at issue if the information were to be released;
• General indicators such as the age of the information, whether the issues addressed by the information are still live issues or likely to arise again which would affect the decision whether or not to disclose the information;
• What role the nature of the privileged information should play in the decision whether or not to release. Is the opinion being sought one that relates to a policy issue or does it relate to an issue which is specific to a particular piece of litigation.

These are the kinds of considerations that must be addressed when deciding whether or not to release solicitor client privileged information. ATIP Coordinators need guidance. Otherwise, if they do not fully understand the nature of the information and what, if any, the implications of releasing it will be, they are unlikely to recommend release.

The Public Interest

As we have seen, there are two approaches that are generally taken in the context of providing for a public interest override to require the release of information which would otherwise be exempt. The legislation can include specific public interest override provisions which apply to specific exempting sections. This is the approach in the Access to Information Act which has specific public interest provisions in Section 19 - personal information, and in Section 20 - third party information. These provisions usually relate to public health, public safety and environmental concerns. The Information Commissioner recommends that the public interest override as it applies to Section 20 should be broadened to include the public interest in matters such as consumer protection. ⁽⁷⁸⁾ This is a good recommendation as this consideration will not always be captured by the current wording.

The second approach is that which is found in the Ontario legislation, as an example. There we have a general public interest override which requires the proactive release of information, even if there is no request for it, in circumstances where there is a public interest in disclosure because the record reveals a grave environmental, health or safety hazard to the public. The Information Commissioner ⁽⁷⁹⁾and the Standing Committee ⁽⁸⁰⁾ have both recommended the inclusion of such an override provision in the Access to Information Act.

A third approach, the New Zealand approach, builds this public interest consideration into every decision, however, it is predicated on there being an access request and is not a proactive release requirement such as the Ontario provision.

There is also a more extensive approach, which would be to enact, either as part of the access legislation, or as a companion statute, legislation which positively requires governments to make certain categories of records public on a regular and ongoing basis, including records which contain information which it is in the public interest, as it relates to health, safety and the environment, to have released. For instance, certain types of government information, such as contracts, are regularly released, with the exception of some proprietary information, when they are requested under the Access to Information Act. Under this type of legislation, these contracts would be released proactively.

A proactive requirement would also have to address liability issues. Would the government be liable for any injury resulting if it failed to release public safety information? Probably such legislation should include a provision absolving government, its employees, servants or agents from any liability - whether arising as a result of a disclosure or as a result of a failure to disclose.

PART V - POTENTIAL RECOMMENDATIONS

Class vs. Injury or Harm-Based Exemptions

On the basis that the class-based exemptions in the Act are reasonably consistent with class-based exemptions in most of the other legislation canvassed, it may be that the Task Force would not make any recommendation to change these exemptions. In particular, if the Task Force is looking at a major revision of the legislation to strengthen the discretionary exemptions or to introduce a specific public interest provision to guide the release of information along the lines of what has been done in New Zealand, and more recently in the United Kingdom, it may not be necessary to re-cast the class exemptions as injury-based exemptions. Indeed, with exemptions such as that for personal information in Section 19 or solicitor client information in Section 23, it is not clear that the exemption could usefully be described as an injury-based one. On the other hand, the Information Commissioner's recommendation that paragraphs 16(1) (a) and (b) be eliminated and replaced with one injury-based exemption is a recommendation with which we agree and which would result in an exempting provision which is common to some of the other jurisdictions which we have reviewed.

On balance, the real concern with class-based exemptions is that they allow for the exemption of information in circumstances where there may in fact be no harm from the release of the information. However, because of the way in which the exemption is framed, there is nothing to prevent the exemption of such information, even in cases where the exemption is also a discretionary one. That being the case, it is the way in which the discretion is exercised that causes the real concern, not the fact that the exemption is described in terms of the class of information. The Task Force should focus on reforms which address the way in which the discretionary exemptions are applied.

Cabinet confidence

As is discussed above, there is no basis for continuing to provide that the Access to Information Act does not apply to Cabinet confidences. The Task Force should consider making recommendations for reform along the lines of the recommendations which were made by the Standing Committee. It needs to be recognized, however, that any reform to the Access to Information Act will need to involve similar reform to the certificate process currently provided for in the Canada Evidence Act, presumably to allow for the review of a certificate by the Federal Court in the same way that it currently reviews certificates for claims of public interest privilege based on international relations, national defence or security.

Mandatory vs. Discretionary Exemptions

In this writer's view, the most productive area for reform would be in a consideration of whether or not there are exemptions which are currently drafted as mandatory ones which could be recast as discretionary ones and whether there is a better way to ensure that the discretion is exercised more consistently in favour of the release of information rather than in favour of the exemption of information. As pointed out above, the use of mandatory exemptions in the Access to Information Act is not excessive and is fairly consistent with the approach in other jurisdictions. However, the application of discretionary exemptions is an area that is ripe for reform. Reform in this area could include the New Zealand approach and a clear expression in the legislation that records are to be released unless there is good reason shown for withholding them. Another approach would be to make the exercise of discretion a more formal process by which the head of an institution must not only state that information is being withheld on the basis of a specific exemption, but must also articulate why he or she has exercised the discretion in favour of exemption. In other words, the decision refusing relief would demonstrate how the public interest in disclosure was balanced as against the public interest in exempting the information.

To facilitate a more transparent exercise of the discretion, Treasury Board, perhaps with the co-operation of the Information Commissioner, could develop more detailed policies and procedures to guide access co-ordinators and heads of institutions in their balancing of the two public interests and exercise of discretion.

The Public Interest

In this writer's view, the Task Force should consider a two-pronged approach to dealing with both the issue of release of information that would otherwise be exempt because there is a compelling public interest in release and the issue of proactive release of information.

Firstly, the approach used in some provinces, such as Ontario, where there is an overriding obligation in all cases to release certain information in the public interest should be considered. However, given the second prong of this recommendation, the Ontario requirement that information be released, even if there has been no request for it may be unnecessary. The current provision in section 20(6) could be replaced with a provision along the following lines:

The head of a government institution shall disclose any record requested under this Act, or any part thereof, if that disclosure would be in the public interest as it related to public health, public safety, the protection of the environment or consumer protection if the public interest in such disclosure clearly outweighs the public interest in exempting such record as identified in the relevant exempting section or sections.

Consideration should also be given to investing the Federal Court with the power to overrule any decision that the head of a government institution might make regarding the balancing of these two competing public interests.

Secondly, consideration should be given to legislation which would mandate the release of information proactively, even if there has not been a request for the record. For instance, contracts which the government enters into must be released, with certain proprietary information and competitive commercial information exempted. Such legislation would mandate publication of such information within a specified time of the execution of the contract. The contractor would be aware, up front, that the contract is to be released and would work with the government institution to identify exemptions. Such an initiative would, no doubt, have a bit of a rocky start, however, as contractors adjusted their expectations, the release procedures would become more routine. The need for requests for this kind of information would decrease.

Clearly there are a number of other categories of government records which could also be made subject to this type of legislative provision.

Conclusion

In conclusion, it is the writer's view that there are four main areas in which the Task Force should consider making significant recommendations for reform of the provisions of the Access to Information Act dealing with exemptions. Most importantly, a new process has to be considered for the implementation of discretionary exemptions. With more guidance and with more thoughtful consideration of the factors that ought to be considered when deciding whether to invoke a discretionary exemption or not, it should be possible to construct a process whereby more information is made available and, when discretionary exemptions are claimed, there is a clearer understanding of why they have been claimed and that there really is a public interest in maintaining the exemption which outweighs the public interest in disclosing the information.

Secondly, specific reforms should be considered in respect of the exemptions for policing information and Third Party commercial information. Paragraphs 16(1)(a) and (b) and all of section 20 should become injury-based exemptions.

Thirdly, there should be reforms to provide for a much broader obligation to release information in the public interest, and particular consideration should be given to providing a general obligation to release certain types of government information on a regular proactive basis.

Finally, the time has come to revisit the whole area of Cabinet confidences. Cabinet confidences should not be excluded from the operation of the Act or from the investigative and supervisory jurisdiction of the Information Commissioner and the Courts.

Return to Table of Contents

Next Page

⁴¹ See, Select Committee on Public Administration, "Public Administration, Third Report", House of Commons, 1997-98. available online www.parliament.the-stationery-office.co.uk/pa/cm199798/cmselect/cmp…/39812.htm

⁴² The New Zealand model - The Official Information Act, 1982, a paper presented at the conference "FOI and the Right to Know" held in Melbourne on August 19-20. Sir Brian Elwood, CBE, D.Litt (Hon) LLB, Chief Ombudsman of New Zealand

⁴³ The Kiwi Paradox, Supra.

⁴⁴ The New Zealand model - The Official Information Act, 1982, Supra.

⁴⁵ Info Source Bulletin, 1999-2000

⁴⁶ Chart 6, Annex G

⁴⁷ See Annex F, Chart 5 tracing the number of requests processed and the rates of disclosure.

⁴⁸ Alasdair Roberts, Monitoring Performance by Federal Agencies: A Tool for Enforcement of the Access to Information Act, Working Paper, School of Policy Studies, Queen's University, March 1999 at p. 2.

⁴⁹ Paper presented to mark the celebration of Ireland's first anniversary of the operation of the Freedom of Information Act, 1997.

⁵⁰ Information Commissioner of Canada, 2000-2001 Annual Report, June, 2001

⁵¹ Open and Shut, Supra., note 18

⁵² Access and Privacy, The Steps Ahead, the Federal Government Response to "Open and Shut"

⁵³ Annual Report, Supra. at page 71

⁵⁴ Annex D

⁵⁵ The Steps Ahead, Supra. at page 39

⁵⁶ Chart 3, Annex D

⁵⁷ Chart 3, Annex D

⁵⁸ Annual Report, Supra. at page 73

⁵⁹ Open and Shut, Supra. at page 26

⁶⁰ Annual Report, Supra. at page 75

⁶¹ Annual Report, Supra. at page 74

⁶² See: Air Atonabee Limited v. Canada (Minister of Transport) (1989), 27 F.T.R. 194 (F.C.T.D.)

⁶³ Open and Shut, Supra. at page 29

⁶⁴ Annual Report, Supra. at pages 76 & 77

⁶⁵ Open and Shut, Supra. at page 29

⁶⁶ Annual Report, Supra. at page 77

⁶⁷ R.S.C. 1985, c. C-5, as amended.

⁶⁸ Section 38.

⁶⁹ Section 39.

70.

⁷⁰ Canada (A.G.) v. Central Cartage Co., [1990] 2 F.C. 641 (C.A.).

⁷¹ [1986] 2 S.C.R. 637.

⁷² (T-1125-99) April 2, 2001.

⁷³ This decision is under appeal to the Federal Court of Appeal.

⁷⁴ Open and Shut, Supra. at p. 29 - 33

⁷⁵ See Annex H for a list of the recommendations of the Standing Committee.

⁷⁶ Treasury Board Manual - Access to Information, Exemptions - General (Chapter 2-7)

⁷⁷ Ibid. Chapter 2-0

⁷⁸ Annual Report, Supra. at page 75

⁷⁹ Annual Report, Supra. at page 72

⁸⁰ Open and Shut, Supra. at pages 68 & 69