Tremblay v. Quebec (Commission des affaires sociales), [1992] 1 S.C.R. 952

Tremblay v. Quebec (Commission des affaires sociales), [1992] 1 S.C.R. 952

Commission des affaires sociales Appellant

v.

Noémie Tremblay Respondent

and

Minister of Manpower

and Income Security Mis en cause

Indexed as: Tremblay v. Quebec (Commission des affaires sociales)

File No.: 21651.

1992: February 27*.

Present: Lamer C.J. and La Forest, L'Heureux‑Dubé, Gonthier and Stevenson JJ.

on appeal from the court of appeal for quebec

Administrative law ‑‑ Natural justice ‑‑ Independence of members of Quebec Commission des affaires sociales ‑‑ Institutionalized consultation procedure ‑‑ Draft decisions systematically sent to Commission's legal counsel for verification and consultation ‑‑ Plenary meeting requested in certain cases by commissioners responsible for making decision or by president of Commission to discuss a given question ‑‑ Meetings held so as to arrive at a consensus: voting, taking of attendance and keeping of minutes ‑‑ Whether consultation process created by Commission consistent with rules of natural justice ‑‑ An Act respecting the Commission des affaires sociales, R.S.Q., c. C‑34, s. 10.

Administrative law ‑‑ Natural justice ‑‑ Appearance of bias ‑‑ Audi alteram partem -- Quebec Commission des affaires sociales ‑‑ Unanimous draft decision prepared by two commissioners present at hearing reviewed by Commission president who proposed a contrary opinion ‑‑ Convening of plenary meeting of Commission to discuss question of law raised ‑‑ Disagreement among commissioners responsible for making decision following meeting ‑‑ Question decided by president ‑‑ Whether active role played by president violates rules of natural justice ‑‑ An Act respecting the Commission des affaires sociales, R.S.Q., c. C‑34, s. 10.

Courts ‑‑ Administrative tribunals ‑‑ Confidentiality of deliberations ‑‑ Objections by Quebec Commission des affaires sociales to questions from social aid recipient concerning formal consultation process set up by Commission ‑‑ Whether objections to evidence based on deliberative secrecy should be dismissed.

Following the refusal of the Ministère de la Main‑d'{oe}uvre et de la Sécurité du revenu of Quebec to reimburse the cost of certain dressings and bandages, the respondent, who was receiving social aid, appealed this decision to the Commission des affaires sociales. The issue was whether the dressings and bandages came within the definition of "medical equipment" within the meaning of s. 10.04 of the Regulation on Social Aid. The appeal was heard by two commissioners and the parties argued in writing. At the close of the hearing, a draft decision favourable to the respondent was signed by the commissioners and sent to the Commission's legal counsel for verification and consultation in accordance with established practice at the Commission. As the legal counsel was on vacation, it was the president of the Commission who reviewed the draft. He then sent the two commissioners a memorandum in which he explained his contrary position. Further to this memorandum, and at the request of a commissioner, the point of law raised was submitted to the "consensus table" machinery of the Commission. At that meeting, a majority of members present expressed their disagreement with the position adopted in the draft decision and, shortly afterwards, one of the commissioners changed her mind and wrote an opinion unfavourable to the respondent. The commissioners were then divided on the question and the matter was submitted to the president of the Commission pursuant to s. 10 of the Act respecting the Commission des affaires sociales. The president decided the matter in the way he had already indicated to the commissioners in his memorandum. The respondent's appeal was accordingly dismissed. Alleging a breach of the rules of natural justice, the respondent challenged the Commission's decision by an action in nullity and asked that the "first draft decision" be declared the Commission's true decision. The Superior Court concluded that the Commission's decision contravened the rules of natural justice and allowed the action, but it refused to regard the first draft of the decision as the Commission's true decision. The Court of Appeal, in a majority decision, upheld the trial judgment.

Held: The principal appeal and the incidental appeal should be dismissed.

By the very nature of the control exercised over their decisions, administrative tribunals cannot rely on deliberative secrecy to the same extent as judicial tribunals. Secrecy remains the rule, but it may nonetheless be lifted when the litigant can present valid reasons for believing that the process followed did not comply with the rules of natural justice. In this case, the objections by the Commission to the questions raised by the respondent concerning the process for dealing with draft decisions within the Commission should be dismissed. These questions did not touch on matters of substance or the decision makers' thinking on such matters. They were directed instead at the formal process established by the Commission to ensure consistency in its decisions. The questions were concerned first with the institutional setting in which the decision was made and how it functioned, and second with its actual or apparent influence on the intellectual freedom of the decision makers.

The consultation machinery created by the Commission is not consistent with the rules of natural justice. While a consultation process by plenary meeting designed to promote adjudicative coherence may prove acceptable for an administrative tribunal, such a process must not however impede the ability or freedom of the members of the tribunal to decide according to their consciences and opinions, or create an appearance of bias in the minds of litigants. Here, the evidence depicts a system in which constraint seems to have outweighed influence. The "consensus tables" held by the Commission, although optional in theory, are in practice compulsory when the legal counsel determines that the proposed decision is contrary to previous decisions. Moreover, the rules for holding plenary meetings of the Commission disclose a number of points which taken together could create an appearance of bias. In particular, a plenary meeting may be requested not only by the commissioners responsible for making the decision but also by the president of the Commission. The mere fact that the president can of his own motion refer a matter for plenary discussion may in itself be a constraint on decision makers. Since the statute clearly provides that it is the decision makers who must decide a matter, they must retain the right to initiate consultation; if they do not wish to consult, they must be free not to do so. Compulsory consultation creates an appearance of a lack of independence, if not actual constraint. In cases of new subject‑matter, compulsory consultation circumvents the will of the legislature by seeking to establish a prior consensus by persons not responsible for deciding the case. There are other facts which support this conclusion of an apparent lack of independence. Plenary meetings of the Commission are held so as to arrive at a consensus: the members present vote by a show of hands, attendance is taken and minutes are kept, These mechanisms may exert undue pressure on decision makers and are not to be recommended. The Commission's decision, as a product of this system of internal consultation, thus seems to have been made in breach of the rules of natural justice. Certain aspects of the system established by the Commission create an appearance of "systemic pressure".

Even if the formal consultation machinery had been in keeping with the rules of natural justice, the fact that the president of the Commission expressed his opinion to the commissioners responsible for making the decision, inviting them to reconsider it, and then became a decision maker is hardly consistent with these rules. The Act respecting the Commission des affaires sociales gives the president the power to settle disputes but, in view of the active part he took in the discussion, he should have delegated this task to one of his vice‑presidents, pursuant to s. 10 of the Act. The active part played by the president in this matter is likely to create a reasonable apprehension of bias in an informed observer. Although the president had not heard the parties when he finally decided the matter, however, the procedure used in this case does not infringe the audi alteram partem rule. The question on which the Commission had to rule was a point of law and the parties pleaded in writing. There is nothing to indicate that new arguments of law were raised at the "consensus table" or that the president considered new points at the decision‑making stage. Since he in fact decided on the basis of the written file as prepared by the commissioners present at the hearing, there was no breach of the audi alteram partem rule.

The first "decision" rendered by the commissioners was in their minds only a draft, a provisional opinion, and cannot be regarded as the Commission's true decision. The intent of the decision makers must be analyzed in terms of the institutionalized consultation process that existed at the time the decision was made, even though that process now proves to have contravened the rules of natural justice. It is therefore the second "decision" which is the Commission's true decision.

Cases Cited

Applied: IWA v. Consolidated‑Bathurst Packaging Ltd., 1990 132 (SCC), [1990] 1 S.C.R. 282; referred to: MacKeigan v. Hickman, 1989 40 (SCC), [1989] 2 S.C.R. 796.

Statutes and Regulations Cited

Act respecting the Commission des affaires sociales, R.S.Q., c. C‑34, s. 10 [am. 1980, c. 33, s. 4].

Regulation on Social Aid, (1975) 107 O.G. II 6455, s. 10.04.

Social Aid Act, R.S.Q. 1977, c. A‑16.

Authors Cited

De Smith, S. A. Judicial Review of Administrative Action, 4th ed. By J. M. Evans. London: Stevens & Sons Ltd., 1980.

APPEALS from a judgment of the Quebec Court of Appeal, 1989 907 (QC CA), [1989] R.J.Q. 2053, 25 Q.A.C. 169, 42 Admin. L.R. 234, affirming a judgment of the Superior Court, [1985] C.S. 490, [1985] C.A.S. 153, quashing a decision of the Commission des affaires sociales, [1983] C.A.S. 713 (sub nom. Aide sociale ‑‑ 86). Principal and incidental appeals dismissed.

William J. Atkinson, Chantal Masse and Murielle Lahaye, for the appellant.

Paul Faribault and André Collard, for the respondent.

//Gonthier J.//

English version of the judgment of the Court delivered by

Gonthier J. -- The case at bar provides an opportunity for the Court to apply the rules already stated by it in IWA v. Consolidated‑Bathurst Packaging Ltd., 1990 132 (SCC), [1990] 1 S.C.R. 282, on so‑called "institutional" decisions. The Court must accordingly decide whether the decision of the appellant, the Commission des affaires sociales ("the Commission"), which refused to reimburse the respondent Noémie Tremblay for certain dressings and bandages was made contrary to the rules of natural justice. This decision of the Commission was the end result of an internal consultation process established by the Commission to ensure consistency in its decisions.

At the hearing, the Court dismissed the principal and incidental appeals from the bench with costs. The reasons that follow are in support of that disposition.

I -‑ Statement of Facts

At the relevant time, the respondent Noémie Tremblay was receiving social aid. The Ministère de la Main‑d'{oe}uvre et de la Sécurité du revenu denied her claim to be reimbursed for the cost of certain dressings and bandages. The respondent appealed this decision to the Commission in accordance with the Social Aid Act, R.S.Q. 1977, c. A‑16. This appeal is governed by the Act respecting the Commission des affaires sociales, R.S.Q., c. C‑34.

On January 20, 1983 the appeal was heard in the social aid and allowances division by a "quorum" consisting of a member of the Commission, Mr. Claude Pothier, and an assessor, Mrs. Dolorès Landry. The point at issue was whether the dressings and bandages came within the definition of "medical equipment" within the meaning of s. 10.04 of the Regulation on Social Aid then in effect. As this point was purely one of law, the parties proceeded by admissions and argued in writing. No witnesses were heard.

At the close of the hearing, Mr. Pothier undertook to draft a decision which he then sent to Mrs. Landry for comments and approval. This draft decision was favourable to the respondent. Mrs. Landry signed the draft, which was then sent to the Commission's legal counsel for verification and consultation in accordance with established practice at the Commission. As the legal counsel was on vacation, it was the president of the Commission, Hon. Gilles Poirier, who reviewed the draft. He then sent the decision makers a memorandum dated March 8, 1983 in which he explained his position, which was contrary to their own. On receiving this memorandum, Mr. Pothier asked that the point of law raised by the case be submitted to the "consensus table" machinery of the Commission. The respondent's case was accordingly placed on the agenda for the next plenary meeting of the Commission.

At that meeting, a majority of members present supported the viewpoint opposed to that originally taken by Mr. Pothier and Mrs. Landry. Shortly after this meeting, Mrs. Landry changed her mind and decided to write an opinion unfavourable to the respondent. As the quorum was thereby in disagreement, the matter was submitted to the president of the Commission, Judge Poirier, as required by the Act respecting the Commission des affaires sociales. Judge Poirier then decided the matter in the way he had already indicated to the decision makers in his memorandum of March 8, 1983. The Commission accordingly dismissed the respondent's appeal: [1983] C.A.S. 713 (sub nom. Aide sociale ‑‑ 86).

The respondent then challenged the Commission's decision by an action in nullity: she alleged a breach of the rules of natural justice. The respondent further asked that the first draft decision written by the members of the Commission who heard her appeal be declared the Commission's true decision. In the Superior Court, Dugas J. concluded that the Commission's decision contravened the rules of natural justice, but he refused to regard the first draft of the decision as the Commission's true decision: [1985] C.S. 490, [1985] C.A.S. 153. The Court of Appeal upheld the trial judgment: 1989 907 (QC CA), [1989] R.J.Q. 2053, 25 Q.A.C. 169, 42 Admin. L.R. 234. On the principal appeal, Jacques and Mailhot JJ.A. concluded that the Commission's decision was made in breach of the rules of natural justice, Monet J.A. dissenting; on the incidental appeal, Monet and Mailhot JJ.A. refused to regard the first version of the Commission's decision as the true one. Jacques J.A. differed on this point.

II -‑ Relevant Legislation

An Act respecting the Commission des affaires sociales, R.S.Q., c. C‑34

10. A matter shall be decided by the majority of the members and assessors having heard it.

When opinions are equally divided on a question, it shall be decided by the president or the vice‑president he designates.

III -‑ Judgments of Courts Below

Superior Court, [1985] C.S. 490

In the Superior Court, Dugas J. agreed with the respondent that there had been a breach of natural justice. He described the function of the legal counsel to whom the draft decision was sent as follows, at p. 494:

[translation] The legal counsel thus has a critical part to play in the verification process. He is asked to verify not only drafting errors, the accuracy of references to legislation and to regulations and of citations -‑ as to which little need be said -‑ but also whether the decision is in accordance with other decisions of the Commission. If the question is a new one, he attempts to determine whether a consensus can be arrived at based on the proposed rule...

The guidance given to the decision makers by the legal counsel gives cause for reflection. The fact that the decision maker is free to object to the legal counsel's suggestions does not take away the procedure's chilling effect. The same is true of the fact that the decision maker may ask for the matter to be brought before a general meeting.

The trial judge then made the following observations, at pp. 495‑96, regarding the "consensus table" procedure established by the Commission:

[translation] The reference to the general meeting is no less constraining because the decision maker does not necessarily have to accept it.

The question the decision makers had to decide was submitted to the other members of the Commission for their consideration. Of the 21 persons eligible to vote, three approved the decision makers' joint opinion, 13 rejected it and five abstained from voting. Faced with a majority like this, "clearly, the quorum may be somewhat more strongly influenced", as the president recognized.

Dugas J. concluded, at p. 496:

[translation] To the extent that the established procedure requires the decision makers to submit their opinions to systematic review by the legal counsel, and in some cases to their colleagues for approval, the procedure followed creates systemic pressure on the decision makers and interferes with their independence.

It cannot be argued that this is simply an internal consultation procedure: it is a compulsory consultation.

Accordingly, the court cannot find that the disagreement between the decision makers, which was necessary for the president to exercise his deciding powers, was the result of a valid decision‑making process.

At page 497 the judge acknowledged that the objective of this consultation procedure, that of ensuring adjudicative coherence, was valid but objected to the particular features of the system set up here:

[translation] This judgment should not be read as a condemnation of any internal consultation procedure in collegiate bodies. On the contrary, it is desirable that such consultations should take place freely whenever a decision maker feels the need to sound out a colleague's view. It should not be taken as suggesting that it is not possible for the members of a collegiate body to reach a consensus.

However, the court is of the view that to the extent that the consultations are compulsory and that they are meant to control the decision, the system imposing them does not provide a litigant with a "public and fair hearing by an independent and impartial tribunal" as guaranteed by s. 23 of the Charter of Human Rights and Freedoms.

He further noted, at pp. 496‑97, the fact that when the president of the Commission was called on to resolve the disagreement between the parties he had already ruled on the point in his letter to the decision makers:

[translation] The plaintiff, who has a right not only that justice shall be done but also that it shall be seen to be done, might well believe that the president had already decided the matter before assuming jurisdiction over it again so as to resolve the decision makers' disagreement.

The judge refused to recognize the first document as the Commission's true decision, noting at p. 497 that the decision makers [translation] "never regarded the joint opinion as their final decision".

Court of Appeal, [1989 907 (QC CA)](https://www.canlii.org/fr/qc/qcca/doc/1989/1989canlii907/1989canlii907.html), [1989] R.J.Q. 2053

The Court of Appeal dismissed the principal appeal brought by the Commission and the incidental appeal brought by the respondent Noémie Tremblay.

Jacques J.A. -‑ dissenting on the incidental appeal

Jacques J.A. concurred with the trial judge, at p. 2075, in his analysis of the internal consultation and verification process:

[translation] This procedure is more than a merely optional consultation, one which is purely voluntary and solely in the discretion of the decision maker, like that existing between judges of the same jurisdiction or between judges of different jurisdictions. It is compulsory. Each person must observe it when it is sought.

Although the vote is not binding on the decision makers, it is unwarranted pressure which deprives them of their intellectual independence and gives litigants the feeling that their case is being decided by persons other than their judges and for unknown reasons. It institutionalizes pressure. Clearly, any pressure is suspect from the outset.

As to the objection to the evidence on grounds of deliberative secrecy, taken under reserve at the hearing in the Superior Court, Jacques J.A. wrote that the trial judge was implicitly of the view that it was without basis. However, he went on, at pp. 2074‑75:

[translation] I consider that the deliberations of a judicial or quasi‑judicial tribunal or of a collegiate court are confidential. Research and discussion with colleagues on points of law are preliminary steps. This process, just as it may lead to a decision, may not lead anywhere. If it leads to a decision it is explained in that decision; if not, it is only preparatory. The process is part of the intellectual route taken to arrive at the decision: various rules of law are studied and rejected as not relevant, or adopted and applied with or without modifications to the problem under consideration. What is crucial in this process is that the judge or decision maker should act with intellectual freedom. This freedom is one aspect of impartiality, just like the absence of any interest in the outcome.

However, this confidentiality yields to application of the rules of natural justice, as observance of these rules is the bedrock of any legal system.

In exceptional cases, therefore, the confidentiality requirement may be lifted when good grounds for doing so are first submitted to the tribunal.

On the incidental appeal, Jacques J.A. would have regarded the first "decision" as that of the Commission, since this was the only one arrived at by a process consistent with the rules of natural justice.

Mailhot J.A.

Like Jacques J.A., Mailhot J.A. considered that the institutionalized consultation process imposed in the case at bar interfered with the independence of the decision makers. In particular, she noted at p. 2077 the part played by the president of the Commission:

[translation] Because of the special function conferred by the statute on the president of the body in the event of a disagreement between two members, I feel that he could not place himself in a situation where he was on his own initiative and by direct intervention raising a point that could create a disagreement between two previously unanimous decision makers. The reason is that, because of the function conferred on him by s. 10 of the statute as final decision maker responsible for resolving the impasse, the president would then be likely to decide the point raised at the final level (in the absence of extraordinary remedies, of course).

This is where the decision‑making process taken together with the "consensus table" does not comply with the rules of natural justice. When a litigant learns of such a situation he will, in my opinion rightly, feel that his case was decided by persons other than those who heard it and will have the impression that he was not treated in accordance with generally accepted rules of fairness.

Mailhot J.A. refused to regard the first "decision" as being that of the Commission. She would have referred the matter back to the Commission for the question to be decided again.

Monet J.A. -‑ dissenting on the principal appeal

Monet J.A. dealt first with the objections to the evidence made by the Commission at the trial. He considered that deliberative secrecy could only be lifted in exceptional circumstances. Looking at the statement of claim, he concluded at p. 2061 that none of its allegations justified admitting evidence of the facts relating to the deliberative process, and that the objection to the evidence should therefore have been allowed as soon as it was made.

On Dugas J.'s finding that the consultation procedure exerted "systemic pressure" on the decision makers, Monet J.A. felt there was no support for this in the evidence presented at the trial. [translation] "It is quite clear", he wrote, "that the consultation was not compulsory and that the vote did not constitute undue pressure" (p. 2065). After examining the testimony, Monet J.A. concluded at p. 2067 that he was:

[translation] . . . quite unable to conclude that the decision was invalid . . . on account of undue pressure, constraint or influence on the decision makers that would be likely to interfere with the independence which in our law they must enjoy.

Monet J.A. therefore considered that the Commission's decision did not conflict with the rules of natural justice and there was accordingly no need to decide on the nature of the first "decision". If he had had to do so, however, he would not have regarded the first draft as the Commission's true decision.

IV -‑ Points at Issue

Principal Appeal

1. Should the Superior Court have allowed the objection to the evidence made by counsel for the Commission and based on deliberative secrecy?

2. Does the machinery established by the Commission to ensure adjudicative coherence give rise to a reasonable apprehension of bias?

3. Is the part played by the president in the case at bar a breach of the rules of natural justice?

Incidental Appeal

4. Should document P‑10 (the first "decision") be regarded as the Commission's true decision?

V -‑ Analysis

I will deal with these four questions in order.

1. Confidentiality of Deliberations

At the trial, counsel for the Commission made several objections to the evidence based on the principle of deliberative secrecy. The Commission objected in particular to the Commission secretary answering the questions of counsel for the respondent on the process for dealing with draft decisions within the Commission (approval by legal counsel, discussion at plenary meeting, and so on). In his judgment, Dugas J. did not expressly deal with these objections; however, he dismissed them implicitly by ruling on the internal consultation procedure followed by the Commission.

In my opinion, the objections made by the Commission should be dismissed. The questions raised by the respondent did not touch on matters of substance or the decision makers' thinking on such matters. These questions were directed instead at the formal process established by the Commission to ensure consistency in its decisions. They were concerned first with the institutional setting in which the decision was made and how it functioned, and second with its actual or apparent influence on the intellectual freedom of the decision makers. This distinction was noted by Dugas J. during the interrogatories themselves.

In the case of administrative tribunals, the difficulty of distinguishing between facts relating to an aspect of the deliberations which can be entered in evidence and those which cannot is quite understandable. The institutionalization of the decisions of administrative tribunals creates a tension between on one hand the traditional concept of deliberative secrecy and on the other the fundamental right of a party to know that the decision was made in accordance with the rules of natural justice. The institutionalized consultation process involving deliberation is the subject of rules of procedure designed to regulate the "consensus tables" process. Paradoxically, it is the public nature of these rules which, while highly desirable, may open the door to an action in nullity or an evocation. It may be questioned whether justice is seen to be done. Accordingly, the very special way in which the practice of administrative tribunals has developed requires the Court to become involved in areas into which, if a judicial tribunal were in question, it would probably refuse to venture:

The judge's right to refuse to answer to the executive or legislative branches of government or their appointees as to how and why the judge arrived at a particular judicial conclusion is essential to the personal independence of the judge, one of the two main aspects of judicial independence: Valente v. The Queen, supra; Beauregard v. Canada . . . To entertain the demand that a judge testify before a civil body, an emanation of the legislature or executive, on how or why he or she made his or her decision would be to strike at the most sacrosanct core of judicial independence. [Emphasis added.]

(MacKeigan v. Hickman, 1989 40 (SCC), [1989] 2 S.C.R. 796, at pp. 830‑31.)

Additionally, when there is no appeal from the decision of an administrative tribunal, as is the case with the Commission, that decision can only be reviewed in one way: as to legality by judicial review. It is of the very nature of judicial review to examine inter alia the decision maker's decision‑making process. Some of the grounds on which a decision may be challenged even concern the internal aspect of that process: for example, was the decision made at the dictate of a third party? Is it the result of the blind application of a previously established directive or policy? All these events accompany the deliberations or are part of them.

Accordingly, it seems to me that by the very nature of the control exercised over their decisions administrative tribunals cannot rely on deliberative secrecy to the same extent as judicial tribunals. Of course, secrecy remains the rule, but it may nonetheless be lifted when the litigant can present valid reasons for believing that the process followed did not comply with the rules of natural justice. This is indeed the conclusion at which the majority of the Court of Appeal arrived, at pp. 2074‑75:

[translation] However, this confidentiality yields to application of the rules of natural justice, as observance of these rules is the bedrock of any legal system.

In exceptional cases, therefore, the confidentiality requirement may be lifted when good grounds for doing so are first submitted to the tribunal.

I would therefore dismiss this first ground of appeal.

2. Legality of the "Institutionalized" Decision‑making Process Established by the Commission

Of the four questions raised by this appeal, the second is clearly the central one. The Commission is arguing that the consultation machinery which it has created is consistent with the rules of natural justice. It describes this consultation machinery not as a compulsory process of consultation but rather as an "automated" process, the purpose of which is not to impose any particular viewpoint but to assist the decision maker by informing him of the existence of precedents.

The following internal directives are significant. They are taken from the Directives concernant le fonctionnement des réunions générales et la constitution du comité de lecture ([translation] "Directives on the holding of general meetings and creation of a reading committee") adopted by the Commission in September 1984:

[translation]

4.A unanimous quorum may also suggest that a given problem be discussed at a general meeting, whether the decision has already been issued or not.

5.The president may suggest discussion at a general meeting of a unanimous decision, in cases where:

(a)the principles stated in that decision or its application contravene or depart from a consensus or precedent decided on by the Commission;

(b)the decision is a ruling in principle on a new point, or develops a new interpretation which sets an important precedent for the Commission.

34.Discussions will generally develop as follows:

‑presentation by the persons concerned (members of the quorum or, if applicable, the legal counsel, a member of the reading committee, the president and so on) of the problem and arguments on either side, and this presentation shall be made without becoming involved in arguments on either side of the issue;

‑questions by the meeting to the authors;

‑additional comments by members and assessors on the point;

‑(possible roundtable of views);

‑brief final comments by the persons concerned;

‑ensuring that the meeting fully understands the question;

‑re‑reading of the question by the president;

‑vote by the meeting (show of hands).

35.A vote is not regarded as the necessary outcome of a discussion: it is only required if the points for consideration have validly emerged from the discussion and a sufficient number of those present feel well enough informed to make a decision.

39.A formal voting process or vote by show of hands is used in place of the "roundtable" process.

41.The president may decide, when he considers that the matter discussed is of great importance, to extend the consultation to members and assessors absent from the general meeting.

44.Any vote taken at a general meeting must be compiled and its result announced at the meeting and entered in the minutes.

45.Consensus is intended to ensure greater consistency in Commission decisions. It will be obtained following thorough discussion and by means of an unambiguous vote; but in the last analysis the quorum retains control of its decision.

46.The consensus is entered in the minutes of the general meeting, with reference to the decision it will be reflected in and the breakdown of the number of votes.

It is true that the system for verifying decisions established in the case at bar was created at the request of the decision makers themselves. In view of the large number of decisions made by the Commission (on the evidence, 2,871 decisions for 1983), members and assessors very soon felt the need to consult their colleagues to ensure consistent and carefully reasoned decisions. As the Commission noted, the objective of consistency responds to litigants' need for stability but also to the dictates of justice. As the Commission's decisions are not subject to appeal, it is the Commission itself which has the duty of preventing inconsistent decision‑making.

However, that does not mean that the actual structure of the machinery created to promote collegiality is unimportant. Clearly, by its very nature administrative law encompasses a wide variety of types of decision‑making. Nonetheless, these must be in keeping with natural justice: accordingly, they should not impede the ability of the members of an administrative tribunal to decide as they see fit nor should they create an appearance of bias in the minds of litigants.

In IWA v. Consolidated‑Bathurst Packaging Ltd., supra, the Court has already had occasion to state the guidelines for creating an "institutionalized" consultation process in administrative tribunals. In IWA, the Court examined the consultation process established by the Ontario Labour Relations Board ("OLRB") to promote consistency and quality in its decisions. In response to the immense task it faced (3,189 decisions in 1982‑83), the OLRB adopted a practice consisting of holding plenary meetings on important questions of policy. The process created was described as follows by the Chairman of the OLRB at pp. 316‑17:

After deliberating over a draft decision, any panel of the Board contemplating a major policy issue may, through the Chairman, cause a meeting of all Board members and vice‑chairmen to be held to acquaint them with this issue and the decision the panel is inclined to make. These "Full Board" meetings have been institutionalized to facilitate a maximum understanding and appreciation throughout the Board of policy developments and to evaluate fully the practical consequences of proposed policy initiatives on labour relations and the economy in the Province. But this institutional purpose is subject to the clear understanding that it is for the panel hearing the case to make the ultimate decision and that discussion at the "Full Board" meeting is limited to the policy implications of a draft decision. The draft decision of a panel is placed before those attending the meeting by the panel and is explained by the panel members. The facts set out in the draft are taken as given and do not become the subject of discussion. No vote is taken at these meetings nor is any other procedure employed to identify a consensus. The meetings invariably conclude with the Chairman thanking the members of the panel for outlining their problem to the entire Board and indicating that all Board members look forward to that panel's final decision whatever it might be. No minutes are kept of such meetings nor is actual attendance recorded.

Relying on this description, the Court weighed the advantages and disadvantages of the OLRB practice in holding such full Board meetings. Since any process of consultation may have the effect of "influencing" decision‑makers, the Court concluded at p. 333 that what should be looked at is not the question of influence but that of constraint:

. . . the relevant issue in this case is not whether the practice of holding full board meetings can cause panel members to change their minds but whether this practice impinges on the ability of panel members to decide according to their opinions. There is nothing in the Labour Relations Act which gives either the chairman, the vice‑chairmen or other Board members the power to impose his [sic] opinion on any other Board member. However, this de jure situation must not be thwarted by procedures which may effectively compel or induce panel members to decide against their own conscience and opinions.

Taking the position of an informed person, the Court held that the OLRB practice contained an adequate number of safeguards of the judicial independence of members of the Board responsible for making the decision. Accordingly, the system established was not likely to cause the litigant to entertain a reasonable apprehension of bias (at p. 334):

A full board meeting set up in accordance with the procedure described by Chairman Adams is not imposed: it is called at the request of the hearing panel or any of its members. It is carefully designed to foster discussion without trying to verify whether a consensus has been reached: no minutes are kept, no votes are taken, attendance is voluntary and presence at the full board meeting is not recorded. The decision is left entirely to the hearing panel. . . . On the other hand, it is true that a consensus can be measured without a vote and that this institutionalization of the consultation process carries with it a potential for greater influence on the panel members. However, the criteria [sic] for independence is not absence of influence but rather the freedom to decide according to one's own conscience and opinions.

In IWA, the Court further noted at p. 326 the frequent policy nature of the decisions of the OLRB and the tripartite makeup of the panels of that board as factors justifying an institutionalized consultation procedure. These factors are not conclusive, but are additional indications. A plenary meeting may perhaps be the only practical means of gathering members from various backgrounds; clearly, it does not mean that only tripartite agencies may set up such consultation machinery. We have seen that the justification for institutionalizing decisions lies primarily in the need to ensure consistency in decisions rendered by administrative tribunals. Whether the latter make decisions with a high policy component or not, those decisions must be consistent with the requirements of justice. A consultation process by plenary meeting designed to promote adjudicative coherence may thus prove acceptable and even desirable for a body like the Commission, provided this process does not involve an interference with the freedom of decision makers to decide according to their consciences and opinions. The process must also, even if it does not interfere with the actual freedom of the decision makers, not be designed so as to create an appearance of bias or lack of independence.

The institutionalized decision‑making process in the case at bar is rather different from that considered by the Court in IWA. Although the "consensus tables" held by the Commission are optional in theory, it appeared from the testimony of the member Claude Pothier that these collegiate discussions are in practice compulsory when the legal counsel determines that the proposed decision is contrary to previous decisions:

[translation]

A.. . . So this time, literally in order not to hold up the case, we sent -‑ it was the only way -‑ sent the Noémie Tremblay file to the discussion table.

Q.When you say ‑ in order not to hold up the case?

A.Well, listen, it is difficult because we were ‑ at least I for one was in a difficult situation in which the quorum had disposed of a matter which was not contrary to principle or the previous decisions of the Commission, because there were none, I think there was one and it was more favourable to the decision being taken than the other possibility. There was Judge Poirier's memo, which was in the file, so we could hardly not take it into account. The only way was to have a general discussion around the table to get the file moving, to move it forward.

Q.Could you have sent the original of document P‑10 to the Commission's secretary?

A.Listen, the administrative procedure did not authorize us to do that. I think I would have been squarely blamed if I had gone over their heads, especially the administrative office, and rendered a decision directly from my office; the discussion took place in my office at the time, and we decided to make this decision for a very good reason.

Q.I did not ask you whether you could have sent it directly to the parties; my question was whether you could have sent the decision to the Commission's secretary for it to be issued?

A.I could not do that either, because the procedure set up was compulsory, all files of whatever kind had to go through the legal counsel. So the only route was to send my file to my secretary who sent it on to the legal counsel's office, and then it went from there to be issued. If anything held it up, such as the legal counsel finding an inadvertent error in the citation of a regulation . . . the file was sent back to us . . .

Additionally, in other cases where the decision was contrary to earlier decisions of the Commission . . . or contrary to the consensus established around the table by my colleagues together, the legal counsel still held up the file, to my personal knowledge of the matter, and if the quorum did not change its opinion, the file to my personal knowledge -‑ in which I was involved in any case -‑ the files went either to the office of the president or the vice‑president and eventually came back to the general discussion table. That is the procedure in the Commission as I have known it for nine (9) years. [Emphasis added.]

Dugas J., who heard the parties and was therefore in a better position to assess the specific concrete aspects of the case, concluded from the testimony that there was undeniable "compulsory consultation" and "systemic pressure". In such circumstances, the fact that at the end of his testimony Mr. Pothier admitted that the vote taken at the plenary meeting had not prevented him from abiding by his decision in no way shows absence of constraint.

The Commission argued that under directive 45, the quorum still retains full control of its decision. As the Court observed in IWA, mere "influence" is to be distinguished from "constraint"; but what is crucial is to determine the actual situation prevailing in the body in question. In the case at bar on the facts do

[t]he methods used at those meetings to discuss policy issues reflect the need to maintain an atmosphere wherein each attending Board member retains the freedom to make up his mind on any given issue and to preserve the panel members' ultimate responsibility for the outcome of the final decision [?]

(IWA, supra, at p. 316.)

I do not think so. The testimony of the member Claude Pothier depicts a system in which in actual fact constraint seems to have outweighed influence, regardless of any internal directive to the contrary.

Additionally, reading the rules for holding plenary meetings of the Commission discloses a number of points which taken together could create an appearance of bias. In my opinion, the key indicator in this regard is to be found in directive 5, which provides that a plenary meeting may be requested not only by the quorum responsible for making the decision but also by the president of the Commission.

The fact that under directive 5 the president of the Commission can raise a question at a plenary meeting without the approval of the quorum responsible for deciding the matter presents a particular problem in light of the following passage from the judgment in IWA, supra, at p. 332:

It is obvious that no outside interference may be used to compel or pressure a decision maker to participate in discussions on policy issues raised by a case on which he must render a decision.

In my view, the mere fact that the president can of his own motion refer a matter for plenary discussion may in itself be a constraint on decision makers. In such circumstances, they may not feel free to refuse to submit a question to the "consensus table" when the president suggests this. Further, the statute clearly provides that it is the decision makers who must decide a matter. Accordingly, it is those decision makers who must retain the right to initiate consultation; imposing it on them amounts to an act of compulsion towards them and a denial of the choice expressly made by the legislature.

The Commission apparently wishes by this machinery to make the expertise of the Commission as a whole available to its members and to inform them of existing precedents. This is a praiseworthy motive. If the quorum has the advantage of the experience and opinions of its colleagues it may be in a position to render a more thoughtful decision. However, it is the quorum, and only the quorum, which has the responsibility of rendering the decision. If it does not wish to consult, it must be truly free not to do so. This constraint, which is subjective for the decision makers, may also cause litigants to have an impression of objective bias. Compulsory consultation creates at the very least an appearance of a lack of independence, if not actual constraint.

The referral process mentioned in directive 5 in cases of new subject‑matter also circumvents the will of the legislature by seeking to establish a prior consensus by persons not responsible for deciding the case. Ordinarily, precedent is developed by the actual decision makers over a series of decisions. The tribunal hearing a new question may thus render a number of contradictory judgments before a consensus naturally emerges. This of course is a longer process; but there is no indication that the legislature intended it to be otherwise. Bearing this in mind, I consider it is particularly important for the persons responsible for hearing a case to be the ones to decide it.

There are other facts which support this conclusion of an apparent lack of independence. For example, plenary meetings of the Commission are held so as to arrive at a consensus: a vote by a show of hands is generally taken, as well as attendance; minutes are kept (directives 34, 35, 39, 44 and 46). The process created by the Commission thus contains very few of the protective devices which led this Court to conclude that the practice of the OLRB was in keeping with natural justice. Such protective devices are important when, as here, what is at issue is also to determine whether there was an appearance of bias or lack of independence. Certain aspects of the system established by the Commission create at the very least an appearance of "systemic pressure", to use the words of Dugas J.

Accordingly the Commission's decision, as a product of this system of internal consultation, seems to me to have been made in breach of the rules of natural justice. The present practice of the Commission of holding plenary meetings without members of a quorum having requested them, as well as the voting procedure and the keeping of minutes, may exert undue pressure on decision makers. Such pressure may be an infringement of a litigant's right to a decision by an independent tribunal. I consider that the institutionalized consultation process currently being used by the Commission may also give rise to a reasonable apprehension of bias in an informed litigant.

I would accordingly dismiss the principal appeal for this reason.

It should not be concluded from all this that the Court does not regard the objective sought by the Commission, of ensuring adjudicative coherence, as important. On the contrary, it has already recognized the manifest benefits which may be obtained from an institutionalized consultation process. The Court is also aware of the breadth of the task which has been entrusted to the Commission by the legislature and of the difficulties which the Commission may face in performing these quasi‑judicial duties.

As it said earlier in IWA, supra, plenary meetings may be a consultation tool which is entirely in keeping with the rules of natural justice. However, they should not be imposed on decision makers and should be held in such a way as to leave decision makers free to decide according to their own consciences and opinions. Voting, the taking of attendance and the keeping of minutes are therefore not to be recommended. There are in any case a number of other methods which can be used to inform members of the Commission of applicable adjudicative trends or to prompt discussion on points of importance; the task of devising these may be left to the Commission.

3. Part Played by President in the Case at Bar and Appearance of Bias

Though this question does not have to be decided in order to dispose of the principal appeal, I will still make certain observations on the part played by Judge Poirier in the case at bar.

The Act respecting the Commission des affaires sociales gives the president of the Commission the power to settle disputes that may arise within a quorum:

10. A matter shall be decided by the majority of the members and assessors having heard it.

When opinions are equally divided on a question, it shall be decided by the president or the vice‑president he designates.

What part did the president of the Commission play here? In the case at bar, it is the president who raised the question by sending the quorum a memorandum in which he indicated the interpretation he would have given to the regulation at issue. This led to engaging the consultation process which eventually led to the disagreement between the two previously unanimous decision makers. Once the disagreement emerged, it was the president again who resolved the matter in the way he had indicated in his first intervention.

I should stress first of all that the Court is not in any way questioning the good faith or impartiality of the Commission's president in the case at bar; the question which concerns it here is one of an appearance of bias, not of actual bias.

I will leave aside for discussion purposes the question of the formal consultation machinery already considered above. Even if that machinery had been in keeping with natural justice, I feel that the fact that the president expressed his opinion to members of the quorum, inviting them to reconsider the decision, and then became a decision maker is hardly consistent with the rules of natural justice. The case is comparable to the typical "prior commitment" situation which de Smith describes as follows:

Disqualification for bias may exist where a member of a tribunal has an interest in the issue by virtue of his identification with one of the parties, or has otherwise indicated partisanship in relation to the issue.

Most of the cases are concerned with magistrates who have adjudicated after having substantially committed themselves by actively opposing or supporting the cause of a party or applicant before them.

(De Smith's Judicial Review of Administrative Action (4th ed. 1980), at pp. 270‑71.)

De Smith recognizes that by the very nature of their duties certain members of administrative tribunals may have to combine various roles. This is what he says at p. 271 about the function of members of boards who issue licences:

In certain respects, however, licensing justices stand in a peculiar position. They are entitled to adopt a general policy, based on their own investigations and private knowledge and their assessment of local requirements, with regard to the granting and renewal of public‑house licences in their area; they are sometimes members of the local authority, which may have adopted town planning policies that will affect the future location of public‑houses; they may themselves initiate an objection to the renewal of a licence and then proceed to sit and vote on the application as members of the compensation authority. But they are disqualified if they have already taken such active steps to oppose the renewal as to give rise to a real likelihood that they will not be capable of hearing and determining the application in a judicial spirit. [Emphasis added.]

The justices to whom de Smith refers combine these various duties as a result of the legislation giving them their powers. The demands of natural justice must therefore be reconciled with the deliberate intent of the legislature to give an administrative tribunal several overlapping duties. In the case at bar, the internal consultation procedure used by the Commission was not created by the legislature; and even if it had been, it does not contemplate the president taking control of cases in place of the legal counsel. There is accordingly less reason to tolerate the president playing several parts within the decision‑making process.

Moreover, s. 10 of the Act respecting the Commission des affaires sociales expressly authorizes the president to designate a vice‑president to resolve disputes between the members of a quorum. In view of the active part he took in the discussion, the president should have delegated the decision to one of his vice‑presidents. He did not do so. The active part played by Mr. Poirier in this matter thus seems to me likely to create a reasonable apprehension of bias in an informed observer.

The respondent further argued that the procedure used infringes the audi alteram partem rule in that the president did not hear the parties when he finally decided the matter. The Court has already considered this point in IWA. At pages 335 to 338 it emphasized the importance of distinguishing between discussions bearing on questions of fact and those relating to questions of law:

The determination and assessment of facts are delicate tasks which turn on the credibility of the witnesses and an overall evaluation of the relevancy of all the information presented as evidence. As a general rule, these tasks cannot be properly performed by persons who have not heard all the evidence and the rules of natural justice do not allow such persons to vote on the result. Their participation in discussions dealing with such factual issues is less problematic when there is no participation in the final decision. However, I am of the view that generally such discussions constitute a breach of the rules of natural justice because they allow persons other than the parties to make representations on factual issues when they have not heard the evidence.

Policy issues must be approached in a different manner because they have, by definition, an impact which goes beyond the resolution of the dispute between the parties. While they are adopted in a factual context, they are an expression of principle or standards akin to law. Since these issues involve the consideration of statutes, past decisions and perceived social needs, the impact of a policy decision by the Board is, to a certain extent, independent from the immediate interests of the parties even though it has an effect on the outcome of the complaint.

. . . It is now necessary to consider the conditions under which full board meetings must be held in order to abide by the audi alteram partem rule. In this respect, the only possible breach of this rule arises where a new policy or a new argument is proposed at a full board meeting and a decision is rendered on the basis of this policy or argument without giving the parties an opportunity to respond.

The question on which the Commission had to rule was clearly a point of law, namely whether "dressings and bandages were included in the definition of medical equipment" within the meaning of s. 10.04 of the Regulation on Social Aid then in effect. Furthermore, the parties chose to plead in writing and so far as one can tell made no representations at the hearing.

In the case at bar, there is no evidence that new arguments of law were raised at the "consensus table". The consultation process therefore did not infringe the audi alteram partem rule. Turning to the next stage, it also seems that no new points were considered by the president at the decision‑making stage. He in fact decided on the basis of the written file as prepared by the quorum. As the Court observed in IWA, supra, at p. 339:

. . . the rule with respect to legal or policy arguments not raising issues of fact is somewhat more lenient because the parties only have the right to state their case adequately and to answer contrary arguments. This right does not encompass the right to repeat arguments every time the panel convenes to discuss the case.

I therefore do not feel the facts of the instant case established a breach of the audi alteram partem rule. The Court moreover notes that the Commission has subsequently altered its practice and has taken the sensible step of giving parties an opportunity to be heard by the president or vice‑president responsible for resolving a disagreement. In the case at bar, the only blame which can attach to the president is thus of having resolved the disagreement between the decision makers when he had already spoken on the matter.

4. Nature of First "Decision"

I concur on this point with the disposition chosen by Dugas J., namely that the only decision in the case at bar is that contained in document P‑9 (the "second" decision, [1983] C.A.S. 713). Like Dugas J., I consider that the first "decision" rendered by the members of the quorum was in their minds only a draft, a provisional opinion.

In this regard, the intent of the decision makers must be analysed in terms of the institutionalized consultation process that existed at the time the decision was made, even though that process now proves to have contravened the rules of natural justice. The Court cannot disregard the setting in which the decision was made in deciding whether it was conclusive.

Finally, I would note that the procedure of early signature of draft decisions by members and assessors followed in the case at bar seems to me unadvisable. Although this procedure may be practical, it only adds to the appearance of bias when a decision maker decides to alter his opinion after free consultation with his colleagues. A litigant who sees a "decision" favourable to him changed to an unfavourable one will not think that there has been a normal consultation process; rather, he will have the impression that external pressure has definitely led persons who were initially favourable to his case to change their minds.

VI -‑ Conclusions

For these reasons, I would dismiss the principal and incidental appeals with costs.

Principal and incidental appeals dismissed with costs.

Solicitors for the appellant: McCarthy, Tétrault, Québec.

Solicitors for the respondent: Gauthier, Bergeron & Faribault, Magog.

• Reasons delivered April 16, 1992.