The High Court of Australia today, in Comcare v Banerji [2019] HCA 23 (7 August 2019), upheld as “reasonable”, and not unconstitutional, the decision of the Department of Immigration and Citizenship to dismiss an employee who had made anonymous political comments about migration matters and government policies, contrary to various codes of conduct. The case provides interesting insights into the operation of the implied constitutional freedom of political communication. Many will see similarities with the dismissal of footballer Israel Folau for comments he shared about the Bible’s view of morality, but as we will see, while somewhat factually similar, the cases raise quite different issues.

The Facts

Ms Banerji was an employee of the Department of Immigration and Citizenship. While so employed (usually after hours but also, on at least one occasion, while at work) she shared some 9000 tweets on the social media platform Twitter under the “handle” @LaLegale. Of those tweets, the plurality decision (Kiefel CJ, Bell, Keene and Nettle JJ) said:

many … were variously critical of the Department, other employees of the Department, departmental policies and administration, Government and Opposition immigration policies, and Government and Opposition members of Parliament.

Para [2]

Eventually her identity became known to the Department, and she was disciplined for breaching the Australian Public Service (APS) Code of Conduct, part of which required that an APS employee “(11)…at all times behave in a way that upholds the APS Values and the integrity and good reputation of the APS”. Guidelines explaining how this was intended to operate included that:

it was not appropriate for a Department employee to make unofficial public comment that is, or is perceived as, compromising the employee’s ability to fulfil his or her duties professionally in an unbiased manner (particularly where comment is made about Department policy and programmes) … “[a]s a rule of thumb, irrespective of the forum, anyone who posts material online should make an assumption that at some point their identity and the nature of their employment will be revealed”.

Para [17]

After having been officially reprimanded and been through relevant hearings, Ms Banerji’s employment was terminated. She then sought compensation from the Government for a psychological condition she said had been caused by the manner of her dismissal. The Government relied on a provision of the compensation legislation that said that no such claim was possible if the psychological harm was caused by “reasonable administrative action taken in a reasonable manner in respect of the respondent’s employment “. The issue at stake proved to be whether the rules under which she was dismissed were not “reasonable” because they breached the implied freedom of political speech under the Constitution- see [18].

Legal Issues

I have previously discussed this “implied freedom” in relation to other cases. It is an implication of the Australian Constitution, rather than an explicit “free speech” right as appears in other constitutional documents in Western countries. Indeed, the High Court is regularly at pains to point out that it is not a “personal right”, but rather a limitation on Commonwealth laws- see eg paras [19]-[20]. The plurality comment:

even if a law significantly restricts the ability of an individual or a group of persons to engage in political communication, the law will not infringe the implied freedom of political communication unless it has a material unjustified effect on political communication as a whole. (emphasis added)

Para [20]

To be frank, I am not sure whether this somewhat restricted view of the principle is completely shared by the other members of the court (Gordon, Gagelar and Edelmann JJ all gave concurring but separate opinions). But it didn’t seem to be crucial in this case- it was accepted that the APS code of conduct did have the effect of imposing a burden on the political speech of public servants in general- see eg [29].

Once legislation is seen to impose such a burden, then the current form of the test to determine whether it breaches the “implied freedom” was sent out as follows by the plurality:

The question is whether that burden is justified according to the two part test of whether the impugned law is for a legitimate purpose consistent with the system of representative and responsible government mandated by the Constitution and, if so, whether that law is reasonably appropriate and adapted to the achievement of that objective.

Para [29]

The second part of that test, “appropriate and adapted”, is to be determined by asking if the law is “suitable, necessary and adequate in its balance”- para [32]. Those issues are addressed by the following tests applied by the plurality:

(i) Suitability [33] A law is suitable in that sense if it exhibits a rational connection to its purpose, and a law exhibits such a connection if the means for which it provides are capable of realising that purpose…

(ii) Necessity...[35]…a law is not ordinarily to be regarded as lacking in necessity unless there is an obvious and compelling alternative which is equally practicable and available and would result in a significantly lesser burden on the implied freedom…

(iii) Adequacy in balance [38] If a law presents as suitable and necessary in the senses described, it is regarded as adequate in its balance unless the benefit sought to be achieved by the law is manifestly outweighed by its adverse effect on the implied freedom.

 Applying this carefully structured test to the provisions of the APS Code of Conduct, the plurality held that it was an important and central feature of the public service that it be “apolitical” (hence this was a “legitimate purpose”), and the rules in question had a rational connection to that goal; there were no obvious alternatives, and their benefit was not outweighed by the costs to the public servants, since there was a carefully structured set of possible penalties, such that not every minor breach would lead to automatic dismissal.

An “alternative” rule that Ms Banerji’s counsel put forward was one which allowed “anonymous” political content to be published by public servants. But the plurality noted that it would be very hard to define exactly whether a comment was “anonymous” or not, and in any event they ruled (following a remark noted above made in the guidelines) that: “anyone who posts material online, particularly on social media websites, should assume that, at some point, his or her identity and the nature of his or her employment will be revealed” (at [24]). 

There are subtle differences of approach in the separate but concurring judgments of the other members of the court, but in this brief note I can’t explore them in detail.

Implications for the Folau case?

Readers of this blog in particular will naturally see connections between this case and the issues surrounding Israel Folau, which have been noted previously here and here. Factually there are similarities: an employee dismissed after making controversial comments on social media, allegedly contrary to a code of conduct, and allegedly in part at least on the grounds that the comments would bring the employer into disrepute.

But that is pretty well as far as the connections go, legally. The Banerji case involves a public servant whose very duties required her to be involved in immigration policy; by contrast, Mr Folau’s job was to play football, and his comments did not relate to his primary duties. Ms Banerji was employed by the Commonwealth government, whose rules are clearly governed by the implied freedom from the Constitution; Mr Folau by a private organisation not subject to that restriction in the rules it makes for its employees.

In particular the legal basis for Ms Banerji’s claim was quite different to that which it seems has been put forward in Mr Folau’s case. Mr Folau is relying, not on protection for “political” speech, but on principles forbidding religious discrimination under s 772 of the Fair Work Act2009 (Cth). Despite the failure of Ms Banerji’s claim, Mr Folau’s case remains to be tested by the courts under quite different principles.

Other implications

That is not to say that the Banerji case may not be important for other religious freedom issues in the future. A recurring issue in this area is precisely the extent to which Parliaments in Australia may limit free speech through “vilification” laws. I have previously argued in some detail that religious vilification laws, and other vilification laws which do not recognise religious freedom, have a number of dangers and should be avoided.

One particularly objectionable law from this perspective is s 17 of the Tasmanian Anti Discrimination Act 1998, which forbids speech which “offends, humiliates, intimidates, insults or ridicules” on the ground of certain attributes, one of which is sexual orientation. It was this law which was used to sue a Roman Catholic archbishop in Hobart for teaching the doctrines of the Roman Catholic church on the matter.

It has long been argued that s 17 breaches the implied freedom of political communication under the Commonwealth Constitution (which has been held to apply to State laws as well as to Federal laws.) The analysis offered here in Banerji does nothing to weaken that case. While it might be argued by some that the law has the “legitimate purpose” of protecting same sex attracted persons from harm, it seems strongly arguable that the law as framed goes well beyond what is “necessary” to do so. In particular the law as framed does not seem “adequate in balance” as set out in Banerji, as the cost it imposes in chilling religious free speech can be said to be out of all proportion to the possible benefits.

Of course the two-fold test with its various sub-tests is to some extent “fuzzy” and its application in different cases not always clear. It may be some time before we see whether the High Court’s protection of free speech will be extended to religious views which have become “unorthodox”. (It sadly did not extend so far as to invalidate laws forbidding “buffer zones” restricting free speech around abortion clinics, as noted here.) But as has often been said, it is not until the views protected are unpopular, that laws on free speech become necessary.

For the moment, however, the result of Banerji will be to make public servants much more reluctant to comment publicly on their perceptions of problems with their Departments. Perhaps this ought to lead the Government to make sure that a decent anti-corruption body is in place to deal with possible malpractice and corruption in the Federal public service, where “whistle-blowers” will have a chance to seek change without being penalised.