Just a week before Anthony Albanese announced his government would introduce “world-first” legislation to ban teens from social media, there was still one big detail that hadn’t been decided: what ages were they banning?
Communications Minister Michelle Rowland was still telling people the government was “finalising” the age at which Australian kids would be banned from having accounts on TikTok, Instagram, Snapchat and other social media platforms, according to her talking points.
Meanwhile, Department of Communication staff were working under the assumption of a specific age limit — one that appears to be different to what the government ultimately chose. When the government later published this public servant-prepared analysis showing evidence supporting its law, little else had changed other than the age for the ban. Essentially, the government had kept the same working for its equation but now said it added up to a different answer.
This insight into the last-minute decisions in the government’s flagship tech policy, the Online Safety Amendment (Social Media Minimum Age) Bill 2024, comes from internal documents Crikey obtained through freedom of information requests, interviews with sources with knowledge of the process, and even details seemingly accidentally left in public documents.
While it’s common for legislation to change up until it has passed both houses, the eleventh-hour alterations have bolstered the criticism from academics, youth and mental health advocacy groups, and tech companies that the law was rushed through and its consequences were not fully fleshed out before it was passed.
A different minimum age
In response to Crikey’s questions, a spokesperson for Rowland didn’t dispute the last-minute changes to the law, including regarding the minimum age. “Keeping children safe — wherever they are — is a collective responsibility, and the Albanese government is stepping up to play our role,” they said in an email.
Despite the prime minister publicly supporting a campaign that called for a minimum age of 16 to access social media in May, it was not official policy until much later. In early September, Albanese committed to introducing teen social media ban legislation but said that his government hadn’t yet decided on the minimum age.
As late as October 31 — four days before cabinet would meet to approve the teen social media ban law — Rowland was prepped for “targeted consultation” meetings with former High Court chief justice Robert French, who had prepared a report about the legality of banning social media for the South Australian government, eSafety commissioner Julie Inman Grant and mental health groups including the Black Dog Institute, Orygen and Project ROCKIT. Rowland’s talking points specified that the government had not yet decided on an age between 14-16.
In fact, Rowland was prepared to preempt an argument from the unspecified mental health groups about the lack of evidentiary basis for any age as a cut-off: “We are aware that there is no robust evidence that supports a definitive answer on a single age,” her talking points said.
Rowland’s spokesperson said the age was set after speaking to various stakeholder groups. “Our decision to set the minimum age at 16 years is based on extensive consultation with young people, parents and carers, academics and child development experts, community, industry and civil organisations, and First Nations youth and state and territory governments,” they said.
Inside the government, public servants in the Department of Communications had been preparing documents based on the assumption of a specific age since at least mid-October. Draft documents from then, released under freedom of information, have the proposed minimum age redacted, citing an exemption for cabinet deliberation or decisions unless they already have been disclosed.
This suggests that staff were using an age other than 16, given that the cabinet decision to use 16 as the minimum age has since been disclosed. Rowland’s spokesperson did not deny this when put to the minister.
Other language in the document backs this up. After discussing a study that suggested social media use is predictive of a decrease in life satisfaction for teens in periods up to age 15, the draft states, “A minimum age [redacted] allows access to social media after most adolescents are outside this stage.” The use of “most” rather than “all” is consistent with an age of 15 or below, not 16.
A missing exemption framework
These draft analysis documents and talking points also contained references to something called the “exemption framework”. This was part of the legislation that would allow tech companies to apply to the eSafety commissioner for an exemption from the ban requirements — essentially allowing children of all ages to use their platforms.
The point of the exemption framework was to allow social media services that didn’t present a risk to Australian children. According to one set of talking points, it was supposed to help platforms that don’t have “harmful features” such as “algorithmic feeds” and to encourage tech companies to make “age-appropriate versions of their apps”.
Not only was the exemption framework repeatedly referred to in these documents, but it was also briefed to the media and stakeholders, including social media companies, that this would happen. As late as November 15 — a week before the bill was introduced to Parliament and close to two weeks after it had passed cabinet — an ABC report included a communications minister spokesperson speaking about the exemption framework.
But when the bill was first introduced to Parliament on November 21, the exemption framework was nowhere to be seen. One person from a social media company that had taken part in the government’s consultation, who spoke on the condition of anonymity, was shocked — where was this framework that they had been reassured would be there?
The change had happened so late that some of the government’s own documents still refer to it. The impact analysis document, released alongside the bill, still referred to an exemption framework that no longer existed.
While the government didn’t answer questions on the record about this, the communications minister’s office did point to its commitment to an as-yet-unlegislated digital duty of care, which could serve a similar function.
Bypassing normal legislation scrutiny
Other emails and drafts reveal how the government avoided a typical internal assessment of the impacts of its law.
In September, Department of Communication staff from its Age Assurance Trial Taskforce indicated they would seek to avoid doing a full impact analysis of the law, as is typically done with new legislation. Instead, they would seek — successfully — to do an alternative “impact analysis equivalent”, which instead relies on previously done analysis. The point of this equivalent is to avoid the unnecessary duplication of analysis.
This alternative process meant the government’s dedicated body for assessing the evidence supporting proposed laws, the Office of Impact Analysis (OIA), was not required to “assess the adequacy of the analysis contained in an [impact analysis equivalent]”, according to a September 27 email sent by an OIA staff member. The OIA wouldn’t consider how good the case was for the bill, just that it had been assessed by others.
What if there were “unforeseen impacts of the legislation” that weren’t picked up because of the expedited analysis process? Rowland wrote to the prime minister that these could be picked up and addressed in a “post-implementation review” within two years of implementation, meaning sometime at the end of 2027.
The teen social media ban would go on to be passed in late November after a handful of small changes, including bringing the deadline for its implementation earlier to no longer than 12 months after its royal ascent. The ban is set to come into effect in mid-December.
Have something to say about this article? Write to us at letters@crikey.com.au. Please include your full name to be considered for publication in Crikey’s Your Say. We reserve the right to edit for length and clarity.
