When should a PM sack a Minister?

Prime Ministers inevitably face a difficult question, at some point in their term: what to do with a problematic minister?

It’s a question we’re asking all over again with the latest mishandling by Minister Andrew Giles.

Hyperactive journalists love a ministerial scalp, because it’s an easy story to cover. Like elections, it’s easier to cover who’s up and who’s down. And if you’re the one who takes one out, you’re like the hunter that bagged themself a trophy kill. You get yourself a Walkley, you get to walk a few inches taller for a while.

The real story about ministerial sackings isn’t about integrity or accountability – it’s about how modern politics has broken the traditional feedback loop between scandal and consequence, and how this breakdown has fundamentally changed when and why Prime Ministers finally pull the trigger.

These days, it pays to wait it out

Before the 1990s, ministerial scandals followed a predictable pattern: sustained media coverage would lead to declining polls, which would threaten marginal seats, which would make backbenchers nervous, which would force the PM’s hand. The threat of losing government was immediate and measurable.

But three things changed this calculus:

First, the collapse of traditional media meant fewer journalists doing detailed policy reporting. What looks like intense coverage is often just a burst of identical stories recycling the same basic facts. Without new revelations, scandals struggle to maintain momentum.

Politicians used to worry because the endless drumbeat of stories and scandals would distract from the ‘main message’ they wanted to have cut through. But the mainstream media is less and less relevant to what messages cut through: you can just go around them. They can talk about it if they want to, but you don’t have to talk to them so much anymore.

Second, polling became more volatile but also less predictive of election outcomes. The connection between ministerial scandals and electoral threat became fuzzy. Howard discovered this accidentally during the travel rorts affair – he held firm despite terrible polls, and the issue eventually evaporated.

How many voters are deciding their vote on a scandal anyway?

Third, party room dynamics shifted. As party membership declined, MPs became more dependent on factional support than local branches. A minister’s value to the PM became more about internal party management than public perception.

This explains why seemingly career-ending scandals (like Robodebt) no longer end careers — unless you’re a department secretary, that is.

The feedback mechanisms that once turned public outrage into political consequences have broken down. Modern PMs don’t protect controversial ministers out of loyalty – they do it because the cost of sacking them now often outweighs the increasingly temporary benefit.

But there is still a tipping point – just not where most people think it is. The conventional wisdom says PMs should sack ministers when the scandal becomes “too damaging” – but that’s backward. Modern PMs actually face their greatest risk not from keeping damaged ministers, but from miscalculating when to cut them loose.

The key metric isn’t the size of the scandal itself, but whether it’s started metastasising in one of two specific ways:

First, when it begins infecting unrelated issues. If journalists start every unrelated press conference with questions about the troubled minister, that’s a warning sign. But the real danger point is when the public starts reflexively connecting other problems back to the original scandal. Once “Robodebt mentality” became shorthand for any harsh government policy, for instance, that was the point of no return.

Second, when it starts corroding party discipline. Backbenchers will tolerate a lot of bad press. What they won’t tolerate is having their own constituencies turn toxic. When local branch members start quitting, when friendly community groups start canceling meetings, when previously supportive local media turns hostile – that’s when backbenchers start making private phone calls to the PM’s office.

This explains why seemingly smaller scandals sometimes claim ministerial scalps while bigger ones don’t. It’s not about the scale of the wrongdoing – it’s about whether the scandal has developed these specific metastatic properties.

The truly skilled PMs recognize these warning signs early and act before they’re forced to. The less skilled ones wait until the damage is already done, then belatedly sack the minister anyway – getting all of the factional blowback but none of the political benefit of looking decisive.

The only exception…

There’s one exception to this rule: if lifestyle bloggers and fitness influencers are writing about your scandal-plagued minister, it’s too bad to salvage. Cut them loose.

Because at that point, this isn’t a ‘bubble’ story. Your scandal has spread. It’s made its wait out of news media and into pop culture. At this point, you’re taking on water. Because now it is genuinely sucking oxygen out of the room.

You might take that as a depressing sign of the times, and maybe it is. But at least there’s solutions.

It’s not that standards have declined; it’s that the system that enforced those standards no longer functions as it once did. The solution isn’t more calls for “integrity” – it’s about rebuilding the broken feedback loops between public disapproval and political consequences. Until then, ministerial sackings will remain rare, not because PMs have become more tolerant of scandal, but because the traditional mechanisms that made them necessary have largely ceased to function.