How AI will reshape Australia’s taxation system

David Bradbury:
AI, there are all sorts of new business models emerging. We will need to be creative in thinking about how the current tax rules apply to those and whether or not new rules need to apply. When you think about the impact on labour taxation, it is inevitable that if nothing changes in the system, that we will be observing declining revenue over time.

Elinor Kasapidis:
Welcome to CPA Australia's With Interest podcast. I'm Elinor Kasapidis, Chief of Policy, Standards and External Affairs at CPA Australia. What happens to a tax system designed for an industrial economy when artificial intelligence starts reshaping how work is performed, how businesses generate value and where economic growth comes from? As AI accelerates across industries, governments around the world are grappling with a new set of questions: who benefits from the productivity gains, what happens to traditional sources of tax revenue, and how should tax policy respond?

In this special With Interest episode, recorded live at CPA Australia's Tax Forum in August 2026, we examine the future of taxation in an AI-driven economy. From the sustainability of current tax bases to the opportunities and challenges AI presents for tax administration, compliance and economic growth, our panel explores what policymakers should be thinking about today to prepare for tomorrow.

Joining the discussion are Viva Hammer, Research Fellow at the Crawford School of Public Policy at the Australian National University; David Bradbury, former Chair of the Board of Taxation; and Michelle de Niese, Executive Director of the Corporate Tax Association. The conversation is moderated by CPA Australia's Tax Lead, Jenny Wong. Let's dive in.

Jenny Wong:
David, I just wanted to start with you. We've spent decades debating how to tax multinational profits, but AI raises a different question altogether. If economies become less dependent on labour, is this actually the biggest tax challenge for the next decade?

David Bradbury:
Look, Jenny, I think it certainly is one of the biggest tax challenges that we're going to face in the years ahead. If you think about a country like Australia where our dependence on personal income tax is very high, across our tax mix it accounts for the largest share, and it continues to be an even more significant share over time. Now, of course, we've got the ageing population, which is one set of challenges that will affect that personal income tax and, in particular, the labour tax component of that. But if automation driven by AI means that there are fewer people in the paid workforce earning incomes and paying taxes, then I think that is going to present a really significant challenge for revenue moving forward.

Perhaps just to circle back to your earlier introductory remarks, if you think about AI and globalisation and international fragmentation, if you connect the two things up, I think at the one level, anything that involves cross-border activity, if you don't have cross-border cooperation from jurisdictions, the result is either going to be no tax being paid or less tax than should be paid or more tax than should be paid, so either double tax or no tax. That was one of the things that the BEPS project was initially initiated to address. So I think that where there's cross-border activity, there's a risk.

But in addition to that, AI, there are all sorts of business models, new business models emerging. We will need to be creative in thinking about how the current tax rules apply to those and whether or not new rules need to apply. When you think about the impact on labour taxation, it is inevitable that if nothing changes in the system, that we will be observing declining revenue over time.

Jenny Wong:
Michelle, I might turn to you next. Whenever AI is discussed, the conversation quickly turns to jobs disappearing. But from what you're seeing in business, is AI replacing workers, changing the type of work being done, and is businesses recruiting less junior staff?

Michelle de Niese:
Right, I'll address those one at a time, Jenny. I just wanted to quote the RBA last week, and it's always important to quote the RBA correctly, so I'm going to read it out, generally a quote. They did report last week that, "AI is not widely being cited as a factor in headcount reduction. Instead, most companies view it as a tool to support revenue growth without a commensurate increase in their cost base." I think this is an accurate reflection of what we're seeing in the large business market.

We're not seeing wholesale replacement of staff in any way, shape, or form. What we are seeing is companies starting to think about what they're looking for in junior staff and what they're looking for from mid-level staff as well. So I think the questions that I've come up with that I'm hearing a lot of in terms of internal questions: How can AI be used, which employees should be using it, and what level of performance is required from it? Anyone that reads the AFR would've seen, there's a ream of articles around how various companies are using it.

I think Telstra made an announcement last week that some of their executive pay is now being connected to, what's the term, I should get that right, AI impact. So what they're looking for there is they've invested a lot of money in AI. They want to ensure that that money is filtering down to better customer service. So they'll be looking for benefits internally, but we're not seeing drastic job losses as yet.

I think with corporate tax, it's really important to say, and I'm happy to say it in this room, we should all remember this, it's a very judgement -heavy discipline, and that will never change, so it's really important. I think a lot of us focus on the compliance aspects of our roles, and we think that AI is going to be a threat. But I think if you focus on things like dealing with tax authorities, risk assessment, advisory work within your organisation, those things can't be taken over by AI.

The pace at which the decision can be made can be quicker, but those roles still need to be there and those decisions need to be made by humans. I think the concern I do have, and I think you talked about this, Jenny, actually, at a conference we were at last year. Last year? No, it was [inaudible 00:05:53].

Jenny Wong:
No. No, this year.

Michelle de Niese:
Okay. I don't know what's going on with this year. It's nearly finished. In Singapore, and you were talking about training of young tax professionals. I do worry a little bit that if you're taking away that grunt work, research, data analytics, those sorts of things, how are we going to train our new tax professionals? How are we going to train them to recognise the outcomes that companies need?

So I do have a little bit of a concern there. I did want to give all of you a plug, including my members. For those that don't know, I represent large companies and their tax functions. What we're really encouraging our members to do within their tax function is to focus on their people skills. So don't be afraid of identifying where AI might make your job easier, but do focus on your value within the organisation. I think tax in that sense is safe predominantly. Viva's going to tell me otherwise.

Jenny Wong:
Well, Viva, I'm going to turn to you. You done a lot of work in this space. Historically governments have taxed labour because that is where much of the income was generated. So if AI shifts value towards capital, data, intellectual property, does the tax system eventually have to follow?

Viva Hammer:
Jenny, we're at the dawn of a new era. Artificial general intelligence is still in utero, but when it's born and matures, no aspect of life on Earth will be the same. That's a big statement. Well, my advisors say we're about January 2020 with Corona, not with artificial intelligence. So how do taxes fit into this? Since World War II, governments have relied on taxing labour to support the age, the infirm, and the young. It's a new system, the income tax system. It's not old. It's been very new. Productivity and incomes have grown because we've invested in education and technology that makes labour more valuable.

Artificial general intelligence will change that. Lots of jobs, lots of jobs, desk jobs, screen jobs will be gone. Australia's big moneymakers and employers are in healthcare and mining, and they're probably not the first jobs to go. AI can't comfort the dying or nurse babies. The high-tech sector isn't a big part of the Australian economy, but even so, much clerical work, computer programming, accounting, and bank work will be taken over by AI, not today, but soon. Many middle-class jobs will be gone.

And jobs are not just about money. They're also occupational therapy. Our social life is built around family and work. But work certainly is about money, too. People get paid when they work and the government gets part of that. Governments are already saddled with huge debt. They've borrowed against the hope of an ever-expanding future, and AI might deliver us those dreams of an infinitely expanding wealth that will pay off our promises, or it may not.

So what are some options? Clearly, governments need to control the Golem, the Frankenstein we've made. It might be too late. But Australia has led the way controlling social media for our children, and maybe we can lead the way in controlling the AI Golem, too. In the meanwhile, as long as the tech giants are making the Golem and making them money, lots of money, we can force the people who own it to share the wealth AI has made. If a smaller group of people is working because of AI and taxes fall, then we may need to go back to the old toolkits of consumption tax and excise tax and come up with tools.

Professor Jeremy Bearer-Friend of George Washington University says governments need to stop taxing AI business in the way that we tax everybody else. Governments need to take their stock and share it with the nation, shares of stock from the AI companies. Then the nation will be riding with the fortunes of the high flyers. Pension funds already know how to manage stock portfolios. In America, Professor Bearer-Friend's ideas has been taken up by both the left and the right. It may be the only thing that the left and the right can agree on. I think that taking the shares of stock of AI companies is something that we should be considering and thinking about and not taxing them in the old ways.

Jenny Wong:
David, if I turn to you, what are your perspectives on that? If AI shifts value towards capital, data, intellectual property, does the tax system follow?

David Bradbury:
Look, maybe if I could pick up on a couple of things that Michelle and Viva have said and then come to that as well. I think, Michelle, in terms of the question of is this going to lead to fewer people employed or is it going to have a meaningful effect on the labour market, look, one answer to me is that, yes, it will. The unresolved question in my mind is, like all the other developments we've seen in the past, whether the jobs created in other areas-

Michelle de Niese:
Yeah, exactly.

David Bradbury:
... whether they compensate for the jobs lost. I think that's the question where the jury is still out, but I think it is going to lead to job losses.

Michelle de Niese:
Yes.

David Bradbury:
If I think about just a couple of examples, I was listening to one of the CEOs of a European, large multinational bank, and the CEO was talking about very specific process of writing a home loan. He went through and he basically said in his bank, there were about 23 steps in that process of writing a home loan. He said they went back and they did a mapping exercise and they identified that only two of those steps really needed a person to be involved in the process.

Now, I think that if you think about the things that many people do in their jobs, there will be steps that can be replaced by artificial intelligence. Now, as a society, how we allow that to be rolled out is another set of questions, but I think it is inevitable that it's going to mean job losses in some places. Once again, I don't know whether they'll be compensated for elsewhere. I had a round table discussion where there was an international speaker that has a lot of experience in analysing AI, and she tried to put it simply. She said, "In terms of jobs of the future, if you have a job where you're on your feet, you'll be okay. If you're on your bum, you're in trouble."

Michelle de Niese:
Okay.

David Bradbury:
So when I went back to my office, immediately I demanded-

Michelle de Niese:
A stand-up desk.

David Bradbury:
Now, I don't know if that's going to get around it. But if you think about it ... I've got a leak in my roof at home at the moment that I've got a tradesman fixing. Tradespeople are in enormous demand, and they are going to continue to be in enormous demand. When I think about where I was when I was growing up in Western Sydney, studying in high school, thinking about my future, everyone said, "Go to university. You got to go to university." Maybe that advice will be slightly different, I don't know moving forward. But there will be demand, there will be demand in different places. So I think it's a really interesting one and a difficult one to think about.

When we think about the tax system, and, Viva, you mentioned the suggestion that consumption tax might also be a way to compensate for that, I definitely think that we will have to tax capital more strongly than we have in the past. Clearly, they haven't said it directly, but government's recent reforms have been about trying to tax capital at higher levels. I think as a society, we're going to have to do that. We're probably going to have to tax consumption at higher levels as well. But for all those people that have been telling me for all of my career that consumption is a really hard tax to avoid, look at what's happening with tobacco and alcohol. So it's not straightforward there either.

But with consumption, I think a lot of the debate at the moment is we'll shift to consumption, and that's okay, but only if what we think the business model of AI will actually be what we're currently conceiving of it as. We're sort of thinking of ChatGPT or something like that. So if they charge me a subscription, then I'm a consumer and I pay consumption tax on that. But having spent the best part of a decade at the OECD trying to crack this issue around digital taxation, I spent a lot of time looking at the business models of the digital giants.

If you take Meta or Facebook in its earlier form, the technology emerged, and, in the same way as with AI, huge amounts of money were going into it, and it just became this really big thing. But they didn't know how to make money out of it, and for years, they didn't know how they were going to make money out of all of that. Then it was advertising revenue became the way that they could do it.

So I'm not convinced that artificial intelligence will pay for itself through a direct consumption model. I think there's going to be new business models out there that are going to challenge us to think even more seriously about how we might capture some of that. I think that's the difficulty is we don't know. A lot of money's being spent on AI. If you look at the returns that will be needed to justify that expenditure, these companies are going to have to generate a lot of revenue, and I can't yet see where that revenue's going to come from. I don't think it's all going to be subscription-based.

I think that's where Viva's idea of having a stake in these companies, that might be of some value. If you think about it for Australians, through superannuation, most of us do actually have a stake in those companies. In fact, there's all sorts of debates around whether we've got too much of a stake in some of those companies in the way in which the portfolios of assets might currently be invested. So I think all of these issues are ... The difficulty with them is we are trying to create taxes for business models that we can't yet conceive of. I think in that, the lesson is we need to be as nimble as we can, and we need to be constantly looking for new ways to capture value as these business models emerge.

Jenny Wong:
Just the next question on what new taxes may emerge or gain greater policy attention given the rapid pace of AI, Michelle, I might start with you.

Michelle de Niese:
Yeah, I agree with David's point. It feels a little bit like building the plane while you're flying it, so it makes us all very nervous. I think one thing that's interesting at the moment is looking at taxation of production inputs: energy, electricity, water, those sorts of things. They're quite politically appealing as well because you're taxing the big techs. They also address community concerns around their impact on communities through those two avenues. So I think they're things that they might be immediate levers that you can flick.

I think that the issue is, because this is developing so quickly, although governments like to talk about new taxes, the robot tax, AI tax, whatever it might be, they're very difficult, as David has alluded to, to implement, administer, keep up with the pace of whatever is actually occurring in the AI space. So I think the attraction to existing levers, some unusual ones like electricity, water, but then turning to consumption, obviously base erosion is always an issue around corporate rate, I think they will be very appealing to governments as an immediate kind of taxes that they can flick on with AI. But it'll be interesting to watch on energy, what they do there. That's an immediate one, yeah.

Jenny Wong:
Viva, do you have a perspective?

Viva Hammer:
Yeah, I agree with everything here. I'm still plugging the concept that income tax is a very new idea, all of human history, governments taxed excise tax, consumption tax, they took labour without asking, that sort of thing. Because we have an income tax and that's the whole system, it doesn't mean that we always will be, nor should it be. I'm the advocate here for saying that the AI will change things much more radically than we envisage them right now. As in January 2020, we didn't imagine how corona would change our lives, too.

So I think that we need to be nimble. It's true. I'm disappointed actually in how little our governments, governments globally, are spending time and energy and hiring young people who are at the centre of this and asking them, what will the world look like and how can we govern? I'm not sure that our governments have the power and the resources to address this major issue at the moment and they need them.

So then we need to hire, and I'm going to say this, young people who understand how it's going. My colleagues who have young people in the system say they expect to be unemployed. These are the brilliant ones who are creating this. They expect to be unemployed within a very short time. So I think that we need, as I said, to be nimble and to have the people who understand this at the centre of government guiding us into how we should be thinking about this in the future.

Michelle de Niese:
Just on that point, Viva, I wanted to mention that I read last week that an FOI request to Treasury found that there were no documents at this point looking at, again, I want to make sure I quote this correctly, "No documents on the analysis of the potential impact on the federal budget and/or income tax revenue from an AI-induced disruption to the labour market." So [inaudible 00:19:35]-

Viva Hammer:
Very good quote there.

Michelle de Niese:
Obviously, they're talking about the intergenerational report where they will look at this. But you're right. There's a genuine concerning gap there in terms of [inaudible 00:19:45].

Viva Hammer:
It's not just Australia. I mean, I'm working in the US with this. We also don't have young people at the centre of the system thinking about ... People are banging on my back saying, "Why aren't you thinking about this? Because you need to." So I think the answer is we need to have people whose job it is to be addressing this real time.

Jenny Wong:
Yeah, there needs to be some sort of scenario analysis done.

David Bradbury:
Yeah.

Michelle de Niese:
Yeah.

Viva Hammer:
But also [inaudible 00:20:08], because, as I said, flying the plane while riding it, while analysing it, we're going to do it, I think that we also ... This is changing so rapidly. It's a logarithmic change. So we need to be at the front of that.

David Bradbury:
Just to add on what taxes might look like, and I think maybe we need to take an even broader view. At the OECD, we used to spend a lot of time focused on the definition of a tax, but there's all sorts of charges that governments might impose that need not be taxes. I think what Michelle is saying around energy and water, if you think about, what are those inputs that are required? If you think about the AI ecosystem, semiconductors, you can't do this without semiconductors. You can't do it without data centres, and that involves land, it involves water, it involves energy.

Michelle de Niese:
[inaudible 00:20:58].

Viva Hammer:
And construction.

Michelle de Niese:
[inaudible 00:21:00].

David Bradbury:
And you can't do any of this without data. Data is a really hard one to think about as an input in all of this, but data is a crucial input in all of this. Now, I think governments are going to be thinking more and more about how you can ... Maybe it's not a tax, but it's a charge. You think about something like the News Bargaining Incentive, which is really a response. It's not just about revenue. It's really about trying to shape the outcomes of new technology.

Let me just give you one example that I think is a really good example of where personal income tax, corporate tax, GST, these are things that are economy-wide taxes that we try and come up with a tax that applies to everyone. The day of the specific tax, it had its day and we say, "Well, it's much better to go broad." But maybe the specific tax will start to get more traction because it is just hard to come up with generalised taxes that are going to capture everything across the economy.

If you think about transport, and the International Transport Federation, which is connected to the OECD, a number of years ago they did a review, and they were looking at the impact of self-driving trucks. We're not talking about cars here. We're just talking about trucks in the freight context. They estimated that if you allowed autonomous vehicles, self-driving trucks on our roads, that it would wipe out some ridiculous number of jobs across the global economy, just in OECD countries it was. What they said was, "But there's a set of policy choices that lead to that scenario. You don't just wake up one day and there are driverless trucks on your roads."

One of the ideas, and it was actually very contentious at the time. I've not seen it implemented anywhere, but I think it's worth sharing just to think about how you might think about the new economy, new tax systems. What they said was what you could do is you could say any company involved in freight transport that wants to have autonomous vehicles will need to have them registered to operate on our roads. That registration charge would be a substantial charge, not just like your car registration, it would be a substantial charge, and it would be a de facto form of taxation. Then some of the proceeds of that could go towards job retraining-

Viva Hammer:
Training, yeah.

David Bradbury:
... and industry adjustment. Now, I'm not saying that it's the answer. And even if it is the answer, it's the answer of a small part of a much larger problem. But I think the whole point in me raising this is that we need to be creative. We need to be thinking about, what are the levers? Why do governments tax income? There's lots of good reasons to not tax income. We want to encourage people to work, yet we tax work. Why do we do it?

Because it's a stable source and it's easy to tax, and through withholding, we can collect all of this money. People don't even see it. It goes straight to the government. We've become dependent upon that. So we need to be balancing not just what are the good economically efficient taxes. We need to balance that with, what are the taxes we can actually as governments impose and collect? I think that should be driving much of the thinking here as well.

Jenny Wong:
Just a final question on AI before we move to the geopolitical commentary. Imagine we're sitting here 2035, what is more likely: AI had expanded the tax base through productivity and new industries, or governments are urgently redesigning the tax system because labour and taxes have fallen so much faster than expected?

Michelle de Niese:
I'd love to say that AI quietly grows the revenue base. It's going to be a mad scramble, I think, is what's going to happen. I mean, we think about the things we're talking about, consumption tax, income tax, these have been issues within our system identified as problems in terms of consumption's too low, tax on income is too high. These have been issues that governments have not addressed for many, many years in Australia.

I can't see them addressing those issues in the context of something so uncertain as AI. I think they'll be forced into that decision. I agree with David, they'll be looking at different methods. I like the registration one. I hadn't thought of that. I'm on the wrong side of the fence, though. I think the transition will be more forced rather than gradual.

Even if we do see other industries that come out of AI, which I think we will, I don't know what they'll be, but I think we will, the revenue that those industries generate won't be fast enough, I don't think, to compensate for the loss of revenue they're going to have from other revenue streams. As Viva alluded to before, we've got very debt-heavy countries. I think Australia hit a trillion today. Isn't that fantastic? So that's really going to drive those decisions as well. Yes, exactly. That was a joke, obviously, a terrible joke. It's actually a shocking statistic.

Jenny Wong:
Viva?

Viva Hammer:
I think in 2035, you know what, only fools and prophets make prophecies, and I'm trying to be neither a fool nor a prophet. I made a big prophecy today that AI will change things radically, and I will stand by that, that it will be a radical, unimaginable change as Corona was. So I think this time, this panel, 2035, is going to involve a very large number of changes that we can't imagine. I really hope that we have smart people in the government that can keep up with it. Let's put it like that.

Jenny Wong:
David, final remarks.

David Bradbury:
Well, in 2035, you won't be listening to me. You'll be listening to my AI agent who will be much more articulate, have a full head of hair, big muscles.

Michelle de Niese:
Yeah, you're twin. You have a [inaudible 00:27:02].

David Bradbury:
Unfortunately, you've got me. What can we look forward to? Look, I think that, of course, we're talking about tax because we're tax people. Often you start with a focus on tax, and then you think about the other things. But in a sense, I think what is going to be most important for Australia is that we capture the benefits of this new technology and that we are able to become a place in the world that is a world leader in the way in which this technology is developed and employed.

I think that in some respects, there may be some trade-offs with tax revenue along the way in order to achieve that as well. But I do think that if you're able to build a more prosperous economy, then you will have a range of other options in terms of tax revenue opportunities down the track as well. So in a sense, I think that attracting that activity today is probably the most important thing. If we've done a good job at achieving that-

Michelle de Niese:
And tax it here.

David Bradbury:
... then in 2035, we will have a lot more options, I think, in terms of how we might be able to capture some of that benefit.

Jenny Wong:
We move on to the topic of geopolitics. David, you were here, the last year's Tax Forum, and you said the window for multilateral corporation had largely closed. So 12 months on, has there been any change, or are we still firmly entering an era of tax fragmentation?

David Bradbury:
Yeah, we're definitely even further advanced down the tax fragmentation direction. There have been some good examples, like the OECD has managed to hold the Pillar Two global minimum tax together but in really difficult circumstances. I think that what we've seen in particular from the US and its approach to tariffs, really it is not an approach predicated on global cooperation on these matters. It's really about taking unilateral measures. It's one of the few countries in the world that can take those unilateral measures and perhaps get away with them because of the strength of the US market.

We're seeing that occur. I think that has, on the one hand, meant that cooperation is less likely. But in a bizarre way, it's also made many countries reluctant to pursue unilateral measures of their own, and I'm not sure how long that can last for. Because if you have a situation where everyone is trying to play ball but the US is cranking up tariffs and extracting what they can, particularly when you think about tariffs themselves ... This sort of ties back to the discussion we were just having.

Tariffs are taxes on physical product. They are targeting the physical economy. Much of what we were just talking about is the intangible economy, and so much of the wealth being generated around the globe today is being driven by the intangible economy. Now, the US approach has typically been to say, "Well, we're going to tax physical products, many of which are being produced by others, but don't you ever dare tax intangible products." It's not a sustainable situation. How governments react, I think, remains to be seen. But there will be a whole range of new, innovative, creative approaches that will lead to greater tax fragmentation. I just can't see how this plays out if that's not the direction that it takes.

Jenny Wong:
Michelle, the next question is, if countries increasingly pursue unilateral measures, such as digital taxes, tariffs, investment incentives, what does that mean for businesses trying to invest across borders?

Michelle de Niese:
Well, it means lawyers do well for a start. Look, from what we're seeing, even at this early stage, business investment doesn't stop. It just becomes more cautious and it's slower, so I've started to call it the cost of caution. I think in Australia, we're really seeing that embedded in our corporate tax system because of, to be frank, our government's approach to some corporate tax issues and the administrator's approach to some of the rules as well. We do have a situation ... We've got quite a bit of form in Australia in the unilateral space. America's the big player, obviously, but Australia does have ... There has been a number of examples of late where Australia has strayed from what I would call multilateral agreement and principle and have gone off on their own, which is a concerning development.

I think what we're seeing in terms of business investment, because the caution is there, because there is temperatures rising around unilateral measures. You've got higher planning costs, hence my mention of lawyers. Double taxation and overlap risk, that is a huge issue for companies attempting to engage in activity across borders. The other thing that we're seeing a lot of is politically motivated rather than economically motivated changes to systems, particularly to protect against things like tariffs. They never end up well. Increasing demands of revenue authorities. Revenue authorities have been given the go-ahead by governments. You go out there and you collect as much as you can under the rules that you have. We certainly see that in Australia. There's no doubt about that.

I think the interesting thing that I'm seeing, and I certainly heard this when we were in Singapore earlier this year, Jenny, is this real interest in safe jurisdictions. I think safe jurisdictions now is a slightly different term than what I would've used previously. Safe jurisdictions are those that adhere to the rule of law, really adhere to the rule of law, and respect internationally understood principles, OECD Principles, Pillar Two. So those countries are becoming more attractive in the midst of other countries that are really playing hardball on unilateral measures. Of course, the US has 899 hanging out there. I think that's certainly still on the boil. I think the possibility of that provision coming back and being applied to Australia is real. Sorry, terrible news all around.

Jenny Wong:
Viva?

Viva Hammer:
The OECD has done an extraordinary job of getting diverse countries to agree on the Pillar Two principles and implement them. It's probably unprecedented. Their leaders have convinced countries with carrots and sticks that it's better to cooperate than not to cooperate. Of course, game theorists tell us that there's a risk of defection. As soon as you've got cooperation, there's a great opportunity for defection, which begins with small disobediences and then it snowballs.

There's a race to the bottom. Then when they hit the bottom, what happens? They want to come back together again because being at the bottom is really not that great. This is a cycle that's been going on since the beginning of time, but also since the report of the League of Nations in 1923 that began the international tax regime, the concepts of source and residence, which we still use today and we probably need to think about not using anymore because they're out of date.

The oscillation between cooperation and affection and the race to the bottom will probably continue to the end of time. I don't know if AI is going to change that race. But certainly while they are human beings, it'll be continuing, but they'll continue with different ideas and different challenges. But the oscillation between cooperation and affection is going to be an eternal human concept.

Jenny Wong:
David, Australia has traditionally been seen as a rules-based economy. In a world where some countries increasingly use tax as a strategic weapon, is being predictable still an advantage?

David Bradbury:
I think being predictable is always going to be an important consideration when it comes to attracting foreign investment. For a country like Australia, foreign investment has always been an important ingredient in our economic success. As a small open economy, that's not going to change anytime soon. So I think predictability is always going to be important.

The challenge is about ensuring that you're sufficiently nimble to manage the changing economy and changing business models that you face around you. I think Australia has traditionally done a pretty good job at striking that balance. There has been innovation in the way in which we've sought to protect our revenue base. But by and large, we have presented a tax system that is in general alignment with accepted principles from the OECD, for example. So I think it's a difficult balance, but Australia has generally done a pretty good job.

I do think that, in the space of AI, there is potentially a case for Australia to think about how it might ensure that its tax system is securing the investment in that area that will be necessary for us to truly be able to take advantage of the benefits of that down the track. That's something that every country should be reassessing on a regular basis, but I think that there is an opportunity there for Australia.

Jenny Wong:
Michelle, your perspectives?

Michelle de Niese:
Look, I think Australia has historically been seen as a predictable jurisdiction. I think that has changed in recent times. I hear a lot of our members compare our revenue authority to very, very challenging revenue authorities. I don't think that's something that we should necessarily be proud of. I think we see a lot of strain from what were previously really strongly held views around rule of law. I think the changes that we saw recently around non-resident CGT assets is a really good example. So predictable's great. Australia historically has been. I think we are not anymore from a corporate tax perspective. Predictability is not the same as being passive, and Australia has been quite passive in terms of looking at its competitiveness. Our top 10 companies pay 30% of all of our corporate tax, and their average age is 10. The average age of the top US companies is about 30, if that now. That's probably lower now, 20 maybe.

Jenny Wong:
15 minutes.

Michelle de Niese:
There you go, 15 minutes, yeah, exactly. So we need to be more nimble, I think, in terms of where we go from here. We're very protective of our revenue base. It's the second largest revenue stream we have in Australia, and all of our effort goes to protecting ... We've got this kind of conundrum in Australia. We're so reliant on our corporate revenue base, but we've got a really high tax rate, which puts us at risk in terms of erosion. So I really feel that if Australia's going to continue to be passive and use tax policies as a lever to win votes rather than actually attract investment, we're going to be in trouble in terms of attracting revenue sources from AI going forward.

Jenny Wong:
David, in a more fragmented world, I'm curious to know what role does the OECD or any other multilateral institution play over the next decade.

David Bradbury:
I think what we saw really from the global financial crisis through until about 2022, '23 was the golden days of global tax cooperation. I really think you can look historically and you won't find a period where there was a greater willingness for countries to actually cooperate on tax matters. It led to, for the very first time, global minimum standards. Starting with the automatic exchange of information, the Common Reporting Standard, through to the four mandatory elements of the BEPS project, and then the global minimum tax, they're really the first instances where you have global tax policy that is agreed at a coordinated level. So for all those years before that, we had a Model Tax Convention, and that allowed countries to have a template of what good practise might be. Then on a bilateral basis, they would negotiate what their tax agreements would look like.

I see Ronita Ram here, who has been very much involved in negotiating many of these things in the past. But can I say that the issue here is that we will return to that type of world where Model Tax Convention will be the guide, the template, but I don't think you're going to have globally enforceable standards. So whether or not we have to wait another decade or more than that, let's wait. But I think you're going to have to wait some time before we get back into that type of space.

Jenny Wong:
I want to finish off on this. What is the biggest tax debate nobody's talking about yet? Maybe Viva, start with you.

Viva Hammer:
I'm going to say that the two things in Australia people need to talk about is consumption tax. It needs to be broadened. The rates need to go up inconsistent with our OECD partners. We have a very narrow base. It needs to be broadened. We also need to talk about personal housing. Now, everyone starts getting really angry with me when I talk about that. But at some point when the gain is $100 million on your personal house, you need to pay tax on it. And in Sydney, we have enormous gains. So nobody likes it when I say that, but this is a base that needs to be dealt with, enormous over-investment in individual personal housing to the detriment of the pricing of housing and to the detriment of the tax system, too. So I think consumption tax and capital gains on personal housing are two things that need to be addressed.

Jenny Wong:
Michelle?

Michelle de Niese:
I think from a business perspective, it's all about the base. It's no longer about the rate. I think there's always been this focus on Australia's corporate rate is too high. I think the focus really does need to be, what is the base going to look like going forward? Again, I like that registration idea. So I think that's the debate that's missing at the moment. Often when people talk to me, they're always like, "Oh, do you think the corporate rate's too high?" So it's not even about the corporate rate anymore. No one's even talking about the corporate rate in Australia. It's all about the base and how that will be taxed going forward.

Jenny Wong:
David, any final remarks?

David Bradbury:
I think an issue that we don't talk about enough is tax morale. That is, in a self-assessment system where we basically expect people to declare their income and to pay their tax with only minimal enforcement activity because you can't enforce everyone and all of their activities, in that type of environment, we rely upon people paying their tax. I think that this is something where there are a number of instances where we lose sight of the importance of that. I really do think that if you think about things like in the post-COVID environment, there are all sorts of groups and people that have emerged that are really very resistant to having rules or regulations imposed upon them for a whole range of reasons. But I think this idea of tax morale is really important. We need people to believe in the importance of why taxes are being raised and why it's important for them to pay their fair share. That requires a fair system. But I think tax morale is a really important issue.

Jenny Wong:
Well, thank you very much to Michelle, Viva, and David. David, this is probably the last time you're going to be speaking about tax. I did want to say that you've left quite a legacy, I think, in basically tax policy. You shaped things from inside Parliament, the OECD, the Board of Taxation. I think what I've valued most is that you've been willing to explain the thinking of things. Personally, I sincerely appreciate it as others as well. So on behalf of CPA Australia, thank you for your contribution.

David Bradbury:
Thanks, Jenny. Thank you.

Jenny Wong:
Thank you, Viva, and thank you, Michelle.

Elinor Kasapidis:
We hope you enjoyed listening to Viva, David and Michelle. Their discussion with Jenny Wong, CPA Australia’s Tax Lead, was recorded at CPA Australia’s Tax Forum in August 2026.

And don't forget to subscribe to With Interest and share this episode with your colleagues and friends in the business community. Until next time, thanks for listening.