Jeremy Hirschhorn, Second Commissioner, Compliance and Engagement Group
Speech at the Tax Institute’s Financial Services conference
Sydney, 26 March 2026
(Check against delivery)
Introduction
Thank you for having me here today.
I join the Tax Institute in acknowledging the traditional owners of the country we are on today, the Gadigal people of the Eora Nation. Looking out I can see the Barramattagal lands, where I had the privilege of growing up.
One of the great aspects of my role is that I have responsibility to champion reconciliation within and through the ATO, and I wanted to make some observations on how we and the Financial Services Industry can contribute.
The first tangible thing is simply ensuring that tax obligations are met. By ensuring that tax is paid, we ensure that money is available to Government to pay for the services that support our most vulnerable communities, whether that’s health, education or infrastructure. Here, the financial services industry contributes, paying about $15 billion per year in corporate tax.
The second is careers. Both the ATO and the financial services industry provide high‑quality, skilled employment. Changing the life of even one person by giving them access to a good career (or actively sponsoring their career) is a real and tangible step towards reconciliation.
The third is that both the ATO and the financial services industry buy lots of things from lots of suppliers. And we can choose to support Indigenous-owned businesses.
But critically, and a particular responsibility of both the ATO and the financial services industry, is our role in supporting financial literacy and financial participation (and removing obstacles to financial participation). For the ATO, we are working on some tangible initiatives, and I will mention a couple of our current focuses.
Firstly, how in an increasingly digitalised world do we support people who do not have access to their own smartphone, or experience challenges in obtaining and maintaining identification documents.
Secondly, how do we ensure that superannuation (and perhaps even more importantly, insurance within super) benefits those it is meant to. This is something that I'm going to send out as a challenge to this group about later today.
And a particular challenge for this industry, is how to ensure that financial products are fit for purpose and meet the needs of the community.
Again, thank you for having me here today to deliver the keynote presentation.
As a tax professional at the very start of my career, I attended the Hartford Conference run by Professor Goodman, the predecessor to this Financial Services Conference. I did not imagine that one day I would be delivering the keynote!
In thinking about what I wanted to talk about today, I reviewed the program. It’s very clear there is no shortage of deep subject‑matter expertise in this room on intricate matters of tax interpretation. So rather than trying to add to that, I thought what might be most useful is to offer a bit of perspective on the system as is, a sense of where we see the tax system heading, and some context for the important work that you do within it. Hopefully this will help inform and shape some of the discussion over the course of the conference.
I do want to add one proviso upfront. In my role as Second Commissioner at the ATO, I’m a bricklayer, not the architect. I don’t make tax policy, my responsibility is to administer the law as enacted by Parliament.
So, in the Q&A later, please don’t ask me about the Budget. Budget questions are especially awkward because I either don’t know the answer, or I do know the answer. Either way, it’s unlikely to be insightful for anyone and it’s best we all avoid that situation. As bricklayers, the ATO follow the instructions of our architects, and we do not publicly advocate for or against policy decisions.
What I do want to talk about is the tax system as a whole, and why that broader context matters. As tax and finance professionals in the financial services industry, it’s easy to focus very closely on your own part of the system, which is often the application of complex law to complex scenarios.
Taking stock of the Australian tax system
How much is shared each year?
One way of looking at tax in a broader sense is that it is the overarching set of “sharing rules” that govern how we as citizens share our resources so that we can jointly do the things we cannot do by ourselves.
From that perspective, the overall system is interested much more in the level of sharing, i.e. that there is enough tax to fund Government programs, and less interested in the fine points of specific provisions and some of the intricacy that will be discussed in the next 2 days of the program.
Australia collects somewhere between $600-700 billion in tax each year, depending on how you count it. The single largest component of that, by a very long way, is withholding from salaries and wages, roughly $300 billion each year. While the taxation of wages is not especially exciting for tax professionals, it is absolutely central to how the system works.
Around $100 billion comes from GST. Corporate tax contributes about $150 billion which is actually very large proportion by international standards. Australia relies far more heavily on corporate tax than most other major economies, although that is partly offset when you take the full imputation system into account.
Within that corporate tax base, depending on how you draw the lines in terms of what is considered large, you’re talking about somewhere between $60 billion and $100 billion coming from the larger corporate population. And of that, the financial services sector contributes roughly $15 billion.
While that is still a very significant amount of money, it’s important to keep it in context. It’s $15 billion within a system that raises well over $600 billion in total.
What is the level of tax compliance?
Australia is one of the relatively small number of countries that measures what’s known as the tax gap in a comprehensive way. That is, we estimate the difference between what the ATO expects to collect and the amount that would have been collected if every taxpayer was fully compliant with relevant taxation law. That gap is around $60 billion each year.
There are different ways of thinking about that number. One way is to see it like school results, as in, we collected about 93% of what the law says is payable. Depending on how your parents reacted to report cards, that’s either a high distinction or an uncomfortable conversation at the dinner table about the lost 7 marks!
The gap drives where the ATO is putting its thought and energy. This context can explain why the system behaves the way it does and why some parts of the system attract a level of attention from as administrator that can feel disproportionate when you’re zoomed in on your own corner of it (or sometimes you might feel that your important issue is being ignored).
The compliance framework and the “fifth pillar”
In tax administration, the academic literature often describes compliance as resting on 4 pillars.
The first pillar is registration, are people actually in the system at all, or are they operating entirely outside it?
The second is lodgment, once registered, are they actually lodging the returns they are required to lodge?
The third is payment, having lodged, do they pay the amount of tax they say they owe?
And the fourth pillar is correct reporting, is what’s been reported accurate and consistent with the law?
That fourth pillar is where a lot of this conference quite rightly spends its time. Because this industry is basically fully compliant with the first three pillars, the fourth pillar is where much of the technical complexity sits and where a great deal of professional expertise is applied in determining your own tax obligations.
Increasingly, at least from the ATO’s perspective, we talk about a fifth pillar. The fifth pillar is the reporting and paying of tax obligations on behalf of others. This is the pillar where the financial services industry sits right at the forefront. It is also where we will continue to see the tax system heading into the future.
Much of what underpins modern tax administration depends on third‑party reporting and withholding. Things like the Annual Investment Income Report of interest and dividends, Managed Investment Trust reporting, trust distribution reporting, superannuation data and employer reporting through Single Touch Payroll all involve one party reporting or paying tax obligations that affect another.
If you consider the ATO service offer to individuals, whether through our app or through registered agents, it relies almost entirely on data provided by other organisations. The vast majority of the income side of an individual return is pre‑filled, and that’s only possible because of the reporting provided by third-parties.
We’re genuinely grateful for that, and this industry in particular, because it makes life much easier for individuals. There’s some striking data on this. Before pre‑fill, our surveys indicated that the average Australian spent around fifteen hours preparing their tax return. Today, for someone with a simple salary and wages return, the average time spent is closer to 8 minutes. Now, perhaps as administrators we might like people to spend a little more time than 8 minutes engaging with their tax affairs, but that shift is a powerful illustration of how much the system already relies on third‑party reporting.
And the important point is this: for most Australians, we’re no longer really operating in a self‑assessment system in the traditional sense. We’re operating in a third‑party‑supported system.
A key element of this is that accuracy upstream increasingly determines outcomes downstream. The underlying trust in the system is now reliant on the quality, timeliness and completeness of the data flowing through it. And for you as tax professionals in the financial services industry, it means your role is no longer just about getting your own tax position right, it increasingly involves acting as an intermediary in the tax system itself.
In short, the financial services industry is a core part of that fifth pillar. And that places this industry not at the edge of the tax system, but right at its centre.
The Financial Services industry
Around 2014, when I came to the ATO, there was a very noticeable surge in public interest in large corporate tax. It coincided with the Occupy Wall Street movement overseas, the work of the Tax Justice Network here in Australia, the parliamentary inquiry into Corporate Tax Avoidance, and the establishment of the Tax Avoidance Taskforce, which has been continued and expanded with bipartisan support ever since.
The community was not comfortable with where large-company tax compliance was perceived to be at the time. In our tax gap analysis at the time, we estimated that around 91% of large company obligations were being met. That may sound relatively high, but in the context of large corporates and very large dollar values, it was not where the community expected the system to be (and maybe significantly, was notably below the level of compliance of individuals with simple affairs).
The good news is that over a decade on, compliance across the sector is now significantly higher. For large corporates, corporate tax performance at lodgment is now in the order of 94%, rising to around 96% once compliance activity is taken into account. That represents a material improvement over the life of the Tax Avoidance Taskforce, billions of dollars each year improvement even before compliance activity.
Within this, the financial services industry has been very responsive to the change in community expectation and sentiment. The tax performance of large financial institutions has fundamentally improved over the past decade and that change has been both real and sustained. In many respects, the major banks, together with some other major Australian listed companies have been leaders in the transformation of tax culture in Australia. They have invested heavily in governance, transparency and engagement with the ATO, and they have set benchmarks that others across the system now look to. That is a significant achievement, and it deserves recognition.
Of course, there are still challenges. In the large market, our ambition is clear. We want corporate tax performance to be around 96% correct at lodgment, and 98% once compliance activity is taken into account. That’s the benchmark we are working towards.
The large market prioritising its tax obligations seriously is not just important in revenue terms, it’s important for community trust in the system as a whole. The broader community wants confidence that everyone is participating fairly. The behaviour of large, sophisticated participants sends a powerful signal about the integrity of the system as a whole.
Technical fine tuning and amendments
Taking the moment for a quick aside, I suspect that the need for technical fine tuning of the law will come up in some (or many or all!) of the sessions following this.
I would just say that there are many important questions in tax, but you’re never going to get to a system where the law provides a fully deterministic or algorithmic answer for every issue.
Our policy design teams and legislative drafters are very capable people, but they can’t anticipate every possible scenario. And if the answer were simply more words in the legislation, I suspect we’d already be there, because the Tax Acts are already more than long enough!
There are real limits on how much complexity the system can absorb. There’s only so much legislation Parliament can pass, only so much attention taxpayers and advisers can give, and only so much volume the system as a whole can cope with. Tax already accounts for a very large share of the legislative agenda of government, and the answer to questions of ambiguity is not always going to be passing legislation for further clarity.
This is also part of the reason for the issue of Practical Compliance Guidelines, which provide useful information for taxpayers in determining their tax risk position in areas which are intractably “shades of grey”. Particularly in areas of “it depends on the facts and circumstances”, which may not be particularly useful in a ruling, we hope that having a clearer knowledge of what will trigger our risk flags is of use.
Macro trends in tax administration
It’s useful to step back and reflect on how tax administration has evolved over time – and what that tells us about where the system is heading.
Australia has been through several major waves of tax reform. The first big one in my memory was in 1985 and 1986. Then there was the 1997 attempted rewrite of the Tax Act – still referred to by the (now) oldies (or at least me) as “the new Act”, I suspect much to the mirth of the younger members of the profession and in this audience. And of course, the Ralph Review, RATS, ANTS and the introduction of the GST. That was another genuinely significant shift.
But what’s important to remember is that all of this reform took place in a very different context. It was a pre-digital world. I still remember that in 1996 there was exactly one person in the KPMG tax practice who had access to something called “the internet”, and several times a day he would post up the Olympic results, which at the time felt extraordinary. That was the technological environment in which much of this reform was designed and implemented.
More recently, of course, we had the Henry Review. Like many major reviews of this kind, it wasn’t substantially implemented at the time but it continues to exist as a reference point and a guide for future governments as they think about the direction of the tax system.
So turning now to some of the macro trends in tax administration. You’ll increasingly hear the concept referred to as OECD Tax Administration 3.0. At its core, the idea is a fairly simple one: tax reporting and tax payment should happen as close as possible to the underlying economic event – ideally at the event itself.
The ambition is to move away from lagged systems and periodic reconciliation and instead move tax administration closer to where the data already sits, within the natural systems used by businesses and taxpayers. Done well, that can dramatically simplify administration, increase compliance, and make life easier for taxpayers.
Of course, this immediately raises an important question: what is the natural system for different groups of taxpayers?
For many small businesses, it may be accounting software (and here, the significant market share of a limited number of participants is important). For others, it may be payment systems or payment platforms. In some cases, particularly relevant to this industry, it may be banks and financial institutions. One of the realities we’re grappling with at the moment is that we don’t yet know where those natural systems will finally settle.
There’s another dimension to this trend as well, which runs slightly in the opposite direction. Sophisticated technology (including AI) may allow us to hide, or at least manage, underlying complexity. The example I often use is the smart phone. It’s probably one of the most complex pieces of technology ever created, but it’s simple to use. The complexity hasn’t disappeared, it’s just been designed out of the user experience.
And that’s a useful way to think about where tax administration is heading. Rather than asking how technology, data and analytics can support the existing tax system, the better question is how they allow us to redesign the system itself.
In other words, not how we digitise today’s processes, but how the system can cut with the grain of business activity and embed itself into the natural flow of commercial activity.
That’s where our focus increasingly sits. And it’s why this conversation is particularly relevant for the financial services industry – because if tax is going to operate closer to the event, then the institutions that effectively host those events will inevitably play a more central role.
The current biggest challenges
Closing the tax gap
The ATO have recently articulated a clear purpose and vision. Our purpose is straightforward: to collect tax so governments can provide the services the community expects and relies on.
Our vision is equally simple, but ambitious, that every taxpayer meets their tax obligations, for three reasons:
First, because compliance is simple.
Second, because support is tailored.
And third, because deliberate non-compliance has consequences.
To achieve this means we have to shrink our current $60 billion tax gap. As administrator, the first question is where we focus our effort for maximum effect.
The reality is that the gap is not evenly distributed. Small business accounts for the largest share. Across all taxes, small business makes up around 60% of the measured tax gap – so something like $35 billion out of that $60 billion.
Within that, around 60% again is attributable to shadow-economy behaviour. That is activity where income is concealed, transactions are not reported, and the normal levers of compliance are deliberately bypassed. That’s why addressing the shadow economy remains such an important focus for us.
What’s striking is that many of these issues are not new. They are long-standing, persistent problems. But even actors in the shadow economy still need to participate, at least to some extent, in the legitimate economy. They need bank accounts, payment platforms, suppliers, and customers.
So the question for us is not just how we detect non-compliance after the fact, but how we increase visibility upstream – how we make it progressively harder to operate outside the system while still accessing the parts of the economy that businesses rely on to function.
Debt
The next major challenge for us is debt and in particular debt owed by small businesses.
Our total on-balance sheet debt book, that is, amounts that have been assessed and are currently owed has now exceeded $100 billion. That figure, of course, excludes any notional or off-balance sheet liabilities that may exist but haven’t yet crystallised through assessment (a lot of our tax gap is effectively “off balance sheet debt”).
Not all of that debt is high-quality. Roughly half of it we consider technically collectible. But of the collectible portion, around two-thirds relates to small business. In practical terms, that means somewhere in the order of $35-$40 billion.
And it’s important to understand what that debt actually represents. Very little of it is income tax on profits. The bulk of it is made up of withholding taxes (PAYG withholding) and GST. Some of it is superannuation. A smaller portion is interest and penalties.
This matters because it goes directly to one of our central strategic challenges, how do we get tax paid, not just assessed?
A significant contributor here is timing. There is often a lag between when the underlying tax event occurs, when a business is paid by its customer, and when the tax associated with that transaction is due to be remitted. The longer that lag, the greater the opportunity for error, for cash flow pressure to build and ultimately for debt to accumulate.
From an administrative perspective, that lag is not neutral. It materially increases both the likelihood and the scale of non-payment. And it’s one of the reasons why we are so focused on system design that brings reporting and payment closer to the economic event itself, so that this debt book does not continue to grow. I touch on this more below.
But our immediate challenge is to get the current debt book paid, and this is where our settings interact with the financial services sector. If you think of us like a credit provider, we provide credit without any credit checks (often without even knowing it!), and are often unsecured (although with recourse to directors for some tax debts).
For various good reasons, the ATO must have concessional tax settings for taxpayers in vulnerable positions, both for things occurring at an individual level, but also across broad communities (such as bushfires, floods and pandemics). Currently, we are ensuring we have an appropriate stance for businesses materially affected by the current fuel price situation.
However, we have realised that our stance in practice, particularly in relation to interest remission, has been more concessional than our intended stance, and even our intended stance may have been overly concessional for many taxpayers not in vulnerable positions.
This has meant that some businesses which should be seeking financing for their working capital from the financial services industry, are instead running what is effectively an ATO working capital overdraft, often on the hope or expectation that the overdraft is in practice interest free (due to anticipated remission of the general interest charge).
Over the next period, we are looking to be more explicit and consistent in our stance in relation to charging interest (which is also now non-deductible), with the hoped for result that businesses should obtain their finance from the appropriate source, which is the financial services industry, both on an on-going basis, but also resolving some of the current debt book.
We will also need your support to bank those businesses which are bankable, and to dispel the myth that you do not finance businesses with tax debts.
Superannuation
One of the biggest recent developments and very relevant to this industry, is Payday Super. In some ways it’s a major change, and in other ways it builds on behaviour that already exists.
Around 40% of employees are already effectively being paid on or very close to payday, so a significant part of the economy is already operating in this mode. Payday Super is a practical example of OECD Tax Administration 3.0 in action, moving obligations closer to the economic event. But it is also a very clear example of the fifth pillar: third-party reporting and payment on behalf of others.
Under Payday Super, superannuation funds will be required to report to us key data sets when an employee’s super is paid in close to real time. That data will then be matched against what employers are reporting through payroll systems at the same time.
There are roughly 15 million “jobs” (ie employer-employee pairings) in the Australian economy. If those jobs are paid fortnightly, this is an extraordinary volume of transactions and data-matching events over the course of a year. It is, quite literally, billions and billions of data points.
The ability for the ATO to digest this data at scale is foundational to the integrity of the system. At the moment, we estimate that around $6 billion of superannuation each year is not paid by employers to employees. This has real consequences for people’s retirement outcomes.
For those superannuation funds in the audience, you will play a key part – not only in accurate and timely reporting to the ATO, but by interacting with employers and their systems in real time, particularly where there are glitches in the data.
But in thinking more broadly about superannuation, I am going to set out a challenge to this group. We often hear about product mis-selling focusing on the features of the product. In superannuation though, lost superannuation can be seen as product failure. If you think about lost super, it’s money that people have put aside across their working lives, for their retirement, or for their families if they pass away and the money doesn’t always make its way to them. At that point, the product has fundamentally failed. And even worse if insurance is not claimed.
So I think this is a really important challenge for us, and for this industry: how do we work together to make sure that these products do not fail the people they’re meant to serve?
Fraud and scams
Another area where we’re working very closely with the financial services industry is in addressing fraud and scams across the entire economy. The ATO is the first government member, and at this stage, the only government member, of the Australian Financial Crimes Exchange (AFCX). This work is incredibly important as through that collaboration, we’ve stopped at least hundreds of millions of dollars in fraud.
The reality is that the banking industry knows things about people, simply by virtue of holding bank accounts, that we could never know (or would be extraordinarily difficult and intrusive to acquire and use). Conversely, the ATO knows things about people that banks could never know. By sharing information about suspicious accounts and suspicious activity in close to real time, we’re not only stopping crime being committed through known mule accounts but also using what’s sometimes called blast radius or network analysis to identify and stop many more mule accounts.
Fraud isn’t really a business about fooling people, it’s an illegal business of making money. And if, between the ATO and the banks, we can make fraud more expensive to carry out, more difficult to scale, with lower success rates, and hence less profitable or loss making, then we can make it uneconomic. And when it becomes uneconomic, it largely stops.
So this is a very significant area of collaboration between the ATO and your organisations. Most importantly, it is one where working together genuinely delivers outcomes for the Australian community that none of us could achieve on our own.
Where might the system go next?
To draw together today’s comments, I would like to bring together some of the macro themes and where we see the tax system sitting today and where, perhaps, the tax system and the financial services system may increasingly come together.
Again, I want to be clear about the context. I’m only a bricklayer, not the architect. Some of these ideas would require government decisions or legislative change, and I’m not seeking to advocate for them, or to front-run the upcoming budget or any future budgets. Rather, these are observations from an administrative perspective, about where the system seems to be heading and the questions that naturally follow from that.
Data sharing for citizen benefit
Firstly, I think we’re going to see data-sharing continue to expand. There are continued opportunities in this space to address both the bad actors in the system as well as on the positive side.
Focusing here on the positive.
At the moment, people are often required to aggregate their own information (including tax history), provide it to their mortgage broker or financial institution, and then the financial institution has to work out whether that data is complete, consistent and high quality. So the question here is: is there scope for the ATO to develop a set of standardised, curated data sets that are relevant for major financial product decisions, which the financial system could obtain from us, with the individual’s consent?
We already see this working overseas. In Norway, for example, there’s a standardised, mortgage-related data set that can be shared, and that has led to mortgage approvals being completed often within the same day. It’s supported better pricing, because the data quality is higher, and it is easier for customers to shop around.
There are benefits on both sides. Banks face less risk because they have greater confidence in the data, and less expense in taking the data on (because it’s in a standard format). And from a consumer perspective, there’s more competition because it becomes much easier to apply and compare products, as well as less time and effort in collating paperwork.
Withholding from salaries and wages and payments for services
If we look at the parts of the system that currently don’t really align with the OECD 3.0 principles, a few things stand out.
PAYGW is around $300 billion out of a $600 billion system. Large employers are already, in effect, operate on payday withholding for employees, but others are on a range of different cycles. So that’s an obvious place to look if you think about whether those principles should apply more broadly across all employers.
Then you have individual contractors, people who sit somewhere in that halfway house between being a genuine business and being an employee. Where does real-time reporting and withholding land for them? And what is the natural system in that case? Is it a bank? Is it a payment platform? Or is it the users of those services?
Tax instalments
I’m sure many people in this room deal with monthly or quarterly tax instalments for their income tax obligations.
Those instalment systems were largely designed in the 1990s, and the amount you’re required to pay often bears very little resemblance to your actual profit for that month or quarter, being based on what was possible in 2000 (as a rough proxy of economic activity and profitability) rather than what can be done today.
Again, if you apply the OECD principles, you have to ask whether that’s really aligned to economic reality. What we do know is that the system produces very large wash-ups and refunds, which strongly suggests that instalments are not a particularly good proxy for underlying profitability.
We also know that small businesses in particular find it very hard to understand and plan for instalments under the current system, creating disengagement
We are actively exploring, currently on an elective basis, the ability to base these instalments on current period profitability, rather than as a percentage of turnover, ideally within small business’ natural business software systems.
Individuals’ deductions
On the individual tax experience, it’s now actually pretty easy for most Australians, with one big exception, which is deductions. On the income side, things are now close to complete, largely because of pre-fill and third-party reporting.
At the moment, the fundamental data that deductions are often based on is a piece of paper and frankly, that’s never going to be a very good quality data source. We also have an easy to say, but hard to apply in practice, technical test of “to the extent” incurred in earning income.
So the question is whether we start to move deductions towards data instead (from both substantiation and technical perspectives)? Do we get to a point where, to claim a deduction, there has to be a bank transfer behind it? And if that’s the case, are there then opportunities to use that data to pre-fill certain types of deductions as well? That’s one possible direction.
Another path might be a greater move towards standard deductions in one form or another. And we’re already seeing indications of that, with the Government’s policy of a flat $1000 work-related expense deduction for certain categories.
GST
There is also a question of whether you could design a GST that administratively operates more in real time and potentially reduces administrative burden.
It’s interesting to reflect on why Australia (and pretty much every other country!) chose a multi-stage VAT rather than a retail sales tax back when it was introduced and why that choice still holds today.
At the risk of over-simplifying, I wanted to focus on some of the big advantages and disadvantages of a multi-stage VAT:
The key advantage was that the burden did not rest on the final business-to-consumer (B2C) leg of a supply chain – even if the retailer failed to comply, most of the VAT would be collected.
The key disadvantages were that there is significant “churn” for “good” business-to-business (B2B) transactions, creating unnecessary cash flow “spirals”; and
“Missing trader” fraud or evasion, where an entity with obligations for net GST disappears (with its customer still getting an input tax credit), or even worse where an entity makes up input tax credits, gets a refund and disappears.
It is worth here examining developments overseas in relation to e-invoicing and how countries are using e-invoicing to address missing trader fraud and evasion, and comparing with the current Australian position.
Many countries have now made it compulsory to e-invoice, at least B2B and B2G, and remit that information in real time to the local tax authorities. This does seem to have eliminated a fair bit of missing trader fraud. It is important to note that it does not address all problems in VAT systems, such as evasion by the end trader (such as omitted income, which is possible even in countries with mandated B2C e-invoicing – it just means the transactions don’t go through the till). It is using new capacities to chock up problems in a “pre-digitalisation” VAT.
Australia to date has taken a different view on e-invoicing, being eager to encourage e-invoicing take-up for productivity and anti (commercial) fraud reasons (for example email invoice fraud), but not mandating it (even B2B / B2G), let alone sending the information to the ATO. Part of the thinking underpinning this was that the underlying missing trader fraud and evasion was not as problematic in the Australian economy, due to the general level of voluntary compliance of Australian taxpayers. Importantly, under a non-mandatory system, this data is not particularly useful for the ATO, in either providing a service to taxpayers or compliance activities, due to lack of confidence in completeness (would we be just producing a data set that has to be reconciled to the natural system of the business, its accounting software).
Rather, the Australian system has tried to address these challenges through mechanisms like taxable payments annual reporting and “no-ABN” withholding (with limited reverse charging in problematic industries like gold refining).
Reflecting on this, I think there’s a bit of a cross-roads here as to how to use digitalisation to improve the administration of the GST.
One way, is to decide that the general economic benefits and incidental GST compliance benefits are worth the extra administration burden, follow other countries, and go down the mandated e-invoicing path (at least B2B and B2G).
But there might be another more optimistic path to use e-invoicing and other digitalisation developments (such as the New Payments Platform) in a different way, which would also enhance the integrity of the GST system in terms of missing trader fraud and evasion, but could also help it operate in real time, with real time withholding.
(As an initial aside, it is worth noting that net taxes are very difficult to withhold in real time. How do you know how much to withhold? A flat 10% withholding would be very unfair for a business which has a lot of input tax credits), but a more precise ratio risks revealing commercially sensitive information about profit margins).
So imagine if e-invoiced transactions between “good” taxpayers (B2B and B2G) were reverse charged instead of being subject to GST, with then an input tax credit being claimed by the recipient. Suddenly, instead of the supplier charging $110 and remitting $10 to the ATO (possibly before the invoice has even been paid), and the recipient being charged $110, and waiting for a $10 ITC from the ATO, for 2 typical businesses the invoice would simply be for $100 with no net GST flows.
This could mean that many small (or even large) businesses with solely business clients and suppliers would have their GST compliance massively reduced, and their cash flows much more certain due to the elimination of churn. Wholesale businesses may even have no GST cash flows.
The flip side is that the entire GST burden would now usually fall on the final B2C leg of the chain, much more like a retail sales tax. This puts more compliance pressure on the tax administrator to make sure that obligations are met. But it also means that the GST burden is now a percentage of gross, so much more amenable to a withholding tax administered within the financial system (eg a platform or even a bank).
For those in the financial services system, instead of paying the supplier $110, you would pay the supplier $100, and then subsequently remit $10 (or a lesser amount if there was a partial input tax credit) to the ATO.
Obviously this sort of idea would require a lot of thinking and exploration. But I think it is a worthwhile thought experiment as to how you would design a VAT in 2026 supported by the current state of technology and the digitalisation of the economy, in contrast to 1999.
Conclusion
I hope I’ve provided a bit of context – about the tax system as a whole in Australia, where the financial services industry sits in terms of its own tax performance, and also where I think the tax system will increasingly be embedded in, and reliant on, the financial system more broadly.
I’ve also shared with you my bricklayer musings about where elements of the tax system might head over time. I doubt any of that is imminent, but they’re useful things to think about in terms of where tax administration may evolve.
I’m very happy now to open things up to questions. Thank you, and I am sure you will all get a lot out of the rest of the conference.