Jason Nassios, Deputy Director and Associate Professor, Centre of Policy Studies, Victoria University, James Giesecke, Professor, Centre of Policy Studies and the Impact Project, Victoria University
Australia’s tax system has come under scrutiny again ahead of the government’s reform roundtable later this month. Economists argue we could raise the same revenue with less economic harm by relying more on efficient taxes and less on inefficient ones.
Our new study offers fresh evidence that two of the most inefficient taxes, stamp duty and insurance taxes, remain harmful even at very low rates.
This strengthens the case for abolishing these taxes altogether, rather than merely scaling them back.
What makes a tax inefficient?
When economists call a tax “inefficient”, they mean it distorts economic activity by discouraging work, saving, investment or mobility. That creates an “excess burden”, a cost to society beyond the actual tax dollars collected.
The economic damage inflicted by raising a dollar of extra tax revenue is referred to as the “marginal excess burden”.
To illustrate, if a particular tax has a marginal excess burden of 25 cents, it means for each extra dollar the government raised through this tax, the economy would shrink by 25 cents. The reduction in economic activity comes from disincentives and distortions. The 25 cents is a pure loss: no one receives it.
Broad-based taxes like the GST and personal income tax have relatively low economic costs, especially compared to narrow taxes such as stamp duties and insurance duties.
What we did differently
Past studies have compared the economic damage of different taxes at current tax rates. But major reforms, like replacing stamp duties with higher income or consumption taxes, require large shifts in tax rates, which can alter how damaging each tax is.
Economic damage per dollar of tax revenue generally falls as a tax’s rate falls. This raises a question: could highly distortionary taxes such as stamp duty and insurance taxes become more efficient if they were set at much lower rates?
To test this, we asked: How does a tax’s economic damage change as we raise or lower its rate?
We built a detailed model of Australia’s economy, representing each state and territory separately, with all major taxes included at a granular, regional level. We then simulated how economic welfare changes as revenue from four key taxes is increased or reduced:
- personal income tax
- goods and services tax (GST)
- stamp duties
- insurance duties.
We focused on these four because they are frequently mentioned in Australian tax reform debates, and because national taxes (personal income tax and GST) are sometimes proposed as replacements for state-based ones (stamp duty and insurance duties).
The result: two taxes that remain inefficient at any level
The table below shows the economic damage, measured in cents of lost welfare per dollar raised, for each of the four taxes. In our journal paper, we also modelled a range of different options for each tax.
For example, stamp duty currently raises revenue equivalent to about 1.1% of GDP. At that rate, the last dollar of stamp duty raised causes 76 cents of economic loss. Our results show that even modest reforms will help: reducing stamp duty to raise revenue equal to 1% of GDP lowers the damage, to 73 cents.
However, our results also show that even at low revenue levels – just 0.01% of GDP – stamp duties and insurance duties still impose high marginal excess burdens: 39 cents and 31 cents, respectively.
These taxes have narrow bases and distort behaviour. Stamp duty discourages mobility, such as downsizing property or moving for a new job, and insurance duties discourage insurance uptake, making these taxes poor tools for raising even modest revenue.
In contrast, the GST and personal income tax cause much less economic damage at current rates (about 24 cents per dollar raised) compared to stamp duties or insurance taxes. While their efficiency declines as their revenue share of GDP rises, they still outperform stamp duty and insurance taxes.
This matters. Keeping inefficient taxes but raising only small amounts may seem politically easier, but our results show it still imposes high economic costs per dollar. Narrow based taxes like these should be removed entirely from Australia’s system.
A better tax mix could lift national welfare
We also modelled a revenue-neutral tax swap: abolish stamp duty and insurance duty, replacing the lost revenue with higher GST and personal income tax.
The most efficient mix – 48% of the replacement revenue from GST and 52% from income tax – would deliver a welfare gain equivalent to a one-off payment of about A$935 per household, and lift the economic damage of GST and personal income tax from 24 cents to 25.1 cents per dollar raised. This is much lower than both stamp duty and insurance taxes, even at very low revenue levels.
States could achieve similar gains independently of the federal government. In earlier work, we estimated payroll tax, another broad-based state tax, has an efficiency cost of about 22 cents, similar to personal income tax and GST. If the states acted together, payroll tax could be used to replace stamp and insurance duties.
The Australian Capital Territory is already phasing out stamp duty and Victoria has abolished stamp duty for commercial and industrial properties.
The path forward
Reform is not simple. Inefficient levies such as stamp duty and insurance taxes fund a large share of state budgets. Eliminating them would create major revenue gaps.
Letting Canberra raise the replacement revenue would worsen Australia’s already high fiscal imbalance between the states and federal government. That would deepen the divide between those who spend and those who tax, and weaken democratic accountability.
The alternative, getting all states to raise payroll taxes in unison, would be just as difficult, requiring complex negotiations.
Still, the economic case is clear: some taxes, especially stamp duty and insurance duty, are inefficient at any level. Replacing them sooner means greater long-term gains for Australians.
The authors do not work for, consult, own shares in or receive funding from any company or organisation that would benefit from this article, and have disclosed no relevant affiliations beyond their academic appointment.
