Malcolm Turnbull has called for a dramatic shift in Australia’s model of federalism, specifically seeking to allow the states to levy their own income taxes. Many economists regard this as sensible and much-overdue reform to correct a deeply entrenched problem in Australia’s fiscal constitution – vertical fiscal imbalance.
But it was met with two types of outrage and scepticism.
The first was from those who heard it as more taxation. But the argument is for a shift from a federal income tax to a state income tax. In principle, this can be done in a completely revenue-neutral way.
That would make it tax reform, not a tax grab. And that would, on the whole, benefit Australian taxpayers because a more efficient tax system is a less costly tax system.
The second came from the state premiers. With the exception of Western Australian Premier Colin Barnett, they hated the idea.
It’s important to understand why. This is not because the idea is bad for the citizens of the states, with the premiers being outraged on their behalf. Rather it is because it is bad for the political classes themselves, and the premiers in particular.
To see why, we need to introduce a somewhat obscure economic model from public finance, developed by Charles Tiebout in 1956.
Before Tiebout’s work, the best economic thinking about the provision of public goods argued that a powerful central government should provide these in order to overcome the free-rider problem, which required the use of political force to compel those who would seek to consume the public goods or services but avoid paying for them. No one can hide from a federal income tax, so no citizen can free-ride.
The major flaw in this mechanism is what economists call the “preference revelation problem”. Centralised funding discourages citizens from honestly signalling, through their votes, their true preferences for local public services (such as education and healthcare) and what they are willing to pay for these in taxes, because regardless of what you might want, you pay for the same set of public goods – i.e. there is just one level of income taxes.
The result is that when services are provided at the state level, but paid for at the federal level, every premier has incentive to claim their citizens want the most and best services possible. Indeed, they would be crazy not to when other states are jointly paying for them.
The result is a perpetual war between state and federal treasurers, a lot of economic choices converted into political deal-making, and an eternal problem of vertical fiscal imbalance.
But Tiebout pointed out there was a non-political (or market) solution to this problem: namely federalism, or what came to be called “voting with one’s feet” and can equally be called “shopping for public goods”.
The basic idea is that the states compete with each other by offering bundles of public goods at different prices (i.e. taxes). This is the significance of the state-level income tax. Victoria, for example, may offer very high levels of public services, but also at a high price through high state income taxes. NSW may offer more moderate public services, but also much lighter taxes.
What happens next is that citizens sort themselves over the states according to their preferences. Those who value high levels of public services move to Victoria, where they pay that marginal valuation in high taxes. Citizens with preferences for lower levels of public services and also taxes move to NSW.
This is a market, not a political, model of local public goods. Economists like it because it encourages competition between the states to provide an efficient bundle of public goods and services at a point that voters (as consumers) are willing to pay. This competition tends to produce an efficient outcome. Consumers reveal their true preferences for public goods and services by their choices about what state to live in.
The caveats
Now obviously there are limitations to this process. There are costs to moving, and for many citizens these costs will be prohibitive. But the model does not require everyone to move, just that the incentives work at the margin. In which case, the premiers are also incentivised to seek to provide the bundle of public goods and services that their constituents want and are willing to pay for.
That is not the situation we have now. Premiers are incentivised to represent their citizens as all wanting the maximum amount of public goods and services, because someone else is paying for them.
State income taxes (coupled with reduced federal income taxes) are a way of implementing this mechanism. The main winners from this will be the 7 million or so Australian taxpayers, because it will deliver a much more efficient supply of public goods and services. The main losers will be the state and territory premiers, because they will have to compete in the market for political goods and services. So it’s probably an even fight.