An important speech by the Federal Minister for Major Projects, Paul Fletcher, has given fair notice that the Turnbull government will force the states and territories to capture the increased property values that follow improving infrastructure.

And a good way to achieve "value capture" is via a broad, no-exceptions land tax. Hence the excuse the states need to grow a spine and introduce the most obvious of the major tax reforms that the nation needs.

Of course Fletcher didn't quite put it like that, but it's there between the lines when he said:

"If the Commonwealth were requested to provide grant funding to a state government to support a particular project, we might make it conditional upon the state government demonstrating that the project was supported by a suitable value sharing strategy.

"That might involve the rezoning of land around new stations to permit multi-story apartment blocks, allowing a significant increase in land value which could be tapped under a value sharing approach to contribute towards the cost of the new railway line."

It has been a glaring missed opportunity for states not to capture the increased value created when infrastructure is improved. There are experiments around the edges by some states and local governments – the humble development levy is a form of "value capture" – but nobody has been politically game enough to take it seriously.

As the situation generally stands now, if the government builds a new train line, it provides a windfall capital gain for those people owning land around the new station. Developers tend to capture a disproportional share of subsequent rezoning and, in the case of owner-occupiers, it's a totally tax-free boost courtesy of other taxpayers.

With an estimated $800 billion infrastructure backlog and state budgets somewhere between a problem and dire, capturing a fair share of that windfall is a reasonable thing to do. It also provides some incentive for governments to build the infrastructure the nation needs.

Now the federal government is telling states to capture value as a condition of getting federal money for projects. That's the excuse the states apparently need before doing what should be done, but it's also the method for Canberra to co-ordinate the abolition of stamp duty on real estate transactions – the other obvious reform that flows from a broad land tax.

The political risk for any state government acting alone is that the opposition will start chanting "big new tax" with the extra bonus of the family home's tax haven status being Australia's sacred cow. Co-ordinate the reform, share the political pain, and it becomes possible.

Among the benefits of a no-exceptions land tax are its efficiency – easy to collect and nigh-impossible to avoid – and that it's inherently progressive. The better-off tend to own more valuable land.  They are benefits that increasing the GST can't claim.

Unlike variations on developer levies, land tax keeps collecting. It's a more reliable source of income than stamp duty that waxes and wanes with the cycles. Compared with stamp duty, it's also much more equitable as everyone ends up paying for government services, not just those who buy property.

And an honest land tax encourages more efficient use of land in the longer term. Replacing stamp duty with land tax doesn't have much immediate impact on housing affordability but over time the more efficient land use it promotes should result in greater and more appropriate supply.

The commonwealth and states can get fancy with financial engineering of uplift bonds or whatever, but the underlying factor is the increased value of the land. Go for it and blame the feds.

Michael Pascoe is a BusinessDay contributing editor.