If you want to shock and appal a politician, just suggest Australia join the United States and China in limiting the building of new coal mines.
Think of all the growth we'd be giving up, they protest. All the jobs that wouldn't be created. Some even argue we have a moral duty to sell more coal to the world. How else will the poor countries be able to develop their economies so they become as rich as we are?
Short answer: by relying more on other, less carbon-emitting forms of energy.
Surely the sooner we arrest global warming the better off we'll all be, rich and poor.
The goal of the moratorium on new mines is to hasten the process of decarbonising economic activity.
It's clear the world's growing commitment to action against climate change will see a decline in the demand for coal - the most emissions-intensive way to make electricity - so that much of our huge deposits of coal will stay in the ground.
It's true there's a lot more coal to be burnt before world demand dries up, but total consumption actually fell in 2014-15. Within that, China's consumption fell by 3.7 per cent.
The big fall in coal prices in recent years tells us the supply of coal now exceeds demand. With Australia accounting for 27 per cent of seaborne trade in coal, what happens if we expand our production capacity and start exporting more?
We push the world price down even further. Since the average cost of electricity from renewable sources is, as yet, higher than for coal-based power, this would worsen the comparison further, slowing the shift away from fossil-based electricity.
It would also lower the prices being received by our existing coal exporters, threatening employment in their mines. So a moratorium would benefit our pockets as well as the environment.
But how much would we lose by not building any more coal mines nor extending existing ones?
The Australia Institute set out to answer this question with help from modelling by Professor Philip Adams, of the Centre of Policy Studies at Victoria University, Melbourne.
The study found that, even with a ban on new mines, Australia's coal production would decline only gradually as existing mines reached the end of their economic lives. Existing mines and those already approved could still produce tens of millions of tonnes of coal into the 2040s, assuming other countries still wanted to buy them.
The modelling suggests the nation's economic growth would be barely affected, with the level of gross domestic product being just 0.6 per cent less than otherwise by 2040. Whether we did or we didn't, nominal GDP would roughly have doubled to $3 trillion by then.
Because coal mining is so capital intensive, the effect on national employment would be even smaller. By 2030, the level of employment would be 0.04 per cent lower than otherwise, but by 2040 this difference would have gone away.
Similarly, the value of our total exports of goods and services is projected to be only 1 per cent lower than otherwise by the final years of the period.
But our coal production is concentrated in NSW and Queensland, so the adverse effect on those state economies would be greater. By 2040, the level of gross state product would be, respectively, 1.3 per cent and 3.8 per cent less than otherwise, while the other states' GSP would be a little higher than otherwise.
Now, I trust that by now you've learnt to be cautious about accepting the results of modelling exercises, especially when they've been sponsored by outfits using the results to advance their cause, as is the case here.
The simple truth is that no-one knows what the future holds, and that's just as true for the econometric models economists construct.
Their models of the economy are more comprehensive and logically consistent than the model we hold in our heads. But relative to the intricacy and complexity of the actual economy, models are still quite primitive (this one doesn't have the official data to let it distinguish between steaming coal and coking coal, for instance).
Models are built on a host of assumptions, some based on economic theories about how the economy works and some about what will happen in the future.
The strength of this particular modelling exercise is that it's a lot franker about the model's limitations and about the specific assumptions.
It uses a dynamic "computable general equilibrium" model designed to capture the interrelationships between 79 industries, divided into states and regions.
The model takes account of "resource constraints" - it acknowledges that land, labour and capital are scarce; that everything you do has an opportunity cost.
This means that, unlike much "modelling" produced for the mining lobby, it doesn't assume that the skilled workers needed for a new mine just appear from nowhere rather than having to be attracted from jobs elsewhere, nor that when a new mine isn't built, all the labour and materials that could have been used sit around idle.
As is normal, the modelling starts by establishing a business-as-usual "baseline" projection out to 2040. For instance, real GDP is assumed to grow at an average annual rate of 3 per cent for the first five years, then 2.6 per cent for the remaining 20 years.
Once this baseline or "reference case" is established, the modellers impose the policy change (no new coal mines) and run the model again to see how this changes the baseline results.
That is, it's not a forecast, just an attempt to get an idea of the consequences of banning new coal mines.
The model's modest results make sense. The effects would be small because the coal industry is just a small part of the economy, because the phase-out would be gradual, and because other industries would expand to fill the vacuum it left.
