Are carbon taxes (another) Australian 'magic pudding' policy?

The Government's proposed carbon tax is being asked to do a lot. It's supposed to (i) reduce greenhouse gas emissions, (ii) include 'compensation' for (some) trade-exposed industries, and (iii) finance (some?) compensation for households. Ross Garnaut also wants it, potentially, to finance (iv) other tax reforms included in the Henry Review, and (v) $3 billion a year for low emissions technology innovations.

A cash-strapped Government promising to get the Budget back into surplus shouldn't promise too much for its carbon tax. Its Budget credibility will suffer.

Consider some scenarios.

Assume a modest carbon tax from 1 July 2012, with no initial effect on greenhouse gas emissions, raising $X million per annum. Assume this tax costs nothing to collect. Assume all carbon tax costs, one way or another, feed through to higher consumer prices. Assume all tax revenue is used to finance a cut in the rate of GST.

This revenue-neutral tax change should produce roughly zero change in the Consumer Price Index (CPI) on average. Emissions-heavy product prices rise, others fall.

On average, consumers' incomes (whether earned income or government benefits) and accumulated savings will retain roughly the same real purchasing power, even if consumer behaviour does not change.

All carbon tax revenue is expended providing consumer compensation via a lower GST rate. None is available for the second, fourth and fifth objectives listed above, unless they exactly coincide with the household compensation objective. They don't.

Now progressively remove the assumptions in this scenario.

Administering a carbon tax requires significant additional Budget resources – significantly more than a single-rate GST. These reduce carbon tax revenue available for consumer compensation and other purposes. Now carbon tax revenue can't even pay for 100% consumer compensation without new Budget savings. Other tax reforms are, of course, beyond this carbon tax budget.

Now suppose consumers are compensated for the carbon tax by income tax cuts and increased welfare benefits rather than GST cuts (as seems likely). The carbon tax still delivers just $X million per annum.

Now the CPI increases. All prices increase due to the tax on emissions embedded in all products.

This CPI increase means total compensation costing more than $X million per year is required, to maintain consumers' real purchasing power from accumulated super and other savings, beyond that required to maintain the real value of incomes. More new Budget savings are needed to deliver household compensation plus other reforms.

Now assume the carbon tax actually reduces emissions to some extent (the object of this exercise).

All product prices still increase, either because emissions continue and are taxed, or because higher-cost, lower emissions, energy production is substituted. Total carbon tax revenue is less than $X million per annum as a result of partial avoidance of the carbon tax.

Carbon tax revenue now can finance even less consumer compensation, and even more new Budget savings are needed to deliver full compensation. Other reforms are even further out of reach.

What's going on here?

The reason for trying – again – to price greenhouse gas emissions is to change the way we produce and consume goods and services.

We want to reduce greenhouse gas emissions by making them costly – don't we? The carbon tax isn't just an excuse to raise more revenue to finance other policies – is it?

A carbon tax intended to reduce greenhouse gas emissions shouldn't be used to finance other policy reforms. It cannot do both jobs. Ineffective climate policies raise revenue. Fully effective policies don't.

For Ross Garnaut's latest policy reform financing proposals, neither effective nor ineffective climate policies offer any fiscal comfort.

If emissions are reduced because higher-cost, lower-emissions, technology is used, carbon tax revenue will be reduced to that extent. If emissions are not reduced at all as a result of the carbon tax, then new tax revenue is raised, but not enough even to finance 100% consumer compensation, let alone other policy reforms.

(However, a principled, comprehensive and consistent way of dealing with objective (ii), the trade-exposed sector, is possible. It requires a comprehensive carbon tax based on national consumption of emissions instead of national emissions production.)

This leaves three sets of questions for the Government.

First, who will be compensated for losses in real income arising from imposing a carbon tax, and who will miss out?

Second, do the politics of a new tax militate in favour of an ineffective (ie, very low carbon price) emissions policy? Won't this mean more consumers can be compensated, (and conversely for an effective policy)? If so, why should Australians accept a new ineffective tax?

Third, whatever the effectiveness of the carbon tax, if all consumers will be compensated, how will this be financed, and how will other tax reforms be paid for? What new Budget savings are being developed?

These questions apply to, and also beyond, 1 July 2012.

If real emissions taxes are increased over time, whether by 4% per annum or any other amount, the same questions arise on every such occasion.

Several policy objectives are being mixed up in the 'selling' of the Government's carbon tax. Its primary purpose – reducing emissions – seems in serious danger of being lost to other policy agendas.