A lot of problems have no good solutions – just ones that are bad in different ways.

Pricing greenhouse gas emissions in agriculture is important, and every way of doing it is going to have problems.

This week, the agricultural sector put up its proposed solution. It has problems. But it is worth looking at the thorny mess that it has to deal with.

In a better world, every country would price all greenhouse gas emissions – whether industrial or agricultural.

Every pricing regime brings risk of ‘leakage’ – that companies simply shift high-emission production to places where emissions are not priced.

The Emissions Trading Scheme provides free carbon credits to businesses in industries with substantial leakage risk. So long as companies can profit by reducing their emissions and selling off their allocated credits, they have a strong incentive to reduce their emissions. If their emissions increase above their free allocation, they must buy carbon credits like everyone else, and face a price that’s now close to $80 per tonne emitted.

This kind of allocation regime reduces the risk of carbon leakage while maintaining incentives to reduce emissions. There are ways of improving on how New Zealand does it, but the basic principle is sound.

Leakage risk is particularly high in agriculture. Nobody else puts a price on agricultural emissions. If emissions pricing results in fewer cows and sheep on New Zealand pastures, and more in feedlots overseas instead, global emissions could wind up going up rather than down. And the climate doesn’t care where a cow burps.

So any good system needs a way of dealing with leakage.

It also has to recognise that short-lived gases like methane need to be treated differently from long-term gases like carbon dioxide.

The sector’s proposal to handle the problem with a very low price on methane does not seem like a first-best or even a second-best.

Alternatives might have applied a higher methane fee while providing rebates for products exported to places where the competition does not face a methane price. A scheme could be designed to maintain incentives to reduce emissions – but it would not be simple.

But it gets a price on methane, which is a start. And neither the first-best nor the second-best should be the enemy of that which can practicably be implemented now, with a few more minor improvements.

It is a start. And a start can be a good thing when nobody else has yet even tried.