Our recent submarine purchases have been justified on economic grounds: to protect our trade routes through Asia.

China and Japan have a toxic relationship, but a key reason they haven't gone to war since the end of World War II is that their economies are so integrated through trade and direct investment. War is simply too costly.

The economy is a key variable in the calculus of war. So, it's worth thinking about what the costs of war would be to the Australian economy.

This is a complex question, but some back-of-the-envelope simulations plugged into a sophisticated computable general equilibrium model of the Australian economy give some useful insights.

When it comes to the economic impact of war, there are knowns and unknowns.

Let's start with a known: trade. If war broke out over Taiwan, it's a reasonable assumption that trade flows between China and Australia would come to an abrupt halt, or would at least be seriously impaired.

Australia exports $181 billion of goods and services to China each year, with about $104 billion coming the other way through imports.

But it's not correct to simply subtract these from the Australian economy.

As China's trade restrictions have showed us, prices, exchange rates and markets adjust and we find new customers/sellers for most things.

These adjustments would help buffer the trade shock on Australia in the event of war. But the sheer size of our trading relationship with China means the cost will still be big.

How big? It would put Australia in recession for at least three years, according to my preliminary estimates, with GDP 11 per cent smaller over five years than it otherwise would be.

Wages fall by 20 per cent over 10 years. The mining and manufacturing sectors take the biggest hits. Mining output falls 11 per cent in the first year alone.

Market adjustments help buffer the shock - the Australian dollar falls 12 per cent which helps bring in new customers - but the impact is still huge.

The other known is the federal budget. Putting aside the military cost of war (which depends on our level of involvement), the federal budget loses about $1.3 billion of revenue for every $10 fall in the price of iron ore.

With prices for our mining exports falling by about 17 per cent from a cessation of trade in the first year alone, the impact on the budget would be in the tens of billions.

Investment from China would likely cease as well. Luckily, contrary to popular opinion, the amount of investment that comes from China is relatively small. But it would nevertheless represent a gap that would need to be filled by domestic savings brought about through even higher interest rates.

That's the good news. The unknowns are much scarier.

READ MORE ADAM TRIGGS:

The first unknown is what happens to supply chains. Economic models can't accurately capture the complexity of global supply chains which are constantly changing.

More than 70 per cent of world trade is now in intermediate goods, meaning the things we import and export go into the production processes of other things, which themselves feed into the production processes of other things and so on.

The impact of war on supply chains is like a dangerous game of whack-a-mole: you never know where the consequences will pop up.

The scariest unknown - and the unknown which will most prominently shape the impact on Australia - is how financial markets respond, and how they judge Australia.

A war would see global capital flee to safe haven assets (like gold) and safe haven countries and currencies (like the US and the US dollar).

The question is how markets treat Australia. Markets have treated Australia as a safe haven in recent years as investors buy up Australian dollars, equities and bonds as a safe place to store their money.

Would this happen if there was a war involving China? It's less clear. Markets are acutely aware of China's importance to the Australian economy. And if they see Australia as being an active participant in a war, they are even more likely to treat Australia as a risky country rather than a safe haven.

This is a serious problem. To put a number on it, if global investors demanded a risk premium of just one percent for investing in Australia, Australian GDP would be 1 per cent lower each year.

It hurts China, too. A 1 per cent increase in risk in China reduces Chinese GDP by 7 per cent over 5 years - a serious understatement in the event of a war. A trade freeze with Australia alone reduces Chinese GDP by 4 per cent over 10 years given their reliance on key minerals.

So, what's the lesson here?

The lesson is that war is costly, and that's a wonderful thing. If war was cheap, we would see more of it, and that would be a terrible thing.

We want war to be as expensive as possible. And what this analysis shows is that the thing that makes war very expensive is economic integration.

The more countries trade with each other, the more countries invest in each other, the more immigration we have and the more tourists and students we exchange, the more expensive war becomes.

Those who advocate cutting economic ties with China and advocate isolating China from the world are proposing a world where war between nuclear powers is cheaper.

Are you sure that's a good idea?