Robert Peston's article today about Chinalco highlights an important point which few economics commentators have discussed.
For the last nine months oil prices have been falling substantially. The developed economics spend so much money on oil that this makes a huge difference to the money available in our pockets. As a fair estimate we can say that world oil consumption is about 80 million barrels a day.
Taking the lowest figures it's reasonable to say that last summer, the world was spending $12 billion a day on oil. Now, with a barrel of oil at $44 the figure is less than $4 billion.
This $4 billion every day is now sitting in our pockets - exactly as if it came to us from a VAT or payroll tax cut. This is equivalent to a $3 trillion per annum fiscal stimulus paid for not by our own government but by oil exporters. How nice of them.
(Caveats: some oil is provided on long-term contracts rather than spot price, so the movements won't have such impact; and the oil exporting countries will lose rather than gain from this change. So the net effect is probably closer to $2 trillion).
Second point: the Chinese government, and other cash-rich people, can still spot a good value asset when they see one. Some assets - such as the Rio mining company - have a relatively predictable underlying value, and a party which is not cash-constrained would be crazy not to buy right now. On a smaller scale, there are people buying cheap houses in London and VCs are still investing in companies at lower valuations than last year.
This is nearly as good as an asset-price support program by a central bank: some people would prefer the Bank of England to own their shares than the government of China, but it doesn't really make that much difference. The quantitative effects are not the same because China is buying the asset with existing cash - presumably selling US government bonds to do so - but anything that supports the price of other assets relative to Treasuries is useful.
Finally, the automatic stabilisers that operate in most governments have not been factored into many analyses of the stimulus. The UK government is expecting to run somewhere around a £100 billion deficit this year - and yet they are criticised for only announcing a £20 billion overt stimulus. The other £80 billion is just as stimulative, if not more so, because it is part of people's natural income and more likely to be spent.
Other examples: US stimulus package: $800 billion over several years - government deficit in 2009 alone: $1.5 trillion. French stimulus package: E26 billion - projected deficit for 2009: 5.6% of GDP or E131 billion.
Other governments such as China may not run a deficit in addition to their stimulus package, but can no longer run the same levels of surplus they previously did.
Most of this comes from tax revenues that naturally decline, and to a lesser extent transfers that naturally expand, during an economic slowdown. This automatic stimulus seems likely to be at least at the level of announced stimulus packages - which add up to around $3 trillion according to Kevin Gallagher on Dani Rodrik's blog - and possibly twice as high.
The great thing is that - given appropriate government policy - all of these factors happen automatically, through the miracle of economic exchange. And thus we can have some confidence that they will meet the Summers criteria of being temporary, timely and targeted.
I am hesitant to put a figure on these three effects without being able to back it up, but it is not unreasonable to suggest $2 trillion from oil, at least $1 trillion of asset purchases and $3 trillion of deficit spending - a total of $6 trillion per annum or 12% of global GDP.
I do not suggest that governments should not also continue their own stimuli, to supplement this and to make visible the fact of government intervention. But we can be justified in hoping - even expecting - that demand will pull out of its decline soon.
Update: A confounding factor is that some of this "stimulus" may not show up in GDP figures directly. A $3 trillion reduction in oil prices has a direct contractionary effect of about 6% on world nominal GDP, but should have a positive effect on the consumption of real goods and services. Conversely, an increase in saving could boost investment and thus measured GDP, but without improving (short-run) consumer welfare.
Thus we might have a real recovery in living standards without an increase in GDP, or vice versa. Macroeconomic statisticians have a tough job.