Maya Forstater

This study looks at statistics on FDI and finds that they are unreliable, because ultimate investment (in a factory or a mine) in one country is double counted with on-paper investment in the holding company which owns the company that owns the factory or mine.

While the Danish research trio make reference to BEPS & the potential for holdcos to be used in profit shifting for tax avoidance, this is not what they are studying. What they are studying is why the statistics are unreliable as a measure of whats going on in the real world

But to be clear: what they have found is not 'phantom investment' as in underhand, dodgy, not reported to the tax authorities, but 'phantom' as in the international statistics are misleading.

And the really important point: What they call 'phantom investment' is not capital 'parked' in the Netherlands or Luxembourg or BVI, not creating real value in the real economy somewhere. Its just numbers that are double counted in the international statistics.

So yes, fixing the international statistics is important. But it doesn't mean that this capital could be freed up to do something else, because it is ALREADY doing something else (i.e. in the factory or mine or wherever that is the ultimate asset)

It would make much more sense to do the fraction as a percentage of "genuine FDI". i.e. if actual FDI is $24tn then their estimate is that 62% goes through holding companies in intermediate jurisdictions showing up as double counting in the statistics.