"The Missing Profits of Nations"

Via a Resolution Foundation email newsletter, this fine study which confirms what we knew all along:

By combining new macroeconomic statistics on the activities of multinational companies with the national accounts of tax havens and the world’s other countries, we estimate that close to 40% of multinational profits are shifted to low-tax countries each year. Profit shifting is highest among U.S. multinationals; the tax revenue losses are largest for the European Union and developing countries...

Our findings have implications for policy. First, they suggest that cutting corporate tax rates, as the United States did at the end of 2017, is less likely to generate quick positive effects on wages than textbook economic models suggest. For wages to rise, productive capital needs to increase, which can happen fast if capital flows from abroad, much less so if paper profits—not productive capital—is what moves across countries.

'Rents' get an honourable mention:

Second, profit shifting raises new challenges for tax policy. It reduces the effective rates paid by multinationals corporations compared to what local firms pays. Whatever one’s view about the efficiency cost of capital taxes, this seems difficult to justify—especially if part of the profits of multinationals derive form rents, which standard models suggest should be taxed.

Having examined the evidence, the authors make a surprising claim:

We show theoretically and empirically that in the current international tax system, tax authorities of high-tax countries do not have incentives to combat profit shifting to tax havens. They instead focus their enforcement effort on relocating profits booked in other high-tax places—in effect stealing revenue from each other. This policy failure can explain the persistence of profit shifting to low-tax countries despite the sizable costs involved for high-tax countries.

Their explanation is on page 23:

To ensure profits are taxed where they have been made (i.e., the prevailing internationally agreed rules), tax authorities in high-tax countries routinely audit large companies. They check that intra-group transactions are conducted at arm’s length (i.e., as if the subsidiaries of a given multinational group were independent entities). When they find it is not the case, they can attempt to ask multinationals to correct their transfer prices, which results in a relocation of taxable income across countries.

In the current international tax system, tax authorities have incentives to relocate profits booked in other high-tax countries—not profits shifted to havens. Take the case of France. e1 relocated to France is worth the same to France whether it comes from Germany or from Bermuda. But it is easier for the French tax authority to relocate e1 booked in Germany, for three reasons.

First, it is feasible, because information exists on the profits booked in Germany (from Orbis), while no or little information typically exists on the profits booked in Bermuda. Second, it is more likely to succeed, because firms are unlikely to spend much resources opposing this transfer price correction: for them, whether profits are booked in France or Germany makes little difference to their global tax bill, since the tax rates in France and Germany are similar. Third, if there is a dispute between France and Germany, it is likely to be settled relatively quickly.

Seems plausible to me.

The answer is to simply disallow all expenses paid to businesses abroad unless the company claiming the deduction can:
a) prove that they have no connection with the other company,
b) explain exactly what the payment was for, and
c) show that this is an arm's length, market price.