Earlier this month, I attended a very interesting conference on the taxation of multinational corporations. It included a case study of how SABMiller avoids paying tax in Africa. While many of the points presented are undoubtedly familiar to this blog’s readers, the conference put it all together with a clarity that I attempt to reproduce below.
The Trouble with Transfer Pricing
A significant amount of international trade consists of transfers between subsidiaries of the same corporation. The prices that the corporation charges itself determine how its profits are divided between different countries for tax purposes.
Imagine that a corporation produces an input in country A and then ships it to manufacture a finished product in country B. By changing the price at which subsidiary A sells the input to subsidiary B, the corporation can shift its profits between the two countries.
If country A has a higher tax rate, the corporation might try to lower the input price until subsidiary A just breaks even on producing it. Then all of the corporation’s profits would accrue to subsidiary B and be taxed at country B’s lower rate.
This scheme can be foiled if the same input is also traded between unrelated corporations. Country A’s tax authorities can calculate what subsidiary A’s profits would have been had it sold the input at market prices, and tax the corporation accordingly. That is basically the OECD’s position on transfer pricing: corporations should charge themselves the same prices as would prevail in arm’s-length transactions.
Services, Intangible Assets and Debt
Transfer pricing becomes more problematic for specialized items not readily comparable to identical items traded on an open market.
Imagine that country C has ultra-low tax rates. The corporation could set up subsidiary C in this tax haven and have it sell “business services” to subsidiaries A and B. It is difficult for tax authorities to determine what such services are actually worth or to prove that they are just an excuse to shift profits to country C.
Perhaps the corporation sells its finished product under a popular brand name. The corporation could decide that subsidiary C owns this brand and that subsidiaries A and B must pay royalties to use it. Even if the brand is valued accurately, the profits from it can be allocated wherever the corporation wants.
The corporation could have subsidiary C make loans to subsidiaries A and B. Even if the loans are at market interest rates, the corporation can shift profits between countries in setting the size of these loans.
Formulary Apportionment: A Solution?
An alternative to transfer pricing is to tax each multinational corporation as a single entity. The American federal government does so by taxing the worldwide profits of US-based corporations minus taxes already paid to foreign governments (up to the US federal rate). However, this approach does not stop American multinationals from playing around with transfer pricing before repatriating their profits to the US.
To replace transfer pricing, one would need a global system of formulary apportionment. The worldwide profits of each multinational corporation would be divided among countries using a formula based on tangible measures of where it sells products, where it employs workers, where it has physical assets, etc.
While this idea may seem fanciful, federal countries actually use formulary apportionment to divide corporate profits between sub-federal jurisdictions. For example, if a corporation operates in more than one Canadian province, Ottawa allocates its profits among provinces based on its sales and payrolls.
In the US, Washington allocates corporate profits among states based on sales, payrolls and physical property. The European Union is developing a system of formulary apportionment to distribute profits among member states.
Where there is a governing authority capable of applying formulary apportionment, this system has been chosen over transfer pricing. Globally, there is no such governing authority. Agreeing on a common formula was difficult for Canadian provinces and could be even more difficult for different countries.
Since a global system of formulary apportionment looks to be far away, it seems reasonable for international economic organizations like the OECD to continue trying to make transfer pricing workable. However, such organizations should also start discussing international formulary apportionment as an option rather than rejecting it out of hand. (Travis may contend in the comments section that we need to start discussing nationalization as an option.)
Foreign Takeovers: A Canadian Epilogue
Foreign takeovers, a hot topic in Canada, did not come up at the conference. One problem is that the costs of financing a takeover are often deducted from the profits of the acquired enterprise.
When a multinational mining company buys a Canadian mining company, the multinational company now has the right to deduct the takeover costs in calculating its Canadian taxes. Under formulary apportionment, the same takeover would give the Canadian government the right to tax a portion of the multinational corporation’s global profits.
- Weir vs. Wall on Potash Profits, Dividends and Royalties (December 6th, 2013)
- PEF Session at the House of Commons Finance Committee (December 2nd, 2013)
- A Nuclear Error: Uranium Royalty Cuts (December 1st, 2013)
- The NSA Scandal is all about Economics (November 2nd, 2013)
- The Blackberry mess and what Canada needs (September 24th, 2013)