Offshore tax havens enable not only individuals to dodge taxes—they also enable multinational corporations to do so. Often this tax avoidance is done within the letter of the law: multinational groups exploit the loopholes of current legislation. The fundamental problem is that the corporate tax is not adapted anymore to today’s globalized world and must be reinvented. The spiral is profound, but here, too, solutions exist.
From Mountain View to Bermuda
The reason for the current failure is that the corporate tax is based on a fiction, the idea that one can establish the profits earned by each multinational subsidiary by subsidiary. But this fiction is no longer tenable today, because multinational groups, advised by great auditing and consulting firms, are in practice free to move their profits wherever they want, which is usually wherever it is taxed the least; and large countries have themselves mostly given up taxing the profits booked outside of their territory.
How do companies make their profits appear in tax havens? There are two main techniques. The first, that of intra-group loans, consists of loading with debt branches located in countries that tax profits heavily, such as France and the United States.The goal is to reduce the profits where they are taxed and have them appear in Luxembourg or in Bermuda, where they are taxed very little or not at all. This popular manipulation nevertheless comes up against a sizable problem: it is rather easy to detect.
The second optimization technique, the manipulation of transfer prices, plays a more important role. Transfer prices are the prices at which branches of a given group buy their own products from one another. Within a single company, the branches in Bermuda sell services at a high price to entities located in the United States. Profits thus appear again in the tax havens and losses in the United States, in the large economies of continental Europe, and in Japan. In principle, intragroup transactions should be conducted using as a reference the market price of the goods and services traded, as if the subsidiaries were unrelated, what is known as “arm’s-length pricing.” But arm’s-length pricing faces severe limitations. First, with billions of intragroup transactions every year, tax authorities cannot conceivably check that they are all correctly priced. And indeed there is compelling evidence of transfer mispricing by U.S. firms.
More fundamentally, in many cases the relevant reference prices simply do not exist. In 2003, less than a year before its initial public offering in August 2004, Google US transferred its search and advertisement technologies to “Google Holdings,” a subsidiary incorporated in Ireland, but which for Irish tax purposes is a resident of Bermuda. What was the fair-market value of Google’s technologies at the time of transfer, before the Mountain View firm was even listed as a public company? Google US had an incentive to charge as little as possible for this transfer. We do not know whether it was able to do so: the transfer price is not public information. But journalistic leaks in the fall of 2014, “LuxLeaks,” revealed that in many similar cases, the transfer prices that IT companies are able to charge when they send their intangibles to Bermuda is negligible, sometimes zero. Once that capital has arrived in Bermuda, all the profits that it generates are taxable there, where the corporate income tax rate is a modest 0%.
The issue is growing, as a rising number of international transactions within international divisions of a single company—such as the sale of proprietary trademarks, logos, and algorithms—are not replicated between third parties, hence have no reference price. Firms can sell themselves bananas or shovels at exorbitant prices—we’ve seen this—but the risk is high for companies that engage in such obvious fraud, as they can find themselves caught by the tax authorities. There is nothing less risky, by contrast, than manipulating the prices of patents, logos, labels, or algorithms, because the value of these assets is intrinsically difficult to establish. This is why the giants of tax avoidance are companies of the new economy: Google, Apple and Microsoft. Taxing companies wanes to the same extent as immaterial capital gains in importance.
Tax Avoidance by U.S. Firms: $130 Billion a Year
Quantifying the government revenue losses caused by profit shifting to lower-tax jurisdictions is not straightforward and, as with personal wealth, involves some margin of error. My approach relies on national accounts and balance-of-payments statistics, focusing on U.S. firms. Consider the basic macroeconomic aggregates of the U.S. economy in 2013. Corporate profits (net of capital depreciation and interest payments) account for 14.5% of U.S. national income, or $2.1 trillion. This figure includes $1.7 trillion of domestic profits, plus $650 billion of profits made by foreign firms owned by U.S. residents (mostly subsidiaries of U.S. corporations), minus $250 billion made by domestic firms owned by foreigners. Close to a third of U.S. corporate profits (650/2,100), therefore, are made abroad.
Where do the $650 billion of foreign profits come from? The balance of payments provides a country-by-country decomposition of this total: according to the latest available figures, 55% is made in six low- or zero-tax countries: the Netherlands, Bermuda, Luxembourg, Ireland, Singapore, and Switzerland. Not much production or sale occurs in these offshore centers; very few workers are employed there—profits appear in Bermuda by sheer accounting manipulations. Since foreign profits account for a third of all U.S. corporate profits, and tax havens for 55% of their foreign profits, the share of tax havens in total U.S. corporate profits reaches 18% (55% of a third) in 2013. The use of tax havens by U.S. firms has steadily increased since the 1980s and continues to rise.
By my estimate, the artificial shifting of profits to low-tax locales enables U.S. companies to reduce their tax liabilities, in total, by about $130 billion a year. Surveys of U.S. multinationals conducted by the Bureau of Economic Analysis show that U.S. firms pay a negligible 3% in taxes to foreign governments on the profits booked in the main low-tax jurisdictions. In the United States, contrary to what happens in most other countries, profits become taxable at a rate of 35% when they are repatriated (with a credit for all foreign corporate taxes previously paid). But in practice, there are few incentives to repatriate. The funds retained offshore can be used to purchase foreign companies, secure loans, pay foreign workers, and finance foreign investments, all of this without incurring U.S. taxes. An even more extreme scenario is possible: firms that would like to use their accumulated earnings trapped offshore as they so wish can merge with foreign companies, in order to change their tax residence and avoid the U.S. law, or what is known as a “tax inversion.”
In 2004 Congress granted a repatriation tax holiday, letting multinationals bring their foreign profits back home if they paid a rate of 5.25%. The holiday failed to increase domestic employment, investment, or R&D; it also boosted the foreign profits retained by U.S. firms in tax havens. Today only a tiny fraction of the profits recorded by US firms in Bermuda and similar havens are brought back to the United States, and this share is falling with expectations of new holidays. In the end, not only do the profits made in the main havens bear negligible foreign taxes; they also mostly go untaxed by the Internal Revenue Service. Since they account for almost 20% of all U.S. corporate profits, profit shifting to low-tax jurisdictions reduces the tax bill of U.S. companies by close to 20%—or $130 billion annually.
The Decline in the Effective Corporate Tax Rate of US Firms
A direct consequence of the increased use of tax havens is that the effective tax rate paid by US firms is declining fast. The effective corporate tax rate is the ratio of all the corporate taxes paid by U.S. firms (to U.S. and foreign governments) by U.S. corporate profits. Despite the fact that the nominal income tax rate in the United States has remained constant at 35%, the effective rate has fallen from 30% in the late 1990s to barely 20% today.
Granted, not all that decline should be attributed to increased tax avoidance. Some changes in U.S. laws have narrowed the tax base, like the introduction of a deduction of manufacturing income, or “bonus depreciation” during and in the aftermath of recessions; there is also a growing number of businesses in the United States, known as S-corporations, which are legally exempt from paying any corporate tax at all. But after factoring in all these changes, about two-thirds of the decline in the effective corporate tax rate since 1998 can be attributed to increased tax avoidance through low-tax jurisdictions.
The cost of tax avoidance by U.S. firms is borne by both the United States and other countries’ governments. Much of Google’s profit that is shifted to Bermuda is earned in Europe; absent tax havens, Google would pay more taxes in France and Germany. On the other hand, some U.S. corporations also use tax havens to avoid taxes on their U.S.-source income. With national accounts data, it is hard to know which government loses most. In both cases, U.S. shareholders win. Since equity ownership is very concentrated, even after including the equities owned by broad-based pension funds, so too are the benefits.
Accounting manipulations do not just cost governments a lot. They also cause basic macroeconomic statistics to lose significance, with adverse consequences for financial regulation and stability. The national accounts of Ireland, for example, are seriously contaminated by the trickery of multinationals. First, in the balance of payments: to shift their profits to the island, where they are taxed at only 12.5%, companies have their Irish branches import at low prices and export at artificially elevated prices—which results in an amazing trade surplus for Ireland of 25% of GDP! This surplus has nothing to do with any sort of competitive advantage; it doesn’t benefit the Irish population at all: it is entirely paid back to the foreign owners of the firms that operate in Ireland, so that Irish national income is only 80% of Irish GDP. Manipulations of transfer prices then massively distort the share of each factor of production (capital and labor) in corporate value added: the artificially elevated profits booked by foreign-owned firms make the capital share rise to more than 50% in sectors where immaterial capital is large, as in the pharmaceutical industry.
A Twenty-First-Century Tax on Companies
What is to be done? The current approach of the OECD and G20 countries consists of trying to reform the existing system by strengthening transfer-pricing regulations. The first efforts began in the second half of the 1990s, and yet the trend toward more widespread use of tax havens by U.S. multinational companies has shown no particular sign of slowing down since then. The current approach, therefore, does not seem very promising. When it comes to manipulating transfer prices, companies will always be far ahead of the controllers, because their means are greater: the tax department of General Electric alone employs close to a thousand individuals. More resources granted to tax authorities might help curb tax avoidance. But in the United States, IRS funding is actually on a downward trend, and, besides, there is a real risk that increased spending by tax authorities would trigger even bigger corporate expenses, leading to no extra revenue and a true loss for the collectivity.
We need a radical reform of corporate taxation. A promising solution consists in starting from the global, consolidated profits of firms, which cannot be manipulated.