This book review appears in the Fall 2015 issue of The American Prospect magazine. Subscribe here.

The Hidden Wealth of Nations: The Scourge of Tax Havens

By Gabriel Zucman

200 pp. University of Chicago Press $20.00

Tax havens cost the world’s governments hundreds of billions of dollars a year, promote corruption, and undermine the rule of law. They are part of a larger worrisome pattern in which the world’s corporations outrun the governing capacity of states. A tax haven is a nation that refuses to cooperate with major countries in order to lure multinational corporations and investors to nominally book transactions in its locale. These transactions can be outright illegal, or borderline, but beyond the reach of legitimate tax authorities.

Gabriel Zucman, a young French economist now at the London School of Economics and the University of California at Berkeley, has written a masterful survey of the origins, importance, and dangers of tax havens. The Hidden Wealth of Nations is a tremendously important contribution to the current discussion of how to adjust the world’s income-tax systems, which are over a century old, to the realities of the 21st century.

Zucman makes three crucial contributions. First, he offers an absorbing account of how tax havens came into being in the period between the two world wars, and how they became much more important after exchange controls were relaxed in the 1980s. Second, Zucman delivers a far-reaching and rigorous quantitative evaluation of the scale and dynamics of tax havens in today’s global economy. And third, Zucman proposes remedies. 

The most original part of Zucman’s book is the second, in which he offers the most comprehensive evaluation to date of the scope of the tax-haven problem. Zucman came up with a brilliant new way of calculating the wealth hidden in tax havens by comparing reported financial assets and liabilities in the world’s financial centers. He concludes that around 8 percent of the world’s financial wealth is held in tax havens. But this overall number masks the true importance of the havens, because in developing and emerging countries the percentage can be much higher. Zucman estimates that in Africa, the share of financial wealth held offshore is around 30 percent, while in Russia and the Middle East it is above 50 percent.

Zucman calculates that these numbers translate into about $200 billion in lost income-tax revenue worldwide. This is a large number, but it is significantly below some other estimates. For example, Joseph Guttentag and I estimated in 2007 that the U.S. alone loses $50 billion a year to illegal evasion in tax havens, based on data from the Boston Consulting Group and Merrill Lynch. This number may be lower now because of the enactment of the Foreign Account Tax Compliance Act of 2010 (FATCA), a law that requires foreign banks and other financial institutions to report accounts held by Americans to the IRS or face a 30 percent tax on their U.S. income. FATCA has had a substantial deterrent effect. But as Zucman points out, FATCA has significant built-in loopholes; for example, the penalty can be avoided by using a bank that has no U.S. assets. In any case, his $200 billion is a useful lower estimate, and a big enough number to induce political action.

Zucman proposes several steps. The most important would be to create a worldwide register of financial wealth, recording who owns what in stocks and bonds. This register would act as a central depository coordinated by governments and international organizations, and it would allow national tax administrations to fight tax evasion by levying taxes on capital-income flows and wealth stocks.

Zucman points out that private registers of financial assets already exist, such as those maintained by the Depository Trust Company in the U.S. or by Clearstream in Luxembourg. His idea is to combine them and enlarge the field of data, and to transfer ownership of the data to the public. “Combined with an automatic exchange of information between the banks of all tax havens and foreign tax authorities, a financial register would deal a fatal blow to financial secrecy,” he writes.

But what if some tax havens refuse to cooperate? In that case, Zucman proposes to levy sanctions proportionate to the costs that tax havens impose on other countries. For example, he suggests that a tariff of 30 percent on goods imported by France, Germany, and Italy from Switzerland would suffice to induce the Swiss to cooperate.

In addition, Zucman suggests that corporate use of tax havens (which is legal, and therefore numerical data are more readily available) can be curbed by adopting an apportionment system for global consolidated profits, such as that used by American states and Canadian provinces, and proposed by the European Commission. Allocating profits based on actual commerce would reduce the benefit derived from artificial booking of income in tax havens. He estimates that such a tax would increase corporate tax revenue by about 20 percent on a global basis, and that it would primarily benefit the large European countries and the U.S.

I have some misgivings about Zucman’s solutions. Like Thomas Piketty’s wealth tax, the proposed register seems to require too much global cooperation among governments to be realistic. Moreover, it seems unlikely that a sanctions regime can be applied to all tax havens at once.

As for the corporate tax, while I am a big fan of formulary apportionment, there are serious practical problems to overcome with any proposal, which Zucman breezes by in the few pages he devotes to solving the corporate tax problem.

Fundamentally, I believe that the problem with Zucman’s solutions is that he is not radical enough.

Fundamentally, I believe that the problem with Zucman’s solutions is that he is not radical enough. His proposals are based on the traditional consensus, reached by the League of Nations in the 1920s, that investment income should be taxed by the countries in which individuals reside, while business income should be taxed by the countries in which it is earned. In my opinion, this consensus should be reconsidered in light of current realities.

The original rationale for the consensus (the “benefits principle”) was that economists in the 1920s believed that active (business) income was earned primarily in the country of source, because that is where business activity took place, while passive (investment) income was earned primarily in the country of residence, because that is where the invested capital was accumulated. Economists no longer believe in the validity of this analysis, because most types of income have more than one source.

Nevertheless, Zucman reflects the common view that the so-called benefits principle still makes sense, because most active business income is earned by corporations and most passive investment income is earned by individuals. Since most investors have a country of domicile, individual residence is meaningful as a basis for taxation in a way that corporate residence is not. Thus, it makes sense to tax individuals based on where they reside and corporations based on where their income is earned. However, this strategy has been overtaken by new tax-avoidance techniques.

This system did function reasonably well until the 1980s. For active income, it could generally not be earned without taxation by source countries, because corporate investments by multinationals were generally immobile and tax competition was limited in scope. The 1979 Dupont case is a good illustration of the limits of tax-planning in the pre-globalization era. It involved a U.S.-based multinational manufacturing goods in the U.S. and selling them in Europe. Both activities were subject to high levels of taxation. Dupont tried to avoid this result by routing the sales through a Swiss subsidiary, but the U.S. court rejected this attempt, and Subpart F of the Internal Revenue Code was enacted in 1962, in part to prevent it from happening again. The underlying assumption behind the Subpart F compromise, which taxes U.S.-based multinationals on passive investment income but not on active business income, was that active business income would be taxed at its true source. If that were the case, deferring U.S. taxation would not induce the multinational to shift its income offshore, because the U.S. tax savings would be offset by the foreign tax cost.

In the 1980s, however, multinationals became much more mobile as the focus shifted from heavy manufacturing to services and intangibles. The result was tax competition as multinationals acquired the ability to pit one country against another. By 1995, Intel, which builds a new chip factory about every other year, was able to boast that tax competition enabled it to avoid paying any foreign tax, having pitted Mexico against Costa Rica as new plant locations, and Israel against Ireland in 2014, earning tax holidays where they settled.

On the individual level, the relaxation of exchange controls and the abolition of withholding taxes on interest (beginning with the U.S. portfolio interest exemption in 1984) created a world in which an estimated $7 trillion in income (according to Zucman’s data) can be routed through tax havens and not declared to residence authorities.

The result has been that by the early 21st century, most cross-border income is untaxed. In the case of multinationals, this is because the source jurisdiction either lacks the authority to tax or because taxation at the source is premised on a physical presence in that country. But in the age of electronic commerce, it is easy to avoid the required physical presence in the source country. Alternatively, even if the physical presence exists, source jurisdictions frequently grant tax holidays to multinationals to induce them to invest and create jobs there. Residence jurisdictions, meanwhile, do not currently tax business income for fear that multinationals will move elsewhere, as the current U.S. inversion saga confirms. (Multiple U.S. corporations, like Burger King, nominally moved their residence, but not their headquarters, to lower-tax countries.) In the case of individuals, exchange of information has not proven effective in deterring tax evasion via offshore secrecy jurisdictions, and no withholding taxes are levied by source jurisdictions.

The problem with this state of affairs is that in the absence of taxation of cross-border flows, the progressive income tax cannot be maintained because it is easier for the wealthy to earn cross-border income. The result has been a worldwide shift to taxing consumption rather than income.

Since 2008, there has been an attempt to revive the income tax on cross-border income. This has culminated in two major projects, the Covention on Mutual Administrative Assistance in Tax Matters (MAATM), which was inspired by FATCA but is now signed by more than 80 countries, and the OECD Base Erosion and Profit Shifting (BEPS) project.

However, while both of these projects are helpful, neither is likely to solve the underlying tax-avoidance strategies. In both cases, the problem is that too many countries need to cooperate for the regime to be effective. In the case of MAATM, every tax haven has to sign on because otherwise, all the funds can be routed through the non-cooperating havens. We are far from there, and even the U.S. has not signed MAATM (it has implemented FATCA, but that applies only to U.S. residents and can be avoided by using financial institutions that have no U.S. presence).

In the case of BEPS, the project only addresses artificial profit-shifting, not tax competition, and it only applies to the OECD and G20, not to the many source countries outside these two organizations. Thus, it is likely that multinationals can avoid BEPS by sourcing income in countries that are not subject to it.

Fundamentally, if the income tax is to be preserved in the 21st century, multilateral solutions are needed. Both MAATM and BEPS are such solutions, but they are hampered by the focus on residence jurisdictions for passive investment income and source jurisdictions for active business income. There are too many residence jurisdictions in the first case, since every country has rich people, and too many source jurisdictions in the second, since multinationals operate globally.

Thus, in my opinion, it is time to re-evaluate the benefits principle first propounded in the 1920s. Most of the current issues can be solved if we tax passive investment income primarily at its source, and active income primarily at residence.

Because most individuals are relatively risk-averse, portfolio investment flows overwhelmingly to a small number of countries—the U.S., the European Union, and Japan. Even the so-called BRICS nations (Brazil, Russia, India, China, and South Africa) mostly attract portfolio investment through mutual funds that are relatively easy to tax. Thus, if the “big three” can coordinate to reinstate a withholding tax on interest, dividends, and royalties flowing from them, most of the problem of taxing investment income can be solved. Crucially, money cannot stay in tax havens and earn decent rates of return, so the cooperation of tax havens is not needed (unlike in the case of MAATM or Zucman’s proposal).

For active business income, more than 90 percent of multinationals are headquartered in the G20, and none of those countries have a tax rate below 20 percent. So if they currently taxed their multinationals on a coordinated basis and restricted the ability to move out, most of the problem would be solved. 

This approach also solves the oft-cited problem of double taxation. Once passive income is taxed at source, taxpayers may be able to credit the tax upon declaring it to their residence country. And once active income is taxed at residence, a credit can be given to source-country taxes, if the source country responds to the limitation of tax competition by re-imposing its tax.

Zucman’s book makes a crucial contribution to understanding the scope of the problem of tax havens. He is to be commended especially for his brilliant use of the data to estimate the problem. But his solutions depend on an unrealistic level of global cooperation. A more measured approach that requires cooperation by fewer countries seems to me to be better.

If nothing is done about tax havens, the rich will continue to avoid the progressive income tax.

If nothing is done about tax havens, the rich will continue to avoid the progressive income tax. And if that happens, ordinary middle-class Americans will be reluctant to pay their taxes. Our tax system is built around voluntary cooperation; if most Americans refused to cooperate, the IRS could not force them to do so. As the Greek experience has recently demonstrated, once a tax culture of non--payment is established, it is very hard to change. Zucman’s book is an important call to do something about the scourge of tax havens before it is too late.

Reuven S. Avi-Yonah is the Irwin I. Cohn Professor of Law and director of the International Tax LL.M. Program at the University of Michigan.