Thursday, April 28, 2011
In recent months, a number of press accounts have mentioned the amount of taxes paid by U.S. multinational firms on their foreign income. Some of the more sensational articles would have readers believe that U.S. companies pay little or no tax on their foreign earnings. Unfortunately, such stories are either based upon a misunderstanding of how U.S. international tax rules work, or are simply careless when portraying the way U.S. companies pay taxes on foreign profits.
The U.S. has a complicated "worldwide" system of taxation that requires American businesses to pay the 35% federal corporate tax rate on their income no matter where it is earned—domestically or abroad. When it comes to foreign profits, companies do pay income taxes—not once, but twice.
First, companies pay income taxes to the country in which their profits were earned, and then they pay additional U.S. taxes on any profits they return to this country. For example, if a subsidiary of a U.S. firm earns $100 in profits in England, it pays the British income tax rate of 26%, or $26. Since our system gives companies a credit for the taxes they pay to other countries, the additional U.S. tax the firm is required to pay is equal to the difference between the U.S. rate of 35% and the 26% British rate -- or $9. Between the two nations, the U.S. firm will pay a total of 35% in taxes on those foreign profits.
However, American firms can delay paying the additional U.S. tax on their foreign profits as long as the earnings are reinvested in the ongoing activities of their foreign subsidiaries. The additional U.S. tax is due when the profits are eventually repatriated.
In order to take advantage of the foreign tax credits allowed under our tax system, U.S. companies are required to report on their annual tax returns, on form 1118, the amount they earn in each country where they operate and how much they pay each country in taxes. According to the most recent IRS data for 2007, American companies paid nearly $100 billion in income taxes to foreign governments on foreign taxable income of $392 billion. As Table 1 indicates, U.S. companies paid an average effective tax rate of 25% on that income.
As the global economy has grown, so too have the foreign earnings and tax bills of U.S. companies. Figure 1 shows the foreign taxable earnings of and foreign taxes paid by U.S. companies between 1992 and 2007, every year for which data is available. Over those 16 years, taxable income grew in real terms by 225% while foreign taxes paid grew by 505%.
Much of this growth occurred after the last recession ended in 2002. Indeed, between 2002 and 2007, the taxable income of U.S. subsidiaries abroad more than doubled, after adjusting for inflation, while taxes paid increased by 134%.
Table 1 summarizes the amount of taxable income reported by U.S. firms for each global region along with the total amount of foreign taxes paid and the average effective tax rate.
Taxable Income (Less Loss) before Adjustments
Foreign Taxes Paid, Accrued, and Deemed Paid
Average Effective Tax Rate
All geographic areas
Other W. Hemisphere
S & SE Asia
Edward Kleinbard (USC) critiques the Tax Foundation report:
The paper is designed to create the impression that US firms in fact bear a substantial tax burden in respect of their foreign earnings. Its conclusion is that US firms on average paid a 25% effective foreign tax rate on their foreign income, which demonstrates that US firms are not "avoiding" taxes on their foreign earnings. In keeping with standard DC interest group expository style on all sides of the political spectrum, the paper urges others to dial back their anti-US multinational rhetoric, and avers that this presentation summarizes just the facts, ma'am.
If only life were that simple.
As the report states, its data are drawn from the IRS Statistics of Information Divisions's aggregation of all the Form 1118s that companies file. Form 1118 is the form on which a firm reports its foreign tax credits and foreign taxable income..
If the issue is, are US multinational firms bearing an appropriate level of foreign taxes, relative to their incomes — which I take the Tax Foundation's paper to agree is the relevant question — then here are several reasons why this aggregation of the Form 1118 is the wrong place to look. In particular, the Form 1118 data are not at all helpful for the debate currently under way, which is how should the United States tax the foreign unrepatriated income of US multinationals.
First and foremost, the Form 1118 looks only to repatriated income. It tells us nothing about the effective foreign tax rate imposed on the unrepatriated incomes of foreign subsidiaries, and, as described below, it reflects the application of tax planning, including the deliberate creation of pools of high-tax income that are repatriated to shelter other income, through cross crediting. But the actual debate is whether income that under current law is unrepatriated should now be allowed to be brought home without further US tax? The Form 1118 simply asks and answers the wrong question.
Second, the Form 1118 looks to taxable income, measured under US norms, that is repatriated to the United States. Other countries with more comprehensive tax bases thus appear to impose higher effective tax rates. The right approach is to use a broader measure of income, like earnings and profits, to see what the economic effective tax rate on that income actually is.
Third, the Form 1118 of course includes income from foreign branch activity. Indeed, roughly 20 percent of the gross income included in the SOI chart comes from branch activity, as do more than 25% of the foreign tax credits paid or accrued (I.e., ignoring deemed paid credits, which are irrelevant to branches). (The data don't show the taxable income of this branch activity on a standalone basis.) The same point applies to the even larger amounts of interest and royalty income received in the United States. (That income of course typically bears a very low rate of foreign tax, and is the principal raw feedstock for the tax director to use in his role as the master distiller, blending high tax and low tax foreign income).
Fourth, aggregated Form 1118 data don't translate into effective total tax burdens, because the aggregate data don't show the effect of the "tax distillery" in operation — the artful blending of zero taxed foreign income (dividends and royalties) with deliberately high-tax pools of foreign earnings that are repatriated to shelter the zero taxed income. In other words, the Form 1118 data by themselves, which look only to what has been repatriated, show some of the consequences of tax planning, including in particular the deliberate creation of high-tax pools of foreign income that a tax director dips into as needed to shelter other income. What the Form 1118 doesn't show is the ever-decreasing effective tax rate on unrepatriated income of foreign subsidiaries, whose modest aggregate tax burdens are in turn stripped out and concentrated into high-tax pools to use in the tax distillery's operations.
Fifth, the Form 1118 data include significant amounts of withholding taxes (about 15% of the foreign taxes paid or accrued). But if the question is, what tax rate is imposed on the foreign unrepatriated earnings of a US subsidiary today, those withholding taxes are irrelevant.
The bottom line is that the Form 1118 data really say nothing about the arguments now waging in DC.
But as it happens, another SOI chart does. That's the chart that aggregates data from all the Form 5471's that are filed. (The Form 5471 is the form on which firms record the results of the operation of their controlled foreign corporations.) .
That form is easy to read. Column 9 tells you the earnings and profits of all CFCs with positive E&P for the year. Column 10 tells you what they paid in foreign taxes. Divide the second into the first, and you discover that the effective foreign tax rate paid by foreign subsidiaries of US MNEs in 2006 is in the neighborhood of 16.4%, not the 25% figure reported by the Tax Foundation.