Discussion Papers, Policy Papers, Books & Reports, Bulletin, Newsletter, Economic Policy Lunchtime Meetings, Workshops & Conferences, Events Diary, Previous Events Programme Areas, Current Research Projects, Networks, Vacancies Programme Directors, Researchers Lists, Noticeboard Press Releases, Coverage, Request a Press Release Data?, Resources for Economists, Data on Other sites Membership information Login, Create a Profile, Profile Benefits, Your Profile Settings, Forgot Your Password? Site Map, How to find us, How to Order Publications, Privacy Policy, Feedback How to find us, Frequently Asked Questions, ESRC Site Guide, Frequently Asked Questions, Vacancies, How to Search Site Map, How to find us, How to Order Publications, Privacy Policy, Feedback CEPR Home Page You have items in your shopping cart.  Click to view your cart

European Economic Perspectives 28

Creative Accounting

Changes in relative corporate tax levels across countries may conceivably affect the location of production. But as new CEPR research reveals, they almost certainly have an impact on where multinational firms declare their profits.

As the previous article indicates, policy-makers in industrialized countries have to balance their need for tax revenues and the attractiveness of their jurisdiction for economic activity – and they react to each other’s corporate tax policies in finding the appropriate balance. But international differences in tax rates can affect revenues not only through shifts in economic activity: they can also alter revenues more directly through shifts in firms’ cross-country allocation of accounting profits. A recent CEPR Discussion Paper by Eric Bartelsman and Roel Beetsma examines the effects of tax differences on profit shifts that occur purely on paper and finds evidence that they have a significant impact on where profits are declared.

All things being equal, multinational firms try to allocate costs to countries with high corporate tax rates and revenues to countries with low tax rates. The potential gains from such shifts are largest if the firm is resident in countries like France and the Netherlands, which use the exemption system – profits taxed elsewhere are tax-exempt in the country of residence. Other countries, including the United Kingdom and the United States, use the credit system, which gives a tax credit for corporate taxes already paid elsewhere. Here, the gains from profit shifting arise primarily from deferring the repatriation of profits to a residence country with higher taxes.

There are essentially two ways in which pure profit shifting can take place. One is through the capital structure: a firm might finance new subsidiaries in high-tax countries by establishing debt contracts with branches in low-tax countries. The other is through incorrectly pricing intra-firm deliveries of goods and services. Although the OECD’s guidelines on tax and transfer pricing call for the use of the ‘arm’s length principle’ whereby internal transfers are made at market prices, there are various problems in applying this principle. With many intra-firm transactions, for example, there is no comparable outside market, most notably with intellectual property developed by one branch of the company and used by other branches in other countries.

The extensive misuse of transfer pricing between industrialized countries and tax havens receives considerable attention in a recent OECD report, Towards Global Tax Competition: Progress in Identifying and Eliminating Harmful Tax Practices. Less attention has been paid to transfer pricing within the industrialized world, where the focus has been more on the effects of tax competition on real activity. Yet Bartelsman and Beetsma’s results suggest that the effects of changes in corporate tax rates on transfer pricing between industrialized countries are substantial.

The researchers estimate the extent of this kind of profit shifting by examining the relationship between the ratio of nominal value added to labour compensation and differences in corporate tax rates between individual countries and the OECD average. The idea is that pure profit shifting resulting from high taxes will lead to a reduction in reported value added, conditional on the scale of a firm’s operations. (Taking account of labour compensation controls for the scale of operations and thus filters out the effect of tax rate changes on shifts in real activity.) Hence, the ratio of nominal value added to labour compensation should decline if firms misuse transfer prices to counteract a rise in tax rates.

To assess whether this does indeed happen, the researchers combine data on corporate tax rates with the OECD’s STAN database, which contains information on a large number of industrial sectors for most OECD countries over the period 1970-97. The results confirm that there are significant effects of tax rate differences on the value added/labour compensation ratio.

So what exactly is the effect of a tax increase on corporate tax revenues? Bartelsman and Beetsma’s baseline estimate suggests that, on average, a unilateral one percentage point increase in the corporate tax rate does not lead to an increase in corporate tax revenues, owing to a more than offsetting decline in reported profits. The countries for which the estimated effects are largest are Japan, Portugal and Spain, followed by Denmark, Germany and the Netherlands. The sectors with the strongest effects are ‘Industrial Chemicals’, ‘Other Chemicals’, ‘Iron and Steel’ and ‘Non-Ferrous Metals’. ‘Other Chemicals’ includes pharmaceutical products, for which research and development and hence intellectual property are relatively important.

Since the analysis focuses on only one type of profit shifting – transfer pricing – it seems likely that the results constitute a lower bound on the effect of tax rate changes on reported profits. This suggests that the rewards of stricter enforcement of the rules on transfer pricing might be quite high. Yet the scope for transfer pricing, and profit shifting in general, within a multinational is tightly linked to the scale at which it operates. Tighter enforcement of the rules in a high-tax country means that the net return on investment would fall, hence carrying the risk that real activity is shifted to other countries with lower taxes or more lenient enforcement. In these circumstances, enforcement of transfer pricing rules would probably benefit from international policy coordination, such as exchange of information about multinationals’ activities and agreements about minimum enforcement standards or common transfer prices.

The analysis also has lessons for productivity measurement and comparisons of growth rates across countries, suggesting, for example, that observed growth spurts in countries that have lowered their tax rates – such as Ireland – may be related, at least partly, to mismeasurement. It is possible that the reduction in the corporate tax rate has boosted measured value added by more than the increase in real activity.

This article discusses research reported in ‘Why Pay More? Corporate Tax Avoidance through Transfer Pricing in OECD Countries’, CEPR Discussion Paper  No. 2543 (August 2000) by Eric Bartelsman and Roel Beetsma. Bartelsman is at the Vrije Universiteit Amsterdam; Beetsma is at the Universiteit van Amsterdam and a Research Affiliate in CEPR’s  International Macroeconomics programme.

Download PDF

 
Return to contents

 
Browse Archives

Your current location: Publications > Newsletter > EEP28
Top CEPR, 77 Bastwick St, London EC1V 3PZ
United Kingdom.
Tel: +44 (0)20 7183 8801     Fax: +44 (0)20 7183 8820
Email: cepr@cepr.org     Webmaster: webmaster@cepr.org
Home
With the support of the European Union: Support for bodies active at European level in the field of active European citizenship