This article appears in the following Oxford Review of Economic Policy issue: BUSINESS TAXATION IN A GLOBALIZED WORLD [View the issue table of contents]
Corporate taxation in the OECD in a wider context
* Centre for Business Taxation, Saïd Business School, Oxford University, e-mail: simon.loretz{at}sbs.ox.ac.uk
 |
Abstract |
|---|
Against the background of increased globalization statutory corporate tax rates have shown a clear downward trend over the last two decades. The sharp decline in these rates was accompanied by substantial tax base broadening and a comparable reduction in personal income tax rates only until the early 1990s. This suggests that corporate tax competition is of increasing importance. So far corporate tax revenues remain fairly stable. But an analysis of corporate taxation in the context of the overall tax systems shows that a substantial shift towards value-added taxes has taken place. While the trends so far have been driven by smaller European countries, recent tax reforms indicate that increasing tax competition is inducing a shift towards consumption taxes even for larger economies.
Key Words: corporate taxation • OECD • tax competition
I would like to express my gratitude to Chris Allsopp and the participants of the conference ‘Business Taxation in a Globalized World’ at the Oxford University Centre for Business Taxation, June 2008, for very valuable discussions. I am further indebted to an anonymous referee for helpful comments and to Johannes Voget and Michael Devereux for numerous fruitful debates.
1 See for example the article in the International Herald Tribune on the German tax reform, available at http://www.iht.com/articles/2006/11/02/business/tax.php
2 There is an increasing literature investigating the trends in corporate taxation, ranging from an early contribution of Chennells and Griffith (1997) to a recent analysis by Devereux (2007). See also Nicodème (2007) for an excellent overview of corporate taxation in the European Union.
3 The dataset used for this paper is an updated and extended version of the Devereux et al. (2002) dataset. See Appendix for data sources and definitions.
4 The only exception in our dataset is Ireland, as we use the preferential tax rate of 10 per cent applicable for the manufacturing sector for the years from 1982 to 2003. Subsequently the corporate tax rate in general was lowered to 12.5 per cent and the preferential treatment was abolished. As the introduction of the preferential tax rate falls just outside our sample period, we also report results with the change of the standard corporate tax rate used where appropriate.
5 This implies excluding Poland, Hungary, the Czech Republic, the Slovak Republic, and Turkey, as the tax rates for these countries are only available from 1991 onwards.
6 See Wilson (1999) for a comprehensive survey of the early theoretical tax competition literature.
7 The countries excluded are the Czech Republic, Hungary, Iceland, Poland, the Slovak Republic, Switzerland, and Turkey.
8 For the following countries data are only available from the year in brackets onwards: Czech Republic (1993), Hungary (1991), Poland (1991), Portugal (1989), the Slovak Republic (1998), and Turkey (1991). Further, Mexico is not part of the sample, as it reports no separate statistic for corporate income tax revenues.
9 Note that we use a different notation from Auerbach and Poterba (1987), as we discuss the legal definition of the corporate tax base under the heading ‘corporate tax base’, while they use the term tax base solely for corporate profits.
10 The analysis of Auerbach and Poterba (1987) has led to some follow-up studies focusing on the determinants of the US corporate tax revenues. For a summary of this literature and a recent analysis of US tax receipts see Auerbach (2007).
11 This approach dates back to at least Jorgensen (1963) and Hall and Jorgensen (1967).
12 See the Appendix for the definition.
13 The following countries enter the sample only after the year in brackets: Denmark (1985); the Czech Republic, Luxembourg, New Zealand, Hungary, the Slovak Republic, and Turkey (1990); Iceland and Poland (1991); Mexico (1994); Korea (1995).
14 See, for example, Bucovetsky and Hauer (2008).
15 The data set we use here is an extended version of the Devereux et al. (2002) data. See the Appendix for more information.
16 See De Mooij and Nicodème (2007a,b) for studies on the number of firms incorporating themselves because of tax reasons. Further, Fuest and Weichenrieder (2002) provide an indepth analysis of the incorporated versus unincorporated sector.
17 See also Slemrod (2004) for a discussion of the determinants of corporate tax rates and revenues.
18 For a discussion of the international tax system, see, for example, Huizinga and Voget (2006).
19 The Council Directive 90/435/EEC of 23 July 1990 rules that dividends paid from a subsidiary to its parent company should be exempted from withholding taxes, and the Council Directive 2003/49/EC of 3 June 2003 governs the withholding taxes on Interest and Royalties flows.
20 In absolute numbers this signifies an increase—out of a total of 870 country pairs—from 441 to 803 country pairs.
21 This is more evident for qualified ownership, as most countries only introduce a system of participation exemption. For anecdotal evidence see Huizinga and Voget (2006) or Loretz (2007).
22 See Boulakia (1971) for a discussion of Khaldûn's contribution to the economic literature.
23 The Laffer curve is named after the economist Arthur Laffer, who allegedly sketched the
parabolic relationship between tax rates and revenues on a napkin to explain it to a journalist.
24 Clausing (2007) finds empirical evidence for a relatively high revenue-maximizing corporate tax rate, while Brill and Hassett (2007) argue that the Laffer curve moved over time, and the revenue-maximizing corporate tax rate has fallen over time. Along these lines, Devereux (2007) finds only weak evidence for the existence of a Laffer curve.
25 This includes scenarios where it would be economically feasible to raise other taxes, but it could be politically impossible to do so.
26 The statutory rates use tax information from 2006, whereas the revenues figures are from 2005, the latest year available. For the United States we use the average of general sales taxes, as they come closest to a value-added tax.
27 Excluding Norway, the correlation is –0.37, significant at the 10 per cent level.
28 We use the top marginal tax rate including local taxes and applicable surcharges.
29 See also De Mooij and Nicodème (2008) for an analysis of this phenomenon, i.e. the reduction in personal tax revenues due to reduced corporate tax rates.
30 See Auerbach and Poterba (1987) and Auerbach (2007) for further explanations of changes in the corporate tax base.
31 See Loretz (2007) for direct evidence. Further, the tax competition literature finds that distance-weighted neighbour tax rates outperform other measures of the competitors’ tax rates. See Devereux and Loretz (2008) for a survey.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  R. A. d. Mooij and S. Ederveen Corporate tax elasticities: a reader's guide to empirical findings Oxf. Rev. Econ. Policy, December 1, 2008; 24(4): 680 - 697. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. P. Devereux Business taxation in a globalized world Oxf. Rev. Econ. Policy, December 1, 2008; 24(4): 625 - 638. [Abstract] [Full Text] [PDF]
|
|