VoxEU Column Taxation

Tax competition tames big government

Opponents of international tax harmonisation argue that tax competition can rein in the tax-raising powers of big-government ‘Leviathans’ and thereby act as a force for good. An analysis of taxation across Swiss municipalities lends support to that argument.

Is tax competition good or bad for the well-being of society? As lucrative tax bases are becoming ever more mobile across national borders, this question is fast-rising towards the top of the international policy agenda. The EU has been grappling for years with the harmonisation of value-added taxes and corporate taxes among member states, and the OECD continues to labour over a consensual definition of what constitutes ‘harmful tax competition’.

The issue is now even coloured by a touch of rock ’n’ roll: U2 relocated their business headquarters from Dublin to Amsterdam in order to benefit from favourable tax treatment of royalty income; and the recent moves by James Blunt to Verbier and by French rock icon Johnny Hallyday to Gstaad were probably motivated more by special tax deals in Swiss cantons than by stunning mountain views.

The main opposing arguments in the debate on tax competition are straightforward.

Advocates of tax harmonisation think of politicians as essentially benevolent seekers of the common good, whose ability to offer the desired level of public services is undermined by the disappearance of their tax base and who are forced to shift the tax burden from mobile (rich) to immobile (poor) tax bases. Such arguments were invoked by Nicholas Sarkozy, who as French finance minister in 2004 suggested that limits be imposed on the variations in corporate tax rates across the EU, and that regional development funds from the common budget be paid only to countries whose corporate tax rate exceeded a certain threshold.

Conversely, those who view tax competition as a force for good consider politicians as self-interested ‘Leviathans’, whose appetite for big government may be held in check by tax competition. This line of argument is employed almost ritually by governments of low-tax countries such as Ireland, Switzerland and (with particular piquancy) the formerly communist states of Central Europe. ‘Leviathan-taming’ also frequently features as a rationale for delegating fiscal powers from national governments to regional authorities – the idea being that sub-national regions will have to compete more fiercely over mobile tax bases than entire countries.

It has proven difficult to generate empirical evidence on the implications of tax competition.1 Several studies show that intensified tax competition exerts downward pressure on tax rates, as most theoretical models suggest.2 However, it is impossible to infer from those results whether tax competition is good (by constraining greedy governments) or bad (by depriving governments of needed revenues and by shifting the tax burden towards less mobile tax payers).

New research

We revisit this question, aiming to address the difficulty of distinguishing good from bad tax competition empirically, in a way that is tied rigorously to the theory.3

In a first (theoretical) step, we derive the following prediction, based on a model featuring an overarching federal government and fiscally independent regional governments: if, among regions with relatively benevolent governments, smaller regions have higher tax rates, and if, other things are equal, this same relationship is reversed for regions that have relatively less benevolent governments (so that smaller regions have lower taxes), then the latter effect can be interpreted as evidence of welfare-increasing ‘Leviathan taming’.

In a second (empirical) step, we take the theoretical prediction to data on taxation in Swiss municipalities. Switzerland offers a propitious laboratory for research on tax competition, thanks to the exceptional institutional diversity and independence of Swiss cantons and municipalities.4 Switzerland represents a sort of ‘miniature EU’, featuring 26 fiscally sovereign cantons and hundreds of municipalities, all engaged in vigorous competition for lucrative taxpayers. The small size and proximity of Swiss cantons and municipalities implies that they are relatively similar in most other respects – unlike countries, whose myriad institutional, geographical and cultural differences make it difficult to derive clean-cut empirical results from international comparisons.

A further useful feature of the Swiss system for our purpose is that a sizeable sub-sample of municipalities set tax rates by direct democratic participation of the citizenry via annual town-hall meetings. Executives of these municipalities are arguably more strongly forced to behave ‘benevolently’ than municipal governments without such direct-democratic tax setting. Thus, we can take the degree of direct democracy in municipalities’ fiscal decision-making systems as a proxy measure for the benevolence of government (i.e. as an inverse measure of the ability of local governments to act as Leviathans).

We observe that, other things being equal, less direct-democratic municipalities indeed have higher tax rates. Our proxy measure thus seems to pick up what we want it to.

Moreover, we find that, among the ‘benevolent’ municipalities, relatively smaller ones set higher equilibrium tax rates. This relationship is reversed in jurisdictions with greater scope for Leviathan government. Hence, the estimation results coincide with the theoretical prediction, and our finding can be interpreted as evidence that tax competition lowers equilibrium taxes because governments are Leviathans. The underlying theory identifies this as evidence of beneficial tax competition.

Our study overcomes some of the interpretational ambiguity of prior empirical work: to the extent that the theory captures the main relevant forces, the finding on Leviathan municipalities can be considered as a manifestation of desirable Leviathan-taming tax competition. Since regional and central governments of other nations may exhibit considerably greater scope for revenue-maximising government behaviour, this result should apply well beyond the Swiss case.

It is important to note that ours is a typical “second-best” argument: faced with two distortions, we find that one of those distortions (cross-border tax externalities) may offset the other distortion (voracious governments) in such a way that the economy ends up closer to the social optimum. Clearly, it would be better to curtail the distortions directly rather than to play them off against each other. Our study implies that tax harmonisation (reducing cross-border externalities) would become more desirable if government actions could be tied more closely to citizen preferences, for example through increased recourse to fiscal referenda. However, direct democracy may not be as workable in larger and more heterogeneous countries than Switzerland. Our results thus imply a case for fiscal decentralisation in those countries.


While our research offers some suggestive evidence on a highly contentious political issue, we need to caution against excessive generalisation. Our analysis cannot cover all the economically relevant dimensions of tax competition. For instance, a diverse fiscal landscape offers people and firms an opportunity to ‘shop around’ for their preferred combination of tax burdens and publicly-provided goods. Similarly, institutional diversity allows for experimentation with different fiscal systems. These are further arguments in favour of decentralised tax setting. Conversely, the fact that small jurisdictions and mobile tax payers are generally better-placed to benefit from tax competition than large jurisdictions and immobile tax payers implies a kind of political inequity that our analysis abstracts from, and that would make tax competition rather less desirable. One can also easily conceive scenarios whereby majority voting on tax rates may not maximise utilitarian social welfare.

The debate on tax competition will continue to rage. Our research suggests that tax competition can be a force for good by constraining excessively big government. How to weight this argument against the potential inequities related to tax competition remains the key political question.


Lars P. Feld and Gebhard Kirchgässner (2001) “The political economy of direct legislation: direct democracy and local decision–making”, Economic Policy, Volume 16, Issue 33, 153-201.
Michael P. Devereux, Rachel Griffith and Alexander Klemm (2002) “Corporate income tax reforms and international tax competition”, Economic Policy, Volume 17, Issue 35, 449-495.
John D. Wilson and David E. Wildasin (2004) “Capital tax competition: bane or boon?” Journal of Public Economics, Volume 88, Issue 6, 1065-1091.




1 For a recent survey, see Wilson and Wildasin (2004).
2 See, e.g., Devereux, Griffith and Klemm (2002).
3Does Tax Competition Tame the Leviathan?”, available as CEPR Discussion Paper No. 6512
4 See Feld and Kirchgässner (2001).


840 Reads