the private labour market ensures that those outsiders who do not get a public
sector job are all able to find private sector employment.
Our categorization of public sector workers into insiders and outsiders is mo-
tivated by the observation that some groups of civil servants are often employed on
long-term contracts providing a high degree of job security whereas other public
sector workers are appointed on short-term contracts offering less job protection.
Our distinction between public sector insiders who have full job security and the
marginal public sector workers who can easily be dismissed captures this observed
difference in the terms of employment in a stylised way. Note that the stronger at-
tachment of insiders to the public sector could explain why this group has formed
a lobby to protect their interests whereas the marginal workers with a looser link
to the public sector do not enter the lobby.
Visser (2006) documents that public sector workers in the OECD area are in
fact better organised than workers in the private sector, as reflected in a much
higher degree of unionisation in the public sector. Our assumption that only pub-
lic sector insiders have formed a lobby seeks to capture this marked difference
across sectors in a simple way. Whether the higher union density implies that
public sector workers actually earn rents is ultimately an empirical issue. As we
show in Appendix 1, if private sector workers are a very tightly knit group in
terms of ideological preferences, they could be politically more influential than
public sector workers even if they have not formed a lobby. However, the prob-
abilistic voting model considered below does imply that public sector workers
generally earn rents under autarky. Our assumption that (organised) public sec-
tor voters constitute a strong interest group capable of extracting rents is made
because it seems to be implicit in the reasoning of many of those who advocate
tax competition as a remedy against rent seeking.7 Our purpose is to investigate
whether fiscal competition could indeed be an appropriate means of curbing an
excessive political influence of public sector workers. While we do not wish to
pass a verdict on whether such an excess influence actually exists, we note that
7 If private sector workers were politically more influential than those employed in the public
sector, the model set up below implies that an unconstrained political candidate would ideally
want to keep the public sector wage rate below that in the private sector in order to keep taxes
low. However, this scenario would not illustrate the interdependence between public sector
rents and tax competition which is the focus of the present paper. Moreover, with flexible wage
adjustment preventing involuntary unemployment in the private sector, the public sector would
face a recruitment problem if it offered a lower wage rate than the private sector.
11