number of employees might affect the impact of immigration which the em-
ployer anticipates. Small employers, in particular those who employed only
themselves, might not gain from intensified competition in the labor mar-
ket. Since large employers may gain from it, small employers may even
lose because of intensified competition in the output market. In addition,
migrants might directly increase the number of producers in some sectors:
OECD (2006: 56-58) shows that self-employment among immigrants in those
countries has increased since 1999.28 Those migrants’ small businesses are
typically restaurants, cleaning services, groceries and the like (Jandl et al.
2003: 37-40).29
Since ESS allows us to distinguish between employers by the number of
employees, we re-estimated all the specifications presented above by replac-
ing employ with two variables. In one case, we used a dummy variable,
employself which is equal to 1 if a respondent employed only him-/herself
and 0 otherwise, together with a variable that records the number of em-
ployees excluding him-/herself. In the other case, we used employself and
employother which is equal to 1 if the number of employees excluding him-
/herself is at least one and 0 otherwise. These variables are also summarized
in Table 3.
Using these variables instead of employ, we found mixed results across
countries as well as across sectors at the Union level. We did not find
28See Clark and Drinkwater (1998; 2000) and Blanchflower and Shadforth (2007) for
the trend in the UK.
29Menz (2002) describes how companies in high-wage countries took advantage of cheap
posted labor of subcontractors in low-wage countries within the Union. This also implies
that small self-employed subcontractors would have been competing with foreign labor,
while large employers benefited from access to foreign labor.
31