significant as that of historic and economic factors. Moreover, we find that
the measures analysed appear to be quite short-term in their effect. When
we lag the deterrence index by more than one year, it ceases to have any
significance. Finally, if one disaggregates the measures included in the
deterrence index (see Table 4), one finds that the index’s significance is
due to the strong effect of only two of the five deterrence measures
analysed here: (1) not allowing asylum seekers to work until their
application has been successful or until they have been allowed to stay in
the host-country more permanently on the basis of a subsidiary protection
status (2) granting protection status to a smaller percentage of asylum
seekers (in relation to the total number of applications) than other host-
states. Each of these two measures on its own is significant at the 0.05
level. Their combined significance in the deterrence index is even stronger.
Table 4: Impact of Individual Deterrence Measures
Deterrence Measure______________ |
Expected Sign |
Coef______ |
Z______ |
Prohibition to Work |
- |
-.4578465 |
-2.31** |
Below Average Recognition Rate |
- |
-.2714122 |
-2.40** |
Safe Third Country Provisions |
- |
-.1900568 |
-1.13 |
No Freedom of Movement |
- |
-.0748292 |
-0.30 |
Non-Cash Benefit Payments______ |
- |
-.1256857 |
-1.40 |
N= 227; R-squared = 0.41; **p<0.05
No significant effect in reducing asylum applications, however, could be
found for the other three measures which have dominated the public
policy debate on asylum in recent years. These are: (1) measures that
allow states to turn asylum seekers back at their borders and return them
to so-called ‘safe third countries’; (2) measures that deny asylum seekers
freedom of movement within a host country, i.e. the introduction of
dispersal schemes and (3) measures which have meant the end to cash
benefit payments to asylum seekers, e.g. through the introduction of a
voucher scheme. The following will provide some initial suggestions as to
28