This article investigates the effects of inclusion and exclusion on the policy agendas of radical right parties. Radical right parties face diverging political opportunity structures in Western Europe. In some countries, the prospect of office has become a realistic option since the beginning of this century; in other countries radical right parties remain ostracised by mainstream parties. Research has focused mainly on the electoral effects of inclusion and exclusion strategies. Systematic analyses of the effects on policy agendas are scarce. This article focuses on the policy positions of radical right parties with respect to the key issues of immigration and integration. The finding is that ostracised parties have not changed much over time, but there is no evidence that cordons sanitaires have a freezing effect. Contrary to expectations, non-ostracised parties have not become more moderate over time. After the turn of the millennium, non-ostracised radical right parties have become just as radical as their ostracised cousins. 5 However, in order to test the robustness of our findings, we have also conducted our analysis with the CMP data. This leads to the same substantive results. See the results section. 6 The manifestos have been coded by two persons. After various try-outs, followed by intensive discussions and corrections of the index, two tests have been done of ten manifestos each. All immigration and integration pledges in these manifestos were coded by the two coders independently. With respect to the total scores, the tests resulted in, respectively, 92.3 and 88.6 per cent overlap -percentages which are generally accepted as providing sufficient reliability. With regard to the identification of immigration/ integration pledges, the tests resulted in, respectively, 85.5 and 86.9 per cent overlap. 7 We have conducted various robustness checks. First, we have estimated the same models with the absolute policy positions of radical right parties (i.e. not taking into account the position of the mainstream competitor). This leads to the same substantive results. Second, we have estimated the models while clustering the manifestos by countries. This did not change the results either. 8 It could be the case that small radical right parties score lower because they are less policy-oriented and consequently devote less attention to pledges. We have checked whether this is the case by dropping six manifestos that contained very few pledges. This did not change the results substantively.