Despite being bound by the same international conventions, different countries appear to have very different practices for assessing asylum applications. This empirical observation is the starting point for my thesis, which uses a statistical model to investigate the causes for discrepancies in recognition rates for refugees between the Scandinavian countries.Norway, Denmark and Sweden are all bound by the Refugee Convention and have exactly the same national legislation for granting asylum seekers refugee status under this convention. However, the probability of gaining protection as a refugee for people who flee from the same country still vary greatly. This gap between the expected similarity of recognition rates and the reality of very different rates is puzzling. Few previous studies have attempted to explain it, possibly due to a lack of available data. To overcome this problem, I construct a new dataset of recognition rates for refugees in the three Scandinavian countries, covering all origin countries for the years 1995-2011. Using this dataset, I estimate a statistical model with the origin-specific recognition rate as the dependent variable.I examine three possible causes for the discrepancy. First, by way of partisan theory, I hypothesise that recognition rates will vary according to the preferences of the voters of the governing parties and these parties’ ideology. Second, I test for the option that antiimmigration parties in the three countries cause the other parties to resort to lower recognition rates in the face of voter flight. Third, I examine the possibility that the three Scandinavian countries emphasise conditions in the country of origin of the asylum seekers differently, which to my knowledge has not been investigated in previous studies. To this end, I construct a new index that measures the level of persecution and conflict in origin countries.My findings point to an effect of politics on recognition rates. I find that recognition rates are lower under centre-right governments than under left-leaning governments in Scandinavia. At the same time, I find no effect of anti-immigration parties on recognition rates. In previous studies, Denmark has been singled out as the strictest Scandinavian country in terms of asylum policy, while Sweden has been considered the most liberal. I find evidence that it is relatively much more important to be from a country in turmoil when seeking asylum in Denmark, where more emphasis is put on conditions in the country of origin. In short, my findings indicate that there are differences between the Scandinavian countries in how asylum cases are decided, and that at least part of these differences can be explained by politics.