Robert Braun, a Dutch Ph.D student at Cornell University, looks at this issue from a different angle. Rather than comparing survival rates across borders, he examines the records of over 125,000 individual Jews in the Netherlands. Building on the efforts of others (most notably Marnix Croes, Peter Tammes and Herman van Rens) he has geo-coded records on where Jews lived before the deportations started, whether they were deported, whether they died of natural causes, and whether they succeeded in emigrating or appealing their Jewish identity legally.
Braun posits that in order to survive the Holocaust you needed help from a network of individuals who could supply fake identification papers, rationing cards, food, shelter and transportation. This network could only operate if the people within it fully trusted each other. One weak link and the entire network might be exposed. Such trust, he argues, is more likely to develop among a religious minority.
To better grasp this point, the image below plots the distribution of religious affiliations in the Netherlands before World War II. The south is largely Catholic and the north largely Protestant (Calvinist), with some communities dominated by orthodox Protestants. These different religious communities did not just have their own churches but also their own political parties, schools, soccer clubs, newspapers and so on. Thus, religious affiliation shaped your social and political life, a concept usually referred to as pillarization.
Braun argues that it is easier for individuals to free-ride in majority communities. In minority communities, everyone is expected to participate and there are mechanisms for reinforcing this participation because group commitment is higher. This may be especially true in settings such as the Netherlands in this era where religious affiliation also shaped your social life. This led to the development of networks of committed individuals who trusted each other and this trust could be employed in rescue networks. He gives examples of instances where even if some within a minority community did not agree with a rescue operation, they would remain silent and even help out when called upon.
He then pairs the geo-coded data with information on church locations and shows quantitatively that proximity to Catholic churches increased evasion in predominantly Protestant areas by more than 20 percent. A similar effect holds for proximity to Protestant churches in Catholic parts of the country. Go read the full paper here.
This is surely not the only factor that matters (and Braun never says that it is) but there are interesting broader implications. As Braun puts it:
[..] whether altruists can turn into heroes depends on the networks in which they are embedded.
The findings imply that it is not something inherent to either Protestantism or Catholicism that makes people more or less likely to assist their threatened Jewish neighbors. Instead, what matters is whether their environment induced either Protestant or Catholic communities to form tight networks that embedded enough trust relationships that allowed them to rescue Jews.
Edit: The first published version erroneously omitted the 5th paragraph.
ps. Updated to add Western in the first sentence.