When is intermarriage a good measure of tolerance? That is the question that Nico Voigtländer and I are asking in our little paper prepared for the ASSA meetings in San Diego (forthcoming in the AER p+p issue this year). Here is part of our conclusion:
Remarkably, attitudes towards intermarriage are stable over more than half a century – in towns and cities where intermarriage rates between Jews and gentiles were high before 1939 because of positive attitudes among the German population, respondents today are much more comfortable with the idea of having a Jewish family member. In contrast, in places where out-marriage rates for Jews were high because the Jewish community was small, attitudes today are also sharply more critical today – not least because the local population was often more hostile, reducing the number of Jews in a location.
... Only the part of the variation [of intermarriage rates] driven by attitudes directly is valuable in explaining cultural preferences. The share of the variation reflecting the “tightness” of the marriage market can confound the result; in extreme cases – such as Germany before 1939 – it may even induce an inverse correlation.