Reply to Comment on “The Effect of Same-Sex Marriage Laws on Different-Sex Marriage: Evidence From the Netherlands”
- 301 Downloads
In my previous article titled “The Effect of Same-Sex Marriage Laws on Different-Sex Marriage: Evidence From the Netherlands” (Trandafir 2014), I studied the evolution of the (different-sex) marriage rate in the Netherlands following the legalization of registered partnership (available to both same-sex and different-sex couples) in 1998 and of same-sex marriage in 2001. In a first step, an aggregate-level analysis produced no evidence of significant negative effects of the two laws on the different-sex marriage rate. Similarly, an individual-level analysis showed no significant negative effects of the two laws on the propensity to marry for the average Dutch person. However, the marriage rate of persons with more-liberal characteristics (in terms of residential location or ethnicity) was significantly lower after each law, consistent with them learning over time about registered partnership. At the same time, persons with more-conservative characteristics tended to marry significantly more after each law. In conclusion, I argue that the evidence in Trandafir (2014) indicates that the legalization of same-sex marriage had no significant negative effects on the Dutch different-sex marriage rate.
In her comment, Dinno argues that this conclusion is invalid because of incomplete inference. In particular, she argues that finding no evidence of change in the marriage rate is not proof of absence of change. Therefore, she suggests using equivalence tests to verify that the same-sex marriage law had no effect on the different-sex marriage rate. As the argument goes, a combination of this type of tests and classical statistical tests of the presence of effects would provide a more complete answer to the question, Are rates of different-sex marriage affected by legal recognition of same-sex marriages?
Although it may be interesting to test whether the different-sex marriage rate is unchanged after the legalization of same-sex marriage, the conclusions in my previous research are valid even without conducting such a test. In Trandafir (2014), I aimed to assess several theories predicting either a negative or a positive effect of same-sex marriage laws on different-sex marriage (for details, see the overview in the Conceptual Framework section in that article). The leading theories, and arguably the ones most relevant to public policy, predict that different-sex marriage is negatively affected by the legal recognition of same-sex marriage.1 The burden of proof is then on finding evidence of a decline in the different-sex marriage rate after the enactment of the Dutch same-sex marriage law. This can be achieved through classical statistical tests of differences between the marriage rate in the presence and in the absence of the same-sex marriage laws. A test for the equivalence of marriage rates before and after the legalization of same-sex marriage, as suggested by Dinno, is a test of a particular alternative hypothesis: that of a zero effect. However, this is just one of an infinite number of alternative hypotheses, and there is no prior reason to favor this particular alternative over any other one.
In conclusion, statistical tests of the type suggested by Dinno are not suitable for answering the arguably more policy-relevant question, Is the rate of different-sex marriage negatively affected by legal recognition of same-sex marriage? Because this question is the main focus of Trandafir (2014), the conclusions obtained therein using classical statistical tests remain valid.
Indeed, several same-sex marriage bans were justified based on the claim that different-sex marriage is negatively affected by the legalization same-sex marriage. For example, in the case of California’s Proposition 8, see document 678 from case number 09-CV-2292, or the transcript of David Blankenhorn's testimony, available online (http://www.afer.org/our-work/hearing-transcripts/perry-trial-day-11-transcript/). For the Federal Defense of Marriage Act, see Rep. Henry Hyde's intervention in House of Representatives Report 104-664, 1996.