The proportion of children born to unmarried parents has risen dramatically over the past three decades; in 2010 just over three in ten of all live births in England and Wales were born to cohabiting couples. This significant change raises a number of important questions that cut across government policy areas and academic disciplines, perhaps the most fundamental of which is whether being born to cohabiting rather than married parents matters for children’s well-being. In this article we assess whether there are differences in early measures of cognitive and socio-emotional development between children born to cohabiting and married couples, and if so, whether marriage is the cause of these differences. We show that children born to married parents exhibit higher cognitive and socio-emotional development at ages 3 and 5, on average, than children born to cohabiting couples. We then adopt a systematic empirical approach to try to identify whether marriage is the cause of these differences, or whether they are in fact accounted for by other characteristics of the parents which happen to be correlated with marital status. As this empirical strategy involves some judgment on our part, we cannot claim a definitive answer. We do, however, regard our results as a strong indication that marriage plays a relatively small role, if any, in promoting children’s early cognitive or socio-emotional development.