Association for Public Policy Analysis & Management

This paper explores differences in perinatal inequality between migrant- and native- origin newborns in Spain and, more specifically, differences in birth weight (BW), an outcome that is highly correlated with health and education in childhood and adulthood. Previous literature has shown that, in line with the "healthy immigrant paradox", the children of immigrant mothers tend to have a lower risk of low BW.

For our analysis we built a dataset combining two sources of administrative data, one for Spain, the receiving country, and one for Ecuador, the sending country. The Ecuadorian Register of Births made available by the National Institute of Ecuadorian Statistics, which results from the systematization of basic information of every birth that took place in the country. The Spanish births came from the Population Movement Statistics in the Childbirth Statistics Bulletin provided by the National Statistics Institute. The Spanish data also have an administrative nature (registers) and cover the entire population of newly born in Spain in a complete natural year. In both cases we analyze data from 2013, the most recent year available simultaneously in both countries. We compare three groups of newborns: those from Spanish-born mothers (N=265,968), from Ecuadorian-born mothers settled in Spain (N=166,079), and from Ecuadorian mothers living in Ecuador (N=4,612). The comparison of migrants to Spain with non-migrants in the country of origin provides an interesting scenario to test the validity of the potential existence of positive selection in the migrant population.

Figure 1. Estimated average BW for the three groups

Our paper does not only confirm that the well-known epidemiological regularity of healthier babies among migrants in advanced economies, namely an advantage of immigrant-origin babies in terms of avoiding low BW, also applies to Spain (see the Figure above). By using quintile regression, it confirms the weight advantage throughout the whole distribution of weight. In the other extreme, when the baby’s weight is above 4,000 grams, a threshold associated with increased health risks for the newborn, migrant-origin babies weigh about 80 grams more than native-origin ones and over 250 grams more than newborns in Ecuador. Our analysis includes relevant controls including indirect  measures of differences, at the province level, in the coverage and quality of health care.

We contribute to the literature by confirming the consistently higher BW enjoyed by the children of migrant mothers in a country such as Spain, where immigration is a recent phenomenon and access to quality healthcare is universal and free of charge. In addition, the comparison between babies from Ecuadorian mothers born in Spain and Ecuador, respectively, provides evidence consistent with the existence of positive selection among migrants. Last, the use of quantile regression has allowed to find out that the higher average BW from immigrant mothers becomes a disadvantage at the highest end of the distribution of weight.