**Magdalena Bendini and Lelys Dinarte \***

The World Bank, Washington, DC 20433, USA; mbendini@worldbank.org **\*** Correspondence: ldinartediaz@worldbank.org

Received: 14 August 2020; Accepted: 1 October 2020; Published: 3 October 2020

**Abstract:** This paper studies the effect of maternal depression on early childhood cognition in Peru, where rates of depression are around 50%. By using an instrumental variables approach, this study exploits variation in the exogeneity of the exposure to shocks during early life to instrument for maternal depression. The empirical strategy exploits a novel longitudinal data—the Young Lives survey—that includes information on cognitive outcomes of children and variation in their mothers' mental health status between rounds of data collection. Results suggest that maternal depression is detrimental to a child's vocabulary at age 5, but effects fade out by age 8. Effects do not vary by maternal education but are significant only for children living in disadvantaged households. Estimations indicate that the presence of a partner worsens the effect of maternal depression on vocabulary development, results that are driven mainly by households with heavy-drinking partners. Our findings make a strong case for recognizing maternal mental health problems as disorders of public health significance and guide maternal and infant health policies in Peru.

**Keywords:** child development; child vocabulary; maternal mental health; Peru
