Association for Public Policy Analysis & Management

New York City is the largest school district in the nation, and it has always attracted waves of immigrants seeking a better life in the US. Approximately, 15 percent of children in New York City public schools were born abroad. They come from all over the world. Most are from Latin America, and in particular from countries such as the Dominican Republic, Haiti, Guyana, and Mexico. In 2005 Guyana experienced extensive flooding that affected about 36 percent of the country’s population, and Bangladesh suffered natural disasters that cumulatively affected over 20% of the country’s residents. In 2010, Haiti suffered an earthquake that affected approximately 37% of its population. Over the course of a year, in 2007, the Dominican Republic and Mexico suffered at least 5 natural disasters between storms and flooding. Many immigrant groups have strong ties to their home countries. Are there spillovers of these natural disasters on immigrant communities in the United States? More specifically, are there spillovers on the academic performance of immigrant children enrolled in US schools?

The notion that such spillovers would exist; that immigrant families and their children would be affected by natural disasters, or other disruptive events, in their home countries makes intuitive sense. Immigrants may have family members or friends still living in the origin country. Often, immigrants sent remittances back home to support family members. So far, however, there is no empirical evidence of these spillovers. In this paper, I provide the first estimates that home country natural disasters have consequences for immigrant students’ academic success in their new country. To do so, I use rich administrative data of two cohorts of immigrant students enrolled in NYC public schools in grades 3 through 8 matched with natural disaster occurrences using student’s country of birth.

To preview initial results, student fixed effects models suggest immigrant students exposed to home country natural disasters that affected at least 10 percent of the home country population in the year before taking a standardized test score 0.05 standard deviations lower on the English Language Arts (ELA) test. The largest test score losses are concentrated on the group of students exposed to home country disasters that affected at least 20% of the home country population. For these children, the negative consequences of a large scale home country natural disaster is a decrease in ELA test scores of almost a 10^th of a standard deviation.

In the rest of the paper, I test the robustness of these results, investigate other outcomes, and focus my attention on exploring the role of schools in moderating the effects of home country natural disasters on the academic outcomes of immigrant students.