Estimating the Impact of Missing Totalization Agreements

The number of international social security agreements (totalization agreements) signed by the United States is significantly smaller than the number signed by other countries such as Canada and the U.K. This paper estimates the impact of the missing totalization agreements \f1\emdash the agreements that other countries such as Mexico and Turkey have signed with Canada but not the U.S. We find suggestive evidence that international social security agreements increase labor mobility. Using a difference-in-differences framework, we estimate that, on average, the social security agreements between Canada and countries that have no social security agreement with the U.S. reduced Canadian (relative to U.S.) exports to and increased Canadian imports from those countries. Moreover, we find the impact on exports increases over time and is concentrated among the agreements enacted either before 1995 or after 2010 with countries that have a low real GDP per capita, a low exports-to-GDP ratio, and a medium imports-to-GDP ratio, while the impact on imports is concentrated in the first three years following the social security agreements with middle-sized countries at the upper end of the distribution of real GDP per capita.