Abstract This paper investigates the influence of the spatial dimension on financial contagion in the subprime crisis based on adjusted and local correlation measures. Daily series of stock indexes of American and Asian countries are used from January 1, 2003, to December 30, 2011. We consider two groups of countries: the first group includes the United States and countries that are geographically close: Brazil, Argentina, Mexico, and Canada. The second group includes countries that are geographically distant from the United States: Hong Kong, India, Australia, Indonesia, Malaysia, South Korea, China, and Singapore. The results show that simple and adjusted correlations are not enough to explain the spatial effect of contagion. Using local correlations and polynomial regressions, the results show the existence of spatial contagion between the United States and all countries in the American region. As for countries that are geographically distant from the United States, we prove the existence of spatial contagion between only some groups of countries (United States/India, United States/Australia, United States/Indonesia, United States/Malaysia, United States/China). These results have international diversification, and within-industry implications.