“…This is partly because most previous empirical research on deindustrialization (see e.g. Wood, 1994Wood, , 1995Saeger, 1997;Ramaswamy, 1997, 1999;Alderson, 1997Alderson, , 1999Kucera and Milberg, 2003;Rowthorn and Coutts, 2004;Whitford, 2005;Brady and Denniston, 2006;Kollmeyer, 2009;Kang and Lee, 2011;Coutts, 2004, 2013;Rodrik, 2015) has either ignored one or more of the explanations of deindustrialization altogether or has simply failed to test all of them simultaneously. In addition to this potential omitted variable bias, which has already been stressed by Kollmeyer (2009), we argue that the common measure of trade as a factor likely to account for deindustrialization in previous studies is inaccurate and may lead to misleading results.…”