Suppose that data are assembled on (1) research and development expenditures as a fraction of industry costs across U.S. industries (ranked from highest to lowest), and (2) the export success of U.S. industries (ranked from highest to lowest). If the "product cycle theory" is useful as an explanation for the pattern of U.S. exports, then an analyst would expect that statistical association (rank correlation coefficient) between these two series of data would be ? Explain.