Panel data and Hsiao's version of Granger non-causality tests are used to revisit the relationship between GDP and aggregate health care spending, their growth rate series and de-trended series. The possible causality is assumed to be valid in either or in both directions. For the sample of 34...