Properly identify the relationship between export and innovation is key to designing efficient policy instruments. However, there is still no consensus on how has to be the direct causal relationship between innovation and export. In this paper the authors develop an empirical analysis for the Chilean manufacturing sector which seeks to unravel this relationship. Specifically, an econometric model for various industrial sub-sectors of the Chilean economy for the period 1995-2010, controlling for the level of productivity of each signature, among other variables is presented. Unlike other studies, we consider the total expenditure on innovation rather than just R&D, which can cover a wider range of activities that seek productive within the firm improvements spectrum, while increasing the number of companies that are studied. The results show that in most subsectors, the export propensity or put another way: participation in global market cause, in Granger terms, an increased effort in innovation. Specifically, for the subsectors of minerals and base metals, and metal-mechanical, mutual Granger causality is presented. Finally, for the subsectors wood and paper no statistical significance was evident of Granger causality between innovation and export in any sense. With these results, some policy conclusions for Chile are discussed in relation to instruments of export promotion and innovation.