These results may explain at least in part why previous
studies on horizontal spillovers from FDI to domestic firms have been ambiguous results.
Therefore, the differentiation of the level from rate effect of spillovers in empirical studies is
needed, especially when the dataset used covers a relatively short time span. Overall, it is applied virtually to all empirical studies.