Abstract:
We test for the presence of externalities in UK manufacturing industry, seeking to identify the channels through which they operate. Using survey data on average variable cost available by industry, we estimate a translog cost function, storing the coefficients on time dummies for a second stage regression in which measures of external activity are entered to capture omitted spillover effects. We carry out the analysis for total manufacturing and for a panel of ten sub-sectors. We find weak evidence that fixed investment represents one significant channel; there is stronger evidence for an effect stemming from high utilisation in the mechanical engineering sector. This appears to be a combination of both thick market effects and knowledge based externalities.