Aggregate Productivity Growth and Firm Dynamics in Korean Manufacturing 2007-2017

We study aggregate productivity growth of the Korean manufacturing industry for the 2007-2017 period. We find that the nature of such growth was quite different for two measures of productivity. For labor productivity, most of growth comes from productivity changes among surviving firms. On the other hand, for TFP, most of the productivity growth comes from that of new entrants in recent years. Our work illustrates the different nature of two productivity measures in terms of their growth paths. We also show interesting industry dynamics for both productivity measures, as exiting firms contributed positively to aggregate productivity growth with increasing trends, which suggests that the market had gradually eliminated firms of lower productivity. Using the dynamic Olley and Pakes (1996) decomposition, we also find that for both productivity measures, a substantial productivity growth after the 2008 global financial crisis was due to market share reallocations between firms, but this between-firm contribution has somewhat slowed or been decreasing since then. Our industry sector level study also shows that there has been fundamentally different heterogeneous productivity growth patterns and components across manufacturing sectors. Finally, we find that the wage level also plays a role in moderating or as an accelerating factor for different productivity growth paths among surviving, entering, and exiting firms. We find that higher wage groups had disproportionately higher entry and exit rates, and that the contributions of these industry dynamics to aggregate productivity growth were largest for the highest wage group while the productivity growth from the between firm component was substantially higher for lower wage groups. Therefore, we find that not only a timely change in input and output, but also in the wage, is a necessary ingredient for the pace and magnitude of reallocation to be effective in aggregate productivity growth.