The impact of organisational and technical innovations on productivity: the case of Korean firms and sectors

Abstract

The aim of this study is to provide both qualitative and rigorous empirical evidence regarding the positive and contemporaneous impact of organisational and technical innovations on productivity in Korean firms and sectors. Econometric methods (fixed effects and generalised method of moments) are used to measure the relationship between the just-in-time process (mostly an organisational type of process), technology (R&D) and productivity. In these regressions we also include the control variables of wage gap incentive, capital-to-labour ratio, and education expenses for sales training to accommodate for other important influences. From the overall and quantitative analyses, we provide evidence to the hypothesis that both organisation and technology greatly improve productivity of Korean manufacturing firms and sectors. In addition, not all firms and industrial sectors are influenced in the same way by all these factors. This variety of organisational and technological ‘behaviour’ is clearly indicated in our empirical findings. The contemporaneous impact of both organisational and technical innovations has not been tested before.

Keywords:

• organisational and technical innovations
• technology
• JIT
• fixed effects
• GMM

Acknowledgements

An SNU (Department of Economics) grant for this paper is gratefully acknowledged. We also thank Anna Gunnthorsdottir for reading and commenting on our paper. Finally, we appreciate the referees' suggestions for a more rigorous and precise way to express our thoughts in this paper.

Notes

Usually, process innovations are not split into technological and organizational (Ha 2007).

Some of these organizational innovations, as per Sanidas (2005), are: craft, factory, mass, lean and other types of production systems; linear versus U-shaped machines layout, time and motion studies in scientific management, just-in-time and quality control processes, and so on.

In 1990, for example, Inman and Mehra reported that over 700 papers on the topic of JIT were published in the 5-year period prior to 1990.

The tendency for lower levels of inventories in JIT/QC is important. However, instead of using the level of inventories, we use total inventories to sales ratio as a proxy for JIT/QC in order to control for the size of firms effect.

Newspapers in the 1990s were also checked and thus provide further evidence for this gradual spreading of JIT.

We take this opportunity to thank the author Jung for his assistance to provide to us and comment on the data he generated for his own studies.

When we say OIs we mean the generic form of JIT/QC as explained above.

On the other hand the K/L's impact on labor productivity (Y/L) is expected to be positive.

The random effects model was also estimated and provided no better results than the FE model. The Hausman test usually supports the FE case.

For a good treatment of these methods see Wooldridge (2002).

The results for FE with instrumental variables are not shown here but are available on request like all other results.

We also used labor productivity (LP) for robustness, which provided similar results to TFP (results not reported here).

The FE model with instrumental variables agrees with the GMM model, thus confirming our adoption of the latter model for the transport sector.

We examined the JIT proxy for all 19 sectors. Only these three sectors clearly show this continuous downward trend. To a much lesser extent also printing and primary metals; finally transportation has both an upward and a downward (recently) trends.