Agglomeration, productivity, and high-growth firms in the manufacturing sector of South Korea

ABSTRACT

Previous empirical studies have highlighted the significant contribution of a small fraction of high-growth firms to job creation in many countries. As a result, local economic development policy has promoted the creation of high-growth firms. Despite the impact of high-growth firms on local employment, few studies have examined the spatial aspect of such firms on a regional scale. This study focuses on the relationship among agglomerations, productivity, and high employment growth in the manufacturing sector of South Korea. Our analysis provides empirical evidence for the mutually positive impact of agglomeration, productivity, and high employment growth in the manufacturing sector of this nation. Results indicate that high-growth firms tend to be more productive, and firms with higher productivity are more likely to experience strong growth in employment. Furthermore, localization, concentration of firms within one industry, reinforces these positive relationships by enhancing productivity that subsequently leads to high employment growth. Our findings imply that spatial policy may play a complementary role in nurturing a greater number of firms with a potential for strong growth in employment.

KEYWORDS:

• High-growth firm
• productivity
• localization
• manufacturing

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

1 The diversity index is calculated using 2-digit KSIC at the provincial level.

2 Source: Chapter 8, High Growth Enterprise, Eurostat-OECD Manual Business Demography, page 61.

3 The interaction variables of TFP*localization and TFP*Urbanization indicated that the effect of the TFP on the employment growth varies from place to place in which the benefit of localization and urbanization differ. For example, if a sign of TFP*Localization is positive, the effect of TFP on the employment growth increases in the place where the extent of localization is greater.

4 As the models include the interaction variables, we have tested the data for multi-collinearity and found that this is not excessive in the present set of variables. All variance inflation factors (VIFs) are less than 2.

5 We are grateful for an anonymous reviewer for this point.

6 We ran TFP models on only a high-tech sector to identify the positive effect of urbanization during three periods. Effects of urbanization are still mixed, and all are not statistically significant. It needs further analysis with more disaggregated geographical areas and industries.