Industrial concentration and competition in Indonesian manufacturing

Abstract

This thesis examines the interrelationship between market structures, firm rivalry, and government intervention in the Indonesian manufacturing sector over the period 1975 to 1995. Two empirical methodologies are used in this study. First, the Structure. Conduct - Performance (S-C-P) approach to industrial organisation provides the basic framework for our analysis in the first part of the study. The second research methodology is indus•ry case studies. Two industry studies presented in this study are cigarettes and cement. The study begins with an overview of the evolution of Indonesian trade and industrial policies and the salient features of the manufacturing sector. An analysis of the trends in seller concentration over the period 1975 to 1993 is also provided. The descriptive analysis on seller concentration shows that there is a long-term declining trend in industrial concentration, particularly in those industries that were highly concentrated in the mid-1970s. This provides the backdrop into subsequent statistical analysis of the interrelationship between concentration and other aspects of structure and performance. For this purpose, we specify and estimate a simultaneous equations model of industry structure, conduct and performance. The interrelationships among six key variables (concentration, profitability, foreign ownership, export intensity, import penetration and trade policy) are estimated for two different policy periods; (1) pre-1986 period of heavy trade and industry policy intervention, and (2) post-1986 period of trade and industry liberalisation. A number of interesting findings emerged from the analysis. Trade protection, interacted with seller concentration, was a major determinant of high profitability in concentrated industries in the pre-liberalisation period. This relationship significantly weakened as a result of the late 1980s trade liberalisation. High concentration had a positive influence on effective trade protection, providing some support for the interest group model, which asserts that highly concentrated industries find it easier to lobby government for protection. However, the level of concentration had no significant influence on export intensity or import penetration over and above the other variables considered in the analysis. Finally, our results suggest that market structure factors (economies of scale, capital costs, product differentiation, market size and regional market segmentation) are the major determinants of industrial concentration in Indonesian manufacturing. We could not find a direct relationship between trade policy, regulation and concentration. These insignificant results probably arise due to limitations inherent in cross-sectional tests of the effect of trade policy and regulation on market structure. We tested the effect of international influences (export intensity, import penetration and foreign ownership) on concentration. However, the results turned out to be statistically insignificant in both the pre - and post-liberalisation periods. The study extends the analysis to include the determinants of changes in leading firms' market shares. Changes in market shares are a good indicator of firm rivalry. The results show that regulations in several industries are associated with stable market shares, suggesting that they have reduced competition in these industries. The second part of thesis presents two industry case studies. The focus of the case studies is on the nature of competition among firms, and the influence of regulation on this process. In cement, until the industry was deregulated in 1998, government regulations - distribution and price controls - had created a market structure that closely resembled a cartel. The second case study is cigarettes. This study estimates the effect of advertising expenditure on seven leading firms' market shares, using monthly advertising and market share data. Our results show that advertising competition reallocates sales between leading firms. An important finding of our study is that an equal percentage increase in advertising by all firms will change the distribution of market shares in favour of the more successful, larger firms in the long-run. This arises because larger firms have an 'image advantage' over the smaller firms. This image advantage is an asymmetry that constitutes a barrier to upward mobility of smaller, less-favoured existing firms. Show more