Page: 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
analysis five forces of framework
Japanese competitiveness, but was in fact an explanation for why certain
industries in Japan were uncompetitive. Industries where competition was limited by
Japanese government policy were uncompetitive. We also collected data on all the legal
cartels in post-World War II Japan, and found that the industries in which cartels
occurred were, with few exceptions, uncompetitive. We also collected data on all
government-sponsored cooperative research projects, which involved several if not most
industry competitors. We found that those industries in which cooperative research
projects occurred were no more likely than the average industry to be competitive, and
many cooperative research projects actually worked against industry competitiveness.
There have been many collaborative projects in the West involving multiple industry
competitors growing out of the efforts to emulate the Japanese case, such as the electric
vehicle project. With few if any exceptions, these have proven disappointing. The
notion that Japan was competitive because of weak antitrust is resoundingly rejected.
Figure 2 highlights some additional data drawn from our study of Japan. We
explored the relationship between the intensity of domestic competition and world export
share in a broad sample of Japanese industries. All of the industries considered were
global in scope. Industries able to command a high world export share were decreed to
be highly productive.
Instead of relying on market structure measures such as seller concentration to proxy
the intensity of competition, we used the extent of fluctuations in domestic market share
among leading firms over an 18-year period. The fluctuation in market share among
leading competitors – controlling for outside shocks – provides a direct and far more
compelling indication of the intensity of competition.9 We found that domestic market
share variability was by far the most powerful influence on Japanese world export share,
dominating conventional measures of comparative advantage such as skilled labor
intensity and capital intensity. The intensity of competition at home, then, was the
strongest influence on Japanese competitiveness abroad. These statistical findings are
consistent with hundreds of industry case studies that have been conducted on the
determinants of competitiveness at the country level, as well as research on national and
regional economic development.10
Interestingly, we found that seller concentration had no significant relationship with
Japanese world export share.11 Nor was it significantly correlated with the extent of
8 Id. See also M. Sakakibara & M.E. Porter, “Competing at Home to Win Abroad: Evidence from
Japanese Industry,” 83 Review of Economics and Statistics 310 (2001).
9 See generally R. Caves & M. Porter, “Market Structure, Oligopoly, and Stability of Market Shares,” 26
Journal of Industrial Economics 289 (1978). For a detailed application to Japan, including definitions,
sources of data, cause and effect issues, see Sakakibara & Porter, supra note 8.
10 See, e.g., “Clusters and Competition: New agendas for Companies, Governments, and Institutions” in
M.E. Porter, On Competition (1998), which contains an extensive bibliography.
11 Sakakibara & Porter, supra note 8.
DRAFT VERSION: 07/22/02
Page 7
domestic market share fluctuations. These results are consistent with other research
which raises doubts about the use of seller concentration as a proxy for the vitality of
competition.12
Figure 2 Competition and International Competitiveness: Evidence from Japanese
Industry
Competitiveness Competitiveness
Local Competition Local Competition
• Measured by World Export Share
• Measured by Fluctuations in
Domestic Market Share
Sakakibara/Porter:
“We find a positive and highly
significant relationship [next page]