313 /342
臺大管理論叢
第
27
卷第
2S
期
313
GDP:
GDP growth rate
u:
individual effect, with ε as the error term
3. Results
Table 1 presents the results of fixed-effect regression models. The dependent variable is
the competition measure Lerner index (LI); a higher value of the Lerner index refers to
greater market power of the firm and a lower level of market competition. The results of
technical and profit efficiency are shown in models (1) and (2), respectively. Our results in
Table 1 first show that the coefficients of the HHI variable in both models are significantly
positive at the 1% level. Consistent with the SCP hypothesis, our results imply that the
concentration level in the Japanese non-life-insurance industry has a positive effect on the
market power of the firm. In addition, we used data envelopment analysis (DEA) to calculate
the technical (TE) and profit efficiency (PE) measures. Our results show that the coefficients
of these two measures are significantly positive at the 10% and 1% levels, respectively. In
other words, if a firm has a higher efficiency score, it has a higher market power and a lower
market competition level. This is consistent with both the efficiency hypothesis and the
results in Casu and Girardone (2006) and Fernández de Guevara and Maudos (2007).
The effects of the other independent variables are as follows. The coefficient of total
assets (lnTA) and the square of total assets are significantly positive and negative, both at the
1% level. This implies that the asset size of insurers has a nonlinear relationship with firm
market competition. We also find that the coefficient of diversification measure (DIV) is
significantly negative at the 10 percent level. This implies that a lower diversification level
raises the market power of the firm. We already find that the firms in the non-life insurance
industry in Japan have started to specialize since the “Big Bang” financial system reforms in
1996. Our results support that specialization in specific product lines helps increase the
market power of firms. For example, when firms specialize only in certain lines, they can
manage themselves more efficiently and thus increase their market power. Our results are
also consistent with the results in Elango et al. (2008), who examined the property-liability
insurance industry in U.S.