A Comparison of the Financial Characteristics of Japanese "Keiretsu-Affiliated" and "Independent" Firms
用方差分析比较224家日本制造业企业，发现系列关联企业因能从主银行获得优惠贷款而负债率更高、流动性更低，但两类企业的净资产收益率无显著差异。
ABSTRACT The Japanese keiretsu system has received considerable attention in the U.S. business literature. A number of studies have investigated various aspects of the keiretsu practice in Japan and the impact of this system on the characteristics of Japanese firms. However, there are no studies that compare the overall characteristics of the keiretsu-affiliated firms and the firms. In this study we attempt such a comparison by using the parametric and nonparametric ANOVA (analysis of variance) techniques and a sample of 224 Japanese firms from 13 manufacturing industries. Due to their ability to borrow from the main bank of the keiretsu with liberal terms, the keiretsu-affiliated firms appear to have significantly higher debt ratios and lower liquidity ratios compared with the firms. The keiretsu-affiliated firms are generally larger than the firms with more institutional ownership. Although the firms have significantly higher operating profit margin and return on assets ratios, the return on equity ratios of the two groups of firms are not significantly different. There are considerable similarities between the two groups of firms in terms of turnover ratios, growth ratios, and market ratios. INTRODUCTION Japanese firms have received considerable attention in the U.S. business literature because of the unusual institutional arrangements under which they operate. Many Japanese firms belong to corporate groups called financial keiretsu. A keiretsu includes a main-bank which not only provides debt financing to member firms with favorable terms but also owns large amounts of their common stock [see: Hoshik et al. (1990)]. Although there are many studies that investigate various effects of the Japanese keiretsu system on member firms [see, e.g., Berglof and Perotti (1994)], there are no studies that compare the overall characteristics of the keiretsu-affiliated firms and the firms. The objective of this study is to attempt such a comparison. DATA AND METHODOLOGY Comparing the characteristics of different groups of firms using ratios has been a widely employed research methodology in finance [see, e.g., Altman (1968) and Belkaoui (1978)). In this study, we also use ratios to compare the characteristics of the keiretsu-affiliated firms and the firms. The variables used in the study as various measures of the firms' characteristics are presented in Table 1. The data for all of the variables, except the institutional ownership ratio figures which are from the 1997 Analysts Guide published by Diawa Institute of Research Ltd., are obtained from the DISCLOSURE/Worldscope database. All ratios are computed with data from the 1997 year-end statements of the firms. The annual total assets growth rate and annual sales growth rate are computed with data for the 1993-97 period. The sample includes 112 keiretsu-affiliated firms and 112 firms, a total of 224 firms. The firms are classified as keiretsu-- affiliated or independent using the classification in Industrial Groupings in Japan: 1994-95 published by Dodwell Marketing Consultants. Industry classification can affect the characteristics of firms. Therefore, to avoid any bias in the comparisons, an equal number of keiretsu-affiliated firms and firms are included in the study from 13 different manufacturing industries with the three-digit SIC industry codes. The one-way analysis of variance (ANOVA) method is used to compare the characteristics of the keiretsu-affiliated firms and the firms. The parametric ANOVA method assumes that the variables compared have normal distributions. However, a study by Constand (1994) finds that most Japanese ratios are not normally distributed. …