By the above-explained examination the result is that the property of economies of scale is hardly recognizable in Japan's life insurance business. This differs from the results Johnson obtained for Britain and Houston for America. So we should like to refer to this diversity as the conclusion of this report. As the cause for such diverse results, firstly differences in method are conceivable. On this point we must admit some lack of strictness on our side compared with the foreign studies, as mentioned already. Secondly, since diversity of business scale is smaller in Japan than in foreign countries, a possibility is that economies of scale, if actually working, might have been dissolved into a sector of the cost curve (straight-line sector). Lastly, by our former study of the same theme for nonlife insurance the property of economies of scale was realized. Then, if the scale-cost relation differs between life and nonlife in the same country of Japan, possibly the reason lies in the institutional particularity of life insurance business. And if so, the biggest factor may be ascribable to the system of policy sale. Nonlife employs a system of agent while life that of fieldman. This system, may possibly affects costs, and hence on the scale-cost relation. Final conclusions on the result of this study should wait solution of these problematic points.