banks held large excess reserves. Our assumption of a constant reserve ratio is therefore
irrelevant, which may be the reason we could not obtain reasonable results for city banks.
5.3 Adding a constant term
One might question why equation (8) has no constant term. Although the theory does not
allow for a constant term, it is not unusual to include it in empirical analysis. The fact that
there is no constant term may make θ critically dependent on the level of interest rates in
equation (8).
The estimation with the addition of a constant term to equation (8) reveals that this
speculation is wrong. The basic results, presented in Tables 3 and 4, and in Figures 2 and 3,
are confirmed by the estimation with a constant term.
6. What Affects the Degree of Competition?
In this section, we investigate what factors relate to the degree of competition. In
particular, clarifying which element of financial liberalization promoted competition in loan
markets is an interesting agenda. As described in section 2, three important events are
recognized in the financial liberalization in Japan, i.e., the emergence of a huge government
bond market, the deregulation of deposit interest rates, and the deregulation of the
segregation of business lines in the financial sector. We regress the degree of competition
obtained in section 4 on variables representing these elements in order to find their
17