A simple enquiry on heterogeneous lending rates and lending
behaviour
1. Introduction
According to a well-known statement of the credit view, the monetary policy propagation process
takes place both trough banks’ assets (loans) and banks’ liabilities (money). A great deal of empirical
research in that field focuses therefore on the behaviour of interest rates spreads as well as credit supply in
different phases of the business cycle to different categories of borrowers. The more common empirical
tools to investigate these issues are either VAR models analysing the behaviour of credit aggregates,
interest rate spreads and the way these variables affect economic activity, or, more recently, dynamic
panel data estimates based on individual firms (or individual banks) observations. The former directly
focuses on the behaviour of macroeconomic and policy variables, the latter have the advantage of
modelling individual agents and accounting for disaggregated behaviour and heterogeneity. The use of
“mesoeconomic” dataset, on the contrary, might allow to analyse the interactions between some
macroeconomic policy variables and the behaviour of variables associated to different categories of
agents. The purpose of this paper is twofold: first, performing a simple analysis (both descriptive and
econometric) on the possible disturbances affecting the link between policy rate and interest rates on bank
credit. Since we are interested in heterogeneity, the second purpose of the paper aims at investigating a
very relevant element of heterogeneity, i.e. verifying whether the behaviour of the largest size class of
loans is demand-determined. To do that we will focus on the share of the largest size class of loans over
the total: if the behaviour of this variable appears to be demand-determined, then, for a given level of
credit supply, the share of the other size classes is a sort of “residual” variable, and bank credit would be
allocated according to a sort of implicit hierarchy according to the size class of loans.
All the analyses are performed with a “mesoeconomic” data set acquired from the Bank of Italy,
whose main characteristics are explained in detailed in the appendix. The next section briefly discusses
the motivations for this methodological choice, after presenting a brief survey on the empirical research
on agents’ heterogeneity in the transmission mechanism of monetary policy for Italy. Section 3 and its
subsections contains a preliminary and descriptive analysis of the dataset employed here. Section 4 and
its subsections contain the econometric analyses. Section 5 contains a few concluding remarks.
2. Empirical research on monetary policy transmission in Italy
A common feature of the Italian empirical analyses on monetary policy transmission is the
emphasis on agents’ heterogeneity, credit market structure and, to a lesser extent, geographic differences