on nominal, monetary policy shocks. Unlike Kiyotaki and Moore [16], we motivate
a demand for nominal units of account through a cost of information gathering
argument similar to the one proposed by Lucas [18]. As in Cooley and Quadrini
[10], we assume imperfect capital markets in the form of an endogenous borrowing
limit but we do not need a limited participation constraint for the monetary au-
thority to affect the nominal interest rate. Instead, the central bank introduces a
reserve requirement which regulates the amount of intermediation in the economy.
We want to stress that by carefully modelling the nominal side of the economy,
there is no need to impose stickiness on any variable since the economy propagates
shocks by itself. Furthermore, we provide our agents with an occupational choice
to decide whether to become a worker or an entrepreneur. This modelling strategy
allows us to examine the role that endogenizing the extensive margin of investment
decisions has for the monetary transmission mechanism.
The paper is organized as follows. Section 2 presents some new evidence on the
effects of monetary shocks on economic activity. Section 3 describes the model.
Section 4 presents the numerical simulations and section 5 concludes.
2 the effects of monetary shocks on economic activity
The empirical literature on the monetary transmission mechanism has concluded
that an expansionary monetary shock drives down nominal interest rates which
makes output rise. The price level responds slowly to these changes and ends up
increasing in the long run. Although there is a general consensus about these
qualitative effects, the profession has yet to agree on a single set of identification
restrictions to isolate monetary shocks in the data and, therefore, the quantitative
estimations of these effects vary widely across papers.
Figure 1 presents the responses of the capacity utilization rate, the growth
rate of the GDP deflator and the 3-month T-bill rate to a monetary shock that
occurs at period 1. These responses are estimated using the scheme employed
in Menner and Rodnguez Mendizabal [19].1 The identification strategy is based
on the sign restrictions of the responses of these variables to three different types
of shocks (demand, supply and monetary) imposed by theory. Because all three
variables are percentages, the responses are measured as differences with respect to
average values. So, for example, an expansionary monetary shock that reduces the
interest rate by 50 basis points, raises capacity utilization from its average of 81.15
percent to 81.40 percent or 0.25 percentage points. For comparison purposes, the
figure also includes the responses estimated with the identification scheme used in
Christiano, Eichenbaum and Evans [8] (CEE) which represents a method widely
1 The sample covers from the first quarter of 1960 to the first quarter of 2003
(173 observations).