Location via proxy:   [ UP ]  
[Report a bug]   [Manage cookies]                

Monetary Economics Credit Cycles - The Kiyotaki Moore Model: Nicola Viegi

Download as pdf or txt
Download as pdf or txt
You are on page 1of 15

Monetary Economics

Credit Cycles - The Kiyotaki Moore Model

Nicola Viegi

University of Pretoria

March 2010

Viegi (University of Pretoria) Credit Cycles Lecture 5 March 2010 1 / 15


Introduction

Crisis shows centrality of …nancial market


Objective of this lecture - introduce …nancial intermediaries in a macro
model
The Microeconomics of Banking (to be done in reading groups)
Diamond Dybvig Model - A Model of Bunk Runs
Holstrom and Tirole Model - The Role of the State in Providing
Liquidity

This lecture
Stiglitz and Weiss Model - Imperfect Information and Financial
Transactions
Kiyotaki Moore Model - The Business Cycle E¤ects of Credit Market
Imperfections

Viegi (University of Pretoria) Credit Cycles Lecture 5 March 2010 2 / 15


Credit Market Imperfection - the Stiglitz Weiss Model

Expected Value of Investment Project

pi Ris + (1 pi ) R f = R
Banks lend B = K W and get (1 + r )B in case of success and R f in
case of failure.
Assume that Ris > (1 + r )B > R f for all i.
Asymmetric info: entrepreneur knows his pi the bank does not.
Pro…ts

Firm : E (π i ) = pi [Ris (1 + r )B ]
Z p Z p
Bank : E ( π b ) = (1 + r )B pi g (pi ) dpi + R f (1 pi ) g (pi ) dpi
0 0

Viegi (University of Pretoria) Credit Cycles Lecture 5 March 2010 3 / 15


Credit Market Imperfections

Express the expected payo¤ π i as :


h i
f f
E (π i ) = R R pi (1 + r )B R

which is decreasing in pi :
high risk investors are willing to pay more for a loan, but then dp/dr < 0.
Impact on Banks
Z p
dE (π b )
= B pi g (pi ) dpi +
dr 0
dp h i
+ (1 + r ) Bpg (p ) + R f (1 pi ) g (p )
dr
Optimal Respone of the Bank - Collateral

Viegi (University of Pretoria) Credit Cycles Lecture 5 March 2010 4 / 15


Credit Market Imperfections

Credit Demand

Credit Supply
Interest Rate

Expected Bank Return

Interest rate signal of riskness


Viegi (University of Pretoria) Credit Cycles Lecture 5 March 2010 5 / 15
Credit Market Imperfection and the Business Cycle - the
Kyotaki Moore model

Credit Market Imperfections in a general equilibrium model - Kiyotaki


and Moore
it produces comovement of amount of credit, asset prices and
aggregate output,
it creates a propagation mechanism that produces persistence and
ampli…cation of a shock,
it produces procyclical productivity even if technology does not change,
it is able to explain cross-industry comovements.

Quite Complex - Need di¤erentiated agents

Viegi (University of Pretoria) Credit Cycles Lecture 5 March 2010 6 / 15


Credit Cycles

Viegi (University of Pretoria) Credit Cycles Lecture 5 March 2010 7 / 15


The Model

a constant interest rate (one less variable)


no labor supply decision
no capital accumulation
only one asset that can be used for production, and is available in
…xed supply in the aggregate.

Viegi (University of Pretoria) Credit Cycles Lecture 5 March 2010 8 / 15


The Model

The model includes two kind of agents, productive (farmers) and


unproductive (gatherers). The expected utilities of farmer and gatherers
are respectevely

!

Farmers : E ∑ βs xt +s
s =0
!

Gatherers : E ∑ β0s xt0+s
s =0

where β0 > β .

Viegi (University of Pretoria) Credit Cycles Lecture 5 March 2010 9 / 15


The Model II

The productive agents use a linear production tecnology in capital only as,

yt = (a + c ) kt 1

a fraction ckt 1 of the produced goods is untradable so that the ”farmer”


must consume it on his own.
The productive agent is subjected to a credit constraint like

Rbt = qt +1 kt

Viegi (University of Pretoria) Credit Cycles Lecture 5 March 2010 10 / 15


The Model

basic expression for the ‡ow of funds

qt (kt kt 1 ) + φ (kt λkt 1 ) + Rbt 1 + xt ckt 1 = akt 1 + bt

where φ (kt λkt 1) denotes an input for reproduction of capital.

Viegi (University of Pretoria) Credit Cycles Lecture 5 March 2010 11 / 15


The Model

Only a fraction π of the population can invest while a fraction (1 π )


cannot.
For a farmer, investment is strictly better than consumption, so that it will
use all the funds available to invest, so that xt 1 ckt 1 and Rbt = qt +1 kt
are binding.
Substituting these in the ‡ow of funds and rearranging we have:
qt +1
qt + φ kt = (α + λφ + qt ) kt 1 Rbt 1
R
On the other hand, non productive capital is just depreciating, so that
kt0 = λkt0 1 .

Viegi (University of Pretoria) Credit Cycles Lecture 5 March 2010 12 / 15


The Model
Combining these two law of motion, we get the aggregate capital law of
motion

π
Kt = ( 1 π ) λKt 1 + q t +1 [(α + λφ + qt ) kt 1 Rbt 1]
qt + φ R
(1)
Aggregate debt follows

Bt = RBt + qt (Kt Kt 1 ) + φ (Kt λKt 1 ) aKt 1


1 (2)
Finally, the Euler equation for consumption (a bit atypical), which
determine the asset price
qt +1
u (Kt ) = qt
R
where u (Kt ) is the user cost of capital = G 0 (K Kt ) /R (explained in
the paper). These equations are a …rst-order non-linear system. There is
an unique steady state (q , K , B ) with associated steady state user cost
Viegi (University of Pretoria) Credit Cycles Lecture 5 March 2010 13 / 15
Business Cycle Dynamics

0.3

0.25

B/B*
0.2

K/K*

0.15

0.1 q/q*

0.05

-0.05
0 10 20 30 40 50 60 70 80 90
time

Viegi (University of Pretoria) Credit Cycles Lecture 5 March 2010 14 / 15


Conclusions

Real Business Cycle with large endogenous ‡uctuations


Credit + Imperfect information = …nancial accelarator mechanism
Next - Introducing …nancial frictions in a monetary model with sticky
prices and monopolistic competition

Viegi (University of Pretoria) Credit Cycles Lecture 5 March 2010 15 / 15

You might also like