Description
The recent global financial crisis was the first in recent history that was triggered by problems in the financial system of the mature economies. Existing work on financial crisis in emerging market countries, however, almost exclusively focus on the role of financial frictions in the domestic economy.
Global Financial Crisis, Financial Contagion,
and Emerging Markets
F. Gulcin Ozkan and D. Filiz Unsal
WP/12/293
© 2012 International Monetary Fund WP/12/293
IMF Working Paper
Strategy, Policy, and Review Department
Global Financial Crisis, Financial Contagion and Emerging Markets
1
Prepared by F. Gulcin Ozkan and D. Filiz Unsal
Authorized for distribution by Catherine Pattillo
December 2012
Abstract
The recent global financial crisis was the first in recent history that was triggered by problems
in the financial system of the mature economies. Existing work on financial crisis in emerging
market countries, however, almost exclusively focus on the role of financial frictions in the
domestic economy. In contrast, we propose a two-country DSGE model to investigate the
transmission of a global financial crisis that originates from financial frictions in the rest of the
world. We find that the scale of financial spillovers from the global to the domestic economy
and trade openness are key determinants of the severity of the financial crisis for the domestic
economy. Our results also suggest that the welfare ranking of alternative monetary policy
regimes is determined by the degree of financial contagion, the degree of trade openness as
well as the scale of foreign currency denominated debt in the domestic economy.
JEL Classification Numbers: E5, F3, F4
Keywords: sudden stops, financial crisis, emerging markets
Author’s E-Mail Address:[email protected] and [email protected]
1
We are grateful to an anonymous referee for comments and suggestions which have helped substantially improve
the paper. We also would like to thank Christopher Adam, Lynne Evans, Hugh Metcalf, Marcus Miller, Joe
Pearlman, Peter Sinclair, Mike Wickens and seminar participants at Newcastle University, participants at the
Money, Macro and Finance Group (MMF) Annual Conference at the University of Bradford, the conference on
“Monetary policy before, during and after the crisis” held at Heriot-Watt University, 29-30 September, 2011 and
the Emerging Markets Group - ESRC Workshop on Global Linkages and Financial Crises at the Cass Business
School, University of London, 27 April 2012. The usual disclaimer applies.
This Working Paper should not be reported as representing the views of the IMF.
The views expressed in this Working Paper are those of the author(s) and do not necessarily
represent those of the IMF or IMF policy. Working Papers describe research in progress by the
author(s) and are published to elicit comments and to further debate.
Contents Page
1. Introduction ....................................................................................................................4
2. The Model ......................................................................................................................6
2.1 Households .........................................................................................................8
2.2 Firms ..................................................................................................................9
2.2.1 Production Firms .....................................................................................10
2.2.2 Importing Firms ......................................................................................11
2.2.3 Unfinished Capital Producing Firms ......................................................11
2.3 Entrepreneurs ...................................................................................................12
2.4 Monetary Policy ...............................................................................................15
2.5 General Equilibrium and Balance of Payments Dynamics ..............................16
3. Solution and Parametrization .......................................................................................17
3.1 Consumption, Production and Monetary Policy ..............................................17
3.2 Entrepreneurs ...................................................................................................18
4. Financial Crisis and the Domestic Economy ...............................................................19
4.1 Model Dynamics ..............................................................................................19
4.2 Financial Crisis Originating in the Domestic Economy ..................................20
4.3 Financial Crisis in the Global Economy ..........................................................21
4.3.1 The Impact of the financial shock on the foreign ecocnomy ..................21
4.3.2 The transmission of the foreign shock onto the domestic economy .......22
4.3.3 The role of trade openness ......................................................................25
4.4 Monetary Policy Options and Welfare Analysis .............................................26
5. Conclusions ..................................................................................................................29
References ................................................................................................................................31
Tables
1. Parameter Values for Consumption, Production and Monetary Policy ...............................45
2. Parameter Values for the Entrepreneurial Sector .................................................................45
3. Business Cycles in Emerging Economies: Data vs. Model .................................................46
4. Business Cycles in Advanced (Big) Economies: Data vs. Model .......................................47
5. Cross-Country Correlations .................................................................................................48
6. Welfare Results ....................................................................................................................49
Figures
1. Responses to a Financial Crisis in Domestic Economy .......................................................50
2. Responses to a Financial Crisis in a Foreign Economy .......................................................51
3. Responses to a Financial Crisis in Foreign Economy with Financial Contagion ................52
4. Responses to a Financial Crisis in Foreign Economy without Financial Contagion ...........53
3
5. Responses to a Financial Crisis in Foreign Economy with Financial Contagion: Domestic
Economy-The Impact of Openness ..........................................................................................54
6. Responses to a Financial Crisis in Foreign Economy without Financial Contagion:
Domestic Economy-The Impact of Openness .........................................................................55
7. Responses to a Financial Crisis in Foreign Economy with Financial Contagion: Domestic
Economy-The Role of Monetary Policy Strategy ....................................................................56
8. Responses to a Financial Crisis in Foreign Economy without Financial Contagion:
Domestic Economy—The Role of Monetary Policy Strategy .................................................57
Appendixes
A. Optimal Contracting Problem .............................................................................................35
B. Model Equations .................................................................................................................39
B.2 Model Equations: Foreign Economy ....................................................................42
1 Introduction
The global …nancial turmoil that has gripped the world economy since August 2007
has been widely viewed as unprecedented, at least since the Great Depression of the
1930s. The turbulence in …nancial systems was followed by a signi…cant reduction
in real economic activity in a large number of countries. For emerging market and
developing economies, …nancial crisis is not a new phenomenon. Indeed, since the
early 1990s countries as diverse as Mexico, Russia, a number of East Asian countries,
Brazil, Turkey and Argentina have all been hit by either currency or …nancial crises,
or both. Although country experiences have varied with regard to the source of
di¢culty in each episode, the pro…le of crises has been fairly similar. A "sudden
stop" of capital in‡ows is almost always followed by a sharp contraction in economic
activity. Furthermore, many countries witnessed substantial losses in the value of
their currencies, which greatly helped in recovering from the crises. In essence,
these countries were able to expand net exports to alleviate or compensate for the
contractionary e¤ects of foreign currency liabilities following devaluations – the so-
called balance sheet e¤ects. For instance, sharp current account reversals were a
common feature of recovery processes in most Asian countries following the 1997
crisis.
When set against this background, from the viewpoint of an emerging market
economy, the recent global …nancial crisis has been di¤erent in two major ways.
1
First, the source of the sharp reduction in capital in‡ows on this occasion has been
the severe liquidity squeeze in the …nancial markets of developed economies, unlike
in any of the previous experiences. Indeed, the slowdown in …nancial ‡ows to the
emerging market economies followed from the virtual standstill in the credit markets
in the US and the UK spreading to other major …nancial markets. Second, emerging
market countries have also witnessed a substantial fall in their exports as the …nan-
cial crisis hit consumer spending in the developed world. Hence, given the strong
downturn in the global economy, countries have been unable to export their way
out of the crisis even though a large number of countries experienced substantial
devaluations of their currencies.
The above evaluation suggests that both …nancial and trade channels have been
1
See, Reinhart and Rogo¤ (2009) for a comprehensive evaluation of the di¤erences between the
current and previous experiences of …nancial crisis.
4
crucial in the transmission of the global …nancial crisis to emerging market coun-
tries. Motivated by this observation, in this paper, we develop a two-country dy-
namic stochastic general equilibrium model with an explicit treatment of both trade
and …nancial linkages between the countries. This enables us to investigate pos-
sible spillover e¤ects of a …nancial crisis originating in the global economy on to
a domestic, small open economy. There are three features of our model economy
that are representative of emerging market economies. First, the domestic economy
exhibits …nancial frictions in the form of high leverage; that is, a large share of in-
vestment is …nanced with external resources. Second, the borrowing is taken to be
from abroad in foreign currency terms, as is common in emerging market countries.
In the presence of such foreign currency denominated debt - widely referred to as
liability dollarization - a change in the perception of foreign lenders of the current
state of the economy leads to an endogenous adjustment in the cost of borrowing,
generating a negative feedback loop between real and …nancial sectors. Moreover,
the presence of liability dollarization makes the balance sheets of the …nancial sys-
tem sensitive to the changes in exchange rates. Finally, our model gives explicit
consideration to exchange rate pass-through, the scale of which distinguishes the
experiences of emerging and mature economies.
The main channels through which a global …nancial shock impacts upon the
emerging market economy are as follows. The re-pricing of credit risk increases the
cost of external …nancing, inducing a sharp decline in output and domestic in‡ation,
and a depreciation of the domestic currency. Since the external risk premium is
tied to the leverage ratio, …rms reduce new borrowing in order to decrease the
endogenous component of the risk premium. Moreover, the fall in domestic in‡ation
and the depreciation of the domestic currency increase the real debt burden for
leveraged households, leading to a decrease in consumption. The depreciation of the
domestic currency enables the home country to compensate, at least partly, for the
decline in consumption and investment demand. However, this only applies if the
…nancial crisis originated in the domestic economy. In contrast, when the source
of the …nancial shock is global, the export channel works in the opposite direction
with global contraction leading to a fall in export demand for the domestic output
resulting in a further decline in domestic economic activity.
We also explore the role of …nancial contagion and trade openness on the prop-
5
agation of international …nancial shocks. Our results reveal that the greater the
…nancial contagion, the greater the severity of the …nancial crisis for the domestic
economy. A higher level of …nancial contagion induces a greater increase in the ex-
ternal risk premium and thus a greater fall in capital in‡ows reducing investment,
output and consumption. We also examine the role of the domestic economy’s open-
ness to trade on its response to the global …nancial shock. We …nd that, in contrast
to the existing literature, the greater a country’s trade integration with the rest of
the world, the greater the response of its macroeconomic aggregates to a sudden
stop of capital ‡ows. We then provide a rigorous welfare analysis of two alternative
monetary policy frameworks; an in‡ation targeting regime versus a …xed exchange
rate regime. Our results reveal that the welfare ranking of the two regimes is deter-
mined by the degree of …nancial contagion, the degree of trade openness as well as
the scale of foreign currency denominated debt in the domestic economy.
The remainder of the paper is organized as follows. Section 2 sets out the struc-
ture of our two-country DSGE model by describing household, …rm and entrepre-
neurial behavior with special emphasis on the description of …nancial frictions. Sec-
tion 3 presents the solution and the calibration of the model. Section 4 presents and
discusses our quantitative results. Section 5 provides the concluding remarks.
2 The Model
Based on the …nancial accelerator mechanism developed by Bernanke et al. (1999),
our model shares its basic features with the recent theoretical studies incorporating
the …nancial accelerator in combination with liability dollarization such as Ces-
pedes et al. (2004), Devereux et al. (2006), Gertler et al. (2007) and Elekdag and
Tchakarov (2007). However, our framework di¤ers from the existing general equi-
librium models with …nancial frictions in three important ways. First, we develop
a two-country sticky price DSGE model where both the trade and …nancial link-
ages between the two countries are fully speci…ed. Modelling the foreign economy
explicitly enables us to investigate the propagation of global …nancial crisis on the
domestic economy by considering both the trade and …nancial channels. Given that
our purpose is to explore the transmission of a global …nancial crisis onto an emerg-
ing market economy, we set the characteristics of the two countries in such a way
that domestic (foreign) country exempli…es an emerging market (mature) economy.
6
Having a two-country framework also enables us to consider important transmission
channels which are ignored in earlier studies.
2
Secondly, in our framework sudden
stops are modeled as an unfavorable change in the perception of lenders which cre-
ates a self-ful…lling pessimism about the economy through the enforcement of tighter
credit conditions, similar to Curdia (2008). We argue that this better represents the
source of …nancial crises relative to the existing work on sudden stops which de…nes
the initial shock as either an aggregate structural shock, such as a rise in foreign
interest rates, or an adverse shock to fundamentals. Finally, in contrast to most
existing New-Keynesian DSGE models, we adopt Greenwood, Hercowitz, Hu¤man
(1988) (hereafter GHH) preferences in the speci…cation of household utility.
3
This
is motivated by the fact that the GHH preferences improve the ability of general
equilibrium models in capturing business cycle facts especially in emerging market
countries (see, for example, Neumeyer and Perri, 2005 among others).
The world economy consists of two economies; a domestic economy and a for-
eign economy. The total measure of the world economy is normalized to unity, with
domestic and foreign economies measuring : and (1 ÷
, respectively. Given that
our main purpose is to explore the impact of a global shock on an emerging market
economy, we maintain that the domestic economy is small in size relative to the
foreign one. Otherwise, domestic and foreign countries share the same preferences,
technology and market structure for consumption and capital goods. The model for
the domestic small economy is presented in this section. We use a similar version of
2
Perri and Quadrini (2010) also utilize a two-country model in analyzing business cycle ‡uctua-
tions across countries. However, in contrast to our model, the …nancial (credit) shock in their model
is transmitted to the rest of the world through its e¤ect on the borrowing capacity of …rms and the
resulting cost and thus employment implications. Other channels of international transmission of
domestic shocks have also been investigated in the literature. These include international portfolio
holdings of leverage constrained investors (Devereux and Yetman, 2010); binding enforcement con-
straints on credit supply (Devereux and Sutherland, 2011); syncronization of borrowing costs and
credit spreads across countries (Dedola and Lombardo, 2012); and global banking (Enders et al.
2011). Our paper di¤ers from these analyses in two major respects, in addition to many di¤erences
in the modelling structure. First, in contrast to our analysis where both …nancial and trade linkages
are explicitly modelled and play a key role, …nancial interactions are treated as the main source of
interdependence between countries in these studies. Second, unlike in the above mentioned studies,
we explore the ampli…cation of a global …nancial shock onto an emerging market economy by setting
the model structure accordingly.
Among the existing studies the one that is closest to our analysis is Gilchrist (2004) who explores
the role of leverage in the transmission of shocks from developed to developing countries. Although
the two models share a number of features as both are based on Bernanke et al. (1999), our
model di¤ers from Gilchrist (2004) in a number of important aspects. These include the teatment
of …nancial accelerator mechanism, the modelling of the source of shocks as well as the choice of
preferences among others.
3
Notable exceptions are Monacelli and Perotti (2008) and Gertler and Karadi (2010).
7
the model for the foreign economy with the exception that entrepreneurs in the do-
mestic country borrow in foreign currency while entrepreneurs in the foreign country
borrow in their domestic currency. In what follows, variables without superscripts
refer to the home economy variables, while variables with a star indicate the foreign
economy variables, unless indicated otherwise.
2.1 Households
A representative household is in…nitely-lived and seeks to maximize:
E
0
1
t=0
,
t
1
1 ÷o
(C
t
÷
H
1+,
t
1 ÷,
)
1o
, (1)
where C
t
is a composite consumption index, H
t
is hours of work, 1
t
is the mathemat-
ical expectation conditional upon information available at t, , is the representative
consumer’s subjective discount factor where 0 < , < 1, o 0 is the inverse of the
intertemporal elasticity of substitution and , 0 is the inverse elasticity of labour
supply. Our speci…cation for household’s utility allows for GHH preferences over
hours, which eliminates wealth e¤ects from labor supply.
The composite consumption index, C
t
, is given by:
C
t
=
_
(1 ÷c)
1
C
(¸1)¸¸
1,t
÷ (c)
1
C
(¸1)¸¸
A,t
_
¸¸(¸1)
, (2)
where ¸ 0 is the elasticity of substitution between domestic and imported (foreign
goods), and 0 < c < 1 denotes the weight of imported goods in domestic con-
sumption basket. Equation (2) suggests that the expenditure share of the imported
goods in the consumption basket of households is given by 0 < c < 1. This weight,
c = (1 ÷·, depends on (1 ÷ , the relative size of the foreign economy, and
on ·, the degree of trade openness of the domestic economy. C
1,t
and C
A,t
are
consumption of domestic and foreign goods, represented by:
C
1,t
=
__
1
0
C
1,t
(,)
(A1)¸A
d,
_
A¸(A1)
; C
A,t
=
__
1
0
C
A,t
(,)
(A1)¸A
d,
_
A¸(A1)
,
where , ¸ [0, 1[ indicates the goods varieties and ` 1 is the elasticity of substitution
among goods produced within a country.
The real exchange rate 11A
t
is de…ned as 11A
t
=
S
t
1
t
1
t
, where o
t
is the
nominal exchange rate, domestic currency price of foreign currency, and 1
t
=
_
_
1
0
1
t
(,)
1A
d,
_
1¸(1A)
is the aggregate price index for foreign country’s consump-
tion goods in foreign currency. In contrast to standard small open economy models,
8
our two-country framework allows us to determine 1
t
endogenously in the foreign
economy block.
Households have access to two types of non-contingent one-period debt; one
denominated in domestic currency, 1
t
, and the other in foreign currency, 1
1
t
, with a
nominal interest rate of i
t
and i
t
w
1,t
. Due to imperfect capital mobility, households
need to pay a premium when borrowing from the rest of the world. This premium,
w
1,t
, is de…ned as a function of deviations of aggregate credit over GDP from its
steady state level; w
1,t
=
D
2
[oxp(
S
t
1
H
t+1
+S
t
1
E
t
1
t
G11
t
÷
S1
H
+S1
E
t
1G11
) ÷ 1[
2
where 1
1
t
is
entrepreneurs’ debt (see equation (11) below).
4
Households own all home production
and the importing …rms and thus are recipients of pro…ts, H
t
. Other sources of
income for the representative household are wages \
t
and new borrowing net of
interest payments on outstanding debts, both in domestic and foreign currency.
Then, the representative household’s budget constraint in period t can be written
as follows:
1
t
C
t
÷ (1 ÷i
t1
)1
t
÷ (1 ÷i
t1
)w
1,t1
o
t
1
1
t
= \
t
H
t
÷1
t+1
÷o
t
1
1
t+1
÷ H
t
. (3)
The representative household chooses the paths for ¦C
t
, H
t
, 1
t+1
, 1
1
t+1
¦
1
t=0
in
order to maximize its expected lifetime utility in (1) subject to the budget constraint
in (3).
2.2 Firms
There are three types of …rms in the model. Production …rms produce a di¤eren-
tiated …nal consumption good using both capital and labor as inputs. These …rms
engage in local currency pricing and face price adjustment costs. As a result, …nal
goods’ prices are sticky in terms of the local currency of the markets in which they
are sold. Importing …rms that sell the goods produced in the foreign economy also
have some market power and face adjustment costs when changing prices. Price
stickiness in export and import prices causes the law of one price to fail such that
exchange rate pass through is incomplete in the short run.
5
Finally, there are com-
4
Following Schmitt-Grohe and Uribe (2003), this premium is introduced for technical reasons to
maintain the stationarity in the economy’s net foreign assets. As in Schmitt-Grohe and Uribe, we
assume that the elasticity of the premium with respect to the aggregate debt is very close to zero
(
D
= 0:0075) so that the dynamics of the model are not a¤ected by this friction.
5
There is considerable empirical evidence of pricing-to-market and incomplete exchange rate
pass-through in small open economies. See, for example, Naug and Nymoen (1996) and Campa and
Goldberg (2005).
9
petitive …rms that combine investment with rented capital to produce un…nished
capital goods, that are then sold to entrepreneurs.
2.2.1 Production Firms
Each …rm produces a di¤erentiated good indexed by , ¸ [0, 1[ using the production
function:
1
t
(,) = ¹
t
·
t
(,)
1j
1
t
(,)
j
, (4)
where ¹
t
denotes labor productivity, common to all the production …rms and ·
t
(,) is
the labor input which is a composite of household, H
t
(,), and entrepreneurial labor,
H
1
t
(,); de…ned as ·
t
(,) = H
t
(,)
1
H
1
t
(,)
. 1
t
(,) denotes capital provided by
the entrepreneur, as is explored in the following subsection. Assuming that the price
of each input is taken as given, the production …rms minimize their costs subject to
(4).
Firms have some market power and they segment domestic and foreign markets
with local currency pricing, where 1
1,t
(,) and 1
A,t
(,) denote price in domestic mar-
ket (in domestic currency) and price in foreign market (in foreign currency). Firms
also face quadratic menu costs in changing prices expressed in the units of consump-
tion basket given by
i
2
(
1
i;t
())
1
i;t1
())
÷ 1)
2
for di¤erent market destinations i = H, A.
6
The combination of local currency pricing together with nominal price rigidities im-
plies that ‡uctuations in the nominal exchange rate have a smaller impact on export
prices so that exchange rate pass-through to export prices is incomplete in the short
run.
As domestic …rms are owned by domestic households, the individual …rm maxi-
mizes its expected value of future pro…ts using the household’s intertemporal rate of
substitution in consumption, given by ,
t
l
c,t
. The objective function of …rm , can
thus be written as:
1
c
1
t=0
,
t
l
c,t
1
t
[1
1,t
(,)1
1,t
(,) ÷o
t
1
A,t
(,)1
A,t
(,) ÷'C
t
1
t
(,)
÷1
t
i=1,A
w
i
2
(
1
i,t
(,)
1
i,t1
(,)
÷1)
2
[, (5)
6
This generates a gradual adjustment in the prices of goods in both markets, as suggested by
Rotemberg (1982).
10
where 1
1,t
(,) and 1
A,t
(,) represent domestic and foreign demand for the domesti-
cally produced good ,. We assume that di¤erent varieties have the same elasticities
in both markets, so that the demand for good , can be written as,
1
i,t
(,) = (
1
i,t
(,)
1
i,t
)
A
1
i,t
, for i = H, A, (6)
where 1
1,t
is the aggregate price index for goods sold in domestic market, as is de-
…ned earlier and 1
A,t
is the export price index given by 1
A,t
=[
_
1
0
1
A,t
(,)
1A
d,[
1¸(1A)
.
1
A,t
denotes the foreign aggregate export demand for domestic goods and is
determined endogenously in the foreign country block, given by
1
A,t
= C
A,t
÷C
1
A,t
÷1
A,t
÷ (c
)(
1
A,t
1
t
)
¸
[
i=1,A
w
i
2
(
1
i,t
1
i,t1
÷1)
2
÷
w
A
2
(
1
A,t
1
A,t1
÷1)
2
÷i
t
1
1
t
1
t
Q
t1
1
t
[. (7)
where C
A,t
and C
1
A,t
are import demand of households and entrepreneurs for con-
sumption, 1
A,t
denotes imported investment goods. The rest of the terms on the
right hand side are imported components of price adjustment costs and the dead-
weight loss brought by …nancial frictions.
2.2.2 Importing Firms
There is a set of monopolistically competitive importing …rms, owned by domestic
households, who buy foreign goods at prices 1
A,t
and then sell to the domestic
market at 1
A,t
with some mark-up. They are also subject to a price adjustment
cost with w
A
_ 0, the price adjustment cost parameter, analogous to the production
…rms. This implies that there is some delay between exchange rates changes and
the import price adjustments so that the short run exchange rate pass through to
import prices is also incomplete.
2.2.3 Un…nished Capital Producing Firms
Let 1
t
denote aggregate investment in period t, which is composed of domestic and
…nal goods:
1
t
=
_
(1 ÷c)
1
1
(¸1)¸¸
1,t
÷ (c)
1
1
(¸1)¸¸
A,t
_
¸¸(¸1)
, (8)
11
where the domestic and imported investment goods’ prices are assumed to be the
same as the domestic and imported consumer goods prices, 1
1,t
and 1
A,t
. The new
capital stock requires the same combination of domestic and foreign goods so that
the nominal price of a unit of investment equals the price level, 1
t
. This implies that
1
1,t
= (1 ÷c)(
1
H;t
1
t
)
¸
1
t
and 1
A,t
= (c)(
1
M;t
1
t
)
¸
1
t
.
Competitive …rms use investment as an input, 1
t
and combine it with rented
capital 1
t
to produce un…nished capital goods. Following the existing literature, we
assume that the marginal return to investment in terms of capital goods is decreasing
in the amount of investment undertaken (relative to the current capital stock) due
to the existence of adjustment costs, represented by
I
2
(
1
t
1
t
÷ c)
2
where c is the
depreciation rate. Then, the production technology of the …rms producing un…nished
capital can be represented by ¯
t
(1
t
, 1t) = [
1
t
1
t
÷
I
2
(
1
t
1
t
÷ c)
2
[1
t
which exhibits
constant returns to scale so that the un…nished capital producing …rms earn zero
pro…t in equilibrium. The stock of capital used by the …rms in the economy evolves
according to:
1
t+1
= [
1
t
1
t
÷
w
1
2
(
1
t
1
t
÷c)
2
[1
t
÷ (1 ÷c)1
t
. (9)
The optimality condition for the un…nished capital producing …rms with respect
to the choice of 1
t
yields the following nominal price of a unit of capital Q
t
:
Q
t
1
t
= [1 ÷w
1
(
1
t
1
t
÷c)[
1
. (10)
2.3 Entrepreneurs
As stated earlier, entrepreneurs are key players in our model. They transform un-
…nished capital goods and sell them to the production …rms. They …nance their
investment by using their own net worth and by borrowing from foreign lenders.
7
There is a continuum of entrepreneurs indexed by / in the interval [0,1]. Each en-
trepreneur has access to a stochastic technology in transforming 1
t+1
(/) units of
un…nished capital into .
t+1
(/)1
t+1
(/) units of …nished capital goods. The idio-
syncratic productivity .
t
(/) is assumed to be i.i.d. (across time and across …rms),
drawn from a distribution 1(.), with p.d.f of )(.) and 1(.) = 1.
8
7
See, Mishkin (1998) and Eichengreen and Hausmann (2005) for the importance of foreign cur-
rency borrowing in emerging market countries.
8
The idiosyncratic productivity is assumed to be distributed log-normally; log(!
t
(k)) s
N(
1
2
2
!
;
2
!
). This characterization is similar to that in Carlstrom and Fuerst (1997), Bernanke et
12
At the end of period t, each entrepreneur / has net worth, ·\
t
(/). The budget
constraint of the entrepreneur is de…ned as follows:
1
t
·\
t
(/) = Q
t
1
t+1
(/) ÷o
t
1
1
t+1
(/), (11)
where 1
1
t+1
denotes foreign currency denominated debt. Equation (11) simply states
that capital …nancing is divided between net worth and foreign debt. It is clear that
the entrepreneurs are exposed to exchange rate risk, that is, ‡uctuations in the
nominal exchange rate create balance sheet e¤ects in the model.
Productivity is observed by the entrepreneur, but not by the lenders who have
imperfect knowledge of the distribution of .
t+1
(/). Following Curdia (2008) we
specify the lenders perception of .
t+1
(/) as given by .
t+1
(/) = .
t+1
(/)¸
t
where ¸
t
is the misperception factor over a given interval [0,1].
9
Further, the misperception
factor, ¸
t
, is assumed to follow ln(¸
t
) = j
¸
ln(¸
t1
) ÷¸ ln(¸
t
) ÷-
¸
where j
¸
denotes
the persistence parameter, and ¸ measures the degree of …nancial contagion from
the foreign to the domestic economy. Similarly, we assume that ¸
t
, the perception
of lenders regarding the the foreign entrepreneurs’ productivity, follows an AR(1)
process with persistence parameter j
¸
. We take the origin of the …nancial shock
as a change in lenders’ perception regarding idiosyncratic productivity (-
¸
). We
assume that when there is uncertainty about the underlying distribution, lenders
take the worst case scenario as the mean of the distribution of .
t+1
(/). Appendix A
provides more details on the speci…cation of the ambiguity aversion faced by lenders
and the optimal contracting problem.
The optimal contract identi…es the capital demand of entrepreneurs, 1
t+1
(/)
and a cut o¤ value, .
t+1
(/) such that the entrepreneur will maximize their expected
return subject to the participation constraints of the lender. As shown in Appendix
A, the …rst order conditions yield:
al. (1999), Cespedes et al. (2004) and Gertler et al. (2007).
9
As in Curdia (2008), our speci…cation of the debt contract between investors and entrepreneurs
explicitly takes into account the misperception factor, which makes risk premium sensitive to the
perception of investors. Curdia (2008) adopts a max-min criteria for the misperception factor so
that during the sudden stop episode, the misperception factor is set to a constant value lower than 1.
This implies that, in his calibration, the investors’ perception is constant for 2.5 years at a pre-set,
lower value. In our case, however, we let the misperception factor to be an AR(1) process during
the sudden stop episode such that the abrupt change in the perception of investors gradually goes
back to the pre-shock level, rather than staying constant for a prolonged period.
13
1
t
[1
1
t+1
[ = 1
t
[(1 ÷i
t
)(1 ÷ 1
t+1
)[, (12)
where (1 ÷ 1
t+1
) is the external risk premium de…ned by:
1 ÷ 1
t+1
= [
.
0
(.
t+1
(/))
q(.
t+1
(/); ¸
t
).
0
(.
t+1
(/)) ÷.(.
t+1
(/))q
0
(.
t+1
(/); ¸
t
)
[1
t
¦
o
t+1
o
t
¦. (13)
where .(.) and q(.(/); ¸) are the borrowers’ and lenders’ share of the total
return, respectively. As shown in Bernanke et al. (1999), this external risk premium
depends on the leverage,
S
t
1
E
t+1
Q
t
1
t+1
, of the entrepreneur. A greater use of external
…nancing generates an incentive for entrepreneurs to take on more risky projects,
which raises the probability of default. This, in turn, will increase the external risk
premium. Therefore, any shock that has a negative impact on the entrepreneurs’
net worth increases their leverage, resulting in an upward adjustment in the external
risk premium.
We follow the existing literature in assuming that a proportion of entrepreneurs
die in each period to be replaced by new-comers. This assumption makes borrowing
constraints on debt always binding. Given that .(/) is independent of all other
shocks and identical across time and across entrepreneurs, all entrepreneurs are
identical ex-ante. Then, each entrepreneur faces the same …nancial contract speci…ed
by the cut o¤ value and the external …nance premium. This allows us to specify the
rest of the model in aggregate terms.
At the beginning of period t, the entrepreneurs collect revenues and repay their
debt contracted in period t ÷1. Denoting the fraction of entrepreneurs who survive
each period by 0, the net worth can be expressed as follows:
1
t
·\
t
= 0[1
1
t
Q
t1
1
t
.(.
t
)[ ÷\
1
t
. (14)
Equation (14) indicates that the entrepreneur’s net worth is made up of the return
on investment and the entrepreneurial wage income. Given that the borrower’s and
the lender’s share of total return should add up to .(.
t
)÷q(.
t
, ¸
t
) = 1÷i
t
(where i
t
is the cost of monitoring, a deadweight loss associated with …nancial frictions) and
by using the participation constraint of the lenders, we can rewrite the net worth of
the entrepreneur as:
14
1
t
·\
t
= 0[1
1
t
Q
t1
1
t
(1 ÷i
t
) ÷(1 ÷i
t1
)o
t
1
1
t
[ ÷\
1
t
. (15)
It is clear from (15) that unanticipated changes in the nominal exchange rate
increase the debt burden of the entrepreneur, and therefore decrease its net worth.
This, in turn, increases the leverage of the entrepreneur and raises the external risk
premium, implying a higher cost of …nancing. This is an additional mechanism that
magni…es the role of the …nancial accelerator in the economy through transmitting
‡uctuations in the nominal exchange rate to the balance sheets of entrepreneurs.
The entrepreneurs leaving the scene at time t consume their return on capital.
The consumption of the exiting entrepreneurs, C
1
t
, can then be written as:
10
1
t
C
1
t
= (1 ÷0)[1
1
t
Q
t1
1
t
(1 ÷i
t
) ÷(1 ÷i
t1
)o
t
1
1
t
[. (16)
Because of investment adjustment costs and incomplete capital depreciation,
entrepreneurs’ return on capital, 1
1
t
, is not identical to the rental rate of capital,
1
t
. The entrepreneurs’ return on capital is the sum of the rental rate on new and
used capital, and the value of the non-depreciated capital stock, after the adjustment
for the ‡uctuations in the asset prices
_
Q
t+1
Q
t
_
:
1
t
1
1
t+1
= 1
t
[
1
t+1
Q
t
÷
Q
t+1
Q
t
¦(1 ÷c) ÷ w
1
(
1
t+1
1
t+1
÷c)
1
t+1
1
t+1
÷
w
1
2
(
1
t+1
1
t+1
÷c)
2
¦[. (17)
2.4 Monetary Policy
We adopt a standard formulation for the structure of monetary policy framework
in the baseline calibration. We assume that the interest rate rule is of the following
form:
1 ÷i
t
= (1 ÷i) (¬
t
)
c
(1
t
,1 )
c
Y
, (18)
where i and 1 denote the steady-state level of nominal interest rate and output,
and ¬
t
is the CPI in‡ation.
Under the …xed exchange rate regime, the monetary policy rule is given by:
o
t
o
t1
= 1. (19)
10
It is assumed that the entrepreneurs consume an identical mix of domestic and foreign goods
in their consumption basket as is given by the composite consumption index in equation (2).
15
2.5 General Equilibrium and Balance of Payments Dynamics
Market clearing in the …nal good sector requires that total domestic output be equal
to domestic consumption, domestic investment and exports to the rest of the world.
Also, given that frictions such as adjustment and monitoring costs are expressed in
terms of the …nal composite good, part of the output is taken up with the price
adjustment costs for …nal consumption goods as well as those for imported and
exported goods and the monitoring costs. Thus the overall resource constraint faced
by the domestic economy can be written as:
1
t
= 1
1,t
÷1
A,t
, (20)
where
1
1,t
= C
1,t
÷C
1
1,t
÷1
1,t
÷ (1 ÷c)(
1
1,t
1
t
)
¸
[
i=1,A
w
i
2
(
1
i,t
1
i,t1
÷1)
2
÷
w
A
2
(
1
A,t
1
A,t1
÷1)
2
÷i
t
1
1
t
1
t
Q
t1
1
t
[, (21)
where 1
A,t
is the foreign demand for domestic goods (see equation (7)), C
1,t
and
C
1
1,t
are the household and entrepreneur’s consumption demand for domestic goods,
and 1
1,t
is the domestic investment goods used by the un…nished capital producing
…rms. In (21), w
1
, w
A
and w
A
denote price adjustment costs for domestic, exported
and the imported good, respectively and i
t
is the cost of monitoring for the lenders
that is passed on to the domestic economy through the external …nance premium.
Equilibrium in the labor market requires that:
·
t
= H
t
1
, (22)
where H
t
1
is the labor demand for non-entrepreneurial labour.
Substituting (20) and the pro…ts of both the …nal good producing and the im-
porting …rms into the budget constraints of the households and the entrepreneurs
yields the following balance of payments condition after aggregation:
o
t
1
A,t
1
A,t
÷1
A,t
1
A,t
= o
t
(1 ÷i
t1
)(1
1
t
w
1,t1
÷1
1
t
) ÷o
t
(1
1
t+1
÷1
1
t+1
), (23)
where 1
A,t
is the price of imports and 1
A,t
is the aggregate import demand. The
…rst and the second terms on the left are exports and imports, and on the right is
16
simply the change in the net foreign asset position aggregated over households and
entrepreneurs.
In Appendix B, we present the equilibrium model equations for both domestic
and the foreign economy.
3 Solution and Parametrization
We …rst transform the model to reach a stationary representation where a steady
state exists. The model is then solved numerically up to a second order approxima-
tion using Sims (2005).
11
3.1 Consumption, Production and Monetary Policy
We maintain that the structure of consumption, production and monetary policy
is identical in the two economies and thus use the same parametrization with the
exception of the size parameter, :, which is set to 0.1 for domestic economy. We set
the discount factor, , at 0.00, implying a riskless annual return of approximately
4 per cent in the steady state (time is measured in quarters). The inverse of the
elasticity of intertemporal substitution is taken as o = 1, which corresponds to log
utility. The inverse of the elasticity of labour supply , is set to 2, which implies that
1,2 of time is spent working. In the baseline case, we set the degree of openness,
·, to be 0.8ò which is within the range of the values used in the literature.
12
The
share of capital in production, j, is taken to be 0.8ò, consistent with other studies.
13
Following Devereux et al. (2006), the elasticity of substitution between di¤erentiated
goods of the same origin, `, is taken to be 11, implying a ‡exible price equilibrium
mark-up of 1.1, and price adjustment cost is assumed to be 120 for all sectors. The
quarterly depreciation rate c is taken to be 0.02ò, a conventional value used in the
literature. Similar to Gertler et al. (2007), we set the share of entrepreneurs’ labour,
\, at 0.01, implying that 1 per cent of the total wage bill goes to the entrepreneurs.
Regarding monetary policy, we use the original Taylor estimates and set c
¬
= 1.ò
and c
Y
= 0.ò in the baseline calibration. We assume j
¸
to be 0.ò, so that it takes
11
The non-stochastic steady state of the model is solved numerically in MATLAB, and then the
second order approximation of the model and the stochastic simulations are computed using Michel
Juillard’s software Dynare. Details of the computation of the non-stochastic steady state and the
stationary model equations are available upon request.
12
The values for openness in the existing literature range between 0.25 (Cook, 2004; Elekdag and
Tchakarov, 2007) and 0.5 (Gertler et al., 2007). We chose to set a middle value of this range, but
we conduct a sensitivity analysis regarding the value of openness in Section 4.2.2.
13
See, for example, Cespedes et al. (2004) and Elekdag and Tchakarov (2007).
17
0 quarters for the shock to die away. Table 1 summarizes the parametrization of
the model for consumption, production, and monetary policy used in the baseline
calibration.
14
3.2 Entrepreneurs
The parameter values for the entrepreneurial sector in the domestic and foreign
economies are set to re‡ect their de…ning characteristics and are listed in Table 2.
We set the steady state leverage ratio, ¸, the ratio of debt to net worth, and the
value of quarterly external risk premium in the domestic economy, 1, respectively
at 0.8 and 200 basis points, re‡ecting the historical average of emerging market
economies within the last decade.
15
The monitoring cost parameter, j, is taken as
0.2 for the domestic economy as in Devereux et al. (2006). These parameter values
imply a survival rate, 0, of approximately 00.88 per cent in the domestic economy.
For the foreign economy, we closely follow Bernanke et al. (1999). The leverage
ratio is set to 0.ò. The higher leverage ratio for entrepreneurs in foreign economy re-
‡ects the fact that advanced economies have deeper and more sophisticated …nancial
markets, and therefore better …nancing opportunities, leading to a higher economy-
wide leverage. Moreover, having experienced dramatic …nancial crises at the turn
of the century, emerging market economies have been more vigilant towards lend-
ing activities through tighter …nancial regulation, which in many cases has helped
to contain leverage ratios in these economies.
16
The risk spread of 2 per cent in
the steady state is reported for the US economy so we set a quarterly external risk
premium, 1
t
, of 0.00ò. The cost of monitoring in the foreign economy, denoted by
j
, is taken to be 0.12, lower than the domestic economy re‡ecting the presence of
better institutions. Given these parameter values, the implied survival rate is 00.66
per cent in the foreign economy.
14
We carry out a series of sensitivity analyses in order to asses the robustness of our results under
the benchmark calibration. In Section 4, we report the results regarding the degree of …nancial
contagion, trade openness, and monetary policy parameters. We also conduct sensitivity checks
with regard to the degree of exchange rate pass-through, and …nd that our main results are not
sensitive to changes in these parameters. These are not reported due to space limitations.
15
These …gures are decade averages for emerging Americas, emerging Asia, and emerging Europe
between 2000-2010. Worldscope data (debt as a percentage of assets - data item WS 08236) are used
for the leverage ratio. External risk premium is calculated as the di¤erence between the lending and
the policy rate for emerging market countries, where available, using data from Haver Analytics for
the same time period. Variations in these parameters a¤ect our results only quantitatively, but not
qualitatively.
16
See Kalemli-Ozcan et al. (2011) for stylized facts on bank and …rm leverage for 2000- 2009 for
both advanced and emerging economies.
18
4 Financial Crisis and the Domestic Economy
4.1 Model Dynamics
In this section, we …rst report a number of business cycle statistics for the model and
compare them with those observed in the data from both emerging and advanced
economies. In our analysis, we focus only on the response of macroeconomic aggre-
gates to two types of …nancial shock: one originating in the domestic economy and
the other in the foreign (global) economy. In both cases, the source of the …nancial
shock is the change in lenders’ perception of the entrepreneurs’ productivity. Given
the speci…c nature of the source of the shock in our analysis, one would not expect
that the model to match the data in all dimensions. Nonetheless, in what follows we
compare movements and comovements of some key variables in the data and in the
model to generate con…dence in the model’s ability to correctly capture dynamics
and in the proposed calibration of the parameters values.
Table 3 presents business cycle statistics coming from the data as well as the
simulation results under our benchmark parametrization. In the top panel of Table 3,
standard deviations of output, consumption, investment, current account, in‡ation,
interest rate and exchange rate for Argentina, Brazil, Korea, Mexico and Philippines
are listed, while the middle panel reports standard deviations of this set of variables
relative to the standard deviation of output. In both panels, the simulation results
are reported under two scenarios; a domestic shock and a global shock. A glance at
the top two panels of Table 3 suggests that in most cases moments from the data
are well-matched by moments from the model except in the case of in‡ation and the
exchange rate. This is particularly the case for volatility of output, consumption,
current account as well as of interest rate in terms of both individual volatilities and
volatilities relative to output.
The correlations of consumption, investment and current account and autocor-
relations of these variables are presented in the bottom panel of Table 3. Similar to
above, it can be seen that, model results match the business cycle statistics quite
well for the same group of countries.
Table 4 repeats the same exercise for an advanced economies sample which we
represent by EU and the US. Similar to that in Table 3, Table 4 suggests that our
simulation results match the business cycle statistics pretty well with the exception
of current account and output correlations. In contrast to the case with emerging
19
market countries, volatility of in‡ation in the model is also quite close to that in the
data for advanced economies.
Finally, Table 5 presents cross-country correlations of output, consumption and
investment from our model with those from the same sample of emerging market
countries as above for two separate sample periods; 1999-2011 and 2007-2011 for the
global crisis period. As can be seen from Table 5, our simulation results match the
cross-correlations in the data for this group of countries quite well especially over
the global …nancial crisis period which is our main focus in this paper.
We now turn to examining the responses of both the domestic and the foreign
country in response to two types of …nancial shock by assessing the impulse responses
in our two-country model.
4.2 Financial Crisis Originating in the Domestic Economy
We …rst investigate the case of a domestic …nancial crisis by tracing how an unan-
ticipated (temporary) shock to investors’ perception of entrepreneurs’ productivity
is transmitted to the rest of the economy. Such a perception shock leads to a re-
versal of capital ‡ows out of the domestic country, which we refer to as the sudden
stop.
17
The response of the domestic economy to the sudden stop is presented in
Fig.1. When the investors’ perception about the distribution of the entrepreneurs’
productivity changes, lending to domestic entrepreneurs becomes more risky, lead-
ing to a rise in the external risk premium on impact. As the cost of borrowing rises,
entrepreneurs reduce their use of external …nancing by undertaking fewer projects.
This decline in leverage causes a downward adjustment in the risk premium, mitigat-
ing the initial impact of a sudden stop to some extent. Lower borrowing, however,
decreases the future supply of capital and hence brings about a decrease in invest-
ment in the economy. The fall in the in‡ow of capital also lowers the demand for
domestic currency, leading to its depreciation. Since the entrepreneurs’ borrowing is
denominated in foreign currency, this unanticipated change in the exchange rate also
creates balance sheet e¤ects through a rise in the real debt burden. The outcome is
lower investment and output in the economy following the sudden stop, in line with
the experience of several emerging market countries during the 1990s.
Although the rise in the nominal exchange rate puts an upward pressure on the
17
The magnitude of the shock is 1 percent, which causes out‡ow of capital by about 1.5 percent
of GDP on impact.
20
CPI based in‡ation, the decrease in the domestic price level more than o¤sets this
e¤ect, bringing about a fall in the CPI. In spite of this lower price level, however,
aggregate demand falls due to lower investment and output, resulting in lower labor
demand and thereby lower real wages.
There is an additional channel through which the e¤ect of the shock is trans-
mitted to the rest of the economy, working through the export demand. Following
the depreciation of the domestic currency, the foreign demand for domestic goods
increases. On the other hand, imports decline on account of both income and ex-
change rate e¤ects. The trade balance improves, but this e¤ect is not strong enough
to o¤set the decline in domestic demand in our simulations, and hence output con-
tracts. In practice, the export channel is generally highly e¤ective for countries that
are hit by …nancial crises and experience a sizable loss of value of their currencies.
For instance, most East Asian countries bene…tted from signi…cant improvements
in their exports following the 1997 Asian crisis, which has been widely viewed to be
an important factor in their swift recoveries (see, for example, Bleaney, 2005). It is
important to note, however, that this favorable impact of export demand on out-
put recovery not only disappears but starts to work in the opposite direction when
the …nancial crisis originates in the global economy, as is explored in the following
section.
4.3 Financial Crisis in the Global Economy
In this section, we explore the channels through which a …nancial shock originat-
ing in the foreign country is transmitted to the domestic economy and the role of
trade openness in shaping the domestic economy’s responses to the …nancial shock.
The perception shock is now faced by the foreign entrepreneurs, which we take to
represent the case of a global …nancial crisis.
4.3.1 The impact of the …nancial shock on the foreign economy
Fig.2 presents responses of foreign macroeconomic variables to a 1 per cent negative
shock to the perception of investors regarding the productivity of foreign entrepre-
neurs. As is seen from Fig.2, the impact of the shock on the foreign economy is
similar to the sudden stop experience of the domestic economy described above.
Following the shock, lending to foreign entrepreneurs become risky, leading to a rise
in risk premium which, through reducing borrowing and thus supply of capital leads
21
to lower investment and output in the foreign economy. It is also clear from Fig.2
that the falls in investment, output and consumption are smaller in the foreign econ-
omy as compared with the domestic one, given that …nancial frictions are greater
in the domestic economy - domestic entrepreneurs borrow in foreign currency and
monitoring domestic entrepreneurs are more costly.
We now turn to how the foreign (global) …nancial shock is transmitted onto the
domestic economy.
4.3.2 The transmission of the foreign shock onto the domestic economy
The global …nancial shock is transmitted to the domestic economy through three
separate channels involving both …nancial and trade linkages. The …rst of these
is through …nancial spillovers. We postulate that investors’ perception regarding
the true distribution of entrepreneurs’ productivity in the foreign economy and the
domestic economy are inherently related. This is based on the notion that investors
optimally choose the scale and the terms of credit they extend to borrowers in a
forward looking manner. For instance, when faced with credit tightening in the
global economy, investors can anticipate ex-ante that this will be transmitted to
the domestic economy through real and …nancial cross-country linkages, implying
an unfavorable change in their perceptions of the domestic entrepreneurs today.
Also, some asset market linkages such as herding behavior only or mainly exist
during times of crisis, a phenomenon commonly referred to as "pure contagion"
(see, for example, Kaminsky and Reinhart, 2000 and Moser, 2003). Motivated
by these observations, we maintain that one channel through which the …nancial
shock spills over onto the domestic economy is through an unfavourable shift in
investors’ perception of the domestic entrepreneurs. In what follows, we refer to this
mechanism as the …nancial contagion channel.
The second channel through which the global …nancial shock is propagated to
the domestic economy is the export channel. The …nancial crisis in the foreign econ-
omy reduces output and thereby export demand in the foreign economy and thus
net exports of the domestic economy. There is also a third channel of propagation
of the global …nancial shock, which works through a substitution e¤ect. The un-
favourable change in investors’ perception of the foreign entrepreneurs’ productivity
induces investors to look for alternative investment opportunities, leading to an in-
crease in capital in‡ows into the domestic economy, which would, partly o¤set the
22
impact of the …rst two channels. Clearly, the lower the …nancial contagion from the
foreign economy to the domestic economy, the greater the impact of the favourable
substitution e¤ect. We now turn to exploring these channels separately.
Fig.3 and 4 present the domestic economy’s response to a global …nancial shock,
with and without …nancial contagion, respectively.
18
Fig.3 features two alternative
scenarios; full contagion, ¸ = 1 and partial contagion, ¸ = 0.ò. The deterioration in
investors’ perception in the domestic economy following that in the foreign economy
raises the external risk premium, setting in motion the process described above.
Responses presented in Fig.3 reveal that the impact of the global …nancial shock on
the domestic economy is larger than that of a domestic one. Although the change
in the perception of foreign investors about the state of the domestic economy is
identical under the two scenarios, falls in capital, investment and output are greater
with the global …nancial shock and the fall in investment is twice the size. Similarly,
the decrease in foreign borrowing is larger than that of the previous case, leading to
a sharper depreciation of the domestic currency. Likewise, changes in in‡ation and
asset prices are much more pronounced under the global …nancial shock than with
the sudden stop of domestic origin.
One main di¤erence between this case of a foreign …nancial shock and the domes-
tically originated one, as explored in the previous section, is in the way the export
channel works. As is seen above, when the economy is hit by a …nancial crisis of
domestic origin, the depreciation of the currency brings about an improvement in
net exports, which partly o¤sets the initial decline in output. In contrast, when
the …nancial shock is originated in the foreign economy with partial …nancial con-
tagion (¸ = 0.ò), the export channel works in the opposite direction, adding to the
unfavourable impact that works through …nancial contagion, hence worsening the
overall fall in output. This is because the global …nancial crisis reduces net worth,
capital, investment and output in the foreign economy, and therefore decreases the
domestic economy’s exports and therefore output. As output in the foreign econ-
omy returns to its previous level, the export demand improves although the trade
18
Our analysis of the ampli…cation of the foreign …nancial shock onto the domestic economy, as
portrayed in Figures 3-8, is based on a 0.75 % negative shock to the perception of investors regarding
the productivity of foreign enterpreneurs. This shock generates an increase in the risk premium by
about 1 percent quarterly on impact in the domestic economy as in the case of a …nancial crisis of
domestic origin so that the responses are comparable to those in Fig. 1.
23
balance continues to deteriorate owing to the rise in imports following the recovery
in domestic output. When there is full-contagion from the global to the domestic
economy, however, the sharp depreciation of the exchange rate delivers an increase
in exports, as is seen in Fig.3. Yet, the worsening of the risk premium and of entre-
preneurs’ balance sheets is much larger and thus the decline in output is almost as
twice as the decline under partial …nancial contagion.
In order to explore the non-contagion channels more explicitly we now have
a further experiment where we shut o¤ the contagion channel and thus focus on
the export and substitution channels. Fig.4 denotes the changes in the domestic
economy following the …nancial crisis in the foreign economy, without contagion
in the form of a further …nancial shock in the domestic economy. There are now
two opposing e¤ects of the foreign economy’s …nancial crisis on the macroeconomic
outcomes in the domestic economy. The …rst is through the reduced net exports due
to the contraction in the foreign economy following the …nancial crisis. The second,
in contrast, is a favourable impact leading to an increase in capital in‡ows. This is
because lending to domestic entrepreneurs is now perceived to be more pro…table
as the domestic economy is in a better …nancial position.
19
The overall response is
determined by the balance of the two e¤ects. As is seen from Fig.4, in the absence of
…nancial contagion, the increase in risk premium is considerably lower than that in
the above two cases and that of the foreign economy. As a result, capital in‡ows, net
worth and investment all increase following the …nancial shock, and thus contraction
in output in the domestic economy is smaller.
The heterogeneous nature of the pro…le of crisis across countries, as demonstrated
by Fig.2 and Fig.3, has also been highlighted by Milesi-Ferretti and Tille (2011) in
their analysis of capital ‡ows during the 2007-2009 global …nancial crisis episode.
A glance at the capital ‡ows diagrams, suggests that following the …nancial shock
there is a sharp drop in capital ‡ows into the foreign economy, which remain below
steady-state levels for extended periods of time. On the other hand, capital ‡ows
to the domestic country recover relatively quickly toward the pre-shock levels, as
is seen in Fig.3. Indeed, Milesi-Ferretti and Tille (2011) document that capital
in‡ows to advanced countries exhibited a sharp reversal in the last quarter of 2008
and remained well-below pre-crisis levels until the end of their sample period, 2009.
19
IMF (2011) shows that net capital in‡ows to emerging economies are highly correlated with
global …nancing conditions, with global interest rates and risk aversion playing an important role.
24
They show that, in contrast, the pull-back of capital ‡ows from emerging market
countries had been much shorter lived, and in‡ows to these countries in 2009 reached
magnitudes broadly comparable to the pre-crisis levels.
4.3.3 The role of trade openness
Having established that a global …nancial shock is transmitted to the domestic econ-
omy through both the trade and …nancial channels, it follows that the extent of the
domestic response to a global …nancial tightening will be determined by, among
other factors, the openness of the domestic economy. We now turn to exploring
the role of trade openness in the propagation of the foreign …nancial shock more
explicitly.
Fig.5 and Fig.6 illustrate the domestic responses to a foreign …nancial shock
under varying degrees of trade integration between the domestic economy and the
foreign economy with and without …nancial contagion, respectively. In our simu-
lations the degree of trade integration is measured by ·, which, together with the
size of the economy, :, determines the share of imported good items in the domes-
tic consumption basket. The pro…le of the domestic economy in Fig.5 exhibits the
important role played by the degree of trade openness in the ampli…cation of the
global …nancial shock. The greater the trade integration between the two countries,
the more signi…cant is the impact of the global …nancial crisis on the domestic econ-
omy. This holds regardless of the degree of …nancial contagion as is seen from the
responses in Fig.5 and Fig.6. The relationship between a country’s openness to trade
and its vulnerability to sudden stops has already been the focus of an extensive lit-
erature (see, for example Calvo et al. 2006 and Martin and Rey, 2006). A common
…nding in this literature has been that openness makes countries less vulnerable to
crises. In contrast, we …nd that when the …nancial shock originates in the rest of
the world - when the crisis is a global one - the more open an economy, the greater
the unfavorable consequences of the …nancial crisis for the domestic economy. In-
deed, among the countries that have experienced largest falls in economic activity
during the initial stage of the recent …nancial crisis have been Singapore, Taiwan
and Turkey, all of which are highly open economies.
20
20
The fall in output in the …rst quarter of 2009 as compared with a year earlier was 10.1, 10.2
and 13.8 per cent for Singapore, Taiwan and Turkey, respectively. Similarly, Germany and Japan,
that are among the most open of mature economies, contracted by 6.9 and 8.8 per cent, respectively
over the same period (The Economist, July 4th, 2009).
25
4.4 Monetary policy options and welfare analysis
What is the role of monetary policy in determining the domestic economy’s response
to the global …nancial shock? In what follows, we attempt to answer this question
by exploring the e¤ects of the …nancial shock on the domestic economy under two
separate monetary policy regimes. These are a …xed exchange rate regime and a
‡exible in‡ation targeting (IT) regime where monetary authority responds to both
in‡ation and the output gap.
Before we start assessing the implications of the …xed exchange rate regime
versus a ‡exible one (as under IT), it is important to point out the consequences of
nominal exchange rate changes in our model. A rise in the nominal exchange rate
improves net exports while creating a contractionary balance sheet e¤ect for the
…nancial (entrepreneurial) sector given …nancial frictions and liability dollarization
in the domestic economy. As leverage and the risk premium increases, the ability
of entrepreneurs to borrow and produce capital declines, reducing investment and
output in equilibrium. Overall, which monetary regime does better in insulating
the real economic activity from …nancial shocks will be determined by the relative
magnitude of these two e¤ects.
Fig.7 and Fig.8 illustrate the responses under the two regimes with and without
…nancial contagion, respectively and o¤er a number of interesting insights. As is
seen from Fig.7, in the presence of …nancial contagion, the global …nancial shock
brings about a sharp rise in the external risk premium, a steep fall in capital in‡ows
and thus in investment under both regimes. However, the impact of the shock on
output is more pronounced under the …xed exchange rate regime than under the
IT regime. In our calibration, the decline in exports under the …xed exchange rate
regime is large enough to o¤set the favourable impact of stable exchange rate on the
entrepreneurs’ balance sheets leading to a sharper fall in output in response to the
…nancial shock relative to under IT.
Interestingly, the absence of …nancial contagion alters the ranking of the two
regimes in insulating the domestic economy from …nancial shocks originating in the
foreign economy. In the absence of a direct …nancial channel through which foreign
…nancial shocks are transmitted onto the domestic economy, output in the domestic
economy rises under the …xed exchange rate regime, as is depicted by Fig.8. Since
the nominal exchange rate is constant under this scenario, the increase in the risk
26
premium is lower than under the IT regime. Lower external risk premia, in turn,
lead to a rise in the net worth of …rms, and thus in investment and output without
increasing entrepreneurs’ debt. The real appreciation of the exchange rate is greater
under the IT regime, so is the decline in exports.
Overall, this experiment indicates that a global …nancial shock is likely to be
expansionary under the …xed exchange rate regime when there is no direct …nancial
spillovers from the global economy. However, in order to reach normative conclusions
regarding the choice of the monetary policy framework one would need to carry out
a rigorous welfare analysis, which we present in the next section.
Welfare analysis We now turn to an assessment of alternative monetary pol-
icy frameworks based on the evaluation of welfare, taking the utility function of
consumers as the objective. Following Faia and Monacelli (2007), and Gertler and
Karadi (2010), we start by expressing the household utility function in recursive
form:
\
t
= l(C
t
, H
t
) ÷,1
t
\
t+1
(24)
where \
t
= E
0
1
t=0
,
t
l(C
t
, H
t
) denotes the welfare function. We take a second order
approximation of \
t
around the deterministic steady state. Using the second order
solution of the model, we then calculate \
t
under both the IT and the …xed exchange
rate regimes.
21
We also calculate the fraction of consumption, w, required to equate
welfare under the …xed exchange rate \
1A
t
, to the one under the IT regime, \
1T
t
,
as a welfare measure. Under our speci…cation of utility function, and o = 1 (log
utility), w can be expressed as
w = (
(oxp¦(\
cjt
t
÷\
t
)(1 ÷,)¦(C ÷
1
1+'
1+,
)) ÷
1
1+'
1+,
C
) ÷1 (25)
where C and H denote the steady state values of the corresponding variables.
22
Table 6 presents the calculated values of welfare and w for the two monetary
policy regimes under a number of di¤erent scenarios.
As is seen from Table 6, when the source of the …nancial shock is domestic,
the IT regime is welfare-improving, as is presented in the upper part of the table
21
See Scmitt-Grohe and Uribe (2007) for an in depth discussion on using second order approxi-
mation to utility function to calculate the welfare.
22
This is based on the assumption that the enterpreneurial share of consumption is negligible
(see, Bernanke et al. 1999).
27
and the welfare gains associated with the IT regime relative to the …xed exchange
rate regime is about 0.03 percent of steady state consumption (…rst and second
columns of Table 6). The same holds in the face of foreign …nancial shocks when
…nancial spillovers are strong (¸ = 1 and ¸ = 0.ò), as the expansionary e¤ects of the
depreciation under IT o¤set the contractionary balance sheet e¤ects generated by
the entrepreneurs’ foreign currency borrowing, as explained above. Overall, in our
calibration, …xed exchange rate regime increases welfare losses associated with the
global crisis when there are strong …nancial spillovers between the two countries.
In contrast, when there is no contagion the …xed exchange rate regime is welfare
superior (negative w) by about 0.01 per cent of consumption, which derives from two
factors. On the real side, the smaller appreciation in the real exchange rate under
the …xed exchange rate regime brings about a less dramatic decline in exports, hence
aggregate demand is less a¤ected. On the …nancial side, lower external risk premia
under the …xed exchange rate regime leads to a rise in the net worth of …rms, and
thus in investment and output.
As explained above, the ranking of the …xed exchange rate regime versus the
IT regime is determined by the relative importance of the export channel versus
the balance sheet channel. As potential factors underlying the importance of these
channels, we consider alternative values of openness and leverage to see how the
welfare comparison between the two regimes is a¤ected. Our simulations reveal
that as openness increases, the welfare cost associated with having a …xed exchange
rate regime increases (third and fourth columns in Table 6). This is because the
export channel is more e¤ective in more open economies, making the absence of
depreciation more costly. We also calibrate the model with an alternative level of
leverage and show that the ranking of the regimes is overturned with higher leverage.
For example, when the leverage is higher, ¸ = 0.ò, the …xed exchange rate regime
results in higher welfare in the face of a global …nancial shock regardless of the level
of …nancial contagion, as is seen in the last two columns of Table 6. When the initial
leverage is high, the …nancial accelerator e¤ect ampli…es, so is the impact of the
crisis, leading to the …xed exchange rate regime delivering a better welfare outcome
(negative w), despite losses in domestic demand.
28
5 Conclusions
This paper has developed a two-country DSGE model to investigate the transmis-
sion of a global …nancial crisis to a small open emerging economy. Our framework
fully speci…es both trade and …nancial linkages between the countries to study some
important aspects of the recent global …nancial crisis experience.
We …nd that a small, open economy facing a sudden stop of capital in‡ows aris-
ing from …nancial distress in the global economy is likely to face a more prolonged
crisis than sudden stop episodes of domestic origin. This is largely attributable to
an important source of di¢culty in responding to a global …nancial shock - the in-
ability of countries to export their way out of a crisis due to the slump in world
consumer demand initiated by the global …nancial distress. In contrast, when the
…nancial shock is of domestic origin, the domestic economy bene…ts from the de-
preciation of its currency and the resulting current account reversal, which at least
partly compensates for the fall in economic activity. This bene…cial export channel
disappears and indeed works in the opposite direction when the rest of the world
also faces an unfavorable …nancial disturbance. The resulting contraction in output
in the foreign economy is transmitted to the domestic economy through a fall in ex-
port demand, further reducing aggregate demand for home produced goods. This,
in turn, is likely to increase the duration and the severity of crises for both countries
in question, as mutual reductions in export demand set in motion a vicious circle,
even in the absence of any protectionist policies.
Given the nature of our framework with both …nancial and trade linkages, our
results have clear implications for the degree of both the …nancial and the trade
integration on the ampli…cation of a global …nancial crisis for the domestic econ-
omy. We …nd that the lower the …nancial contagion from the global economy, the
less signi…cant the impact of a global …nancial shock on the domestic economy, as
expected. Low contagion enables the domestic country to recover from the global
…nancial shock rapidly on the back of capital ‡ows ‡eeing from the foreign towards
the domestic economy. We argue that this …nding may shed some light on why the
recent …nancial crisis has been relatively short-lived for a number of emerging market
countries, especially those with limited …nancial exposure to the …nancial distress
in the global economy. As capital in‡ows surged, these economies have experienced
rapid recovery in credit, asset prices, investment, and output. Regarding the role
29
of trade integration, in contrast to the existing literature, we …nd that the greater
a country’s trade integration with the rest of the world, the greater the response of
its macroeconomic aggregates to a sudden stop of capital ‡ows. This is because the
more open an economy the greater the impact of the fall in export demand from the
rest of the world, as was experienced by countries such as Singapore, Taiwan and
Turkey, all highly open economies registering among the highest output contractions
in 2009.
Our results also suggest that the degree of …nancial contagion, the degree of trade
openness and the scale of foreign currency denominated debt are key determinants
of how monetary policy regimes in‡uence the propagation of the global …nancial
shock. In the absence of …nancial spillovers between countries, risk premia are
lower, net worth is higher so are investment, consumption and output under the
…xed exchange rate regime, a result which is overturned in the presence of …nancial
contagion. Moreover, our welfare calculations reveal that the greater the openness
of an economy, the greater the costs associated with the …xed exchange rate regime.
We …nd that, in contrast, at high levels of liability dollarization the …xed exchange
rate regime improves welfare irrespective of the degree of …nancial spillovers, a result
deriving from the balance sheet channel dominating the export channel.
30
References
[1] Bernanke, B. S., M. Gertler, and S. Gilchrist (1999). The …nancial accelerator
in a quantitative business cycle framework. In J. B. Taylor and M. Woodford
(Eds.), Handbook of Macroeconomics, Volume 1C, Chapter 21, 1341–93. Ams-
terdam: North-Holland.
[2] Bleaney, M. (2005). The aftermath of a currency collapse: how di¤erent are
emerging markets? World Economy, 28(1), 79-89.
[3] Calvo, G., A. Izquierdo, and R. Loo-Kung (2006). Relative price volatility under
sudden stops: the relevance of balance sheet e¤ects. Journal of International
Economics, 69(1), 231-254.
[4] Campa, J. and L. Goldberg (2005). Exchange rate pass-through into import
prices. Review of Economics and Statistics, 87(4), 679-690.
[5] Carlstrom, C.T. and T. S. Fuerst (1997). Agency costs, net worth, and busi-
ness cycle ‡uctuations: a computable general equilibrium analysis. American
Economic Review 87, 893–910.
[6] Cespedes, L. F., R. Chang, and A. Velasco (2004). Balance sheets and exchange
rate policy. American Economic Review, 94, 1183–1193.
[7] Christiano, L. J., C. Gust, and J. Roldos (2004). Monetary policy in a …nancial
crisis. Journal of Economic Theory 119(1), 64–103.
[8] Cook, D. (2004). Monetary policy in emerging markets: Can liability dollar-
ization explain contractionary devaluations?Journal of Monetary Economics,
51(6), 1155-1181.
[9] Curdia, V. (2008). Optimal monetary policy under sudden stops. Sta¤ Report
no. 323, Federal Reserve Bank of New York.
[10] Dedola, L. and G. Lombardo (2012). Financial frictions, …nancial integration
and the international propagation of shocks. Economic Policy, 27(70), 319-359.
[11] Devereux, M. B., P. R. Lane, and J. Xu (2006). Exchange rates and monetary
policy in emerging market economies. The Economic Journal, 116, 478–506.
31
[12] Devereux, M. B. and J. Yetman (2010). Leverage constraints and the inter-
national transmission of shocks. Journal of Money, Credit and Banking, 42,
71-105.
[13] Devereux, M. B. and A. Sutherland (2011). Evaluating international …nancial
integration under leverage constraints. European Economic Review 55(3), 427-
442.
[14] Eichengreen, B. and R. Hausmann (2005). Other People’s Money: Debt De-
nomination and Financial Instability in Emerging Market Economies, The Uni-
versity of Chicago Press.
[15] Elekdag, S. and I. Tchakarov (2007). Balance sheets, exchange rate policy, and
welfare. Journal of Economic Dynamics & Control, 31, 3986–4015.
[16] Enders, Z., R. Kollmann and G. Muller (2011). Global banking and interna-
tional business cycles. European Economic Review, 55, 407-426.
[17] Faia, E. and T. Monacelli (2007). Optimal interest rate rules, asset prices and
credit frictions. Journal of Economic Dynamics & Control, 31, 3228-3254.
[18] Gertler, M., S. Gilchrist, and F. Natalucci (2007). External constraints on mone-
tary policy and the …nancial accelerator. Journal of Money, Credit and Banking,
39, 295–330.
[19] Gertler, M. and P. Karadi (2010). A model of unconventional monetary policy.
Journal of Monetary Economics, 58, 17-34.
[20] Gilchrist, S. (2004). Financial markets and …nancial leverage in a two-country
world economy. In Banking Market Structure and Monetary Policy. Central
Bank of Chile.
[21] Greenwood, J., Z. Hercowitz, and G. Hu¤man (1988). Investment, capacity
utilization, and the real business cycles. American Economic Review, 78 402-
417.
[22] International Monetary Fund (2011). World Economic Outlook: Tensions from
the Two-Speed Recovery: Unemployment, Commodities, and Capital Flows.
Washington, April.
32
[23] Kalemli-Ozcan,S., B. Sorensen and S. Yesiltas (2011). Leverage across …rms,
banks and countries. Paper prepared for NBER-MIT Global Financial Crisis
Conference, Bretton Woods, NH.
[24] Kaminsky, G.L. and C.M. Reinhart (2000). On crisis, contagion and confusion.
Journal of International Economics, 51, 145-168.
[25] Martin, P. and H. Rey (2006). Globalization and emerging markets: with or
without crash? American Economic Review, 96(5), 1631-1651.
[26] Mendoza, E. (1991). Real Business Cycles in a Small Open Economy. American
Economic Review, 81, 797–818.
[27] Milesi-Ferretti, G. and C. Tille (2011). Capital ‡ows and the crisis. Economic
Policy 26(66), 291-346.
[28] Monacelli, T. and R. Perotti (2008). Fiscal policy, wealth e¤ects and markups,
NBER Working Paper, No. 14584.
[29] Mishkin, F. S. (1998). The dangers of exchange rate pegging in emerging market
countries. International Finance, 1(1), 81-101.
[30] Moser, T. (2003). What is international …nancial contagion? International Fi-
nance, 6, 157-178.
[31] Naug, B. and R. Nmyoen (1996). Pricing to market in a small open economy.
Scandinavian Journal of Economics, 98(3), 329-350.
[32] Neumeyer, P.A., and F. Perri (2005). Business cycles in emerging markets: the
role of interest rates. Journal of Monetary Economics, 52(2), 345-380.
[33] Perri, F. and V. Quadrini (2008) International recessions. NBER Working Pa-
per, No. 17201.
[34] Reinhart, C. and K. Rogo¤ (2009). This Time is Di¤erent. Eight Centuries of
Financial Folly. The Princeton University Press.
[35] Rotemberg, J. (1982). Sticky prices in the United States. Journal of Political
Economy, 90, 1187–1211.
33
[36] Schmitt-Grohe, S. and M. Uribe (2003). Closing small open economy models.
Journal of International Economics, 61, 163–185.
[37] Schmitt-Grohe, S. and M. Uribe (2007). Optimal simple and implementable
monetary and …scal rules. Journal of Monetary Economics, Elsevier, 54(6),
1702-1725.
[38] Sims, C. (2005). Second-order accurate solution of discrete time dynamic equi-
librium models. Mimeo, Princeton University.
34
Appendix A - Optimal Contracting Problem
Entrepreneurs observe .
t+1
(/) ex post, but the lenders can only observe it at a
monitoring cost which is assumed to be a certain fraction (j) of the return. As shown
by Bernanke et al. (1999), the optimal contract between the lender and the entre-
preneur is a standard debt contract characterized by a default threshold, .
t+1
(/),
such that if .
t+1
(/) _ .
t+1
(/), the lender receives a …xed return in the form of a
contracted interest on the debt. If .
t+1
(/) < .
t+1
(/), then the borrower defaults,
the lender audits by paying the monitoring cost and keeps what it …nds. There-
fore, we can de…ne the expected return to entrepreneur and lender, respectively, as
follows:
1
t
[1
1
t+1
Q
t
1
t+1
(/)(
_
1
.
t+1
(I)
.(/))(.)d. ÷.
t+1
(/)
_
1
.
t+1
(I)
)(.)d.)[
= 1
t
[1
1
t+1
Q
t
1
t+1
(/).(.
t+1
(/))[, (A1)
1
t
[1
1
t+1
Q
t
1
t+1
(/)(.
t+1
(/)
_
1
.
t+1
(I)
)(.
)d.
÷(1 ÷j)
_
.
t+1
(I)
0
.
t+1
(/))(.
)d.
)[
= 1
t
[1
1
t+1
Q
t
1
t+1
(/)q(.
t+1
(/); ¸
t
)[, (A2)
where 1
1
t
denotes the ex-post realization of return to capital and .(.) is the bor-
rowers’ share of the total return. We use the de…nition of the lender’s perception of
productivity shock .
t+1
(/) in Equation (A2) where q(.(/); ¸) represents the lenders’
share of the total return, itself a function of both the idiosyncratic shock and the
perception factor.
We assume that each entrepreneur is subject to an idiosyncratic shock .
t
¸ [0, ·)
with 1(.
t
) = 1 and c.d.f and p.d.f are given by 1(.
t
) and )(.
t
). We de…ne .(.) as
the expected gross share of the proceeds going to the borrower (ignoring the time
subscript t and entrepreneur index / for notational simplicity):
.(.) =
_
1
.
.)(.)d. ÷.
_
1
.
)(.)d. (A3)
= 1 ÷
_
.
0
.)(.)d. ÷.
_
1
.
)(.)d.
= 1 ÷I(.),
35
where I(.) = [1 ÷1(.)[. ÷
_
.
0
.d1(.), following Bernanke et al. (1999).
Let 1
C
t
be the contractual rate speci…ed by the lender. By de…nition, the default
threshold (.
t+1
(/)) is set at the level of returns that is just enough to honor the
debt contract obligations, satisfying the following equation:
.
t+1
(/)1
1
t+1
Q
t
1
t+1
(/)
o
t+1
= 1
C
t+1
(/)1
1
t+1
(/) (A4)
Recall that the misperception of investors regarding the distribution of .
t+1
is rep-
resented by ¸
t
such that .
t+1
(/) = .
t+1
(/)¸
t
. As in Curdia (2008), we write the
participation constraint of investors (in foreign currency):
(1 ÷i
t
)1
1
t+1
(/) = 1
t
[(1 ÷1
(.
t+1
(/)))1
C
t+1
(/)1
1
t+1
(/)[
÷(1 ÷j)1
t
[
_
.
t+1
(I)
0
.
(/)d1
(.
(/))
1
1
t+1
Q
t
1
t+1
(/)
o
t+1
[.(A5)
De…ne:
1
(.) = Ii(.
_ .) (A6)
= Ii(.¸ _ .)
= Ii(. _
.
¸
)
= 1(
.
¸
).
We also de…ne G(.) =
_
.
0
.)(.)d. =
_
.
0
.d1(.) and note that G(.) = 1(.)1(.[. <
.). Then we similarly express:
G
(.) = 1
(.)1(.
[.
< .) (A7)
= 1(
.
¸
)1(.[. <
.
¸
)
= ¸G(
.
¸
).
Noting that the monitoring cost (·
t
) is given by jG
(.) and substituting (A4),
(A6) and (A7) into (A5) we get:
(1 ÷i
t
)1
1
t+1
(/) = 1
t
[
1
1
t+1
Q
t
1
t+1
(/)
o
t+1
¦(1 ÷1(
.
t+1
(/)
¸
t
)).
t+1
(/)
÷(1 ÷j)¸
t
G(
.
t+1
(/)
¸
t
)¦[ (A8)
36
which can be re-arranged to yield:
(1 ÷i
t
)1
1
t+1
(/) = 1
t
[q(.
t+1
(/); ¸
t
)
1
1
t+1
Q
t
1
t+1
(/)
o
t+1
[ (A9)
In (A9) q(.; ¸) is de…ned as
q(.; ¸) = ¸[I(
.
¸
) ÷jG(
.
¸
)[. (A10)
We assume that the aggregate risk in terms of the exchange rate and the return
to capital is borne by lenders such that the participation constraint holds with
expectations as in Cespedes et al. (2004), Elekdag and Tchakarov (2007) and Curdia
(2008). Therefore, it should be clear that return to capital (1
1
) and the cut o¤
value (.) are state contingent and the participation constraint holds ex post with
equality at each possible state.
We can now analyze the optimal contract which determines a state contingent
cut o¤ value .
t
and 1
t+1
(/) solving the following maximization problem:
max 1
t
[1
1
t+1
Q
t+1
1
t+1
(/).(.
t+1
(/))[
subject to the participation constraint (A8). The optimality conditions for this
maximization problem are:
1
t
[1
1
t+1
Q
t
.(.
t+1
(/))[ ÷1
t
[A
t+1
(
1
1
t+1
Q
t
q(.
t+1
(/); ¸
t
)
o
t+1
÷
(1 ÷i
t
)Q
t
o
t
)[ = 0 (A11)
A
t+1
(l) = ÷
¬(l).
0
(.
l
t+1
(/))o
l
t+1
q
0
(.
l
t+1
(/); ¸
t
)
(A12)
where l is the state of the world, ¬(l) is the probability of the state l and A
t+1
is
the Lagrangian multiplier. Substituting (A12) into (A11) yields:
1
t
[1
1
t+1
(
.
0
(.
t+1
(/))q(.
t+1
(/); ¸
t
)
q
0
(.
t+1
(/); ¸
t
)
÷.(.
t+1
(/)))[ (A13)
= (1 ÷i
t
)1
t
[
o
t+1
o
t
(
.
0
(.
t+1
(/))
q
0
(.
t+1
(/); ¸
t
)
)[
Given that entrepreneurs are identical ex ante, each entrepreneur faces the same
…nancial contract. We can then write the external risk premium (1÷1
t+1
) as follows:
37
1 ÷ 1
t+1
= [
.
0
(.
t+1
)
q(.
t+1
, ¸
t
).
0
(.
t+1
÷.(.
t+1
)q
0
(.
t+1
, ¸
t
)
[1
t
¦
o
t+1
o
t
¦. (A14)
Using (A14), (A13) can be re-written as 1
t
[1
1
t+1
[ = 1
t
[(1 ÷ i
t
)(1 ÷ 1
t+1
)[. These
two equations correspond to (13) and (12) in the text.
38
Appendix B - Model Equations
B.1. Domestic Economy
Households
C
1,t
= (1 ÷c)(
1
1,t
1
t
)
¸
C
t
(B1.1)
C
A,t
= c(
1
A,t
1
t
)
¸
C
t
(B1.2)
1
t
= [(1 ÷c)1
1¸
1,t
÷c1
1¸
A,t
[
1¸(1¸)
(B1.3)
H
,
t
= \
t
(B1.4)
(C
t
÷
H
1+,
t
1 ÷,
)
o
= ,(1 ÷i
t
)1
t
[(C
t+1
÷
H
1+,
t+1
1 ÷,
)
o
1
t
1
t+1
[ (B1.5)
(C
t
÷
H
1+,
t
1 ÷,
)
o
= ,(1 ÷i
t
)w
1,t
1
t
[(C
t+1
÷
H
1+,
t+1
1 ÷,
)
o
1
t
1
t+1
o
t+1
o
t
[. (B1.6)
Production Firms
\
t
=
(1 ÷j)(1 ÷\)1
t
'C
t
·
t
(B1.7)
\
1
t
= (1 ÷j)\1
t
'C
t
(B1.8)
1
t
=
j1
t
'C
t
1
t
(B1.9)
'C
t
=
1
j
t
\
1j
t
¹
t
j
j
(1 ÷j)
1j
(B1.10)
1
1,t
=
`
` ÷1
'C
t
÷
w
1
` ÷1
1
t
1
1,t
1
1,t
1
1,t1
(
1
1,t
1
1,t1
÷1)
÷
w
1
` ÷1
1
t
[O
t
1
t+1
1
1,t
1
1,t+1
1
1,t
(
1
1,t+1
1
1,t
÷1)[ (B1.11)
o
t
1
A,t
=
`
` ÷1
'C
t
÷
w
A
` ÷1
1
t
1
A,t
1
A,t
1
A,t1
(
1
A,t
1
A,t1
÷1)
÷
w
A
` ÷1
1
t
[O
t
1
t+1
1
A,t
1
A,t+1
1
A,t
(
1
A,t+1
1
A,t
÷1)[ (B1.12)
39
O
t
= ,
(C
t+1
÷
¸
1+,
H
1+,
t+1
)
o
(C
t
÷
¸
1+,
H
1+,
t
)
o
1
t
1
t+1
(B1.13)
Importing Firms
1
A,t
=
`
` ÷1
o
t
1
A,t
÷
w
A
` ÷1
1
t
1
A,t
1
A,t
1
A,t1
(
1
A,t
1
A,t1
÷1) (B1.14)
÷
w
A
` ÷1
1
t
[O
t
1
t+1
1
A,t
1
A,t+1
1
A,t
(
1
A,t+1
1
A,t
÷1)[
Un…nished Capital Producing Firms
1
1,t
= (1 ÷c)(
1
1,t
1
t
)
¸
1
t
(B1.15)
1
A,t
= c(
1
A,t
1
t
)
¸
1
t
(B1.16)
1
t+1
= [
1
t
1
t
÷
w
1
2
(
1
t
1
t
÷c)
2
[1
t
÷ (1 ÷c)1
t
(B1.17)
Q
t
1
t
= [1 ÷w
1
(
1
t
1
t
÷c)[
1
(B1.18)
Entrepreneurs
1
t
[
1
1
t+1
Q
t
1
t+1
(/)
o
t+1
q(.
t+1
(/); ¸
t
)[ = (1 ÷i
t
)1
t+1
(/) (B1.19)
1
t
·\
t
= 0[1
1
t
Q
t1
1
t
(1 ÷i
t
) ÷(1 ÷i
t1
)o
t
1
t
[ ÷\
1
t
(B1.20)
1
t
C
1
t
= (1 ÷0)[1
1
t
Q
t1
1
t
(1 ÷i
t
) ÷(1 ÷i
t1
)o
t
1
t
[ (B1.21)
C
1
1,t
= (1 ÷c)(
1
1,t
1
t
)
¸
C
1
t
(B1.22)
C
1
A,t
= c(
1
A,t
1
t
)
¸
C
1
t
(B1.23)
1
t
[1
1
t+1
[ = 1
t
[(1 ÷i
t
)(1 ÷ 1
t+1
)[ (B1.24)
40
1 ÷ 1
t+1
= [
.
0
(.
t+1
)
q(.
t+1
; ¸
t
).
0
(.
t+1
) ÷.(.
t+1
)q
0
(.
t+1
; ¸
t
)
[1
t
¦
o
t+1
o
t
¦ (B1.25)
1
t
[1
1
t+1
[ = 1
t
[
1
t+1
Q
t
÷
Q
t+1
Q
t
¦(1÷c)÷w
1
(
1
t+1
1
t+1
÷c)
1
t+1
1
t+1
÷
w
1
2
(
1
t+1
1
t+1
÷c)
2
¦[ (B1.26)
Let .
1t
=
ln .
t
+0.5o
2
!
o
!
and .
2t
=
ln(
!
t
%
t
)+0.5o
2
!
o
!
, and O be a c.d.f of a standard normal
distribution:
oif(r) = 2O(r
_
2) ÷1 (B1.27)
oif c(r) = 2(1 ÷O(r
_
2))
.(.
t
) = 1 ÷(.
1t
÷o
.
) ÷.
t
(1 ÷O(.
1t
)) (B1.28)
q(.
t
; ¸
t
) = ¸
t
[
.
t
¸
t
(1 ÷O(.
2t
)) ÷ (1 ÷j)O(.
2t
÷o
.
)[ (B1.29)
.
0
(.
t
) = ÷(1 ÷O(.
1t
)) (B1.30)
q
0
(.
t
; ¸
t
) = ¸
t
[1 ÷O(.
2t
) ÷
j
_
2¬o
.
oxp(÷
.
2
2t
2
) (B1.31)
Monetary Policy
1 ÷i
t
= (1 ÷i) (¬
t
)
c
(1
t
,1 )
c
Y
(B1.32)
General Equilibrium
1
t
= 1
1,t
÷1
A,t
(B1.33)
1
1,t
= C
1,t
÷C
1
1,t
÷1
1,t
÷ (1 ÷c)(
1
1,t
1
t
)
¸
[
i=1,A
w
i
2
(
1
i,t
1
i,t1
÷1)
2
÷
w
A
2
(
1
A,t
1
A,t1
÷1)
2
÷i
t
1
1
t
1
t
Q
t1
1
t
[ (B1.34)
1
A,t
= C
A,t
÷C
1
A,t
÷1
A,t
÷ (c)(
1
A,t
1
t
)
¸
[
i=1,A
w
i
2
(
1
i,t
1
i,t1
÷1)
2
÷
w
A
2
(
1
A,t
1
A,t1
÷1)
2
÷i
t
1
1
t
1
t
Q
t1
1
t
[ (B1.35)
41
o
t
1
A,t
1
A,t
÷1
A,t
1
A,t
= o
t
(1 ÷i
t1
)(1
1
t
w
1,t1
÷1
t
) ÷o
t
(1
1
t+1
÷1
t+1
) (B1.36)
B.2. Model Equations: Foreign Economy
Households
C
1,t
= (1 ÷c
)(
1
1,t
1
t
)
¸
C
t
(B2.1)
C
A,t
= c
(
1
A,t
1
t
)
¸
C
t
(B2.2)
1
t
= [(1 ÷c
)1
1¸
1,t
÷c
1
1¸
A,t
[
1¸(1¸
)
(B2.3)
H
,
t
= \
t
(B2.4)
(C
t
÷
H
1+,
t
1 ÷,
)
o
= ,(1 ÷i
t
)1
t
[(C
t+1
÷
H
1+,
t+1
1 ÷,
)
o
1
t
1
t+1
[ (B2.5)
Production Firms
\
t
=
(1 ÷j)(1 ÷\)1
t
'C
t
·
t
(B2.6)
\
1
t
= (1 ÷j)\1
t
'C
t
(B2.7)
1
t
=
j1
t
'C
t
1
t
(B2.8)
1
1,t
=
`
` ÷1
'C
t
÷
w
1
` ÷1
1
t
1
1,t
1
1,t
1
1,t1
(
1
1,t
1
1,t1
÷1)
÷
w
1
` ÷1
1
t
[O
t
1
t+1
1
1,t
1
1,t+1
1
1,t
(
1
1,t+1
1
1,t
÷1)[ (B2.9)
o
t
1
A,t
=
`
` ÷1
'C
t
÷
w
A
` ÷1
1
t
1
A,t
1
A,t
1
A,t1
(
1
A,t
1
A,t1
÷1)
÷
w
A
` ÷1
1
t
[O
t
1
t+1
1
A,t
1
A,t+1
1
A,t
(
1
A,t+1
1
A,t
÷1)[ (B2.10)
O
t
= ,
(C
t+1
÷
¸
1+,
H
1+,
t+1
)
o
(C
t
÷
¸
1+,
H
1+,
t
)
o
1
t
1
t+1
(B2.11)
42
Importing Firms
1
A,t
=
`
` ÷1
1
A,t
÷
w
A
` ÷1
1
t
1
A,t
1
A,t
1
A,t1
(
1
A,t
1
A,t1
÷1) (B2.12)
÷
w
A
` ÷1
1
t
[O
t
1
t+1
1
A,t
1
A,t+1
1
A,t
(
1
A,t+1
1
A,t
÷1)[
Un…nished Capital Producing Firms
1
1,t
= (1 ÷c
)(
1
1,t
1
t
)
¸
1
t
(B2.13)
1
A,t
= (c
)(
1
A,t
1
t
)
¸
1
t
(B2.14)
1
t+1
= [
1
t
1
t
÷
w
1
2
(
1
t
1
t
÷c)
2
[1
t
÷ (1 ÷c)1
t
(B2.15)
Q
t
1
t
= [1 ÷w
1
(
1
t
1
t
÷c)[
1
(B2.16)
Entrepreneurs
1
t
[1
1
t+1
Q
t
1
t+1
q
(.
t+1
; ¸
t
)[ = (1 ÷i
t
)1
t+1
(B2.17)
1
t
·\
t
= 0[1
1
t
Q
t1
1
t
(1 ÷i
t
) ÷(1 ÷i
t1
)1
t
[ ÷\
1
t
(B2.18)
1
t
C
1
t
= (1 ÷0)[1
1
t
Q
t1
1
t
(1 ÷i
t
) ÷(1 ÷i
t1
)1
t
[ (B2.19)
C
1
1,t
= (1 ÷c
)(
1
1,t
1
t
)
¸
C
1
t
(B2.20)
C
1
A,t
= c
(
1
A,t
1
t
)
¸
C
1
t
(B2.21)
1
t
[1
1
t+1
[ = 1
t
[(1 ÷i
t
)(1 ÷ 1
t+1
)[ (B2.22)
43
1 ÷ 1
t+1
= [
.
0
(.
t+1
(/))
q
(.
t+1
(/); ¸
t
).
0
(.
t+1
(/)) ÷.
(.
t+1
(/))q
0
(.
t+1
(/); ¸
t
)
[ (B2.23)
Let .
1t
=
ln .
t
+0.5o
2
!
o
!
and .
2t
=
ln(
!
t
%
t
)+0.5o
2
!
o
!
, and O be a c.d.f of a standard
normal distribution as above:
.
(.
t
) = 1 ÷O(.
1t
÷o
.
) ÷.
t
(1 ÷O(.
1t
)) (B2.24)
q
(.
t
; ¸
t
) = ¸
t
[
.
t
¸
t
(1 ÷O(.
2t
)) ÷ (1 ÷j)O(.
2t
÷o
.
)[ (B2.25)
.
0
(.
t
) = ÷(1 ÷O(.
1t
)) (B2.26)
q
0
(.
t
; ¸
t
) = ¸
t
[1 ÷O(.
2t
) ÷
j
_
2¬o
.
oxp(÷
.
2
2t
2
) (B2.27)
1
t
[1
1
t+1
[ = 1
t
[
1
t+1
Q
t
÷
Q
t+1
Q
t
¦(1÷c)÷w
1
(
1
t+1
1
t+1
÷c)
1
t+1
1
t+1
÷
w
1
2
(
1
t+1
1
t+1
÷c)
2
¦[. (B2.28)
Monetary Policy
1 ÷i
t
= (1 ÷i
) (¬
t
)
c
(1
t
,1
)
c
Y
(B2.29)
General Equilibrium
o
t
= 1,o
t
(B2.30)
1
t
= 1
1,t
÷1
A,t
, (B2.31)
1
1,t
= C
1,t
÷C
1
1,t
÷1
1,t
÷ (1 ÷c
)(
1
1,t
1
t
)
¸
[
i=1,A
w
i
2
(
1
i,t
1
i,t1
÷1)
2
÷
w
A
2
(
1
A,t
1
A,t1
÷1)
2
÷i
t
1
1
t
1
t
Q
t1
1
t
[, (B2.32)
1
A,t
= C
A,t
÷C
1
A,t
÷1
A,t
÷c
(
1
A,t
1
t
)
¸
[
i=1,A
w
i
2
(
1
i,t
1
i,t1
÷1)
2
÷
w
A
2
(
1
A,t
1
A,t1
÷1)
2
÷i
t
1
1
t
1
t
Q
t1
1
t
[. (B2.33)
44
Table 1: Parameter Values for Consumption, Production and Monetary Policy
: = 0.1 Size of the domestic economy
, = 0.00 Discount factor
o = 1 Inverse of the intertemporal elasticity of substitution
¸ = 1 Elasticity of substitution between domestic and foreign goods
, = 2 Frisch elasticity of labour supply
· = 0.8ò Degree of openness
j = 0.8ò Share of capital in production
` = 11 Elasticity of substitution between domestic goods
c = 0.02ò Quarterly rate of depreciation
\ = 0.01 Share of entrepreneurial labor
w
1
= 12 Investment adjustment cost
w
1
= 0.007ò Responsiveness of household risk premium to debt/GDP
w
i
, w
A
= 120 Price adjustment costs for i = H, A
c
¬
= 1.ò Coe¢cient of CPI in‡ation in the policy rule
c
Y
= 0.ò Coe¢cient of output gap in the policy rule
j
¸
= 0.ò Persistence of the domestic perception shock
Table 2: Parameter Values for the Entrepreneurial Sector
Domestic Economy
¸ = 0.8 Leverage
1
t
= 0.02 External risk premium
j = 0.2 Monitoring cost
0 = 0.0088 Survival rate
Foreign Economy
¸
= 0.ò Leverage
1
t
= 0.00ò External risk premium
j
= 0.12 Monitoring cost
0 = 0.0066 Survival rate
45
Output Consumption Investment Current account Inflation Interest rate Exchange rate
Argentina 4.58 5.95 12.94 1.01 2.94 2.17 5.77
Brazil 1.94 1.95 4.89 2.19 2.33 2.90 11.65
Korea 2.57 3.52 5.49 3.40 1.27 1.19 9.80
Mexico 2.55 3.57 6.98 5.80 1.70 1.65 9.96
Philippines 2.58 1.93 7.03 4.24 2.13 2.17 6.75
Average 2.84 3.38 7.47 3.33 2.07 2.02 8.78
Model - Domestic shock 3.15 3.24 17.20 3.61 0.60 2.16 3.23
Model- Global shock ?=0.5 2.91 2.23 15.95 2.12 0.67 2.27 1.60
ii) Standard deviations relative to output
Output Consumption Investment Current account Inflation Interest rate Exchange rate
Argentina 1.00 1.30 2.83 0.22 0.64 0.47 1.26
Brazil 1.00 1.01 2.52 1.13 1.20 1.49 6.00
Korea 1.00 1.37 2.14 1.32 0.50 0.46 3.81
Mexico 1.00 1.40 2.74 2.27 0.67 0.65 3.91
Philippines 1.00 0.75 2.72 1.64 0.82 0.84 2.61
Average 1.00 1.16 2.59 1.32 0.77 0.78 3.52
Model - Domestic shock 1.00 1.03 5.46 1.15 0.19 0.69 1.03
Model- Global shock ?=0.5 1.00 0.77 5.48 0.73 0.23 0.78 0.55
iii) Correlations with Output and Autocorrelations
?(C,Y) ?(I,Y) ?(CA,Y) ?(Y(t),Y(t-1)) ?(C(t),C(t-1)) ?(I(t),I(t-1)) ?(CA(t),CA(t-1))
Argentina 0.92 0.83 -0.54 0.83 0.85 0.91 0.93
Brazil 0.77 0.38 -0.03 0.35 0.17 0.55 0.92
Korea 0.87 0.86 -0.72 0.80 0.82 0.79 0.86
Mexico 0.78 0.85 -0.45 0.82 0.80 0.85 0.83
Philippines 0.82 0.10 0.01 0.78 0.71 0.59 0.77
Average 0.83 0.61 -0.35 0.72 0.67 0.74 0.86
Model - Domestic shock 0.93 0.96 -0.93 0.51 0.63 0.51 0.42
Model- Global shock ?=0.5 0.68 0.91 -0.67 0.71 0.92 0.62 0.77
Table 3. Business Cycles in Emerging Economies: Data vs. Model
i) Standard deviations (in %)
46
Output Consumption Investment Current account Inflation Interest rate Exchange rate
E.U. 1.37 0.81 3.80 0.50 0.61 1.02 7.09
U.S. 1.22 0.88 4.26 1.26 0.16 0.28 3.01
Average 1.30 0.85 4.03 0.88 0.39 0.65 5.05
Model- Global shock ?=0.5 0.73 0.56 5.43 0.64 0.12 0.23 1.94
ii) Standard deviations relative to output
Output Consumption Investment Current account Inflation Interest rate Exchange rate
E.U. 1.00 0.59 2.77 0.36 0.45 0.74 5.18
U.S. 1.00 0.72 3.49 1.03 0.13 0.23 2.47
Average 1.00 0.66 3.13 0.70 0.29 0.49 3.82
Model- Global shock ?=0.5 1.00 0.77 7.44 0.88 0.16 0.32 2.66
iii) Correlations with output and autocorrelations
?(C*,Y*) ?(I*,Y*) ?(CA*,Y*) ?(Y*(t),Y*(t-1)) ?(C*(t),C*(t-1)) ?(I*(t),I*(t-1)) ?(CA*(t),CA*(t-1))
E.U. 0.56 0.95 -0.36 0.91 0.67 0.91 0.71
U.S. 0.88 0.91 -0.71 0.88 0.85 0.93 0.75
Average 0.72 0.93 -0.54 0.90 0.76 0.92 0.73
Model- Global shock ?=0.5 0.43 0.80 0.66 0.71 0.76 0.47 0.77
Table 4. Business Cycles in Advanced (Big) Economies: Data vs. Model
i) Standard deviations (in %)
47
With respect to the U.S.
Output Consumption Investment Output Consumption Investment
Argentina 0.49 0.33 0.67 0.89 0.61 0.95
Brazil 0.30 0.02 0.17 0.80 0.57 0.28
Korea 0.11 0.15 -0.14 0.86 0.82 0.16
Mexico 0.66 0.37 0.34 0.96 0.75 0.83
Phillipines 0.42 0.00 -0.12 0.66 0.34 -0.08
Average 0.40 0.17 0.19 0.84 0.62 0.43
Model- Global shock ?=0 0.63 -0.13 -0.75 0.63 -0.13 -0.75
Model- Global shock ?=0.5 0.94 0.35 0.96 0.94 0.35 0.96
Model- Global shock ?=1 0.98 0.61 0.98 0.98 0.61 0.98
Table 5. Cross-Country Correlations
Global crisis period (2007Q1-2011Q3) Full sample (1990Q1-2011Q3)
48
Welfare ? Welfare ? Welfare ?
i) Domestic shock
Inflation targeting -59.1945 ? -124.5435 ? -60.9214 -
Fixed exchange rate regime -59.2231 0.0283 -124.5585 0.0090 -60.9467 0.0146
ii) Foreign shock
a) Contagion (?=1)
Inflation targeting -59.1854 - -124.4821 - -60.8783 -
Fixed exchange rate regime -59.2345 0.0403 -124.5549 0.0434 -60.8744 -0.0025
Inflation targeting -59.1517 ? -124.4877 - -60.8401 -
Fixed exchange rate regime -59.1874 0.0324 -124.5421 0.0325 -60.8066 -0.0197
Inflation targeting -59.1367 - -124.4968 -60.8169 -
Fixed exchange rate regime -59.1358 -0.0121 -124.5298 0.0196 -60.7425 -0.0435
b) Contagion (?=0.5)
c) No contagion (?=0)
Table 6. Welfare Results
Baseline Higher openness Higher leverage
(?= 0.35, ?= 0.3) (?=0.6, ?=0.3) (?= 0.35, ?=0.5)
49
(percent deviations from the steady state)
Figure 1. Responses to a Financial Crisis in Domestic Economy: Domestic Economy*
?1.8
-1.2
-0.6
0.0
0.6
1.2
0 5 10
Output
-9
-6
-3
0
3
6
0 5 10
Investment
-1.5
-1.0
-0.5
0.0
0.5
1.0
0 5 10
Consumption
-3
-2
-1
0
1
2
0 5 10
Asset Prices
-4
-2
0
2
4
6
0 5 10
Risk Premium
-0.9
-0.6
-0.3
0.0
0.3
0.6
0 5 10
Entreprenurs' Debt
-4.5
-3.0
-1.5
0.0
1.5
3.0
0 5 10
Net Worth
-1.0
-0.5
0.0
0.5
1.0
1.5
0 5 10
Exports
-2.4
-1.6
-0.8
0.0
0.8
1.6
0 5 10
Capital Inflows
Real Exchange Rate CPI Inflation Interest Rate
50
*The figures show the impact of a 1% negative shock to the perception of investors regarding the productivity of
domestic entrepreneurs. The variables are presented as log-deviations from the steady state (except for interest
rate), multiplied by 100 to have an interpretation of percentage deviations.
-0.9
-1.0
-2.4
-1.6
-0.8
0.0
0.8
1.6
2.4
0 5 10
Real Exchange Rate
-0.3
-0.2
-0.1
0.0
0.1
0.2
0 5 10
CPI Inflation
-1.2
-0.8
-0.4
0.0
0.4
0.8
0 5 10
Interest Rate
Figure 2. Responses to a Financial Crisis in Foreign Economy: Foreign Economy*
(percent deviations from the steady state)
?0.6
-0.4
-0.2
0.0
0.2
0.4
0 5 10
Output
-4.5
-3.0
-1.5
0.0
1.5
3.0
0 5 10
Investment
-0.21
-0.14
-0.07
0.00
0.07
0.14
0 5 10
Consumption
-1.2
-0.8
-0.4
0.0
0.4
0.8
0 5 10
Asset Prices
-5.0
-2.5
0.0
2.5
5.0
7.5
0 5 10
Risk Premium
-2.7
-1.8
-0.9
0.0
0.9
1.8
0 5 10
Entreprenurs' Debt
-2.4
-1.6
-0.8
0.0
0.8
1.6
0 5 10
Net Worth
-1.4
-0.7
0.0
0.7
1.4
2.1
0 5 10
Exports
-0.45
-0.30
-0.15
0.00
0.15
0.30
0 5 10
Capital Inflows
2.1
Real Exchange Rate
0.6
CPI Inflation
1.0
Interest Rate
51
*The figures show the impact of a 1% negative shock to the perception of investors regarding the productivity of
foreign entrepreneurs. The variables are presented as log-deviations from the steady state (except for interest rate),
multiplied by 100 to have an interpretation of percentage deviations.
1.4 -0.45
-1.4
-0.7
0.0
0.7
1.4
2.1
0 5 10
Real Exchange Rate
-0.9
-0.6
-0.3
0.0
0.3
0.6
0 5 10
CPI Inflation
-1.5
-1.0
-0.5
0.0
0.5
1.0
0 5 10
Interest Rate
Figure 3. Responses to a Financial Crisis in Foreign Economy with Financial Contagion:
(percent deviations from the steady state)
Domestic Economy*
?2.4
-1.6
-0.8
0.0
0.8
1.6
0 5 10
Output
?=0.5
?=1
-15
-10
-5
0
5
10
0 5 10
Investment
-2.1
-1.4
-0.7
0.0
0.7
1.4
0 5 10
Consumption
-9
-6
-3
0
3
6
0 5 10
Net Worth
-8
-4
0
4
8
12
0 5 10
Risk Premium
-4.8
-3.2
-1.6
0.0
1.6
3.2
0 5 10
Asset Prices
-1.8
-1.2
-0.6
0.0
0.6
1.2
0 5 10
Entrepreneurs' Debt
-1.2
-0.6
0.0
0.6
1.2
1.8
0 5 10
Exports
-3
-2
-1
0
1
2
0 5 10
Capital Inflows
3
Real Exchange Rate
0 4
CPI Inflation
1 6
Interest Rate
52
*The figures show the impact of a 0.75% negative shock to the perception of investors regarding the productivity
of foreign entrepreneurs. The variables are presented as log-deviations from the steady state (except for interest
rate), multiplied by 100 to have an interpretation of percentage deviations.
-1.8
-1.2
-3
-2
-1
0
1
2
3
0 5 10
Real Exchange Rate
-0.6
-0.4
-0.2
0.0
0.2
0.4
0 5 10
CPI Inflation
-2.4
-1.6
-0.8
0.0
0.8
1.6
0 5 10
Interest Rate
Figure 4. Responses to a Financial Crisis in Foreign Economy without Financial Contagion:
Domestic Economy*
(percent deviations from the steady state)
?0.09
-0.06
-0.03
0.00
0.03
0.06
0 5 10
Output
-0.6
-0.3
0.0
0.3
0.6
0.9
0 5 10
Investment
-0.3
-0.2
0.0
0.2
0.3
0.5
0 5 10
Consumption
-0.4
-0.2
0.0
0.2
0.4
0.6
0 5 10
Asset Prices
-3
-2
0
2
3
5
0 5 10
Risk Premium
-0.6
-0.3
0.0
0.3
0.6
0.9
0 5 10
Entreprenurs' Debt
-0.4
-0.2
0.0
0.2
0.4
0.6
0 5 10
Net Worth
-1.5
-1.0
-0.5
0.0
0.5
1.0
0 5 10
Net Exports
-0.8
-0.4
0.0
0.4
0.8
1.2
0 5 10
Capital Inflows
Real Exchange Rate CPI Inflation Interest Rate
53
*The figures show the impact of a 0.75% negative shock to the perception of investors regarding the productivity
of foreign entrepreneurs. The variables are presented as log-deviations from the steady state (except for interest
rate), multiplied by 100 to have an interpretation of percentage deviations.
-0.6 -1.5
-0.8
-0.9
-0.6
-0.3
0.0
0.3
0.6
0 5 10
Real Exchange Rate
-0.09
-0.06
-0.03
0.00
0.03
0.06
0 5 10
CPI Inflation
-0.15
-0.10
-0.05
0.00
0.05
0.10
0 5 10
Interest Rate
Figure 5. Responses to a Financial Crisis in Foreign Economy with Financial Contagion:
Domestic Economy- The Impact of Openness*
(percent deviations from the steady state)
?2.4
-1.6
-0.8
0.0
0.8
1.6
0 5 10
Output
? =0.6
? =0.35
? =0.1
-18
-12
-6
0
6
12
0 5 10
Investment
-2.4
-1.6
-0.8
0.0
0.8
1.6
0 5 10
Consumption
-7.5
-5.0
-2.5
0.0
2.5
5.0
0 5 10
Net Worth
-10
-5
0
5
10
15
0 5 10
Risk Premium
-6.0
-4.0
-2.0
0.0
2.0
4.0
0 5 10
Asset Prices
-2.4
-1.6
-0.8
0.0
0.8
1.6
0 5 10
Entrepreneurs' Debt
-1.6
-0.8
0.0
0.8
1.6
2.4
0 5 10
Exports
-3
-2
-1
0
1
2
0 5 10
Capital Inflows
3
Real Exchange Rate
1 0
CPI Inflation
2 0
Interest Rate
54
*The figures show the impact of a 0.75% negative shock to the perception of investors regarding the productivity
of foreign entrepreneurs. The variables are presented as log-deviations from the steady state (except for interest
rate), multiplied by 100 to have an interpretation of percentage deviations.
-2.4 -1.6
-3
-2
-1
0
1
2
3
0 5 10
Real Exchange Rate
-1.5
-1.0
-0.5
0.0
0.5
1.0
0 5 10
CPI Inflation
-3.0
-2.0
-1.0
0.0
1.0
2.0
0 5 10
Interest Rate
Figure 6. Responses to a Financial Crisis in Foreign Economy without Financial Contagion:
Domestic Economy- The Impact of Openness*
(percent deviations from the steady state)
?0.45
-0.30
-0.15
0.00
0.15
0.30
0.45
0 5 10
Output
? =0.6 ? =0.35
? =0.1
-0.9
-0.6
-0.3
0.0
0.3
0.6
0.9
1.2
0 5 10
Investment
-0.4
-0.2
0.0
0.2
0.4
0.6
0 5 10
Consumption
-0.6
-0.3
0.0
0.3
0.6
0.9
0 5 10
Net Worth
-3
-2
0
2
3
5
0 5 10
Risk Premium
-0.30
-0.15
0.00
0.15
0.30
0.45
0 5 10
Asset Prices
-0.8
-0.4
0.0
0.4
0.8
1.2
0 5 10
Entrepreneurs' Debt
-2.1
-1.4
-0.7
0.0
0.7
1.4
0 5 10
Exports
-1
0
0
0
1
1
0 5 10
Capital Inflows
1 60
Real Exchange Rate
0 08
CPI Inflation
0 2
Interest Rate
55
*The figures show the impact of a 0.75% negative shock to the perception of investors regarding the productivity
of foreign entrepreneurs. The variables are presented as log-deviations from the steady state (except for interest
rate), multiplied by 100 to have an interpretation of percentage deviations.
-0.8 -2.1
-1
-2.40
-1.60
-0.80
0.00
0.80
1.60
0 5 10
Real Exchange Rate
-0.12
-0.08
-0.04
0.00
0.04
0.08
0 5 10
CPI Inflation
-0.3
-0.2
-0.1
0.0
0.1
0.2
0 5 10
Interest Rate
Figure 7. Responses to a Financial Crisis in Foreign Economy with Financial Contagion:
Domestic Economy- The Role of Monetary Policy Strategy*
(percent deviations from the steady state)
?5.1
-3.4
-1.7
0.0
1.7
3.4
0 5 10
Output
Fixed exchange rate regime
Inflation targeting
-21
-14
-7
0
7
14
0 5 10
Investment
-5.4
-3.6
-1.8
0.0
1.8
3.6
0 5 10
Consumption
-12
-8
-4
0
4
8
0 5 10
Net Worth
-10
-5
0
5
10
15
0 5 10
Risk Premium
-7.5
-5.0
-2.5
0.0
2.5
5.0
0 5 10
Asset Prices
-1.8
-1.2
-0.6
0.0
0.6
1.2
0 5 10
Entrepreneurs' Debt
-1.2
-0.6
0.0
0.6
1.2
1.8
0 5 10
Exports
-5
-3
-2
0
2
3
0 5 10
Capital Inflows
3
Real Exchange Rate
0 6
CPI Inflation
1 6
Interest Rate
56
*The figures show the impact of a 0.75% negative shock to the perception of investors regarding the productivity
of foreign entrepreneurs. The variables are presented as log-deviations from the steady state (except for interest
rate), multiplied by 100 to have an interpretation of percentage deviations.
-1.8
-1.2 -5
-2
-1
0
1
2
3
0 5 10
Real Exchange Rate
-0.9
-0.6
-0.3
0.0
0.3
0.6
0 5 10
CPI Inflation
-2.4
-1.6
-0.8
0.0
0.8
1.6
0 5 10
Interest Rate
Figure 8. Responses to a Financial Crisis in Foreign Economy without Financial Contagion:
Domestic Economy- The Role of Monetary Policy Strategy*
(percent deviations from the steady state)
?1.0
-0.5
0.0
0.5
1.0
1.5
0 5 10
Output
Fixed exchange rate regime
Inflation targeting
-2
-1
0
1
2
3
0 5 10
Investment
-1.2
-0.6
0.0
0.6
1.2
1.8
0 5 10
Consumption
-1.0
-0.5
0.0
0.5
1.0
1.5
0 5 10
Net Worth
-3
-2
0
2
3
5
0 5 10
Risk Premium
-0.6
-0.3
0.0
0.3
0.6
0.9
0 5 10
Asset Prices
-0.6
-0.3
0.0
0.3
0.6
0.9
0 5 10
Entrepreneurs' Debt
-1.5
-1.0
-0.5
0.0
0.5
1.0
0 5 10
Exports
-1.0
-0.5
0.0
0.5
1.0
1.5
0 5 10
Capital Inflows
1 0
Real Exchange Rate
0 3
CPI Inflation
0 8
Interest Rate
57
*The figures show the impact of a 0.75% negative shock to the perception of investors regarding the productivity
of foreign entrepreneurs. The variables are presented as log-deviations from the steady state (except for interest
rate), multiplied by 100 to have an interpretation of percentage deviations.
-0.6
-1.5 -1.0
-1.5
-1.0
-0.5
0.0
0.5
1.0
0 5 10
Real Exchange Rate
-0.5
-0.3
-0.2
0.0
0.2
0.3
0 5 10
CPI Inflation
-1.2
-0.8
-0.4
0.0
0.4
0.8
0 5 10
Interest Rate
doc_978290818.pdf
The recent global financial crisis was the first in recent history that was triggered by problems in the financial system of the mature economies. Existing work on financial crisis in emerging market countries, however, almost exclusively focus on the role of financial frictions in the domestic economy.
Global Financial Crisis, Financial Contagion,
and Emerging Markets
F. Gulcin Ozkan and D. Filiz Unsal
WP/12/293
© 2012 International Monetary Fund WP/12/293
IMF Working Paper
Strategy, Policy, and Review Department
Global Financial Crisis, Financial Contagion and Emerging Markets
1
Prepared by F. Gulcin Ozkan and D. Filiz Unsal
Authorized for distribution by Catherine Pattillo
December 2012
Abstract
The recent global financial crisis was the first in recent history that was triggered by problems
in the financial system of the mature economies. Existing work on financial crisis in emerging
market countries, however, almost exclusively focus on the role of financial frictions in the
domestic economy. In contrast, we propose a two-country DSGE model to investigate the
transmission of a global financial crisis that originates from financial frictions in the rest of the
world. We find that the scale of financial spillovers from the global to the domestic economy
and trade openness are key determinants of the severity of the financial crisis for the domestic
economy. Our results also suggest that the welfare ranking of alternative monetary policy
regimes is determined by the degree of financial contagion, the degree of trade openness as
well as the scale of foreign currency denominated debt in the domestic economy.
JEL Classification Numbers: E5, F3, F4
Keywords: sudden stops, financial crisis, emerging markets
Author’s E-Mail Address:[email protected] and [email protected]
1
We are grateful to an anonymous referee for comments and suggestions which have helped substantially improve
the paper. We also would like to thank Christopher Adam, Lynne Evans, Hugh Metcalf, Marcus Miller, Joe
Pearlman, Peter Sinclair, Mike Wickens and seminar participants at Newcastle University, participants at the
Money, Macro and Finance Group (MMF) Annual Conference at the University of Bradford, the conference on
“Monetary policy before, during and after the crisis” held at Heriot-Watt University, 29-30 September, 2011 and
the Emerging Markets Group - ESRC Workshop on Global Linkages and Financial Crises at the Cass Business
School, University of London, 27 April 2012. The usual disclaimer applies.
This Working Paper should not be reported as representing the views of the IMF.
The views expressed in this Working Paper are those of the author(s) and do not necessarily
represent those of the IMF or IMF policy. Working Papers describe research in progress by the
author(s) and are published to elicit comments and to further debate.
Contents Page
1. Introduction ....................................................................................................................4
2. The Model ......................................................................................................................6
2.1 Households .........................................................................................................8
2.2 Firms ..................................................................................................................9
2.2.1 Production Firms .....................................................................................10
2.2.2 Importing Firms ......................................................................................11
2.2.3 Unfinished Capital Producing Firms ......................................................11
2.3 Entrepreneurs ...................................................................................................12
2.4 Monetary Policy ...............................................................................................15
2.5 General Equilibrium and Balance of Payments Dynamics ..............................16
3. Solution and Parametrization .......................................................................................17
3.1 Consumption, Production and Monetary Policy ..............................................17
3.2 Entrepreneurs ...................................................................................................18
4. Financial Crisis and the Domestic Economy ...............................................................19
4.1 Model Dynamics ..............................................................................................19
4.2 Financial Crisis Originating in the Domestic Economy ..................................20
4.3 Financial Crisis in the Global Economy ..........................................................21
4.3.1 The Impact of the financial shock on the foreign ecocnomy ..................21
4.3.2 The transmission of the foreign shock onto the domestic economy .......22
4.3.3 The role of trade openness ......................................................................25
4.4 Monetary Policy Options and Welfare Analysis .............................................26
5. Conclusions ..................................................................................................................29
References ................................................................................................................................31
Tables
1. Parameter Values for Consumption, Production and Monetary Policy ...............................45
2. Parameter Values for the Entrepreneurial Sector .................................................................45
3. Business Cycles in Emerging Economies: Data vs. Model .................................................46
4. Business Cycles in Advanced (Big) Economies: Data vs. Model .......................................47
5. Cross-Country Correlations .................................................................................................48
6. Welfare Results ....................................................................................................................49
Figures
1. Responses to a Financial Crisis in Domestic Economy .......................................................50
2. Responses to a Financial Crisis in a Foreign Economy .......................................................51
3. Responses to a Financial Crisis in Foreign Economy with Financial Contagion ................52
4. Responses to a Financial Crisis in Foreign Economy without Financial Contagion ...........53
3
5. Responses to a Financial Crisis in Foreign Economy with Financial Contagion: Domestic
Economy-The Impact of Openness ..........................................................................................54
6. Responses to a Financial Crisis in Foreign Economy without Financial Contagion:
Domestic Economy-The Impact of Openness .........................................................................55
7. Responses to a Financial Crisis in Foreign Economy with Financial Contagion: Domestic
Economy-The Role of Monetary Policy Strategy ....................................................................56
8. Responses to a Financial Crisis in Foreign Economy without Financial Contagion:
Domestic Economy—The Role of Monetary Policy Strategy .................................................57
Appendixes
A. Optimal Contracting Problem .............................................................................................35
B. Model Equations .................................................................................................................39
B.2 Model Equations: Foreign Economy ....................................................................42
1 Introduction
The global …nancial turmoil that has gripped the world economy since August 2007
has been widely viewed as unprecedented, at least since the Great Depression of the
1930s. The turbulence in …nancial systems was followed by a signi…cant reduction
in real economic activity in a large number of countries. For emerging market and
developing economies, …nancial crisis is not a new phenomenon. Indeed, since the
early 1990s countries as diverse as Mexico, Russia, a number of East Asian countries,
Brazil, Turkey and Argentina have all been hit by either currency or …nancial crises,
or both. Although country experiences have varied with regard to the source of
di¢culty in each episode, the pro…le of crises has been fairly similar. A "sudden
stop" of capital in‡ows is almost always followed by a sharp contraction in economic
activity. Furthermore, many countries witnessed substantial losses in the value of
their currencies, which greatly helped in recovering from the crises. In essence,
these countries were able to expand net exports to alleviate or compensate for the
contractionary e¤ects of foreign currency liabilities following devaluations – the so-
called balance sheet e¤ects. For instance, sharp current account reversals were a
common feature of recovery processes in most Asian countries following the 1997
crisis.
When set against this background, from the viewpoint of an emerging market
economy, the recent global …nancial crisis has been di¤erent in two major ways.
1
First, the source of the sharp reduction in capital in‡ows on this occasion has been
the severe liquidity squeeze in the …nancial markets of developed economies, unlike
in any of the previous experiences. Indeed, the slowdown in …nancial ‡ows to the
emerging market economies followed from the virtual standstill in the credit markets
in the US and the UK spreading to other major …nancial markets. Second, emerging
market countries have also witnessed a substantial fall in their exports as the …nan-
cial crisis hit consumer spending in the developed world. Hence, given the strong
downturn in the global economy, countries have been unable to export their way
out of the crisis even though a large number of countries experienced substantial
devaluations of their currencies.
The above evaluation suggests that both …nancial and trade channels have been
1
See, Reinhart and Rogo¤ (2009) for a comprehensive evaluation of the di¤erences between the
current and previous experiences of …nancial crisis.
4
crucial in the transmission of the global …nancial crisis to emerging market coun-
tries. Motivated by this observation, in this paper, we develop a two-country dy-
namic stochastic general equilibrium model with an explicit treatment of both trade
and …nancial linkages between the countries. This enables us to investigate pos-
sible spillover e¤ects of a …nancial crisis originating in the global economy on to
a domestic, small open economy. There are three features of our model economy
that are representative of emerging market economies. First, the domestic economy
exhibits …nancial frictions in the form of high leverage; that is, a large share of in-
vestment is …nanced with external resources. Second, the borrowing is taken to be
from abroad in foreign currency terms, as is common in emerging market countries.
In the presence of such foreign currency denominated debt - widely referred to as
liability dollarization - a change in the perception of foreign lenders of the current
state of the economy leads to an endogenous adjustment in the cost of borrowing,
generating a negative feedback loop between real and …nancial sectors. Moreover,
the presence of liability dollarization makes the balance sheets of the …nancial sys-
tem sensitive to the changes in exchange rates. Finally, our model gives explicit
consideration to exchange rate pass-through, the scale of which distinguishes the
experiences of emerging and mature economies.
The main channels through which a global …nancial shock impacts upon the
emerging market economy are as follows. The re-pricing of credit risk increases the
cost of external …nancing, inducing a sharp decline in output and domestic in‡ation,
and a depreciation of the domestic currency. Since the external risk premium is
tied to the leverage ratio, …rms reduce new borrowing in order to decrease the
endogenous component of the risk premium. Moreover, the fall in domestic in‡ation
and the depreciation of the domestic currency increase the real debt burden for
leveraged households, leading to a decrease in consumption. The depreciation of the
domestic currency enables the home country to compensate, at least partly, for the
decline in consumption and investment demand. However, this only applies if the
…nancial crisis originated in the domestic economy. In contrast, when the source
of the …nancial shock is global, the export channel works in the opposite direction
with global contraction leading to a fall in export demand for the domestic output
resulting in a further decline in domestic economic activity.
We also explore the role of …nancial contagion and trade openness on the prop-
5
agation of international …nancial shocks. Our results reveal that the greater the
…nancial contagion, the greater the severity of the …nancial crisis for the domestic
economy. A higher level of …nancial contagion induces a greater increase in the ex-
ternal risk premium and thus a greater fall in capital in‡ows reducing investment,
output and consumption. We also examine the role of the domestic economy’s open-
ness to trade on its response to the global …nancial shock. We …nd that, in contrast
to the existing literature, the greater a country’s trade integration with the rest of
the world, the greater the response of its macroeconomic aggregates to a sudden
stop of capital ‡ows. We then provide a rigorous welfare analysis of two alternative
monetary policy frameworks; an in‡ation targeting regime versus a …xed exchange
rate regime. Our results reveal that the welfare ranking of the two regimes is deter-
mined by the degree of …nancial contagion, the degree of trade openness as well as
the scale of foreign currency denominated debt in the domestic economy.
The remainder of the paper is organized as follows. Section 2 sets out the struc-
ture of our two-country DSGE model by describing household, …rm and entrepre-
neurial behavior with special emphasis on the description of …nancial frictions. Sec-
tion 3 presents the solution and the calibration of the model. Section 4 presents and
discusses our quantitative results. Section 5 provides the concluding remarks.
2 The Model
Based on the …nancial accelerator mechanism developed by Bernanke et al. (1999),
our model shares its basic features with the recent theoretical studies incorporating
the …nancial accelerator in combination with liability dollarization such as Ces-
pedes et al. (2004), Devereux et al. (2006), Gertler et al. (2007) and Elekdag and
Tchakarov (2007). However, our framework di¤ers from the existing general equi-
librium models with …nancial frictions in three important ways. First, we develop
a two-country sticky price DSGE model where both the trade and …nancial link-
ages between the two countries are fully speci…ed. Modelling the foreign economy
explicitly enables us to investigate the propagation of global …nancial crisis on the
domestic economy by considering both the trade and …nancial channels. Given that
our purpose is to explore the transmission of a global …nancial crisis onto an emerg-
ing market economy, we set the characteristics of the two countries in such a way
that domestic (foreign) country exempli…es an emerging market (mature) economy.
6
Having a two-country framework also enables us to consider important transmission
channels which are ignored in earlier studies.
2
Secondly, in our framework sudden
stops are modeled as an unfavorable change in the perception of lenders which cre-
ates a self-ful…lling pessimism about the economy through the enforcement of tighter
credit conditions, similar to Curdia (2008). We argue that this better represents the
source of …nancial crises relative to the existing work on sudden stops which de…nes
the initial shock as either an aggregate structural shock, such as a rise in foreign
interest rates, or an adverse shock to fundamentals. Finally, in contrast to most
existing New-Keynesian DSGE models, we adopt Greenwood, Hercowitz, Hu¤man
(1988) (hereafter GHH) preferences in the speci…cation of household utility.
3
This
is motivated by the fact that the GHH preferences improve the ability of general
equilibrium models in capturing business cycle facts especially in emerging market
countries (see, for example, Neumeyer and Perri, 2005 among others).
The world economy consists of two economies; a domestic economy and a for-
eign economy. The total measure of the world economy is normalized to unity, with
domestic and foreign economies measuring : and (1 ÷
, respectively. Given thatour main purpose is to explore the impact of a global shock on an emerging market
economy, we maintain that the domestic economy is small in size relative to the
foreign one. Otherwise, domestic and foreign countries share the same preferences,
technology and market structure for consumption and capital goods. The model for
the domestic small economy is presented in this section. We use a similar version of
2
Perri and Quadrini (2010) also utilize a two-country model in analyzing business cycle ‡uctua-
tions across countries. However, in contrast to our model, the …nancial (credit) shock in their model
is transmitted to the rest of the world through its e¤ect on the borrowing capacity of …rms and the
resulting cost and thus employment implications. Other channels of international transmission of
domestic shocks have also been investigated in the literature. These include international portfolio
holdings of leverage constrained investors (Devereux and Yetman, 2010); binding enforcement con-
straints on credit supply (Devereux and Sutherland, 2011); syncronization of borrowing costs and
credit spreads across countries (Dedola and Lombardo, 2012); and global banking (Enders et al.
2011). Our paper di¤ers from these analyses in two major respects, in addition to many di¤erences
in the modelling structure. First, in contrast to our analysis where both …nancial and trade linkages
are explicitly modelled and play a key role, …nancial interactions are treated as the main source of
interdependence between countries in these studies. Second, unlike in the above mentioned studies,
we explore the ampli…cation of a global …nancial shock onto an emerging market economy by setting
the model structure accordingly.
Among the existing studies the one that is closest to our analysis is Gilchrist (2004) who explores
the role of leverage in the transmission of shocks from developed to developing countries. Although
the two models share a number of features as both are based on Bernanke et al. (1999), our
model di¤ers from Gilchrist (2004) in a number of important aspects. These include the teatment
of …nancial accelerator mechanism, the modelling of the source of shocks as well as the choice of
preferences among others.
3
Notable exceptions are Monacelli and Perotti (2008) and Gertler and Karadi (2010).
7
the model for the foreign economy with the exception that entrepreneurs in the do-
mestic country borrow in foreign currency while entrepreneurs in the foreign country
borrow in their domestic currency. In what follows, variables without superscripts
refer to the home economy variables, while variables with a star indicate the foreign
economy variables, unless indicated otherwise.
2.1 Households
A representative household is in…nitely-lived and seeks to maximize:
E
0
1
t=0
,
t
1
1 ÷o
(C
t
÷
H
1+,
t
1 ÷,
)
1o
, (1)
where C
t
is a composite consumption index, H
t
is hours of work, 1
t
is the mathemat-
ical expectation conditional upon information available at t, , is the representative
consumer’s subjective discount factor where 0 < , < 1, o 0 is the inverse of the
intertemporal elasticity of substitution and , 0 is the inverse elasticity of labour
supply. Our speci…cation for household’s utility allows for GHH preferences over
hours, which eliminates wealth e¤ects from labor supply.
The composite consumption index, C
t
, is given by:
C
t
=
_
(1 ÷c)
1
C
(¸1)¸¸
1,t
÷ (c)
1
C
(¸1)¸¸
A,t
_
¸¸(¸1)
, (2)
where ¸ 0 is the elasticity of substitution between domestic and imported (foreign
goods), and 0 < c < 1 denotes the weight of imported goods in domestic con-
sumption basket. Equation (2) suggests that the expenditure share of the imported
goods in the consumption basket of households is given by 0 < c < 1. This weight,
c = (1 ÷
·, depends on (1 ÷ , the relative size of the foreign economy, andon ·, the degree of trade openness of the domestic economy. C
1,t
and C
A,t
are
consumption of domestic and foreign goods, represented by:
C
1,t
=
__
1
0
C
1,t
(,)
(A1)¸A
d,
_
A¸(A1)
; C
A,t
=
__
1
0
C
A,t
(,)
(A1)¸A
d,
_
A¸(A1)
,
where , ¸ [0, 1[ indicates the goods varieties and ` 1 is the elasticity of substitution
among goods produced within a country.
The real exchange rate 11A
t
is de…ned as 11A
t
=
S
t
1
t
1
t
, where o
t
is the
nominal exchange rate, domestic currency price of foreign currency, and 1
t
=
_
_
1
0
1
t
(,)
1A
d,
_
1¸(1A)
is the aggregate price index for foreign country’s consump-
tion goods in foreign currency. In contrast to standard small open economy models,
8
our two-country framework allows us to determine 1
t
endogenously in the foreign
economy block.
Households have access to two types of non-contingent one-period debt; one
denominated in domestic currency, 1
t
, and the other in foreign currency, 1
1
t
, with a
nominal interest rate of i
t
and i
t
w
1,t
. Due to imperfect capital mobility, households
need to pay a premium when borrowing from the rest of the world. This premium,
w
1,t
, is de…ned as a function of deviations of aggregate credit over GDP from its
steady state level; w
1,t
=
D
2
[oxp(
S
t
1
H
t+1
+S
t
1
E
t
1
t
G11
t
÷
S1
H
+S1
E
t
1G11
) ÷ 1[
2
where 1
1
t
is
entrepreneurs’ debt (see equation (11) below).
4
Households own all home production
and the importing …rms and thus are recipients of pro…ts, H
t
. Other sources of
income for the representative household are wages \
t
and new borrowing net of
interest payments on outstanding debts, both in domestic and foreign currency.
Then, the representative household’s budget constraint in period t can be written
as follows:
1
t
C
t
÷ (1 ÷i
t1
)1
t
÷ (1 ÷i
t1
)w
1,t1
o
t
1
1
t
= \
t
H
t
÷1
t+1
÷o
t
1
1
t+1
÷ H
t
. (3)
The representative household chooses the paths for ¦C
t
, H
t
, 1
t+1
, 1
1
t+1
¦
1
t=0
in
order to maximize its expected lifetime utility in (1) subject to the budget constraint
in (3).
2.2 Firms
There are three types of …rms in the model. Production …rms produce a di¤eren-
tiated …nal consumption good using both capital and labor as inputs. These …rms
engage in local currency pricing and face price adjustment costs. As a result, …nal
goods’ prices are sticky in terms of the local currency of the markets in which they
are sold. Importing …rms that sell the goods produced in the foreign economy also
have some market power and face adjustment costs when changing prices. Price
stickiness in export and import prices causes the law of one price to fail such that
exchange rate pass through is incomplete in the short run.
5
Finally, there are com-
4
Following Schmitt-Grohe and Uribe (2003), this premium is introduced for technical reasons to
maintain the stationarity in the economy’s net foreign assets. As in Schmitt-Grohe and Uribe, we
assume that the elasticity of the premium with respect to the aggregate debt is very close to zero
(
D
= 0:0075) so that the dynamics of the model are not a¤ected by this friction.
5
There is considerable empirical evidence of pricing-to-market and incomplete exchange rate
pass-through in small open economies. See, for example, Naug and Nymoen (1996) and Campa and
Goldberg (2005).
9
petitive …rms that combine investment with rented capital to produce un…nished
capital goods, that are then sold to entrepreneurs.
2.2.1 Production Firms
Each …rm produces a di¤erentiated good indexed by , ¸ [0, 1[ using the production
function:
1
t
(,) = ¹
t
·
t
(,)
1j
1
t
(,)
j
, (4)
where ¹
t
denotes labor productivity, common to all the production …rms and ·
t
(,) is
the labor input which is a composite of household, H
t
(,), and entrepreneurial labor,
H
1
t
(,); de…ned as ·
t
(,) = H
t
(,)
1
H
1
t
(,)
. 1
t
(,) denotes capital provided by
the entrepreneur, as is explored in the following subsection. Assuming that the price
of each input is taken as given, the production …rms minimize their costs subject to
(4).
Firms have some market power and they segment domestic and foreign markets
with local currency pricing, where 1
1,t
(,) and 1
A,t
(,) denote price in domestic mar-
ket (in domestic currency) and price in foreign market (in foreign currency). Firms
also face quadratic menu costs in changing prices expressed in the units of consump-
tion basket given by
i
2
(
1
i;t
())
1
i;t1
())
÷ 1)
2
for di¤erent market destinations i = H, A.
6
The combination of local currency pricing together with nominal price rigidities im-
plies that ‡uctuations in the nominal exchange rate have a smaller impact on export
prices so that exchange rate pass-through to export prices is incomplete in the short
run.
As domestic …rms are owned by domestic households, the individual …rm maxi-
mizes its expected value of future pro…ts using the household’s intertemporal rate of
substitution in consumption, given by ,
t
l
c,t
. The objective function of …rm , can
thus be written as:
1
c
1
t=0
,
t
l
c,t
1
t
[1
1,t
(,)1
1,t
(,) ÷o
t
1
A,t
(,)1
A,t
(,) ÷'C
t
1
t
(,)
÷1
t
i=1,A
w
i
2
(
1
i,t
(,)
1
i,t1
(,)
÷1)
2
[, (5)
6
This generates a gradual adjustment in the prices of goods in both markets, as suggested by
Rotemberg (1982).
10
where 1
1,t
(,) and 1
A,t
(,) represent domestic and foreign demand for the domesti-
cally produced good ,. We assume that di¤erent varieties have the same elasticities
in both markets, so that the demand for good , can be written as,
1
i,t
(,) = (
1
i,t
(,)
1
i,t
)
A
1
i,t
, for i = H, A, (6)
where 1
1,t
is the aggregate price index for goods sold in domestic market, as is de-
…ned earlier and 1
A,t
is the export price index given by 1
A,t
=[
_
1
0
1
A,t
(,)
1A
d,[
1¸(1A)
.
1
A,t
denotes the foreign aggregate export demand for domestic goods and is
determined endogenously in the foreign country block, given by
1
A,t
= C
A,t
÷C
1
A,t
÷1
A,t
÷ (c
)(
1
A,t
1
t
)
¸
[
i=1,A
w
i
2
(
1
i,t
1
i,t1
÷1)
2
÷
w
A
2
(
1
A,t
1
A,t1
÷1)
2
÷i
t
1
1
t
1
t
Q
t1
1
t
[. (7)
where C
A,t
and C
1
A,t
are import demand of households and entrepreneurs for con-
sumption, 1
A,t
denotes imported investment goods. The rest of the terms on the
right hand side are imported components of price adjustment costs and the dead-
weight loss brought by …nancial frictions.
2.2.2 Importing Firms
There is a set of monopolistically competitive importing …rms, owned by domestic
households, who buy foreign goods at prices 1
A,t
and then sell to the domestic
market at 1
A,t
with some mark-up. They are also subject to a price adjustment
cost with w
A
_ 0, the price adjustment cost parameter, analogous to the production
…rms. This implies that there is some delay between exchange rates changes and
the import price adjustments so that the short run exchange rate pass through to
import prices is also incomplete.
2.2.3 Un…nished Capital Producing Firms
Let 1
t
denote aggregate investment in period t, which is composed of domestic and
…nal goods:
1
t
=
_
(1 ÷c)
1
1
(¸1)¸¸
1,t
÷ (c)
1
1
(¸1)¸¸
A,t
_
¸¸(¸1)
, (8)
11
where the domestic and imported investment goods’ prices are assumed to be the
same as the domestic and imported consumer goods prices, 1
1,t
and 1
A,t
. The new
capital stock requires the same combination of domestic and foreign goods so that
the nominal price of a unit of investment equals the price level, 1
t
. This implies that
1
1,t
= (1 ÷c)(
1
H;t
1
t
)
¸
1
t
and 1
A,t
= (c)(
1
M;t
1
t
)
¸
1
t
.
Competitive …rms use investment as an input, 1
t
and combine it with rented
capital 1
t
to produce un…nished capital goods. Following the existing literature, we
assume that the marginal return to investment in terms of capital goods is decreasing
in the amount of investment undertaken (relative to the current capital stock) due
to the existence of adjustment costs, represented by
I
2
(
1
t
1
t
÷ c)
2
where c is the
depreciation rate. Then, the production technology of the …rms producing un…nished
capital can be represented by ¯
t
(1
t
, 1t) = [
1
t
1
t
÷
I
2
(
1
t
1
t
÷ c)
2
[1
t
which exhibits
constant returns to scale so that the un…nished capital producing …rms earn zero
pro…t in equilibrium. The stock of capital used by the …rms in the economy evolves
according to:
1
t+1
= [
1
t
1
t
÷
w
1
2
(
1
t
1
t
÷c)
2
[1
t
÷ (1 ÷c)1
t
. (9)
The optimality condition for the un…nished capital producing …rms with respect
to the choice of 1
t
yields the following nominal price of a unit of capital Q
t
:
Q
t
1
t
= [1 ÷w
1
(
1
t
1
t
÷c)[
1
. (10)
2.3 Entrepreneurs
As stated earlier, entrepreneurs are key players in our model. They transform un-
…nished capital goods and sell them to the production …rms. They …nance their
investment by using their own net worth and by borrowing from foreign lenders.
7
There is a continuum of entrepreneurs indexed by / in the interval [0,1]. Each en-
trepreneur has access to a stochastic technology in transforming 1
t+1
(/) units of
un…nished capital into .
t+1
(/)1
t+1
(/) units of …nished capital goods. The idio-
syncratic productivity .
t
(/) is assumed to be i.i.d. (across time and across …rms),
drawn from a distribution 1(.), with p.d.f of )(.) and 1(.) = 1.
8
7
See, Mishkin (1998) and Eichengreen and Hausmann (2005) for the importance of foreign cur-
rency borrowing in emerging market countries.
8
The idiosyncratic productivity is assumed to be distributed log-normally; log(!
t
(k)) s
N(
1
2
2
!
;
2
!
). This characterization is similar to that in Carlstrom and Fuerst (1997), Bernanke et
12
At the end of period t, each entrepreneur / has net worth, ·\
t
(/). The budget
constraint of the entrepreneur is de…ned as follows:
1
t
·\
t
(/) = Q
t
1
t+1
(/) ÷o
t
1
1
t+1
(/), (11)
where 1
1
t+1
denotes foreign currency denominated debt. Equation (11) simply states
that capital …nancing is divided between net worth and foreign debt. It is clear that
the entrepreneurs are exposed to exchange rate risk, that is, ‡uctuations in the
nominal exchange rate create balance sheet e¤ects in the model.
Productivity is observed by the entrepreneur, but not by the lenders who have
imperfect knowledge of the distribution of .
t+1
(/). Following Curdia (2008) we
specify the lenders perception of .
t+1
(/) as given by .
t+1
(/) = .
t+1
(/)¸
t
where ¸
t
is the misperception factor over a given interval [0,1].
9
Further, the misperception
factor, ¸
t
, is assumed to follow ln(¸
t
) = j
¸
ln(¸
t1
) ÷¸ ln(¸
t
) ÷-
¸
where j
¸
denotes
the persistence parameter, and ¸ measures the degree of …nancial contagion from
the foreign to the domestic economy. Similarly, we assume that ¸
t
, the perception
of lenders regarding the the foreign entrepreneurs’ productivity, follows an AR(1)
process with persistence parameter j
¸
. We take the origin of the …nancial shock
as a change in lenders’ perception regarding idiosyncratic productivity (-
¸
). We
assume that when there is uncertainty about the underlying distribution, lenders
take the worst case scenario as the mean of the distribution of .
t+1
(/). Appendix A
provides more details on the speci…cation of the ambiguity aversion faced by lenders
and the optimal contracting problem.
The optimal contract identi…es the capital demand of entrepreneurs, 1
t+1
(/)
and a cut o¤ value, .
t+1
(/) such that the entrepreneur will maximize their expected
return subject to the participation constraints of the lender. As shown in Appendix
A, the …rst order conditions yield:
al. (1999), Cespedes et al. (2004) and Gertler et al. (2007).
9
As in Curdia (2008), our speci…cation of the debt contract between investors and entrepreneurs
explicitly takes into account the misperception factor, which makes risk premium sensitive to the
perception of investors. Curdia (2008) adopts a max-min criteria for the misperception factor so
that during the sudden stop episode, the misperception factor is set to a constant value lower than 1.
This implies that, in his calibration, the investors’ perception is constant for 2.5 years at a pre-set,
lower value. In our case, however, we let the misperception factor to be an AR(1) process during
the sudden stop episode such that the abrupt change in the perception of investors gradually goes
back to the pre-shock level, rather than staying constant for a prolonged period.
13
1
t
[1
1
t+1
[ = 1
t
[(1 ÷i
t
)(1 ÷ 1
t+1
)[, (12)
where (1 ÷ 1
t+1
) is the external risk premium de…ned by:
1 ÷ 1
t+1
= [
.
0
(.
t+1
(/))
q(.
t+1
(/); ¸
t
).
0
(.
t+1
(/)) ÷.(.
t+1
(/))q
0
(.
t+1
(/); ¸
t
)
[1
t
¦
o
t+1
o
t
¦. (13)
where .(.) and q(.(/); ¸) are the borrowers’ and lenders’ share of the total
return, respectively. As shown in Bernanke et al. (1999), this external risk premium
depends on the leverage,
S
t
1
E
t+1
Q
t
1
t+1
, of the entrepreneur. A greater use of external
…nancing generates an incentive for entrepreneurs to take on more risky projects,
which raises the probability of default. This, in turn, will increase the external risk
premium. Therefore, any shock that has a negative impact on the entrepreneurs’
net worth increases their leverage, resulting in an upward adjustment in the external
risk premium.
We follow the existing literature in assuming that a proportion of entrepreneurs
die in each period to be replaced by new-comers. This assumption makes borrowing
constraints on debt always binding. Given that .(/) is independent of all other
shocks and identical across time and across entrepreneurs, all entrepreneurs are
identical ex-ante. Then, each entrepreneur faces the same …nancial contract speci…ed
by the cut o¤ value and the external …nance premium. This allows us to specify the
rest of the model in aggregate terms.
At the beginning of period t, the entrepreneurs collect revenues and repay their
debt contracted in period t ÷1. Denoting the fraction of entrepreneurs who survive
each period by 0, the net worth can be expressed as follows:
1
t
·\
t
= 0[1
1
t
Q
t1
1
t
.(.
t
)[ ÷\
1
t
. (14)
Equation (14) indicates that the entrepreneur’s net worth is made up of the return
on investment and the entrepreneurial wage income. Given that the borrower’s and
the lender’s share of total return should add up to .(.
t
)÷q(.
t
, ¸
t
) = 1÷i
t
(where i
t
is the cost of monitoring, a deadweight loss associated with …nancial frictions) and
by using the participation constraint of the lenders, we can rewrite the net worth of
the entrepreneur as:
14
1
t
·\
t
= 0[1
1
t
Q
t1
1
t
(1 ÷i
t
) ÷(1 ÷i
t1
)o
t
1
1
t
[ ÷\
1
t
. (15)
It is clear from (15) that unanticipated changes in the nominal exchange rate
increase the debt burden of the entrepreneur, and therefore decrease its net worth.
This, in turn, increases the leverage of the entrepreneur and raises the external risk
premium, implying a higher cost of …nancing. This is an additional mechanism that
magni…es the role of the …nancial accelerator in the economy through transmitting
‡uctuations in the nominal exchange rate to the balance sheets of entrepreneurs.
The entrepreneurs leaving the scene at time t consume their return on capital.
The consumption of the exiting entrepreneurs, C
1
t
, can then be written as:
10
1
t
C
1
t
= (1 ÷0)[1
1
t
Q
t1
1
t
(1 ÷i
t
) ÷(1 ÷i
t1
)o
t
1
1
t
[. (16)
Because of investment adjustment costs and incomplete capital depreciation,
entrepreneurs’ return on capital, 1
1
t
, is not identical to the rental rate of capital,
1
t
. The entrepreneurs’ return on capital is the sum of the rental rate on new and
used capital, and the value of the non-depreciated capital stock, after the adjustment
for the ‡uctuations in the asset prices
_
Q
t+1
Q
t
_
:
1
t
1
1
t+1
= 1
t
[
1
t+1
Q
t
÷
Q
t+1
Q
t
¦(1 ÷c) ÷ w
1
(
1
t+1
1
t+1
÷c)
1
t+1
1
t+1
÷
w
1
2
(
1
t+1
1
t+1
÷c)
2
¦[. (17)
2.4 Monetary Policy
We adopt a standard formulation for the structure of monetary policy framework
in the baseline calibration. We assume that the interest rate rule is of the following
form:
1 ÷i
t
= (1 ÷i) (¬
t
)
c
(1
t
,1 )
c
Y
, (18)
where i and 1 denote the steady-state level of nominal interest rate and output,
and ¬
t
is the CPI in‡ation.
Under the …xed exchange rate regime, the monetary policy rule is given by:
o
t
o
t1
= 1. (19)
10
It is assumed that the entrepreneurs consume an identical mix of domestic and foreign goods
in their consumption basket as is given by the composite consumption index in equation (2).
15
2.5 General Equilibrium and Balance of Payments Dynamics
Market clearing in the …nal good sector requires that total domestic output be equal
to domestic consumption, domestic investment and exports to the rest of the world.
Also, given that frictions such as adjustment and monitoring costs are expressed in
terms of the …nal composite good, part of the output is taken up with the price
adjustment costs for …nal consumption goods as well as those for imported and
exported goods and the monitoring costs. Thus the overall resource constraint faced
by the domestic economy can be written as:
1
t
= 1
1,t
÷1
A,t
, (20)
where
1
1,t
= C
1,t
÷C
1
1,t
÷1
1,t
÷ (1 ÷c)(
1
1,t
1
t
)
¸
[
i=1,A
w
i
2
(
1
i,t
1
i,t1
÷1)
2
÷
w
A
2
(
1
A,t
1
A,t1
÷1)
2
÷i
t
1
1
t
1
t
Q
t1
1
t
[, (21)
where 1
A,t
is the foreign demand for domestic goods (see equation (7)), C
1,t
and
C
1
1,t
are the household and entrepreneur’s consumption demand for domestic goods,
and 1
1,t
is the domestic investment goods used by the un…nished capital producing
…rms. In (21), w
1
, w
A
and w
A
denote price adjustment costs for domestic, exported
and the imported good, respectively and i
t
is the cost of monitoring for the lenders
that is passed on to the domestic economy through the external …nance premium.
Equilibrium in the labor market requires that:
·
t
= H
t
1
, (22)
where H
t
1
is the labor demand for non-entrepreneurial labour.
Substituting (20) and the pro…ts of both the …nal good producing and the im-
porting …rms into the budget constraints of the households and the entrepreneurs
yields the following balance of payments condition after aggregation:
o
t
1
A,t
1
A,t
÷1
A,t
1
A,t
= o
t
(1 ÷i
t1
)(1
1
t
w
1,t1
÷1
1
t
) ÷o
t
(1
1
t+1
÷1
1
t+1
), (23)
where 1
A,t
is the price of imports and 1
A,t
is the aggregate import demand. The
…rst and the second terms on the left are exports and imports, and on the right is
16
simply the change in the net foreign asset position aggregated over households and
entrepreneurs.
In Appendix B, we present the equilibrium model equations for both domestic
and the foreign economy.
3 Solution and Parametrization
We …rst transform the model to reach a stationary representation where a steady
state exists. The model is then solved numerically up to a second order approxima-
tion using Sims (2005).
11
3.1 Consumption, Production and Monetary Policy
We maintain that the structure of consumption, production and monetary policy
is identical in the two economies and thus use the same parametrization with the
exception of the size parameter, :, which is set to 0.1 for domestic economy. We set
the discount factor, , at 0.00, implying a riskless annual return of approximately
4 per cent in the steady state (time is measured in quarters). The inverse of the
elasticity of intertemporal substitution is taken as o = 1, which corresponds to log
utility. The inverse of the elasticity of labour supply , is set to 2, which implies that
1,2 of time is spent working. In the baseline case, we set the degree of openness,
·, to be 0.8ò which is within the range of the values used in the literature.
12
The
share of capital in production, j, is taken to be 0.8ò, consistent with other studies.
13
Following Devereux et al. (2006), the elasticity of substitution between di¤erentiated
goods of the same origin, `, is taken to be 11, implying a ‡exible price equilibrium
mark-up of 1.1, and price adjustment cost is assumed to be 120 for all sectors. The
quarterly depreciation rate c is taken to be 0.02ò, a conventional value used in the
literature. Similar to Gertler et al. (2007), we set the share of entrepreneurs’ labour,
\, at 0.01, implying that 1 per cent of the total wage bill goes to the entrepreneurs.
Regarding monetary policy, we use the original Taylor estimates and set c
¬
= 1.ò
and c
Y
= 0.ò in the baseline calibration. We assume j
¸
to be 0.ò, so that it takes
11
The non-stochastic steady state of the model is solved numerically in MATLAB, and then the
second order approximation of the model and the stochastic simulations are computed using Michel
Juillard’s software Dynare. Details of the computation of the non-stochastic steady state and the
stationary model equations are available upon request.
12
The values for openness in the existing literature range between 0.25 (Cook, 2004; Elekdag and
Tchakarov, 2007) and 0.5 (Gertler et al., 2007). We chose to set a middle value of this range, but
we conduct a sensitivity analysis regarding the value of openness in Section 4.2.2.
13
See, for example, Cespedes et al. (2004) and Elekdag and Tchakarov (2007).
17
0 quarters for the shock to die away. Table 1 summarizes the parametrization of
the model for consumption, production, and monetary policy used in the baseline
calibration.
14
3.2 Entrepreneurs
The parameter values for the entrepreneurial sector in the domestic and foreign
economies are set to re‡ect their de…ning characteristics and are listed in Table 2.
We set the steady state leverage ratio, ¸, the ratio of debt to net worth, and the
value of quarterly external risk premium in the domestic economy, 1, respectively
at 0.8 and 200 basis points, re‡ecting the historical average of emerging market
economies within the last decade.
15
The monitoring cost parameter, j, is taken as
0.2 for the domestic economy as in Devereux et al. (2006). These parameter values
imply a survival rate, 0, of approximately 00.88 per cent in the domestic economy.
For the foreign economy, we closely follow Bernanke et al. (1999). The leverage
ratio is set to 0.ò. The higher leverage ratio for entrepreneurs in foreign economy re-
‡ects the fact that advanced economies have deeper and more sophisticated …nancial
markets, and therefore better …nancing opportunities, leading to a higher economy-
wide leverage. Moreover, having experienced dramatic …nancial crises at the turn
of the century, emerging market economies have been more vigilant towards lend-
ing activities through tighter …nancial regulation, which in many cases has helped
to contain leverage ratios in these economies.
16
The risk spread of 2 per cent in
the steady state is reported for the US economy so we set a quarterly external risk
premium, 1
t
, of 0.00ò. The cost of monitoring in the foreign economy, denoted by
j
, is taken to be 0.12, lower than the domestic economy re‡ecting the presence of
better institutions. Given these parameter values, the implied survival rate is 00.66
per cent in the foreign economy.
14
We carry out a series of sensitivity analyses in order to asses the robustness of our results under
the benchmark calibration. In Section 4, we report the results regarding the degree of …nancial
contagion, trade openness, and monetary policy parameters. We also conduct sensitivity checks
with regard to the degree of exchange rate pass-through, and …nd that our main results are not
sensitive to changes in these parameters. These are not reported due to space limitations.
15
These …gures are decade averages for emerging Americas, emerging Asia, and emerging Europe
between 2000-2010. Worldscope data (debt as a percentage of assets - data item WS 08236) are used
for the leverage ratio. External risk premium is calculated as the di¤erence between the lending and
the policy rate for emerging market countries, where available, using data from Haver Analytics for
the same time period. Variations in these parameters a¤ect our results only quantitatively, but not
qualitatively.
16
See Kalemli-Ozcan et al. (2011) for stylized facts on bank and …rm leverage for 2000- 2009 for
both advanced and emerging economies.
18
4 Financial Crisis and the Domestic Economy
4.1 Model Dynamics
In this section, we …rst report a number of business cycle statistics for the model and
compare them with those observed in the data from both emerging and advanced
economies. In our analysis, we focus only on the response of macroeconomic aggre-
gates to two types of …nancial shock: one originating in the domestic economy and
the other in the foreign (global) economy. In both cases, the source of the …nancial
shock is the change in lenders’ perception of the entrepreneurs’ productivity. Given
the speci…c nature of the source of the shock in our analysis, one would not expect
that the model to match the data in all dimensions. Nonetheless, in what follows we
compare movements and comovements of some key variables in the data and in the
model to generate con…dence in the model’s ability to correctly capture dynamics
and in the proposed calibration of the parameters values.
Table 3 presents business cycle statistics coming from the data as well as the
simulation results under our benchmark parametrization. In the top panel of Table 3,
standard deviations of output, consumption, investment, current account, in‡ation,
interest rate and exchange rate for Argentina, Brazil, Korea, Mexico and Philippines
are listed, while the middle panel reports standard deviations of this set of variables
relative to the standard deviation of output. In both panels, the simulation results
are reported under two scenarios; a domestic shock and a global shock. A glance at
the top two panels of Table 3 suggests that in most cases moments from the data
are well-matched by moments from the model except in the case of in‡ation and the
exchange rate. This is particularly the case for volatility of output, consumption,
current account as well as of interest rate in terms of both individual volatilities and
volatilities relative to output.
The correlations of consumption, investment and current account and autocor-
relations of these variables are presented in the bottom panel of Table 3. Similar to
above, it can be seen that, model results match the business cycle statistics quite
well for the same group of countries.
Table 4 repeats the same exercise for an advanced economies sample which we
represent by EU and the US. Similar to that in Table 3, Table 4 suggests that our
simulation results match the business cycle statistics pretty well with the exception
of current account and output correlations. In contrast to the case with emerging
19
market countries, volatility of in‡ation in the model is also quite close to that in the
data for advanced economies.
Finally, Table 5 presents cross-country correlations of output, consumption and
investment from our model with those from the same sample of emerging market
countries as above for two separate sample periods; 1999-2011 and 2007-2011 for the
global crisis period. As can be seen from Table 5, our simulation results match the
cross-correlations in the data for this group of countries quite well especially over
the global …nancial crisis period which is our main focus in this paper.
We now turn to examining the responses of both the domestic and the foreign
country in response to two types of …nancial shock by assessing the impulse responses
in our two-country model.
4.2 Financial Crisis Originating in the Domestic Economy
We …rst investigate the case of a domestic …nancial crisis by tracing how an unan-
ticipated (temporary) shock to investors’ perception of entrepreneurs’ productivity
is transmitted to the rest of the economy. Such a perception shock leads to a re-
versal of capital ‡ows out of the domestic country, which we refer to as the sudden
stop.
17
The response of the domestic economy to the sudden stop is presented in
Fig.1. When the investors’ perception about the distribution of the entrepreneurs’
productivity changes, lending to domestic entrepreneurs becomes more risky, lead-
ing to a rise in the external risk premium on impact. As the cost of borrowing rises,
entrepreneurs reduce their use of external …nancing by undertaking fewer projects.
This decline in leverage causes a downward adjustment in the risk premium, mitigat-
ing the initial impact of a sudden stop to some extent. Lower borrowing, however,
decreases the future supply of capital and hence brings about a decrease in invest-
ment in the economy. The fall in the in‡ow of capital also lowers the demand for
domestic currency, leading to its depreciation. Since the entrepreneurs’ borrowing is
denominated in foreign currency, this unanticipated change in the exchange rate also
creates balance sheet e¤ects through a rise in the real debt burden. The outcome is
lower investment and output in the economy following the sudden stop, in line with
the experience of several emerging market countries during the 1990s.
Although the rise in the nominal exchange rate puts an upward pressure on the
17
The magnitude of the shock is 1 percent, which causes out‡ow of capital by about 1.5 percent
of GDP on impact.
20
CPI based in‡ation, the decrease in the domestic price level more than o¤sets this
e¤ect, bringing about a fall in the CPI. In spite of this lower price level, however,
aggregate demand falls due to lower investment and output, resulting in lower labor
demand and thereby lower real wages.
There is an additional channel through which the e¤ect of the shock is trans-
mitted to the rest of the economy, working through the export demand. Following
the depreciation of the domestic currency, the foreign demand for domestic goods
increases. On the other hand, imports decline on account of both income and ex-
change rate e¤ects. The trade balance improves, but this e¤ect is not strong enough
to o¤set the decline in domestic demand in our simulations, and hence output con-
tracts. In practice, the export channel is generally highly e¤ective for countries that
are hit by …nancial crises and experience a sizable loss of value of their currencies.
For instance, most East Asian countries bene…tted from signi…cant improvements
in their exports following the 1997 Asian crisis, which has been widely viewed to be
an important factor in their swift recoveries (see, for example, Bleaney, 2005). It is
important to note, however, that this favorable impact of export demand on out-
put recovery not only disappears but starts to work in the opposite direction when
the …nancial crisis originates in the global economy, as is explored in the following
section.
4.3 Financial Crisis in the Global Economy
In this section, we explore the channels through which a …nancial shock originat-
ing in the foreign country is transmitted to the domestic economy and the role of
trade openness in shaping the domestic economy’s responses to the …nancial shock.
The perception shock is now faced by the foreign entrepreneurs, which we take to
represent the case of a global …nancial crisis.
4.3.1 The impact of the …nancial shock on the foreign economy
Fig.2 presents responses of foreign macroeconomic variables to a 1 per cent negative
shock to the perception of investors regarding the productivity of foreign entrepre-
neurs. As is seen from Fig.2, the impact of the shock on the foreign economy is
similar to the sudden stop experience of the domestic economy described above.
Following the shock, lending to foreign entrepreneurs become risky, leading to a rise
in risk premium which, through reducing borrowing and thus supply of capital leads
21
to lower investment and output in the foreign economy. It is also clear from Fig.2
that the falls in investment, output and consumption are smaller in the foreign econ-
omy as compared with the domestic one, given that …nancial frictions are greater
in the domestic economy - domestic entrepreneurs borrow in foreign currency and
monitoring domestic entrepreneurs are more costly.
We now turn to how the foreign (global) …nancial shock is transmitted onto the
domestic economy.
4.3.2 The transmission of the foreign shock onto the domestic economy
The global …nancial shock is transmitted to the domestic economy through three
separate channels involving both …nancial and trade linkages. The …rst of these
is through …nancial spillovers. We postulate that investors’ perception regarding
the true distribution of entrepreneurs’ productivity in the foreign economy and the
domestic economy are inherently related. This is based on the notion that investors
optimally choose the scale and the terms of credit they extend to borrowers in a
forward looking manner. For instance, when faced with credit tightening in the
global economy, investors can anticipate ex-ante that this will be transmitted to
the domestic economy through real and …nancial cross-country linkages, implying
an unfavorable change in their perceptions of the domestic entrepreneurs today.
Also, some asset market linkages such as herding behavior only or mainly exist
during times of crisis, a phenomenon commonly referred to as "pure contagion"
(see, for example, Kaminsky and Reinhart, 2000 and Moser, 2003). Motivated
by these observations, we maintain that one channel through which the …nancial
shock spills over onto the domestic economy is through an unfavourable shift in
investors’ perception of the domestic entrepreneurs. In what follows, we refer to this
mechanism as the …nancial contagion channel.
The second channel through which the global …nancial shock is propagated to
the domestic economy is the export channel. The …nancial crisis in the foreign econ-
omy reduces output and thereby export demand in the foreign economy and thus
net exports of the domestic economy. There is also a third channel of propagation
of the global …nancial shock, which works through a substitution e¤ect. The un-
favourable change in investors’ perception of the foreign entrepreneurs’ productivity
induces investors to look for alternative investment opportunities, leading to an in-
crease in capital in‡ows into the domestic economy, which would, partly o¤set the
22
impact of the …rst two channels. Clearly, the lower the …nancial contagion from the
foreign economy to the domestic economy, the greater the impact of the favourable
substitution e¤ect. We now turn to exploring these channels separately.
Fig.3 and 4 present the domestic economy’s response to a global …nancial shock,
with and without …nancial contagion, respectively.
18
Fig.3 features two alternative
scenarios; full contagion, ¸ = 1 and partial contagion, ¸ = 0.ò. The deterioration in
investors’ perception in the domestic economy following that in the foreign economy
raises the external risk premium, setting in motion the process described above.
Responses presented in Fig.3 reveal that the impact of the global …nancial shock on
the domestic economy is larger than that of a domestic one. Although the change
in the perception of foreign investors about the state of the domestic economy is
identical under the two scenarios, falls in capital, investment and output are greater
with the global …nancial shock and the fall in investment is twice the size. Similarly,
the decrease in foreign borrowing is larger than that of the previous case, leading to
a sharper depreciation of the domestic currency. Likewise, changes in in‡ation and
asset prices are much more pronounced under the global …nancial shock than with
the sudden stop of domestic origin.
One main di¤erence between this case of a foreign …nancial shock and the domes-
tically originated one, as explored in the previous section, is in the way the export
channel works. As is seen above, when the economy is hit by a …nancial crisis of
domestic origin, the depreciation of the currency brings about an improvement in
net exports, which partly o¤sets the initial decline in output. In contrast, when
the …nancial shock is originated in the foreign economy with partial …nancial con-
tagion (¸ = 0.ò), the export channel works in the opposite direction, adding to the
unfavourable impact that works through …nancial contagion, hence worsening the
overall fall in output. This is because the global …nancial crisis reduces net worth,
capital, investment and output in the foreign economy, and therefore decreases the
domestic economy’s exports and therefore output. As output in the foreign econ-
omy returns to its previous level, the export demand improves although the trade
18
Our analysis of the ampli…cation of the foreign …nancial shock onto the domestic economy, as
portrayed in Figures 3-8, is based on a 0.75 % negative shock to the perception of investors regarding
the productivity of foreign enterpreneurs. This shock generates an increase in the risk premium by
about 1 percent quarterly on impact in the domestic economy as in the case of a …nancial crisis of
domestic origin so that the responses are comparable to those in Fig. 1.
23
balance continues to deteriorate owing to the rise in imports following the recovery
in domestic output. When there is full-contagion from the global to the domestic
economy, however, the sharp depreciation of the exchange rate delivers an increase
in exports, as is seen in Fig.3. Yet, the worsening of the risk premium and of entre-
preneurs’ balance sheets is much larger and thus the decline in output is almost as
twice as the decline under partial …nancial contagion.
In order to explore the non-contagion channels more explicitly we now have
a further experiment where we shut o¤ the contagion channel and thus focus on
the export and substitution channels. Fig.4 denotes the changes in the domestic
economy following the …nancial crisis in the foreign economy, without contagion
in the form of a further …nancial shock in the domestic economy. There are now
two opposing e¤ects of the foreign economy’s …nancial crisis on the macroeconomic
outcomes in the domestic economy. The …rst is through the reduced net exports due
to the contraction in the foreign economy following the …nancial crisis. The second,
in contrast, is a favourable impact leading to an increase in capital in‡ows. This is
because lending to domestic entrepreneurs is now perceived to be more pro…table
as the domestic economy is in a better …nancial position.
19
The overall response is
determined by the balance of the two e¤ects. As is seen from Fig.4, in the absence of
…nancial contagion, the increase in risk premium is considerably lower than that in
the above two cases and that of the foreign economy. As a result, capital in‡ows, net
worth and investment all increase following the …nancial shock, and thus contraction
in output in the domestic economy is smaller.
The heterogeneous nature of the pro…le of crisis across countries, as demonstrated
by Fig.2 and Fig.3, has also been highlighted by Milesi-Ferretti and Tille (2011) in
their analysis of capital ‡ows during the 2007-2009 global …nancial crisis episode.
A glance at the capital ‡ows diagrams, suggests that following the …nancial shock
there is a sharp drop in capital ‡ows into the foreign economy, which remain below
steady-state levels for extended periods of time. On the other hand, capital ‡ows
to the domestic country recover relatively quickly toward the pre-shock levels, as
is seen in Fig.3. Indeed, Milesi-Ferretti and Tille (2011) document that capital
in‡ows to advanced countries exhibited a sharp reversal in the last quarter of 2008
and remained well-below pre-crisis levels until the end of their sample period, 2009.
19
IMF (2011) shows that net capital in‡ows to emerging economies are highly correlated with
global …nancing conditions, with global interest rates and risk aversion playing an important role.
24
They show that, in contrast, the pull-back of capital ‡ows from emerging market
countries had been much shorter lived, and in‡ows to these countries in 2009 reached
magnitudes broadly comparable to the pre-crisis levels.
4.3.3 The role of trade openness
Having established that a global …nancial shock is transmitted to the domestic econ-
omy through both the trade and …nancial channels, it follows that the extent of the
domestic response to a global …nancial tightening will be determined by, among
other factors, the openness of the domestic economy. We now turn to exploring
the role of trade openness in the propagation of the foreign …nancial shock more
explicitly.
Fig.5 and Fig.6 illustrate the domestic responses to a foreign …nancial shock
under varying degrees of trade integration between the domestic economy and the
foreign economy with and without …nancial contagion, respectively. In our simu-
lations the degree of trade integration is measured by ·, which, together with the
size of the economy, :, determines the share of imported good items in the domes-
tic consumption basket. The pro…le of the domestic economy in Fig.5 exhibits the
important role played by the degree of trade openness in the ampli…cation of the
global …nancial shock. The greater the trade integration between the two countries,
the more signi…cant is the impact of the global …nancial crisis on the domestic econ-
omy. This holds regardless of the degree of …nancial contagion as is seen from the
responses in Fig.5 and Fig.6. The relationship between a country’s openness to trade
and its vulnerability to sudden stops has already been the focus of an extensive lit-
erature (see, for example Calvo et al. 2006 and Martin and Rey, 2006). A common
…nding in this literature has been that openness makes countries less vulnerable to
crises. In contrast, we …nd that when the …nancial shock originates in the rest of
the world - when the crisis is a global one - the more open an economy, the greater
the unfavorable consequences of the …nancial crisis for the domestic economy. In-
deed, among the countries that have experienced largest falls in economic activity
during the initial stage of the recent …nancial crisis have been Singapore, Taiwan
and Turkey, all of which are highly open economies.
20
20
The fall in output in the …rst quarter of 2009 as compared with a year earlier was 10.1, 10.2
and 13.8 per cent for Singapore, Taiwan and Turkey, respectively. Similarly, Germany and Japan,
that are among the most open of mature economies, contracted by 6.9 and 8.8 per cent, respectively
over the same period (The Economist, July 4th, 2009).
25
4.4 Monetary policy options and welfare analysis
What is the role of monetary policy in determining the domestic economy’s response
to the global …nancial shock? In what follows, we attempt to answer this question
by exploring the e¤ects of the …nancial shock on the domestic economy under two
separate monetary policy regimes. These are a …xed exchange rate regime and a
‡exible in‡ation targeting (IT) regime where monetary authority responds to both
in‡ation and the output gap.
Before we start assessing the implications of the …xed exchange rate regime
versus a ‡exible one (as under IT), it is important to point out the consequences of
nominal exchange rate changes in our model. A rise in the nominal exchange rate
improves net exports while creating a contractionary balance sheet e¤ect for the
…nancial (entrepreneurial) sector given …nancial frictions and liability dollarization
in the domestic economy. As leverage and the risk premium increases, the ability
of entrepreneurs to borrow and produce capital declines, reducing investment and
output in equilibrium. Overall, which monetary regime does better in insulating
the real economic activity from …nancial shocks will be determined by the relative
magnitude of these two e¤ects.
Fig.7 and Fig.8 illustrate the responses under the two regimes with and without
…nancial contagion, respectively and o¤er a number of interesting insights. As is
seen from Fig.7, in the presence of …nancial contagion, the global …nancial shock
brings about a sharp rise in the external risk premium, a steep fall in capital in‡ows
and thus in investment under both regimes. However, the impact of the shock on
output is more pronounced under the …xed exchange rate regime than under the
IT regime. In our calibration, the decline in exports under the …xed exchange rate
regime is large enough to o¤set the favourable impact of stable exchange rate on the
entrepreneurs’ balance sheets leading to a sharper fall in output in response to the
…nancial shock relative to under IT.
Interestingly, the absence of …nancial contagion alters the ranking of the two
regimes in insulating the domestic economy from …nancial shocks originating in the
foreign economy. In the absence of a direct …nancial channel through which foreign
…nancial shocks are transmitted onto the domestic economy, output in the domestic
economy rises under the …xed exchange rate regime, as is depicted by Fig.8. Since
the nominal exchange rate is constant under this scenario, the increase in the risk
26
premium is lower than under the IT regime. Lower external risk premia, in turn,
lead to a rise in the net worth of …rms, and thus in investment and output without
increasing entrepreneurs’ debt. The real appreciation of the exchange rate is greater
under the IT regime, so is the decline in exports.
Overall, this experiment indicates that a global …nancial shock is likely to be
expansionary under the …xed exchange rate regime when there is no direct …nancial
spillovers from the global economy. However, in order to reach normative conclusions
regarding the choice of the monetary policy framework one would need to carry out
a rigorous welfare analysis, which we present in the next section.
Welfare analysis We now turn to an assessment of alternative monetary pol-
icy frameworks based on the evaluation of welfare, taking the utility function of
consumers as the objective. Following Faia and Monacelli (2007), and Gertler and
Karadi (2010), we start by expressing the household utility function in recursive
form:
\
t
= l(C
t
, H
t
) ÷,1
t
\
t+1
(24)
where \
t
= E
0
1
t=0
,
t
l(C
t
, H
t
) denotes the welfare function. We take a second order
approximation of \
t
around the deterministic steady state. Using the second order
solution of the model, we then calculate \
t
under both the IT and the …xed exchange
rate regimes.
21
We also calculate the fraction of consumption, w, required to equate
welfare under the …xed exchange rate \
1A
t
, to the one under the IT regime, \
1T
t
,
as a welfare measure. Under our speci…cation of utility function, and o = 1 (log
utility), w can be expressed as
w = (
(oxp¦(\
cjt
t
÷\
t
)(1 ÷,)¦(C ÷
1
1+'
1+,
)) ÷
1
1+'
1+,
C
) ÷1 (25)
where C and H denote the steady state values of the corresponding variables.
22
Table 6 presents the calculated values of welfare and w for the two monetary
policy regimes under a number of di¤erent scenarios.
As is seen from Table 6, when the source of the …nancial shock is domestic,
the IT regime is welfare-improving, as is presented in the upper part of the table
21
See Scmitt-Grohe and Uribe (2007) for an in depth discussion on using second order approxi-
mation to utility function to calculate the welfare.
22
This is based on the assumption that the enterpreneurial share of consumption is negligible
(see, Bernanke et al. 1999).
27
and the welfare gains associated with the IT regime relative to the …xed exchange
rate regime is about 0.03 percent of steady state consumption (…rst and second
columns of Table 6). The same holds in the face of foreign …nancial shocks when
…nancial spillovers are strong (¸ = 1 and ¸ = 0.ò), as the expansionary e¤ects of the
depreciation under IT o¤set the contractionary balance sheet e¤ects generated by
the entrepreneurs’ foreign currency borrowing, as explained above. Overall, in our
calibration, …xed exchange rate regime increases welfare losses associated with the
global crisis when there are strong …nancial spillovers between the two countries.
In contrast, when there is no contagion the …xed exchange rate regime is welfare
superior (negative w) by about 0.01 per cent of consumption, which derives from two
factors. On the real side, the smaller appreciation in the real exchange rate under
the …xed exchange rate regime brings about a less dramatic decline in exports, hence
aggregate demand is less a¤ected. On the …nancial side, lower external risk premia
under the …xed exchange rate regime leads to a rise in the net worth of …rms, and
thus in investment and output.
As explained above, the ranking of the …xed exchange rate regime versus the
IT regime is determined by the relative importance of the export channel versus
the balance sheet channel. As potential factors underlying the importance of these
channels, we consider alternative values of openness and leverage to see how the
welfare comparison between the two regimes is a¤ected. Our simulations reveal
that as openness increases, the welfare cost associated with having a …xed exchange
rate regime increases (third and fourth columns in Table 6). This is because the
export channel is more e¤ective in more open economies, making the absence of
depreciation more costly. We also calibrate the model with an alternative level of
leverage and show that the ranking of the regimes is overturned with higher leverage.
For example, when the leverage is higher, ¸ = 0.ò, the …xed exchange rate regime
results in higher welfare in the face of a global …nancial shock regardless of the level
of …nancial contagion, as is seen in the last two columns of Table 6. When the initial
leverage is high, the …nancial accelerator e¤ect ampli…es, so is the impact of the
crisis, leading to the …xed exchange rate regime delivering a better welfare outcome
(negative w), despite losses in domestic demand.
28
5 Conclusions
This paper has developed a two-country DSGE model to investigate the transmis-
sion of a global …nancial crisis to a small open emerging economy. Our framework
fully speci…es both trade and …nancial linkages between the countries to study some
important aspects of the recent global …nancial crisis experience.
We …nd that a small, open economy facing a sudden stop of capital in‡ows aris-
ing from …nancial distress in the global economy is likely to face a more prolonged
crisis than sudden stop episodes of domestic origin. This is largely attributable to
an important source of di¢culty in responding to a global …nancial shock - the in-
ability of countries to export their way out of a crisis due to the slump in world
consumer demand initiated by the global …nancial distress. In contrast, when the
…nancial shock is of domestic origin, the domestic economy bene…ts from the de-
preciation of its currency and the resulting current account reversal, which at least
partly compensates for the fall in economic activity. This bene…cial export channel
disappears and indeed works in the opposite direction when the rest of the world
also faces an unfavorable …nancial disturbance. The resulting contraction in output
in the foreign economy is transmitted to the domestic economy through a fall in ex-
port demand, further reducing aggregate demand for home produced goods. This,
in turn, is likely to increase the duration and the severity of crises for both countries
in question, as mutual reductions in export demand set in motion a vicious circle,
even in the absence of any protectionist policies.
Given the nature of our framework with both …nancial and trade linkages, our
results have clear implications for the degree of both the …nancial and the trade
integration on the ampli…cation of a global …nancial crisis for the domestic econ-
omy. We …nd that the lower the …nancial contagion from the global economy, the
less signi…cant the impact of a global …nancial shock on the domestic economy, as
expected. Low contagion enables the domestic country to recover from the global
…nancial shock rapidly on the back of capital ‡ows ‡eeing from the foreign towards
the domestic economy. We argue that this …nding may shed some light on why the
recent …nancial crisis has been relatively short-lived for a number of emerging market
countries, especially those with limited …nancial exposure to the …nancial distress
in the global economy. As capital in‡ows surged, these economies have experienced
rapid recovery in credit, asset prices, investment, and output. Regarding the role
29
of trade integration, in contrast to the existing literature, we …nd that the greater
a country’s trade integration with the rest of the world, the greater the response of
its macroeconomic aggregates to a sudden stop of capital ‡ows. This is because the
more open an economy the greater the impact of the fall in export demand from the
rest of the world, as was experienced by countries such as Singapore, Taiwan and
Turkey, all highly open economies registering among the highest output contractions
in 2009.
Our results also suggest that the degree of …nancial contagion, the degree of trade
openness and the scale of foreign currency denominated debt are key determinants
of how monetary policy regimes in‡uence the propagation of the global …nancial
shock. In the absence of …nancial spillovers between countries, risk premia are
lower, net worth is higher so are investment, consumption and output under the
…xed exchange rate regime, a result which is overturned in the presence of …nancial
contagion. Moreover, our welfare calculations reveal that the greater the openness
of an economy, the greater the costs associated with the …xed exchange rate regime.
We …nd that, in contrast, at high levels of liability dollarization the …xed exchange
rate regime improves welfare irrespective of the degree of …nancial spillovers, a result
deriving from the balance sheet channel dominating the export channel.
30
References
[1] Bernanke, B. S., M. Gertler, and S. Gilchrist (1999). The …nancial accelerator
in a quantitative business cycle framework. In J. B. Taylor and M. Woodford
(Eds.), Handbook of Macroeconomics, Volume 1C, Chapter 21, 1341–93. Ams-
terdam: North-Holland.
[2] Bleaney, M. (2005). The aftermath of a currency collapse: how di¤erent are
emerging markets? World Economy, 28(1), 79-89.
[3] Calvo, G., A. Izquierdo, and R. Loo-Kung (2006). Relative price volatility under
sudden stops: the relevance of balance sheet e¤ects. Journal of International
Economics, 69(1), 231-254.
[4] Campa, J. and L. Goldberg (2005). Exchange rate pass-through into import
prices. Review of Economics and Statistics, 87(4), 679-690.
[5] Carlstrom, C.T. and T. S. Fuerst (1997). Agency costs, net worth, and busi-
ness cycle ‡uctuations: a computable general equilibrium analysis. American
Economic Review 87, 893–910.
[6] Cespedes, L. F., R. Chang, and A. Velasco (2004). Balance sheets and exchange
rate policy. American Economic Review, 94, 1183–1193.
[7] Christiano, L. J., C. Gust, and J. Roldos (2004). Monetary policy in a …nancial
crisis. Journal of Economic Theory 119(1), 64–103.
[8] Cook, D. (2004). Monetary policy in emerging markets: Can liability dollar-
ization explain contractionary devaluations?Journal of Monetary Economics,
51(6), 1155-1181.
[9] Curdia, V. (2008). Optimal monetary policy under sudden stops. Sta¤ Report
no. 323, Federal Reserve Bank of New York.
[10] Dedola, L. and G. Lombardo (2012). Financial frictions, …nancial integration
and the international propagation of shocks. Economic Policy, 27(70), 319-359.
[11] Devereux, M. B., P. R. Lane, and J. Xu (2006). Exchange rates and monetary
policy in emerging market economies. The Economic Journal, 116, 478–506.
31
[12] Devereux, M. B. and J. Yetman (2010). Leverage constraints and the inter-
national transmission of shocks. Journal of Money, Credit and Banking, 42,
71-105.
[13] Devereux, M. B. and A. Sutherland (2011). Evaluating international …nancial
integration under leverage constraints. European Economic Review 55(3), 427-
442.
[14] Eichengreen, B. and R. Hausmann (2005). Other People’s Money: Debt De-
nomination and Financial Instability in Emerging Market Economies, The Uni-
versity of Chicago Press.
[15] Elekdag, S. and I. Tchakarov (2007). Balance sheets, exchange rate policy, and
welfare. Journal of Economic Dynamics & Control, 31, 3986–4015.
[16] Enders, Z., R. Kollmann and G. Muller (2011). Global banking and interna-
tional business cycles. European Economic Review, 55, 407-426.
[17] Faia, E. and T. Monacelli (2007). Optimal interest rate rules, asset prices and
credit frictions. Journal of Economic Dynamics & Control, 31, 3228-3254.
[18] Gertler, M., S. Gilchrist, and F. Natalucci (2007). External constraints on mone-
tary policy and the …nancial accelerator. Journal of Money, Credit and Banking,
39, 295–330.
[19] Gertler, M. and P. Karadi (2010). A model of unconventional monetary policy.
Journal of Monetary Economics, 58, 17-34.
[20] Gilchrist, S. (2004). Financial markets and …nancial leverage in a two-country
world economy. In Banking Market Structure and Monetary Policy. Central
Bank of Chile.
[21] Greenwood, J., Z. Hercowitz, and G. Hu¤man (1988). Investment, capacity
utilization, and the real business cycles. American Economic Review, 78 402-
417.
[22] International Monetary Fund (2011). World Economic Outlook: Tensions from
the Two-Speed Recovery: Unemployment, Commodities, and Capital Flows.
Washington, April.
32
[23] Kalemli-Ozcan,S., B. Sorensen and S. Yesiltas (2011). Leverage across …rms,
banks and countries. Paper prepared for NBER-MIT Global Financial Crisis
Conference, Bretton Woods, NH.
[24] Kaminsky, G.L. and C.M. Reinhart (2000). On crisis, contagion and confusion.
Journal of International Economics, 51, 145-168.
[25] Martin, P. and H. Rey (2006). Globalization and emerging markets: with or
without crash? American Economic Review, 96(5), 1631-1651.
[26] Mendoza, E. (1991). Real Business Cycles in a Small Open Economy. American
Economic Review, 81, 797–818.
[27] Milesi-Ferretti, G. and C. Tille (2011). Capital ‡ows and the crisis. Economic
Policy 26(66), 291-346.
[28] Monacelli, T. and R. Perotti (2008). Fiscal policy, wealth e¤ects and markups,
NBER Working Paper, No. 14584.
[29] Mishkin, F. S. (1998). The dangers of exchange rate pegging in emerging market
countries. International Finance, 1(1), 81-101.
[30] Moser, T. (2003). What is international …nancial contagion? International Fi-
nance, 6, 157-178.
[31] Naug, B. and R. Nmyoen (1996). Pricing to market in a small open economy.
Scandinavian Journal of Economics, 98(3), 329-350.
[32] Neumeyer, P.A., and F. Perri (2005). Business cycles in emerging markets: the
role of interest rates. Journal of Monetary Economics, 52(2), 345-380.
[33] Perri, F. and V. Quadrini (2008) International recessions. NBER Working Pa-
per, No. 17201.
[34] Reinhart, C. and K. Rogo¤ (2009). This Time is Di¤erent. Eight Centuries of
Financial Folly. The Princeton University Press.
[35] Rotemberg, J. (1982). Sticky prices in the United States. Journal of Political
Economy, 90, 1187–1211.
33
[36] Schmitt-Grohe, S. and M. Uribe (2003). Closing small open economy models.
Journal of International Economics, 61, 163–185.
[37] Schmitt-Grohe, S. and M. Uribe (2007). Optimal simple and implementable
monetary and …scal rules. Journal of Monetary Economics, Elsevier, 54(6),
1702-1725.
[38] Sims, C. (2005). Second-order accurate solution of discrete time dynamic equi-
librium models. Mimeo, Princeton University.
34
Appendix A - Optimal Contracting Problem
Entrepreneurs observe .
t+1
(/) ex post, but the lenders can only observe it at a
monitoring cost which is assumed to be a certain fraction (j) of the return. As shown
by Bernanke et al. (1999), the optimal contract between the lender and the entre-
preneur is a standard debt contract characterized by a default threshold, .
t+1
(/),
such that if .
t+1
(/) _ .
t+1
(/), the lender receives a …xed return in the form of a
contracted interest on the debt. If .
t+1
(/) < .
t+1
(/), then the borrower defaults,
the lender audits by paying the monitoring cost and keeps what it …nds. There-
fore, we can de…ne the expected return to entrepreneur and lender, respectively, as
follows:
1
t
[1
1
t+1
Q
t
1
t+1
(/)(
_
1
.
t+1
(I)
.(/))(.)d. ÷.
t+1
(/)
_
1
.
t+1
(I)
)(.)d.)[
= 1
t
[1
1
t+1
Q
t
1
t+1
(/).(.
t+1
(/))[, (A1)
1
t
[1
1
t+1
Q
t
1
t+1
(/)(.
t+1
(/)
_
1
.
t+1
(I)
)(.
)d.
÷(1 ÷j)
_
.
t+1
(I)
0
.
t+1
(/))(.
)d.
)[
= 1
t
[1
1
t+1
Q
t
1
t+1
(/)q(.
t+1
(/); ¸
t
)[, (A2)
where 1
1
t
denotes the ex-post realization of return to capital and .(.) is the bor-
rowers’ share of the total return. We use the de…nition of the lender’s perception of
productivity shock .
t+1
(/) in Equation (A2) where q(.(/); ¸) represents the lenders’
share of the total return, itself a function of both the idiosyncratic shock and the
perception factor.
We assume that each entrepreneur is subject to an idiosyncratic shock .
t
¸ [0, ·)
with 1(.
t
) = 1 and c.d.f and p.d.f are given by 1(.
t
) and )(.
t
). We de…ne .(.) as
the expected gross share of the proceeds going to the borrower (ignoring the time
subscript t and entrepreneur index / for notational simplicity):
.(.) =
_
1
.
.)(.)d. ÷.
_
1
.
)(.)d. (A3)
= 1 ÷
_
.
0
.)(.)d. ÷.
_
1
.
)(.)d.
= 1 ÷I(.),
35
where I(.) = [1 ÷1(.)[. ÷
_
.
0
.d1(.), following Bernanke et al. (1999).
Let 1
C
t
be the contractual rate speci…ed by the lender. By de…nition, the default
threshold (.
t+1
(/)) is set at the level of returns that is just enough to honor the
debt contract obligations, satisfying the following equation:
.
t+1
(/)1
1
t+1
Q
t
1
t+1
(/)
o
t+1
= 1
C
t+1
(/)1
1
t+1
(/) (A4)
Recall that the misperception of investors regarding the distribution of .
t+1
is rep-
resented by ¸
t
such that .
t+1
(/) = .
t+1
(/)¸
t
. As in Curdia (2008), we write the
participation constraint of investors (in foreign currency):
(1 ÷i
t
)1
1
t+1
(/) = 1
t
[(1 ÷1
(.
t+1
(/)))1
C
t+1
(/)1
1
t+1
(/)[
÷(1 ÷j)1
t
[
_
.
t+1
(I)
0
.
(/)d1
(.
(/))
1
1
t+1
Q
t
1
t+1
(/)
o
t+1
[.(A5)
De…ne:
1
(.) = Ii(.
_ .) (A6)
= Ii(.¸ _ .)
= Ii(. _
.
¸
)
= 1(
.
¸
).
We also de…ne G(.) =
_
.
0
.)(.)d. =
_
.
0
.d1(.) and note that G(.) = 1(.)1(.[. <
.). Then we similarly express:
G
(.) = 1
(.)1(.
[.
< .) (A7)
= 1(
.
¸
)1(.[. <
.
¸
)
= ¸G(
.
¸
).
Noting that the monitoring cost (·
t
) is given by jG
(.) and substituting (A4),
(A6) and (A7) into (A5) we get:
(1 ÷i
t
)1
1
t+1
(/) = 1
t
[
1
1
t+1
Q
t
1
t+1
(/)
o
t+1
¦(1 ÷1(
.
t+1
(/)
¸
t
)).
t+1
(/)
÷(1 ÷j)¸
t
G(
.
t+1
(/)
¸
t
)¦[ (A8)
36
which can be re-arranged to yield:
(1 ÷i
t
)1
1
t+1
(/) = 1
t
[q(.
t+1
(/); ¸
t
)
1
1
t+1
Q
t
1
t+1
(/)
o
t+1
[ (A9)
In (A9) q(.; ¸) is de…ned as
q(.; ¸) = ¸[I(
.
¸
) ÷jG(
.
¸
)[. (A10)
We assume that the aggregate risk in terms of the exchange rate and the return
to capital is borne by lenders such that the participation constraint holds with
expectations as in Cespedes et al. (2004), Elekdag and Tchakarov (2007) and Curdia
(2008). Therefore, it should be clear that return to capital (1
1
) and the cut o¤
value (.) are state contingent and the participation constraint holds ex post with
equality at each possible state.
We can now analyze the optimal contract which determines a state contingent
cut o¤ value .
t
and 1
t+1
(/) solving the following maximization problem:
max 1
t
[1
1
t+1
Q
t+1
1
t+1
(/).(.
t+1
(/))[
subject to the participation constraint (A8). The optimality conditions for this
maximization problem are:
1
t
[1
1
t+1
Q
t
.(.
t+1
(/))[ ÷1
t
[A
t+1
(
1
1
t+1
Q
t
q(.
t+1
(/); ¸
t
)
o
t+1
÷
(1 ÷i
t
)Q
t
o
t
)[ = 0 (A11)
A
t+1
(l) = ÷
¬(l).
0
(.
l
t+1
(/))o
l
t+1
q
0
(.
l
t+1
(/); ¸
t
)
(A12)
where l is the state of the world, ¬(l) is the probability of the state l and A
t+1
is
the Lagrangian multiplier. Substituting (A12) into (A11) yields:
1
t
[1
1
t+1
(
.
0
(.
t+1
(/))q(.
t+1
(/); ¸
t
)
q
0
(.
t+1
(/); ¸
t
)
÷.(.
t+1
(/)))[ (A13)
= (1 ÷i
t
)1
t
[
o
t+1
o
t
(
.
0
(.
t+1
(/))
q
0
(.
t+1
(/); ¸
t
)
)[
Given that entrepreneurs are identical ex ante, each entrepreneur faces the same
…nancial contract. We can then write the external risk premium (1÷1
t+1
) as follows:
37
1 ÷ 1
t+1
= [
.
0
(.
t+1
)
q(.
t+1
, ¸
t
).
0
(.
t+1
÷.(.
t+1
)q
0
(.
t+1
, ¸
t
)
[1
t
¦
o
t+1
o
t
¦. (A14)
Using (A14), (A13) can be re-written as 1
t
[1
1
t+1
[ = 1
t
[(1 ÷ i
t
)(1 ÷ 1
t+1
)[. These
two equations correspond to (13) and (12) in the text.
38
Appendix B - Model Equations
B.1. Domestic Economy
Households
C
1,t
= (1 ÷c)(
1
1,t
1
t
)
¸
C
t
(B1.1)
C
A,t
= c(
1
A,t
1
t
)
¸
C
t
(B1.2)
1
t
= [(1 ÷c)1
1¸
1,t
÷c1
1¸
A,t
[
1¸(1¸)
(B1.3)
H
,
t
= \
t
(B1.4)
(C
t
÷
H
1+,
t
1 ÷,
)
o
= ,(1 ÷i
t
)1
t
[(C
t+1
÷
H
1+,
t+1
1 ÷,
)
o
1
t
1
t+1
[ (B1.5)
(C
t
÷
H
1+,
t
1 ÷,
)
o
= ,(1 ÷i
t
)w
1,t
1
t
[(C
t+1
÷
H
1+,
t+1
1 ÷,
)
o
1
t
1
t+1
o
t+1
o
t
[. (B1.6)
Production Firms
\
t
=
(1 ÷j)(1 ÷\)1
t
'C
t
·
t
(B1.7)
\
1
t
= (1 ÷j)\1
t
'C
t
(B1.8)
1
t
=
j1
t
'C
t
1
t
(B1.9)
'C
t
=
1
j
t
\
1j
t
¹
t
j
j
(1 ÷j)
1j
(B1.10)
1
1,t
=
`
` ÷1
'C
t
÷
w
1
` ÷1
1
t
1
1,t
1
1,t
1
1,t1
(
1
1,t
1
1,t1
÷1)
÷
w
1
` ÷1
1
t
[O
t
1
t+1
1
1,t
1
1,t+1
1
1,t
(
1
1,t+1
1
1,t
÷1)[ (B1.11)
o
t
1
A,t
=
`
` ÷1
'C
t
÷
w
A
` ÷1
1
t
1
A,t
1
A,t
1
A,t1
(
1
A,t
1
A,t1
÷1)
÷
w
A
` ÷1
1
t
[O
t
1
t+1
1
A,t
1
A,t+1
1
A,t
(
1
A,t+1
1
A,t
÷1)[ (B1.12)
39
O
t
= ,
(C
t+1
÷
¸
1+,
H
1+,
t+1
)
o
(C
t
÷
¸
1+,
H
1+,
t
)
o
1
t
1
t+1
(B1.13)
Importing Firms
1
A,t
=
`
` ÷1
o
t
1
A,t
÷
w
A
` ÷1
1
t
1
A,t
1
A,t
1
A,t1
(
1
A,t
1
A,t1
÷1) (B1.14)
÷
w
A
` ÷1
1
t
[O
t
1
t+1
1
A,t
1
A,t+1
1
A,t
(
1
A,t+1
1
A,t
÷1)[
Un…nished Capital Producing Firms
1
1,t
= (1 ÷c)(
1
1,t
1
t
)
¸
1
t
(B1.15)
1
A,t
= c(
1
A,t
1
t
)
¸
1
t
(B1.16)
1
t+1
= [
1
t
1
t
÷
w
1
2
(
1
t
1
t
÷c)
2
[1
t
÷ (1 ÷c)1
t
(B1.17)
Q
t
1
t
= [1 ÷w
1
(
1
t
1
t
÷c)[
1
(B1.18)
Entrepreneurs
1
t
[
1
1
t+1
Q
t
1
t+1
(/)
o
t+1
q(.
t+1
(/); ¸
t
)[ = (1 ÷i
t
)1
t+1
(/) (B1.19)
1
t
·\
t
= 0[1
1
t
Q
t1
1
t
(1 ÷i
t
) ÷(1 ÷i
t1
)o
t
1
t
[ ÷\
1
t
(B1.20)
1
t
C
1
t
= (1 ÷0)[1
1
t
Q
t1
1
t
(1 ÷i
t
) ÷(1 ÷i
t1
)o
t
1
t
[ (B1.21)
C
1
1,t
= (1 ÷c)(
1
1,t
1
t
)
¸
C
1
t
(B1.22)
C
1
A,t
= c(
1
A,t
1
t
)
¸
C
1
t
(B1.23)
1
t
[1
1
t+1
[ = 1
t
[(1 ÷i
t
)(1 ÷ 1
t+1
)[ (B1.24)
40
1 ÷ 1
t+1
= [
.
0
(.
t+1
)
q(.
t+1
; ¸
t
).
0
(.
t+1
) ÷.(.
t+1
)q
0
(.
t+1
; ¸
t
)
[1
t
¦
o
t+1
o
t
¦ (B1.25)
1
t
[1
1
t+1
[ = 1
t
[
1
t+1
Q
t
÷
Q
t+1
Q
t
¦(1÷c)÷w
1
(
1
t+1
1
t+1
÷c)
1
t+1
1
t+1
÷
w
1
2
(
1
t+1
1
t+1
÷c)
2
¦[ (B1.26)
Let .
1t
=
ln .
t
+0.5o
2
!
o
!
and .
2t
=
ln(
!
t
%
t
)+0.5o
2
!
o
!
, and O be a c.d.f of a standard normal
distribution:
oif(r) = 2O(r
_
2) ÷1 (B1.27)
oif c(r) = 2(1 ÷O(r
_
2))
.(.
t
) = 1 ÷(.
1t
÷o
.
) ÷.
t
(1 ÷O(.
1t
)) (B1.28)
q(.
t
; ¸
t
) = ¸
t
[
.
t
¸
t
(1 ÷O(.
2t
)) ÷ (1 ÷j)O(.
2t
÷o
.
)[ (B1.29)
.
0
(.
t
) = ÷(1 ÷O(.
1t
)) (B1.30)
q
0
(.
t
; ¸
t
) = ¸
t
[1 ÷O(.
2t
) ÷
j
_
2¬o
.
oxp(÷
.
2
2t
2
) (B1.31)
Monetary Policy
1 ÷i
t
= (1 ÷i) (¬
t
)
c
(1
t
,1 )
c
Y
(B1.32)
General Equilibrium
1
t
= 1
1,t
÷1
A,t
(B1.33)
1
1,t
= C
1,t
÷C
1
1,t
÷1
1,t
÷ (1 ÷c)(
1
1,t
1
t
)
¸
[
i=1,A
w
i
2
(
1
i,t
1
i,t1
÷1)
2
÷
w
A
2
(
1
A,t
1
A,t1
÷1)
2
÷i
t
1
1
t
1
t
Q
t1
1
t
[ (B1.34)
1
A,t
= C
A,t
÷C
1
A,t
÷1
A,t
÷ (c)(
1
A,t
1
t
)
¸
[
i=1,A
w
i
2
(
1
i,t
1
i,t1
÷1)
2
÷
w
A
2
(
1
A,t
1
A,t1
÷1)
2
÷i
t
1
1
t
1
t
Q
t1
1
t
[ (B1.35)
41
o
t
1
A,t
1
A,t
÷1
A,t
1
A,t
= o
t
(1 ÷i
t1
)(1
1
t
w
1,t1
÷1
t
) ÷o
t
(1
1
t+1
÷1
t+1
) (B1.36)
B.2. Model Equations: Foreign Economy
Households
C
1,t
= (1 ÷c
)(
1
1,t
1
t
)
¸
C
t
(B2.1)
C
A,t
= c
(
1
A,t
1
t
)
¸
C
t
(B2.2)
1
t
= [(1 ÷c
)1
1¸
1,t
÷c
1
1¸
A,t
[
1¸(1¸
)
(B2.3)
H
,
t
= \
t
(B2.4)
(C
t
÷
H
1+,
t
1 ÷,
)
o
= ,(1 ÷i
t
)1
t
[(C
t+1
÷
H
1+,
t+1
1 ÷,
)
o
1
t
1
t+1
[ (B2.5)
Production Firms
\
t
=
(1 ÷j)(1 ÷\)1
t
'C
t
·
t
(B2.6)
\
1
t
= (1 ÷j)\1
t
'C
t
(B2.7)
1
t
=
j1
t
'C
t
1
t
(B2.8)
1
1,t
=
`
` ÷1
'C
t
÷
w
1
` ÷1
1
t
1
1,t
1
1,t
1
1,t1
(
1
1,t
1
1,t1
÷1)
÷
w
1
` ÷1
1
t
[O
t
1
t+1
1
1,t
1
1,t+1
1
1,t
(
1
1,t+1
1
1,t
÷1)[ (B2.9)
o
t
1
A,t
=
`
` ÷1
'C
t
÷
w
A
` ÷1
1
t
1
A,t
1
A,t
1
A,t1
(
1
A,t
1
A,t1
÷1)
÷
w
A
` ÷1
1
t
[O
t
1
t+1
1
A,t
1
A,t+1
1
A,t
(
1
A,t+1
1
A,t
÷1)[ (B2.10)
O
t
= ,
(C
t+1
÷
¸
1+,
H
1+,
t+1
)
o
(C
t
÷
¸
1+,
H
1+,
t
)
o
1
t
1
t+1
(B2.11)
42
Importing Firms
1
A,t
=
`
` ÷1
1
A,t
÷
w
A
` ÷1
1
t
1
A,t
1
A,t
1
A,t1
(
1
A,t
1
A,t1
÷1) (B2.12)
÷
w
A
` ÷1
1
t
[O
t
1
t+1
1
A,t
1
A,t+1
1
A,t
(
1
A,t+1
1
A,t
÷1)[
Un…nished Capital Producing Firms
1
1,t
= (1 ÷c
)(
1
1,t
1
t
)
¸
1
t
(B2.13)
1
A,t
= (c
)(
1
A,t
1
t
)
¸
1
t
(B2.14)
1
t+1
= [
1
t
1
t
÷
w
1
2
(
1
t
1
t
÷c)
2
[1
t
÷ (1 ÷c)1
t
(B2.15)
Q
t
1
t
= [1 ÷w
1
(
1
t
1
t
÷c)[
1
(B2.16)
Entrepreneurs
1
t
[1
1
t+1
Q
t
1
t+1
q
(.
t+1
; ¸
t
)[ = (1 ÷i
t
)1
t+1
(B2.17)
1
t
·\
t
= 0[1
1
t
Q
t1
1
t
(1 ÷i
t
) ÷(1 ÷i
t1
)1
t
[ ÷\
1
t
(B2.18)
1
t
C
1
t
= (1 ÷0)[1
1
t
Q
t1
1
t
(1 ÷i
t
) ÷(1 ÷i
t1
)1
t
[ (B2.19)
C
1
1,t
= (1 ÷c
)(
1
1,t
1
t
)
¸
C
1
t
(B2.20)
C
1
A,t
= c
(
1
A,t
1
t
)
¸
C
1
t
(B2.21)
1
t
[1
1
t+1
[ = 1
t
[(1 ÷i
t
)(1 ÷ 1
t+1
)[ (B2.22)
43
1 ÷ 1
t+1
= [
.
0
(.
t+1
(/))
q
(.
t+1
(/); ¸
t
).
0
(.
t+1
(/)) ÷.
(.
t+1
(/))q
0
(.
t+1
(/); ¸
t
)
[ (B2.23)
Let .
1t
=
ln .
t
+0.5o
2
!
o
!
and .
2t
=
ln(
!
t
%
t
)+0.5o
2
!
o
!
, and O be a c.d.f of a standard
normal distribution as above:
.
(.
t
) = 1 ÷O(.
1t
÷o
.
) ÷.
t
(1 ÷O(.
1t
)) (B2.24)
q
(.
t
; ¸
t
) = ¸
t
[
.
t
¸
t
(1 ÷O(.
2t
)) ÷ (1 ÷j)O(.
2t
÷o
.
)[ (B2.25)
.
0
(.
t
) = ÷(1 ÷O(.
1t
)) (B2.26)
q
0
(.
t
; ¸
t
) = ¸
t
[1 ÷O(.
2t
) ÷
j
_
2¬o
.
oxp(÷
.
2
2t
2
) (B2.27)
1
t
[1
1
t+1
[ = 1
t
[
1
t+1
Q
t
÷
Q
t+1
Q
t
¦(1÷c)÷w
1
(
1
t+1
1
t+1
÷c)
1
t+1
1
t+1
÷
w
1
2
(
1
t+1
1
t+1
÷c)
2
¦[. (B2.28)
Monetary Policy
1 ÷i
t
= (1 ÷i
) (¬
t
)
c
(1
t
,1
)
c
Y
(B2.29)
General Equilibrium
o
t
= 1,o
t
(B2.30)
1
t
= 1
1,t
÷1
A,t
, (B2.31)
1
1,t
= C
1,t
÷C
1
1,t
÷1
1,t
÷ (1 ÷c
)(
1
1,t
1
t
)
¸
[
i=1,A
w
i
2
(
1
i,t
1
i,t1
÷1)
2
÷
w
A
2
(
1
A,t
1
A,t1
÷1)
2
÷i
t
1
1
t
1
t
Q
t1
1
t
[, (B2.32)
1
A,t
= C
A,t
÷C
1
A,t
÷1
A,t
÷c
(
1
A,t
1
t
)
¸
[
i=1,A
w
i
2
(
1
i,t
1
i,t1
÷1)
2
÷
w
A
2
(
1
A,t
1
A,t1
÷1)
2
÷i
t
1
1
t
1
t
Q
t1
1
t
[. (B2.33)
44
Table 1: Parameter Values for Consumption, Production and Monetary Policy
: = 0.1 Size of the domestic economy
, = 0.00 Discount factor
o = 1 Inverse of the intertemporal elasticity of substitution
¸ = 1 Elasticity of substitution between domestic and foreign goods
, = 2 Frisch elasticity of labour supply
· = 0.8ò Degree of openness
j = 0.8ò Share of capital in production
` = 11 Elasticity of substitution between domestic goods
c = 0.02ò Quarterly rate of depreciation
\ = 0.01 Share of entrepreneurial labor
w
1
= 12 Investment adjustment cost
w
1
= 0.007ò Responsiveness of household risk premium to debt/GDP
w
i
, w
A
= 120 Price adjustment costs for i = H, A
c
¬
= 1.ò Coe¢cient of CPI in‡ation in the policy rule
c
Y
= 0.ò Coe¢cient of output gap in the policy rule
j
¸
= 0.ò Persistence of the domestic perception shock
Table 2: Parameter Values for the Entrepreneurial Sector
Domestic Economy
¸ = 0.8 Leverage
1
t
= 0.02 External risk premium
j = 0.2 Monitoring cost
0 = 0.0088 Survival rate
Foreign Economy
¸
= 0.ò Leverage
1
t
= 0.00ò External risk premium
j
= 0.12 Monitoring cost
0 = 0.0066 Survival rate
45
Output Consumption Investment Current account Inflation Interest rate Exchange rate
Argentina 4.58 5.95 12.94 1.01 2.94 2.17 5.77
Brazil 1.94 1.95 4.89 2.19 2.33 2.90 11.65
Korea 2.57 3.52 5.49 3.40 1.27 1.19 9.80
Mexico 2.55 3.57 6.98 5.80 1.70 1.65 9.96
Philippines 2.58 1.93 7.03 4.24 2.13 2.17 6.75
Average 2.84 3.38 7.47 3.33 2.07 2.02 8.78
Model - Domestic shock 3.15 3.24 17.20 3.61 0.60 2.16 3.23
Model- Global shock ?=0.5 2.91 2.23 15.95 2.12 0.67 2.27 1.60
ii) Standard deviations relative to output
Output Consumption Investment Current account Inflation Interest rate Exchange rate
Argentina 1.00 1.30 2.83 0.22 0.64 0.47 1.26
Brazil 1.00 1.01 2.52 1.13 1.20 1.49 6.00
Korea 1.00 1.37 2.14 1.32 0.50 0.46 3.81
Mexico 1.00 1.40 2.74 2.27 0.67 0.65 3.91
Philippines 1.00 0.75 2.72 1.64 0.82 0.84 2.61
Average 1.00 1.16 2.59 1.32 0.77 0.78 3.52
Model - Domestic shock 1.00 1.03 5.46 1.15 0.19 0.69 1.03
Model- Global shock ?=0.5 1.00 0.77 5.48 0.73 0.23 0.78 0.55
iii) Correlations with Output and Autocorrelations
?(C,Y) ?(I,Y) ?(CA,Y) ?(Y(t),Y(t-1)) ?(C(t),C(t-1)) ?(I(t),I(t-1)) ?(CA(t),CA(t-1))
Argentina 0.92 0.83 -0.54 0.83 0.85 0.91 0.93
Brazil 0.77 0.38 -0.03 0.35 0.17 0.55 0.92
Korea 0.87 0.86 -0.72 0.80 0.82 0.79 0.86
Mexico 0.78 0.85 -0.45 0.82 0.80 0.85 0.83
Philippines 0.82 0.10 0.01 0.78 0.71 0.59 0.77
Average 0.83 0.61 -0.35 0.72 0.67 0.74 0.86
Model - Domestic shock 0.93 0.96 -0.93 0.51 0.63 0.51 0.42
Model- Global shock ?=0.5 0.68 0.91 -0.67 0.71 0.92 0.62 0.77
Table 3. Business Cycles in Emerging Economies: Data vs. Model
i) Standard deviations (in %)
46
Output Consumption Investment Current account Inflation Interest rate Exchange rate
E.U. 1.37 0.81 3.80 0.50 0.61 1.02 7.09
U.S. 1.22 0.88 4.26 1.26 0.16 0.28 3.01
Average 1.30 0.85 4.03 0.88 0.39 0.65 5.05
Model- Global shock ?=0.5 0.73 0.56 5.43 0.64 0.12 0.23 1.94
ii) Standard deviations relative to output
Output Consumption Investment Current account Inflation Interest rate Exchange rate
E.U. 1.00 0.59 2.77 0.36 0.45 0.74 5.18
U.S. 1.00 0.72 3.49 1.03 0.13 0.23 2.47
Average 1.00 0.66 3.13 0.70 0.29 0.49 3.82
Model- Global shock ?=0.5 1.00 0.77 7.44 0.88 0.16 0.32 2.66
iii) Correlations with output and autocorrelations
?(C*,Y*) ?(I*,Y*) ?(CA*,Y*) ?(Y*(t),Y*(t-1)) ?(C*(t),C*(t-1)) ?(I*(t),I*(t-1)) ?(CA*(t),CA*(t-1))
E.U. 0.56 0.95 -0.36 0.91 0.67 0.91 0.71
U.S. 0.88 0.91 -0.71 0.88 0.85 0.93 0.75
Average 0.72 0.93 -0.54 0.90 0.76 0.92 0.73
Model- Global shock ?=0.5 0.43 0.80 0.66 0.71 0.76 0.47 0.77
Table 4. Business Cycles in Advanced (Big) Economies: Data vs. Model
i) Standard deviations (in %)
47
With respect to the U.S.
Output Consumption Investment Output Consumption Investment
Argentina 0.49 0.33 0.67 0.89 0.61 0.95
Brazil 0.30 0.02 0.17 0.80 0.57 0.28
Korea 0.11 0.15 -0.14 0.86 0.82 0.16
Mexico 0.66 0.37 0.34 0.96 0.75 0.83
Phillipines 0.42 0.00 -0.12 0.66 0.34 -0.08
Average 0.40 0.17 0.19 0.84 0.62 0.43
Model- Global shock ?=0 0.63 -0.13 -0.75 0.63 -0.13 -0.75
Model- Global shock ?=0.5 0.94 0.35 0.96 0.94 0.35 0.96
Model- Global shock ?=1 0.98 0.61 0.98 0.98 0.61 0.98
Table 5. Cross-Country Correlations
Global crisis period (2007Q1-2011Q3) Full sample (1990Q1-2011Q3)
48
Welfare ? Welfare ? Welfare ?
i) Domestic shock
Inflation targeting -59.1945 ? -124.5435 ? -60.9214 -
Fixed exchange rate regime -59.2231 0.0283 -124.5585 0.0090 -60.9467 0.0146
ii) Foreign shock
a) Contagion (?=1)
Inflation targeting -59.1854 - -124.4821 - -60.8783 -
Fixed exchange rate regime -59.2345 0.0403 -124.5549 0.0434 -60.8744 -0.0025
Inflation targeting -59.1517 ? -124.4877 - -60.8401 -
Fixed exchange rate regime -59.1874 0.0324 -124.5421 0.0325 -60.8066 -0.0197
Inflation targeting -59.1367 - -124.4968 -60.8169 -
Fixed exchange rate regime -59.1358 -0.0121 -124.5298 0.0196 -60.7425 -0.0435
b) Contagion (?=0.5)
c) No contagion (?=0)
Table 6. Welfare Results
Baseline Higher openness Higher leverage
(?= 0.35, ?= 0.3) (?=0.6, ?=0.3) (?= 0.35, ?=0.5)
49
(percent deviations from the steady state)
Figure 1. Responses to a Financial Crisis in Domestic Economy: Domestic Economy*
?1.8
-1.2
-0.6
0.0
0.6
1.2
0 5 10
Output
-9
-6
-3
0
3
6
0 5 10
Investment
-1.5
-1.0
-0.5
0.0
0.5
1.0
0 5 10
Consumption
-3
-2
-1
0
1
2
0 5 10
Asset Prices
-4
-2
0
2
4
6
0 5 10
Risk Premium
-0.9
-0.6
-0.3
0.0
0.3
0.6
0 5 10
Entreprenurs' Debt
-4.5
-3.0
-1.5
0.0
1.5
3.0
0 5 10
Net Worth
-1.0
-0.5
0.0
0.5
1.0
1.5
0 5 10
Exports
-2.4
-1.6
-0.8
0.0
0.8
1.6
0 5 10
Capital Inflows
Real Exchange Rate CPI Inflation Interest Rate
50
*The figures show the impact of a 1% negative shock to the perception of investors regarding the productivity of
domestic entrepreneurs. The variables are presented as log-deviations from the steady state (except for interest
rate), multiplied by 100 to have an interpretation of percentage deviations.
-0.9
-1.0
-2.4
-1.6
-0.8
0.0
0.8
1.6
2.4
0 5 10
Real Exchange Rate
-0.3
-0.2
-0.1
0.0
0.1
0.2
0 5 10
CPI Inflation
-1.2
-0.8
-0.4
0.0
0.4
0.8
0 5 10
Interest Rate
Figure 2. Responses to a Financial Crisis in Foreign Economy: Foreign Economy*
(percent deviations from the steady state)
?0.6
-0.4
-0.2
0.0
0.2
0.4
0 5 10
Output
-4.5
-3.0
-1.5
0.0
1.5
3.0
0 5 10
Investment
-0.21
-0.14
-0.07
0.00
0.07
0.14
0 5 10
Consumption
-1.2
-0.8
-0.4
0.0
0.4
0.8
0 5 10
Asset Prices
-5.0
-2.5
0.0
2.5
5.0
7.5
0 5 10
Risk Premium
-2.7
-1.8
-0.9
0.0
0.9
1.8
0 5 10
Entreprenurs' Debt
-2.4
-1.6
-0.8
0.0
0.8
1.6
0 5 10
Net Worth
-1.4
-0.7
0.0
0.7
1.4
2.1
0 5 10
Exports
-0.45
-0.30
-0.15
0.00
0.15
0.30
0 5 10
Capital Inflows
2.1
Real Exchange Rate
0.6
CPI Inflation
1.0
Interest Rate
51
*The figures show the impact of a 1% negative shock to the perception of investors regarding the productivity of
foreign entrepreneurs. The variables are presented as log-deviations from the steady state (except for interest rate),
multiplied by 100 to have an interpretation of percentage deviations.
1.4 -0.45
-1.4
-0.7
0.0
0.7
1.4
2.1
0 5 10
Real Exchange Rate
-0.9
-0.6
-0.3
0.0
0.3
0.6
0 5 10
CPI Inflation
-1.5
-1.0
-0.5
0.0
0.5
1.0
0 5 10
Interest Rate
Figure 3. Responses to a Financial Crisis in Foreign Economy with Financial Contagion:
(percent deviations from the steady state)
Domestic Economy*
?2.4
-1.6
-0.8
0.0
0.8
1.6
0 5 10
Output
?=0.5
?=1
-15
-10
-5
0
5
10
0 5 10
Investment
-2.1
-1.4
-0.7
0.0
0.7
1.4
0 5 10
Consumption
-9
-6
-3
0
3
6
0 5 10
Net Worth
-8
-4
0
4
8
12
0 5 10
Risk Premium
-4.8
-3.2
-1.6
0.0
1.6
3.2
0 5 10
Asset Prices
-1.8
-1.2
-0.6
0.0
0.6
1.2
0 5 10
Entrepreneurs' Debt
-1.2
-0.6
0.0
0.6
1.2
1.8
0 5 10
Exports
-3
-2
-1
0
1
2
0 5 10
Capital Inflows
3
Real Exchange Rate
0 4
CPI Inflation
1 6
Interest Rate
52
*The figures show the impact of a 0.75% negative shock to the perception of investors regarding the productivity
of foreign entrepreneurs. The variables are presented as log-deviations from the steady state (except for interest
rate), multiplied by 100 to have an interpretation of percentage deviations.
-1.8
-1.2
-3
-2
-1
0
1
2
3
0 5 10
Real Exchange Rate
-0.6
-0.4
-0.2
0.0
0.2
0.4
0 5 10
CPI Inflation
-2.4
-1.6
-0.8
0.0
0.8
1.6
0 5 10
Interest Rate
Figure 4. Responses to a Financial Crisis in Foreign Economy without Financial Contagion:
Domestic Economy*
(percent deviations from the steady state)
?0.09
-0.06
-0.03
0.00
0.03
0.06
0 5 10
Output
-0.6
-0.3
0.0
0.3
0.6
0.9
0 5 10
Investment
-0.3
-0.2
0.0
0.2
0.3
0.5
0 5 10
Consumption
-0.4
-0.2
0.0
0.2
0.4
0.6
0 5 10
Asset Prices
-3
-2
0
2
3
5
0 5 10
Risk Premium
-0.6
-0.3
0.0
0.3
0.6
0.9
0 5 10
Entreprenurs' Debt
-0.4
-0.2
0.0
0.2
0.4
0.6
0 5 10
Net Worth
-1.5
-1.0
-0.5
0.0
0.5
1.0
0 5 10
Net Exports
-0.8
-0.4
0.0
0.4
0.8
1.2
0 5 10
Capital Inflows
Real Exchange Rate CPI Inflation Interest Rate
53
*The figures show the impact of a 0.75% negative shock to the perception of investors regarding the productivity
of foreign entrepreneurs. The variables are presented as log-deviations from the steady state (except for interest
rate), multiplied by 100 to have an interpretation of percentage deviations.
-0.6 -1.5
-0.8
-0.9
-0.6
-0.3
0.0
0.3
0.6
0 5 10
Real Exchange Rate
-0.09
-0.06
-0.03
0.00
0.03
0.06
0 5 10
CPI Inflation
-0.15
-0.10
-0.05
0.00
0.05
0.10
0 5 10
Interest Rate
Figure 5. Responses to a Financial Crisis in Foreign Economy with Financial Contagion:
Domestic Economy- The Impact of Openness*
(percent deviations from the steady state)
?2.4
-1.6
-0.8
0.0
0.8
1.6
0 5 10
Output
? =0.6
? =0.35
? =0.1
-18
-12
-6
0
6
12
0 5 10
Investment
-2.4
-1.6
-0.8
0.0
0.8
1.6
0 5 10
Consumption
-7.5
-5.0
-2.5
0.0
2.5
5.0
0 5 10
Net Worth
-10
-5
0
5
10
15
0 5 10
Risk Premium
-6.0
-4.0
-2.0
0.0
2.0
4.0
0 5 10
Asset Prices
-2.4
-1.6
-0.8
0.0
0.8
1.6
0 5 10
Entrepreneurs' Debt
-1.6
-0.8
0.0
0.8
1.6
2.4
0 5 10
Exports
-3
-2
-1
0
1
2
0 5 10
Capital Inflows
3
Real Exchange Rate
1 0
CPI Inflation
2 0
Interest Rate
54
*The figures show the impact of a 0.75% negative shock to the perception of investors regarding the productivity
of foreign entrepreneurs. The variables are presented as log-deviations from the steady state (except for interest
rate), multiplied by 100 to have an interpretation of percentage deviations.
-2.4 -1.6
-3
-2
-1
0
1
2
3
0 5 10
Real Exchange Rate
-1.5
-1.0
-0.5
0.0
0.5
1.0
0 5 10
CPI Inflation
-3.0
-2.0
-1.0
0.0
1.0
2.0
0 5 10
Interest Rate
Figure 6. Responses to a Financial Crisis in Foreign Economy without Financial Contagion:
Domestic Economy- The Impact of Openness*
(percent deviations from the steady state)
?0.45
-0.30
-0.15
0.00
0.15
0.30
0.45
0 5 10
Output
? =0.6 ? =0.35
? =0.1
-0.9
-0.6
-0.3
0.0
0.3
0.6
0.9
1.2
0 5 10
Investment
-0.4
-0.2
0.0
0.2
0.4
0.6
0 5 10
Consumption
-0.6
-0.3
0.0
0.3
0.6
0.9
0 5 10
Net Worth
-3
-2
0
2
3
5
0 5 10
Risk Premium
-0.30
-0.15
0.00
0.15
0.30
0.45
0 5 10
Asset Prices
-0.8
-0.4
0.0
0.4
0.8
1.2
0 5 10
Entrepreneurs' Debt
-2.1
-1.4
-0.7
0.0
0.7
1.4
0 5 10
Exports
-1
0
0
0
1
1
0 5 10
Capital Inflows
1 60
Real Exchange Rate
0 08
CPI Inflation
0 2
Interest Rate
55
*The figures show the impact of a 0.75% negative shock to the perception of investors regarding the productivity
of foreign entrepreneurs. The variables are presented as log-deviations from the steady state (except for interest
rate), multiplied by 100 to have an interpretation of percentage deviations.
-0.8 -2.1
-1
-2.40
-1.60
-0.80
0.00
0.80
1.60
0 5 10
Real Exchange Rate
-0.12
-0.08
-0.04
0.00
0.04
0.08
0 5 10
CPI Inflation
-0.3
-0.2
-0.1
0.0
0.1
0.2
0 5 10
Interest Rate
Figure 7. Responses to a Financial Crisis in Foreign Economy with Financial Contagion:
Domestic Economy- The Role of Monetary Policy Strategy*
(percent deviations from the steady state)
?5.1
-3.4
-1.7
0.0
1.7
3.4
0 5 10
Output
Fixed exchange rate regime
Inflation targeting
-21
-14
-7
0
7
14
0 5 10
Investment
-5.4
-3.6
-1.8
0.0
1.8
3.6
0 5 10
Consumption
-12
-8
-4
0
4
8
0 5 10
Net Worth
-10
-5
0
5
10
15
0 5 10
Risk Premium
-7.5
-5.0
-2.5
0.0
2.5
5.0
0 5 10
Asset Prices
-1.8
-1.2
-0.6
0.0
0.6
1.2
0 5 10
Entrepreneurs' Debt
-1.2
-0.6
0.0
0.6
1.2
1.8
0 5 10
Exports
-5
-3
-2
0
2
3
0 5 10
Capital Inflows
3
Real Exchange Rate
0 6
CPI Inflation
1 6
Interest Rate
56
*The figures show the impact of a 0.75% negative shock to the perception of investors regarding the productivity
of foreign entrepreneurs. The variables are presented as log-deviations from the steady state (except for interest
rate), multiplied by 100 to have an interpretation of percentage deviations.
-1.8
-1.2 -5
-2
-1
0
1
2
3
0 5 10
Real Exchange Rate
-0.9
-0.6
-0.3
0.0
0.3
0.6
0 5 10
CPI Inflation
-2.4
-1.6
-0.8
0.0
0.8
1.6
0 5 10
Interest Rate
Figure 8. Responses to a Financial Crisis in Foreign Economy without Financial Contagion:
Domestic Economy- The Role of Monetary Policy Strategy*
(percent deviations from the steady state)
?1.0
-0.5
0.0
0.5
1.0
1.5
0 5 10
Output
Fixed exchange rate regime
Inflation targeting
-2
-1
0
1
2
3
0 5 10
Investment
-1.2
-0.6
0.0
0.6
1.2
1.8
0 5 10
Consumption
-1.0
-0.5
0.0
0.5
1.0
1.5
0 5 10
Net Worth
-3
-2
0
2
3
5
0 5 10
Risk Premium
-0.6
-0.3
0.0
0.3
0.6
0.9
0 5 10
Asset Prices
-0.6
-0.3
0.0
0.3
0.6
0.9
0 5 10
Entrepreneurs' Debt
-1.5
-1.0
-0.5
0.0
0.5
1.0
0 5 10
Exports
-1.0
-0.5
0.0
0.5
1.0
1.5
0 5 10
Capital Inflows
1 0
Real Exchange Rate
0 3
CPI Inflation
0 8
Interest Rate
57
*The figures show the impact of a 0.75% negative shock to the perception of investors regarding the productivity
of foreign entrepreneurs. The variables are presented as log-deviations from the steady state (except for interest
rate), multiplied by 100 to have an interpretation of percentage deviations.
-0.6
-1.5 -1.0
-1.5
-1.0
-0.5
0.0
0.5
1.0
0 5 10
Real Exchange Rate
-0.5
-0.3
-0.2
0.0
0.2
0.3
0 5 10
CPI Inflation
-1.2
-0.8
-0.4
0.0
0.4
0.8
0 5 10
Interest Rate
doc_978290818.pdf