# Lectures in Open Economy Macroeconomics1

Mart´ Uribe2 ın This draft, November 24, 2007

Preliminary and incomplete. I would like to thank Javier Garc´ ıa-Cicco, Felix Hammermann, and Stephanie Schmitt-Groh´ for comments and suggestions. Newer e versions of these notes are available at www.econ.duke.edu/∼uribe. Comments welcome. 2 Duke University and NBER. E-mail: uribe@duke.edu.

1

ii

Contents

1 A First Look at the Data 2 An 2.1 2.2 2.3 2.4 Endowment Economy The Model Economy . . . . . . . . . Response to Output Shocks . . . . . Nonstationary Income Shocks . . . Testing the Model . . . . . . . . . . 1 5 5 11 13 18 23 23 26 29 30 32 34 37 38 46 48 49 50 52 54 56 58 61 61 65

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

3 An Economy with Capital 3.1 The Basic Framework . . . . . . . . . . 3.1.1 A Permanent Productivity Shock 3.1.2 A Temporary Productivity shock 3.2 Capital Adjustment Costs . . . . . . . . 3.2.1 Dynamics of the Capital Stock . 3.2.2 A Permanent Technology Shock .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

4 The Real Business Cycle Model 4.1 The Model . . . . . . . . . . . . . . . . . . . . . . . 4.2 The Model’s Performance . . . . . . . . . . . . . . . 4.2.1 The Role of Capital Adjustment Costs . . . . 4.3 Alternative Ways to Induce Stationarity . . . . . . . 4.3.1 External Discount Factor (EDF) . . . . . . . 4.3.2 External Debt-Elastic Interest Rate (EDEIR) 4.3.3 Internal Debt-Elastic Interest Rate (IDEIR) 4.3.4 Portfolio Adjustment Costs (PAC) . . . . . . 4.3.5 Complete Asset Markets (CAM) . . . . . . . 4.3.6 The Nonstationary Case (NC) . . . . . . . . 4.3.7 Quantitative Results . . . . . . . . . . . . . . 4.4 Appendix A: Log-Linearization . . . . . . . . . . . . iii

. . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. .1 An Empirical Model . 131 6. Impulse Response Functions . . . . . . .4 Production of Importable. . . . . . . . .2 Impulse Response Functions . . . . . 145 6. . . . . .4. 5. . . . . . . . . . . . . . . . .3. . Second Moments . . . . . . . . . . .9 4. .4. . . . . . .3. .3. . . . . . .11
CONTENTS Appendix B: Solving Dynamic General Equilibrium Models . . . . . . . . . . . . . .2 Empirical Regularities .iv 4. . 147
. Matlab Code For Linear Perturbation Methods . . . . . . . . . 126 6. . . . . 5. . . . . . . 5. . . . . . Exercise . . . . . . . . . . . . 5. . . . . . . . . .10 How Important Are the Terms of Trade? . . . . . . . . .2 Firms . . . . . . . . . . . . . 5. . .3 Production of Tradable Consumption Goods . . . . . . . . . . . . . . . . . .1 Households . . . . . . . . . .7 Competitive Equilibrium . . . . . . 120 6. . . . . 137 6. . . . .3. . . . 5. . . . . . Exportable. . . .4 Equilibrium. . . . . . . 118 6. . . . . . . . . . . . . . . . . . . . Functional Forms. . . . . . . . . . . . 5. . 5.5 Theoretical and Estimated Impulse Responses . . . . . . . . . . . . . . . . .4. . . . . . . . . . . . . . . 132 6. . . .2. .
6 Interest-Rate Shocks 115 6. . . . . . . . . .8 Calibration . . . . . . . . . . . . . . . . .3. . . . .7 4. . . . . and Nontradable Goods . . . .6 The Endogeneity of Country Spreads .9 Model Performance . . .3. . . . . . .1 TOT-TB Correlation: Two Early Explanations . . . .4 A Theoretical Model . . . . . . . . . 5. .3 Terms-of-Trade Shocks in an RBC Model . . Higher Order Approximations . . . . .6 4. . . . . . . . . . . . . . . . . .3. . . . . . . . . . . . . . . . . 4. . . . . . . . . 5. . . . . . . . . .2 Production of Consumption Goods . . . . . . . .3 Variance Decompositions . .5 Market Clearing .3. . . . . . . . . . . . . . . . . . .1 Households . . . . . . 5. . .4. . .3 Driving Forces . . . . . . . . . .6 Driving Forces .11. . . . . . . . 5. . . . 141 6. . . . . . .5 4. . 5. . . . . . . . . .3. . . . .3. . . . . . . . .8 4. . . . and Parameter Values 142 6. . . . . . . . . . . . . .10 4. . . .1 An RBC Small Open Economy with an internal debtelastic interest-rate premium . Local Existence and Uniqueness of Equilibrium . . . . . . . . . . . . . . . 67 78 79 83 83 84 85 85 87 88 88 90 99 99 102 103 104 106 106 107 107 111 113
5 The Terms of Trade 5. . . . . . . . . .1 Deﬁning the Terms of Trade . . . .

. . 157 .
. . 7. . . . . . . 7. .
.1 Empirical Regularities .3 A Reputational Model of Sovereign Debt . . . . .2 The Cost of Default . . . . 152 . .
. . . . . . . . . . . . 7. . .CONTENTS 7 Sovereign Debt 7. . .4 Saving and the Breakdown of Reputational Lending
v 151 .
. . . . . . . . . . . . . . . . . . 166
. . 159 .

vi
CONTENTS
.

Table 1.Chapter 1
A First Look at the Data
In discussions of business cycles in small open economies. and examples of small open emerging economies are Argentina and Malaysia. The group of developed economies is typically deﬁned by countries with high income per capita. The volatility of detrended output is 4. Periods of economic boom (contraction) are characterized by relatively larger trade 1
. a critical distinction is between developed and emerging economies. Another remarkable diﬀerence between developing and developed countries suggested by the table is that the trade balance-to-output ratio is much more countercyclical in emerging countries than in developed countries. Examples of developed small open economies are Canada and Belgium.6 in Argentina and only 2.8 in Canada. and the group of emerging economies is composed of middle income countries. A striking diﬀerence between developed and emerging economies is that observed business cycles in emerging countries are about twice as volatile as in developed countries.1 illustrates this contrast by displaying key business-cycle properties in Argentina and Canada.

1 2.96 0. GDPt ) 1 1 0. xt−1 ) 0.70
0.2
Mart´ Uribe ın
Table 1.9 4.79 0.76 0.94 0.0 3.70
0.66
0.3 9.5 13.48 0. data on hours and productivity are limited to the manufacturing sector.80 0.7 corr(xt .54
0.31
0.59 0.6 2.37
Source: Mendoza (1991).84 -0.0 1. Kydland and Zarazaga (1997).64 -0.13 0.
.1: Business Cycles in Argentina and Canada Variable GDP Argentina Canada Consumption Argentina Canada Investment Argentina Canada TB/GDP Argentina Canada Hours Argentina Canada Productivity Argentina Canada σx 4. For Argentina.8 5.4 2.61 corr(xt .3 1.8 2.

The diﬀerences between the business cycles of Argentina and Canada turn out to hold much more generally between emerging and developed countries. whereas the reverse is the case in Canada. and productivity. For all countries.
. hours. Also. Two additional diﬀerences between the business cycle in Argentina and Canada are that in Argentina the correlation of the components of aggregate demand with GDP are twice as high as in Canada.Lectures in Open Economy Macroeconomics. A third diﬀerence between Canadian and Argentine business cycles is that in Argentina consumption appears to be more volatile than output at businesscycle frequencies. The data is detrended using a bandpass ﬁlter that leaves out all frequencies above 32 quarters and below 6 quarters. The table averages second moments of detrended data for 13 small emerging countries and 13 small developed countries (the list of countries appears at the foot of the table). emerging countries are signiﬁcantly more volatile and display a much more countercyclical trade-balance share than developed countries. In both countries. consumption is more volatile than output in emerging countries but less volatile than output in developed countries. the time series are at least 40 quarters long. The data shown in the table is broadly in line with the conclusions drawn from the comparison of business cycles in Argentina and Canada. Table 1. and that in Argentina hours and productivity are less correlated with GDP than in Canada. Chapter 1
3
deﬁcits (surpluses) in emerging countries than in developed countries. In particular. these variables move in tandem with output. One dimension along which business cycles in Argentina and Canada are similar is the procyclicality of consumption.2 displays average business cycle facts in developed and emerging economies. investment.

Emerging countries: Argentina.95 0.42 0. Spain. Switzerland.4
Mart´ Uribe ın
Table 1. Denmark. Portugal. South Africa. Brazil. Austria. New Zealand. Ecuador.y ρc.9 0. Slovak Republic. Sweden. Belgium.87 Developed Countries 1.69 0.96 2.
.26 0.y ρi.74 0. Finland. Data are detrended using a band-pass ﬁlter including frequencies between 6 and 32 quarters with 12 leads and lags.23 1.y Emerging Countries 2. Korea. Israel. Developed Countries: Australia.75
Note: Average values of moments for 13 small emerging countries and 13 small developed countries. Netherlands. Canada. Mexico. Philippines. and Turkey.94 3.86 0.58 0. Peru. Malaysia. Thailand.02 1.87 0.04 0.71 -0. Source: Aguiar and Gopinath (2004). Developed Economies Moment σy σ∆y ρy ρ∆y σc /σy σi /σy σtb/y ρtb/y.32 3.09 -0.2: Business Cycles: Emerging Vs.09 0. Norway.

5
(2.1
The Model Economy
Consider an economy populated by a large number of inﬁnitely lived households with preferences described by the utility function
∞
E0
t=0
β t U (ct ).
2.1)
. The model developed in this chapter is simple enough to allow for a full characterization of its equilibrium dynamics using pen and paper.Chapter 2
An Endowment Economy
The purpose of this chapter is to build a canonical dynamic. general equilibrium model of the small open economy capable of capturing some of the empirical regularities of business cycles in small emerging and developed countries documented in chapter 1.

ct − yt . This is because if the allocation {ct .6
Mart´ Uribe ın
where ct denotes consumption and U denotes the single-period utility function. The optimal allocation of consumption and debt will always feature this constraint holding with strict equality. r denotes the interest rate. and excess expenditure over income. This endowment process represents the sole source of uncertainty in this economy. assumed to be constant. dt − dt−1 . then one can choose an alternative allocation {ct .
. dt }∞ that also satisﬁes the no-Ponzi-game cont=0 straint and satisﬁes ct ≥ ct . for all t ≥ 0.2)
where dt denotes the debt position assumed in period t. (1 + r)j
j→∞
lim Et
(2. has two sources. The evolution of the debt position of the representative household is given by dt = (1 + r)dt−1 + ct − yt . (2. and yt is an exogenous and stochastic endowment of goods. which is assumed to be strictly increasing and strictly concave. dt }∞ satisﬁes t=0 the no-Ponzi-game constraint with strict inequality. with ct > ct for at least one date t ≥ 0. The above constraint states that the change in the level of debt. Households are subject to the following borrowing constraint that prevents them from engaging in Ponzi games: dt+j ≤ 0. interest services on previously acquired debt.3)
This limit condition states that the household’s debt position must be expected to grow at a rate lower than the interest rate r. rdt−1 . This alternative allocation is clearly strictly preferred to the original one because the single period utility function is strictly increasing.

This assumption eliminates long-run growth in consumption. and the following Euler condition: U (ct ) = β(1 + r)Et U (ct+1 ). At the optimal allocation. so as to maximize (2. we assume that the period utility index is quadratic and given by 1 U (c) = − (c − c)2 .4)
The interpretation of this expression is simple.
.3). the Euler
1 After imposing this assumption. β and 1/(1 + r). The optimality conditions associated with this problem are (2.
β(1 + r) = 1. We make two additional assumptions that greatly facilitates the analysis. Chapter 2
7
The household chooses processes for ct and dt for t ≥ 0. Second.5)
with c < c.1) subject to (2. 1 + r. If the household sacriﬁces one unit of consumption in period t and invests it in ﬁnancial assets. ¯ 2
(2. Under these assumptions.1 This particular functional form makes it possible to obtain ¯ a closed-form solution of the model. (2. that is. In period t + 1 the household receives the unit of goods invested plus interests.2). yielding β(1 + r)Et U (ct+1 ) utils.2) and (2. First we require that the subjective and pecuniary rates of discount. (2. be equal to each other.Lectures in Open Economy Macroeconomics.3) holding with equality. its period-t utility falls by U (ct ). our model becomes essentially Hall’s (1978) permanent income model of consumption. the cost and beneﬁt of postponing consumption must equal each other in the margin.

(1 + r)j
. j (1 + r) (1 + r)s
Apply expectations conditional on information available at time t and take the limit for s → ∞ using the transversality condition (equation (2.3) holding with equality) to get the following intertemporal resource constraint:
∞
(1 + r)dt−1 = Et
j=0
yt+j − ct+j .2) and the no-Ponzi-scheme constraint (2.8 condition (2.
Mart´ Uribe ın
(2.3) holding with equality—also known as the transversality condition.4) collapses to ct = Et ct+1 . Begin by expressing the sequential budget constraint in period t as
(1 + r)dt−1 = yt − ct + dt . households expect to maintain a constant level of consumption. at each point in time.6)
which says that consumption follows a random walk.
Lead this equation 1 period and use it to get rid of dt : yt+1 − ct+1 dt+1 + .
We now derive an intertemporal resource constraint by combining the household’s sequential budget constraint (2. 1+r 1+r
(1 + r)dt−1 = yt − ct +
Repeat this procedure s times to get
s
(1 + r)dt−1 =
j=0
yt+j − ct+j dt+s + .

1) deﬁnes
the serial correlation of the endowment process. we assume that the endowment process follows an AR(1) process of the form. The larger is ρ the more persistent the endowment process.6) to deduce that Et ct+j = ct . ct .Lectures in Open Economy Macroeconomics.
Using this expression to eliminate expectations of future income from equa-
. yt = ρyt−1 + t . (1 + r)j
(2. this equation says that the country’s initial net foreign debt position must equal the expected present discounted value of current and future diﬀerences between output and absorption.i.
Now use the Euler equation (2. innovation and the parameter ρ ∈ (−1.7)
This expression states that the optimal plan allocates the annuity value of the income stream
r 1+r Et yt+j ∞ j=0 (1+r)j
to consumption. where
t
denotes an i. the j-period-ahead forecast of output in period t is given by
Et yt+j = ρj yt . Use this result to get rid of expected future consumption in the above expression and rearrange to obtain r Et 1+r
∞
rdt−1 + ct =
j=0
yt+j . and to debt ser-
vice.d. To be able to fully characterize the equilibrium in this economy. Chapter 2
9
Intuitively. rdt−1 . Then.

it follows that the equilibrium evolution of the stock of external debt is given by 1−ρ yt . we have that a unit increase in the endowment leads to a less-than-unity increase in consumption.e. Letting tbt ≡ y − ct and cat ≡ −rdt−1 + tbt denote. we obtain r 1+r
∞
Mart´ Uribe ın
rdt−1 + ct = yt =
j=0
ρ 1+r
j
r yt . i. 1+r−ρ
ct =
(2. cat = −(dt − dt−1 ).. we obtain r yt − rdt−1 .10 tion (2. 1+r−ρ
Solving for ct . we have 1−ρ yt 1+r−ρ
tbt = rdt−1 +
and cat = 1−ρ yt . 1+r−ρ
dt = dt−1 −
.7). 1+r−ρ
Note that the current account inherits the stochastic process of the underlying endowment shock. the trade balance and the current account in period t.8)
Because ρ is less than unity. respectively. Because the current account equals the change in the country’s net foreign asset position.

when innovations in the endowment are purely temporary only a small part of the changes in income—a fraction r/(1 + r)—is allocated to current consumption. In this case. ρ = 0. the current account plays the role of a shock absorber.
2. Assume that 0 < ρ < 1. Because income is expected to fall quickly to its long-run level households smooth consumption by eating a tiny part of the current windfall and leaving the rest for future consumption. so that endowment shocks are positively serially correlated. the more volatile is the current account. In the extreme case of purely transitory shocks. an increase in the endowment shock leads to a permanent deterioration in the trade balance. Households borrow to ﬁnance negative income shocks and save in response to positive shocks. the standard deviation of the current account is given by σy /(1 + r).8). Chapter 2
11
According to this expression.Lectures in Open Economy Macroeconomics. Most of the endowment increase—a fraction 1/(1 + r)—is saved. Two polar cases are of interest. According to equation (2. Thus. In the ﬁrst case. A temporary increase in the endowment produces a gradual but permanent decline in the stock of foreign liabilities.2
Response to Output Shocks
Consider the response of our model economy to an unanticipated increase in output. which is close to the volatility of the endowment
. external debt follows a random walk and is therefore nonstationary. the endowment shock is assumed to be purely transitory. It follows that the more temporary are endowment shocks. The longrun behavior of the trade balance is governed by the dynamics of external debt. The intuition for this result is clear.

households allocate all innovations in their endowments to current consumption. Note that the trade balance converges to a level lower than the preshock level. as documented in chapter 1. In this case. That is. households are able to sustain a smooth consumption path by consuming the totality of the current income shock. it improves during expansions and deteriorates during contractions. For. and. This prediction represents a serious problem for this model. As a result. Summarizing. the trade balance and the current account improve. 1)) is illustrated in ﬁgure 2. The intermediate case of a gradually trend-reverting endowment process (ρ ∈ (0.1. Intuitively.12
Mart´ Uribe ın
shock itself for small values of r. the current account is countercyclical in small open economies. This is because in the long-run the economy settles at a lower level of external debt. which requires a smaller trade surplus to be served. an increase in income today is not accompanied by the expectation of future income contractions. external borrowing is conducted under the principle: ‘ﬁnance tem-
. In response to the positive endowment shock. This expansion in domestic absorption is smaller than the initial increase in income. More importantly. which captures the essential elements of what has become known as the intertemporal approach to the current account. especially in developing countries. consumption experiences a once-and-for-all increase. as a result the current account is nil and the stock of debt remains constant over time. when endowment shocks are permanent. these two variables converge gradually to their respective long-run levels. ρ → 1. As a result. in this model. The other polar case emerges when shocks are highly persistent. After the initial increase. the current account is procyclical.

Speciﬁcally.1: Response to a Positive Endowment Shock
porary shocks. rather than its level.Lectures in Open Economy Macroeconomics.’ A central failure of the model is the prediction of a procyclical current account. Chapter 2
13
Figure 2. let
∆yt ≡ yt − yt−1
. displays mean reversion.
2. Fixing this problem is at the heart of what follows in this and the next two chapters. adjust to permanent shocks.3
Nonstationary Income Shocks
Suppose now that the rate of change of output.

2: Stationary Versus Nonstationary Endowment Shocks
denote the change in endowment between periods t − 1 and t. consumption-smoothing households have an incentive to borrow against future income. and ρ ∈ [0.i.2 provides a graphical expression of this intuition. shock with mean zero and variance σ 2 . Figure 2.14
Mart´ Uribe ın
Figure 2. Faced with such an income proﬁle.d. is an i. According to this process. thereby producing a countercyclical tendency in the current account. Suppose that ∆yt evolves according to the following autoregressive process:
∆yt = ρ∆yt−1 + t . the level of income is nonstationary. 1)
where
t
is a constant parameter. This is the basic intuition why allowing for a nonstationary output process can help explain the behavior of the trade balance and the current account at business-cycle frequencies.
. The following model formalizes this story. in the sense that a positive output shock (
t
> 0) produces a
permanent future expected increase in the level of output.

5) subject to the sequential resource constraint (2. we have that Et ∆yt+j = ρj ∆yt . Chapter 2
15
As before. Using these optimality conditions yields the expression for consumption given in equation (2. and the Euler equation (2.7). (1 + r)j
Rearranging.6). Using
. the no-Ponzi-game constraint holding with equality. The ﬁrst-order conditions associated with this problem are the sequential budget constraint.2) and the no-Ponzi-game constraint (2. the model economy is inhabited by an inﬁnitely lived representative household that chooses contingent plans for consumption and debt to maximize the utility function (2. (1 + r)j
This expression states that the current account equals the present discounted value of future expected income decreases. which we reproduce here for convenience r Et 1+r
∞
ct = −rdt−1 +
j=0
yt+j . According to the autoregressive process assumed for the endowment.Lectures in Open Economy Macroeconomics. (1 + r)j
Using this expression and recalling that the current account is deﬁned as cat = yt − ct − rdt−1 . we obtain
∞
cat = −Et
j=1
∆yt+j . we can write r Et 1+r
∞
cat = yt −
j=0
yt+j .3).

because here we analyze changes in consumption and output. the current account ceases to be countercyclical (is procyclical). we can write the current account as: −ρ ∆yt . we wish to ﬁnd
2 2 out conditions under which σ∆c can be higher than σ∆y in equilibrium. Are implied changes in consumption more or less volatile than changes in output? This question is important because. For an unexpected increase in income to induce an increase in consumption larger than the increase in income itself. respectively. whereas in chapter 1 we reported statistics pertaining to the growth rates of consumption and output. or ρ > 0. the current account deteriorates in response to a positive innovation in output. as we saw in chapter 1. Recall that when the endowment level is stationary the current account increases in response to a positive endowment shock. letting σ∆c and σ∆y denote the standard deviations of ∆ct ≡ ct − ct−1 and ∆yt . 1+r−ρ
cat =
According to this formula.2 We
Strictly speaking. it is necessary that future income be expected to be higher than current income. We note that the countercyclicality of the current account in the model with nonstationary shocks depends crucially on output changes being positively serially correlated. This implication is an important improvement relative to the model with stationary shocks. The intuition behind this result is clear. which happens only if ∆yt is positively serially correlated.16
Mart´ Uribe ın
this result in the above expression. In eﬀect. Formally. developing countries are characterized by consumption growth being more volatile than output growth. this exercise is not comparable to the data displayed in chapter 1.
2
. when ρ is zero (negative).

Taking diﬀerences.
The right-hand side of this expression equals unity at ρ = 0. we obtain
cat − cat−1 = ∆yt − ∆ct − r(dt−1 − dt−2 ). Then. we can write the standard deviation of consumption changes as σ∆c 1+r = σ∆y 1+r−ρ
1 − ρ2 . The right hand side of the above expression is increasing in ρ at ρ = 0.
Noting that dt−1 − dt−2 = −cat−1 and solving for ∆ct .Lectures in Open Economy Macroeconomics. namely that when the level of income is a random walk. consumption and income move hand in hand.9)
2 It follows directly from the AR(1) speciﬁcaiton of ∆yt that σ∆y (1−ρ2 ) =
σ 2 . It follows that there are values of
. 1+r−ρ
(2. so their changes are equally volatile. This result conﬁrms the one obtained earlier in this chapter. Chapter 2 start with the deﬁnition of the current account
17
cat = yt − ct − rdt−1 . we obtain:
∆ct = ∆yt − cat + (1 + r)cat−1 = ∆yt + = = ρ ρ(1 + r) ∆yt − ∆yt−1 1+r−ρ 1+r−ρ 1+r ρ(1 + r) ∆yt − ∆yt−1 1+r−ρ 1+r−ρ 1+r t.

the variance of ∆yt becomes inﬁnitely large as changes in income become a random walk. Here we review these restrictions and their empirical validity. This is because as ρ → 1. This property ceases to hold as ∆yt becomes highly persistent. ∆ct follows an i.9) shows. Noting that the current account in period t. Speciﬁcally. Campbell (1987). 1). as expression (2. is given by yt − ct − rdt−1 we can write equation (2. In the context of the open economy. Hall’s work motivated a large empirical literature devoted to testing the empirical relevance of the model described above. denoted cat .
.18
Mart´ Uribe ın
ρ in the interval (0. process with ﬁnite variance for all values of ρ ∈ [0.7) as
∞
−(1 + r)cat = −yt + rEt
j=1
(1 + r)−j yt+j .d. in particular.i.4
Testing the Model
Hall (1978) was the ﬁrst to explore the econometric implication of the simple model developed in this chapter. Campbell’s restrictions are readily expressed in terms of the current account. 1) for which the volatility of consumption changes is indeed higher than that of income changes. deduced and tested a number of theoretical restrictions on the equilibrium behavior of national savings.
2. Hall tested the prediction that consumption must follow a random walk. We start by deriving a representation of the current account that involves expected future changes in income. whereas.

this expression states that the country borrows from the rest of the world (runs a current account deﬁcit) income is expected to grow in the future. Then. from the above VAR system.10)
Intuitively. the country chooses to build its net foreign asset position (runs a current account surplus) when income is expected to decline in the future.
(2. In this case the country saves for a rainy day.
. Chapter 2 Deﬁning ∆xt+1 = xt+1 − xt . cat Consider estimating a VAR system including xt :
xt = Dxt−1 + t .Lectures in Open Economy Macroeconomics. Deﬁne
∆yt xt = .
Let Ht denote the information contained in the vector xt .
Consider now an empirical representation of the time series ∆yt and cat .
Combining the above two expression we can write the current account as
∞
cat = −
j=1
(1 + r)−j Et ∆yt+j . Similarly. we have that the forecast of xt+j given Ht is given by Et [xt+j |Ht ] = D j xt . it is simple to show that
∞ ∞
19
−yt + rEt
j=1
(1 + r)−j yt+j = (1 + r)
j=1
(1 + r)−j Et ∆yt+j .

In computing F . So the model implies the following restriction on the vector F :
F = [0
1].1.04. Now consider running a
regression of the left and right hand side of equation (2. Since xt includes cat as one element. This p-value means that if the null hypothesis was true.20 It follows that
∞
Mart´ Uribe ın
(1 + r)−j Et [∆yt+j |Ht ] =
j=1
1 0
∆yt [I − D/(1 + r)]−1 D/(1 + r) . The Wald statistic associated with null hypothesis that F = [0 1] is 16. The VAR system that Nason and Rogers estimate includes 4 lags.
Consider now an additional testable cross-equation restriction on the
. They estimate the VAR system using Canadian data on the current account and GDP net of investment and government spending. Their data strongly rejects the above cross-equation restriction of the model.1 or higher only 4 out of 100 times.
Nason and Rogers (2006) perform an econometric test of this restriction.7 percent per year. The regression
coeﬃcients of the right-hand side regression is F . which reﬂects the discrepancy of F from [0 1]. then the Wald statistic.
with an asymptotic p-value of 0. we obtain that the regression coefﬁcient for the left-hand side regression is the vector [0 1]. cat
Let F ≡ −
1 0
[I − D/(1 + r)]−1 D/(1 + r). they calibrate r at 3. would take a value of 16.10) onto the vector xt . The estimation sample is 1963:Q1 to 1997:Q4.

From equation (2.
Clearly. one can consider projecting the above expression onto ∆yt−1 and cat−1 .
(2. and µt is an i. Consider. will not hold.11)
According to this expression. Speciﬁcally. where c is still a constant. because in general µt is correlated with cat . for instance. Nason and Rogers (2006) ﬁnd that a test of the hypothesis that all coeﬃcients are zero in a regression of
Consider projecting the left. In particular.
3
. This yields the orthogonality condition [0 1][D − (1 + r)I]D − [1 0]D2 = [0 0]. Chapter 2 theoretical model. replace c ¯ in equation (2.d. in this case we have that cat+1 − (1 + r)cat − ∆yt+1 should be unpredictable given information available in period t − 1 or earlier.and right-hand sides of this expression on the information set Ht .11) becomes
Et cat+1 − (1 + r)cat − Et ∆yt+1 = µt .i. the variable cat+1 − (1 + r)cat − ∆yt+1 is unpredictable in period t. 4 In particular. the orthogonality condition stating that cat+1 − (1 + r)cat − ∆yt+1 be orthogonal to variables dated t or earlier. Nevertheless. This projection yields the orthogonality restriction [0 1][D −(1+r)I]−[1 0]D = [0 0]. equation (2.Lectures in Open Economy Macroeconomics.5) by c + µt . a variation of the model economy where the bliss point is a random variable. if one runs a regression of this variable on current and past values of xt .10) it follows that
21
Et cat+1 − (1 + r)cat − Et ∆yt+1 = 0. all coeﬃcients should be equal to zero. In this environment.3 This restriction is not valid in a more general version of the model featuring private demand shocks. ¯ ¯ shock with mean zero.4 Both of the orthogonality conditions discussed here are strongly rejected by the data.

. The p-value associated with a regression featuring as regressors past values of xt is 0.22
Mart´ Uribe ın
cat+1 − (1 + r)cat − ∆yt+1 onto current and past values of xt has a p-value of 0.01.06.

1)
where ct denotes consumption. assumed to be strictly increasing.
t=0
(3. we show that allowing for capital accumulation can help resolve this problem. Households seek to maximize this utility function subject to the following three 23
. In this chapter. strictly concave.
3. β ∈ (0. and twice continuously diﬀerentiable.Chapter 3
An Economy with Capital
A theme of chapter 2 is that the simple endowment economy fails to predict the countercyclicality of the trade balance. and U denotes the period utility function. 1) denotes the subjective discount factor.1
The Basic Framework
Consider a small open economy populated by a large number of inﬁnitely lived households with preferences described by the utility function
∞
β t U (ct ).

2)
(3. strictly concave.
The ﬁrst-order conditions corresponding to this problem are
U (ct ) = λt . The function F describes the production technology and is assumed to be strictly increasing.
λt = β(1 + r)λt+1 . and it denotes investment. The variable θt denotes an exogenous nonstochastic productivity factor. j→∞ (1 + r)j lim
Mart´ Uribe ın
(3. and bt+j ≥ 0. and to satisfy the Inada conditions. We relax this assumption later. The Lagrangian associated with the household’s problem is
∞
L=
t=0
β t {U (ct ) + λt [(1 + r)bt−1 + kt + θt F (kt ) − ct − kt+t − bt+1 ]} . kt denotes the stock of physical capital.24 constraints: bt = (1 + r)bt−1 + θt F (kt ) − ct − it . kt+1 = kt + it . we assume that the capital stock does not depreciate.
We wish to characterize the response of the economy to a permanent increase in θt . For the sake of simplicity.3)
where bt denotes real bonds bought in period t yielding the constant interest rate r > 0.
.

θt+1 . The ﬁrst of these equilibrium conditions states that households invest in physical capital until the marginal product of capital equals the rate of return on foreign bonds. and a decreasing function of the opportunity cost of holding physical capital. The above optimality conditions can be reduced to the following two expressions: r = θt+1 F (kt+1 ) and ct = rbt−1 + r 1+r
∞
(3. It follows from (3. (1 + r)j
(3. κ2 < 0. Chapter 3 λt = βλt+1 [1 + θt+1 F (kt+1 )].
. we assume that β(1+r) = 1. t→∞ (1 + r)t lim
25
As in the endowment-economy model of chapter 2. is an increasing function of the future expected level productivity. to avoid inessential long-run dynamics. r.
kt+1 = κ(θt+1 .5)
for t ≥ 0. r).Lectures in Open Economy Macroeconomics.
κ1 > 0. kt+1 .4) that next period’s level of physical capital. and the transversality condition bt = 0. Formally. bt = (1 + r)bt−1 + θt F (kt ) − ct − kt+1 + kt . This assumption together with the ﬁrst two of the above optimality conditions implies that consumption is constant over time.4)
j=0
θt+j F (kt+j ) − kt+j+1 + kt+j .

but also the present discounted value of the diﬀerences between output and investment.2) to t t obtain the sequence of asset positions bs = b−1 for all t ≥ 0. Consumption experiences a permanent increase in response to the permanent technology shock.
3. θt = θ for t < 0 and θt = θ for t ≥ 0. ¯ ¯ unexpected increase in the technology factor to θ > θ. Obviously. It follows that the optimal consumption path must also be
. To show that the proposed allocation is feasible.5). follow the same steps as in the derivation of its counterpart for the endowment economy studied in chapter 2. To obtain this expression. b−1 . We have shown the existence of a feasible consumption path that ! is strictly preferred to the pre-shock consumption allocation. consider the suboptimal paths for consumption and investment ¯ ¯ cs = θ F (k) + rb−1 and is = 0 for all t. That is. because θ > θ. Suppose further that in period 0 there is a permanent. t limt→∞ b−1 /(1 + r)t = 0. where k denotes the t t ¯ initial level of capital.3). which includes not only initial ﬁnancial wealth. the consumption path cs is t ¯ ¯ strictly preferred to the pre-shock path.1
A Permanent Productivity Shock
Suppose that up until period -1 inclusive the technology factor θt was con¯ stant and equal to θ. let us plug the consumption and investment paths cs and is into the sequential budget constraint (3. for t ≥ 0. Clearly. says that consumption equals the interest ﬂow on a broad deﬁnition of wealth. To
see this. so the proposed suboptimal allocation satisﬁes the no-Ponzi-game condition (3. given by θF (k) + rb−1 . That is. equation (3.1. ct = c0 > c−1 for all t > 0.26
Mart´ Uribe ın
The second equilibrium condition.

This result together with the fact that the optimal consumption path is constant over time implies that consumption must jump up once and for all in period 0. necessarily goes up in period 1.Lectures in Open Economy Macroeconomics. 1+r 1+r
c0 = rb−1 +
The trade balance is given by tbt = θt F (kt ) − ct − it . Because k0 was chosen in period −1. r). investment experiences a once-and-for-all increase that brings the level of ¯ ¯ capital up from k to a level k ∗ ≡ κ(θ . the trade balance is simply equal to −rb−1 . when households expected θ0 to be ¯ ¯ ¯ ¯ ¯ equal to θ. Chapter 3
27
strictly preferred to the pre-shock consumption path. In period 0. 1+r
tb0 = −rb−1 −
Before period zero.5) and evaluating that equation at t = 0 we get r 1 ¯ ¯ θ F (k) − k ∗ + k + θ F (k ∗ ). given by output minus investment minus consumption. Plugging this path for the capital stock into equation (3. r) > k. This together with the fact that the expression within square brackets in the above equation is unambiguously positive. where k is given by k = κ(θ. Thus. we have that k0 = k. kt = k ∗ for t ≥ 1. note that output increases from θ F (k0 ) in period 0 to θ F (k ∗ ) in period 1. the trade balance. Thus. while consumption remains unchanged. in period 1 investment falls to zero. To see this. implies that in response to the permanent technology shock the trade balance deteriorates in period zero. On the demand side. The trade balance remains constant after
. we have 1 ¯ ¯ θ F (k ∗ ) − θ F (k) + (k ∗ − k) . In period 1 the trade balance improves. Thus.

the entire adjustment in investment occurs in period zero. As a result. We will study the role of adjustment costs more closely shortly. is expected to stay high in the future.28
Mart´ Uribe ın
period 1. Another important factor in generating a decline in the trade balance in response to a positive productivity shock is the assumed absence of capital adjustment costs. A permanent increase in productivity induces a strong response in domestic absorption. For the initial worsening of the external accounts in a context of expansion in aggregate activity is in line with the empirical evidence presented in chapter 1. Note that in response to the increase in future expected productivity. dampening the increase in domestic absorption in the date the shock occurs. it pays to increase investment spending. Indeed. investment falls to zero in period 1 and remains nil thereafter. note ﬁrst that from the vintage point of period zero. the propensity to consume out of current income is high. households face an increasing path for output net of investment. At the same time. investment spending is spread over a number of periods. To see this. θt . because the productivity of capital.
. The current account deteriorates in period zero and is nil from period 1 onward. The initial deteriorations of the trade balance and the current account implied by the model represent a signiﬁcant improvement with respect to the endowment economy studied in chapter 2. In the presence of costs of adjusting the stock of capital. In obtaining a deterioration of the trade balance the assumed persistence of the productivity shock plays a signiﬁcant role.

are unaﬀected by the productivity shock. where k is the level of capital inherited in period 0. This output eﬀect
. however.1. suppose that everybody correctly expects the productivity shock ¯ to be purely temporary. suppose that up until period -1 inclusive the productivity factor θt ¯ was constant and equal to θ. ¯ ¯ where y−1 ≡ θF (k) is the pre-shock level of output. θ0 = θ > θ. But agents have no incentives to have a higher capital stock in the future. a zero probability event happens. This is intuitive. As a result. That is. Namely. and therefore also ¯ investment. The positive productivity shock in period zero does produce an increase ¯ ¯ ¯ in output in that period. it is worth analyzing the eﬀect of a purely temporary shock. because its productivity is expected to go back down to its ¯ historic level θ right after period 0. Investment in period zero can only increase the future stock of capital. But k0 is ﬁxed in period zero. Specifically. That is ¯ ¯ y0 = y−1 + (θ − θ)F (k). In this case. That is. everybody expects θt = θ for all t > 0. Suppose also that in period -1 people assigned ¯ a zero probability to the event that θ0 would be diﬀerent from θ.4) implies that the capital stock. In period ¯ 0. households would like to have more capital in that period. kt = k for all ¯ t ≥ 0. from θF (k) to θ F (k).Lectures in Open Economy Macroeconomics.2
A Temporary Productivity shock
To stress the importance of persistence in productivity movements in inducing a deterioration of the trade balance in response to a positive output shock. The productivity of capital unexpectedly increases in period zero. Chapter 3
29
3. Furthermore. equation (3.

we get that the trade balance in period 0 is given by 1 ¯ (θ − θ)F (k0 ) > 0.g.
3. 1+r
tb0 − tb−1 = (y0 − y−1 ) − (c0 − c−1 ) − (i0 − i−1 ) =
This expression shows that the trade balance improves on impact. households invest the entire increase in output in the international ﬁnancial market and increase consumption by the interest ﬂow associated with that ﬁnancial investment. and consumers save most of the purely temporary increase in income in order to smooth consumption over time. in diﬀerent forms.30
Mart´ Uribe ın
induces higher consumption. e. 1+r
c0 = c−1 +
¯ ¯ where c−1 ≡ −rb−1 + θF (θ) is the pre-shock level of consumption. Capital adjustment costs are typically introduced. Basically. The reason for this counterfactual response is simple: ‘Firms have no incentive to invest.
. In eﬀect.5) we have that r ¯ ¯ (θ − θ)F (k). using equation (3..2
Capital Adjustment Costs
Consider now an economy identical to the one described above but in which changes in the stock of capital come at a cost. Combining the above two expressions and recalling that investment is unaﬀected by the temporary shock. in small open economy models to dumpen the volatility of investment over the business cycle (see. as the increase in the productivity of capital is short lived.

respectively.6)
(3.
The variables λt and qt denote Lagrange multipliers on the sequential budget constraint and the law of motion of the capital stock. both the level and the slope of this function vanish at the steady-state value of investment. 1998). The Lagrangian associated with this optimization problem is
∞
L=
t=0
β t U (ct ) + λt (1 + r)bt−1 + θt F (kt ) − ct − it −
1 i2 t − bt + qt (kt + it − kt+1 ) 2 kt
. and Schmitt-Groh´.7)
. Suppose that the sequential e budget constraint is of the form 1 i2 t .Lectures in Open Economy Macroeconomics.3). 2 kt
bt = (1 + r)bt−1 + θt F (kt ) − ct − it −
Here. the law of motion of the capital stock is given by kt+1 = kt + it . capital adjustment costs are given by i2 /(2kt ) and are a strictly cont vex function of investment.1) subject to the above two restrictions and the no-Ponzi-game constraint (3. 1991. Moreover. Households seek to maximize the utility function given in (3. Chapter 3
31
Mendoza. it = 0. We continue to assume that β(1 + r) = 1. As in the economy without adjustment costs. The ﬁrst-order conditions associated with the household problem are:
it = (qt − 1)kt 1 (1 + r)qt = θt+1 F (kt+1 ) + (it+1 /kt+1 )2 + qt+1 2
(3.

the unit of capital can be sold at a price qt+1 next period. At the optimum both strategies must yield the same return.9)
.
j=0
The Lagrange multiplier qt represents the shadow relative price of capital in terms of consumption goods. This is the left-hand side of equation (3. This unit yields θt+1 F (kt+1 ) units of output next period.8)
qt =
(3. an extra unit of capital reduces tomorrow’s adjustment costs by (it+1 /kt+1 )2 /2.32 kt+1 = kt + it ct = rbt−1 + r 1+r
∞
Mart´ Uribe ın
θt+j F (kt+j ) − it+j − 1 (i2 /kt+j ) 2 t+j (1 + r)j
. thereby increasing it . instead of using qt units of goods to buy one unit of capital. as qt increases agents have incentives to allocate goods to the production of capital. Finally.6). The sum of these three sources of income form the right hand side of (3.1
Dynamics of the Capital Stock
Eliminating it from the above equations we get the following two ﬁrst-order. nonlinear diﬀerence equations in kt and qt :
kt+1 = qt kt θt+1 F (qt kt ) + (qt+1 − 1)2 /2 + qt+1 . In addition.
3.2. Equation (3. 1+r
(3. the agent can engage in a ﬁnancial investment by purchasing qt units of bonds in period t with a gross return of (1 + r)qt . and is known as Tobin’s q.7) compares the rates of return on bonds and physical capital: Adding one unit of capital to the existing stock costs qt .7).7). Alternatively. According to equation (3.

10)
Above the locus KK . qt ) for which kt+1 = kt in equation (3.1: The Dynamics of the Capital Stock
q Q
S a • 1 K • K′
S′
Q′ k0 k* k
The perfect foresight solution to these equations is depicted in ﬁgure 3. The locus QQ corresponds to the pairs (kt .Lectures in Open Economy Macroeconomics. The horizontal line KK corresponds to the pairs (kt . qt ) for which qt+1 = qt in equation (3.1. Chapter 3
33
Figure 3.
(3. the capital stock grows over time and below KK the capital stock declines over time. That is.11)
. q = 1. (3. rq = θF (qk) + (q − 1)2 /2.9).8). That is.

it is clear from equation (3. The system (3. For qt near unity. 1. on the other hand. qt . The value of k ∗ is implicitly determined by the expression r = θF (k). In the period of the shock.2
A Permanent Technology Shock
Suppose now that in period 0 the technology factor θt increases permanently ¯ ¯ from θ to θ > θ.10) that the locus KK is not aﬀected by the productivity shock.11) determine the steady-state value of the capital stock. The price of capital. and the steady-state value of the Tobin’s q. adjustment costs vanish in the steady state.34
Mart´ Uribe ın
Jointly. Consider now the transition to the new steady state. Then the new steady-state values of k and q are given by k ∗ and 1.9) is saddle-path stable. q decreases over time. which is the same value obtained in the economy without adjustment costs. the locus QQ is downward sloping. If the initial capital stock is diﬀerent from its long-run level.11) that the locus QQ shifts up and to the right. the long-run level of capital experiences a permanent increase.2. The locus SS represents the converging saddle path. both q and k converge monotonically to their steady states along the saddle path. equations (3. as noted earlier.10) and (3. q increases over time and below and to the left of QQ . On the other hand.8)-(3. because. Suppose that the steady-state value of capital prior to the innovation in productivity is k0 in ﬁgure 3. Above and to the right of QQ .1. The
.
3. the capital stock does not move. is not aﬀected in the long run. It follows that in response to a permanent increase in productivity. It is clear from equation (3. This is not surprising. which we denote by k ∗ .

Lectures in Open Economy Macroeconomics, Chapter 3

35

price of installed capital, qt , jumps to the new saddle path, point a in the ﬁgure. This increase in the price of installed capital induces an increase in investment, which in turn makes capital grow over time. After the initial impact, qt decreases toward 1. Along this transition, the capital stock increases monotonically towards its new steady-state k ∗ . The equilibrium dynamics of output, investment, and capital in the presence of adjustment costs are quite diﬀerent from those arising in the absence thereof. In the frictionless environment, investment, output, and the capital stock all reach their long-run steady state one period after the productivity shock. Under capital adjustment costs, by contrast, these variables adjust gradually to the unexpected increase in productivity. The more pronounced are adjustment costs, the more sluggish is the response of investment, thereby making it less likely that the trade balance deteriorates in response to a positive technology shock as required for the model to be in line with the data.1 This observation opens the question of what would the model predict for the behavior of the trade balance in response to output shocks when one introduces a realistic degree of adjustment costs. We address this issue in the next chapter.

It is straightforward to see that the response of the model to a purely temporary productivity shock is identical as that of the model without adjustment costs. In particular, capital and investment display a mute response.

1

36

Mart´ Uribe ın

Chapter 4

The Real Business Cycle Model

In the previous two chapters, we arrived at the conclusion that a model driven by productivity shocks can explain the observed countercyclicality of the current account. We also established that two features of the model are important in making this prediction possible. First, productivity shocks must be suﬃciently persistent. Second, capital adjustment costs must not be too strong. In this chapter, we extend the model of the previous chapter by allowing for three features that add realism to the model’s implied dynamics. Namely, endogenous labor supply and demand, uncertainty in the technology shock process, and capital depreciation. The resulting theoretical framework is known as the Real Business Cycle model, or, succinctly, the RBC model. 37

This property is desirable from a purely technical point of view because it makes it possible to rely on linear approximations to characterize equilibrium dynamics. ht )θt t ≥ 0. yt denotes domestic output. The period-by-period budget constraint of the representative household is given by
dt = (1 + rt−1 )dt−1 − yt + ct + it + Φ(kt+1 − kt ).
(4. βh > 0.3)
where βc < 0. This preference speciﬁcation was conceived by Uzawa (1968) and introduced in the small-open-economy literature by Mendoza (1991).1)
θ0 = 1. In particular. ht ).38
Mart´ Uribe ın
4. rt denotes the interest rate at which domestic residents can borrow in period t.2) (4. The reason why we adopt this type of utility function here is that it gives rise to a steady state of the model that is independent of initial conditions.4)
where dt denotes the household’s debt position at the end of period t. θt+1 = β(ct . ct denotes consumption.
(4.
(4. it denotes gross
.1
The Model
Consider an economy populated by an inﬁnite number of identical households with preferences described by the utility function
∞
E0
t=0
θt U (ct . under these preferences the steady state is independent of the initial net foreign asset position of the economy.

1) subject to (4. θt+1 }∞ so as to maxt=0 imize the utility function (4.2)-(4.5)
where At is an exogenous stochastic productivity shock. The restrictions imposed on Φ ensure that adjustment costs are nil in the steady state and that in the steady state the interest rate equals the marginal product of capital net of depreciation.3)-
. it . Chapter 4
39
investment. ht ). As pointed out earlier.
Output is produced by means of a linearly homogeneous production function that takes capital and labor services as inputs.4).
(4.
(4. 1) denotes the rate of depreciation of physical capital. The stock of capital evolves according to kt+1 = it + (1 − δ)kt . ht .6) and a no-Ponzi constraint of the form
j→∞
(4. yt .Lectures in Open Economy Macroeconomics. small open economy models typically include capital adjustment costs to avoid excessive investment volatility in response to variations in the foreign interest rate.3) and (4. The function Φ(·) is meant to capture capital adjustment costs and is assumed to satisfy Φ(0) = Φ (0) = 0. the ﬁrst-order conditions of the household’s maximization problem are (4.7)
Letting θt ηt and θt λt denote the Lagrange multipliers on (4. dt . Households choose processes {ct . where δ ∈ (0. and kt denotes physical capital.
yt = At F (kt .6)
lim Et
dt+j j s=0 (1 + rs )
≤ 0. kt+1 .

11)
λt [1+Φ (kt+1 −kt )] = β(ct .12) These ﬁrst-order conditions are fairly standard. ht+j ).40 (4. yields rt = r. Similarly. ht ) ηt = −Et U (ct+1 . (4. To see this. ht ) = λt At Fh (kt . ηt equals the present discounted value of utility from period t + 1 onward.10) (4. The world interest rate is assumed to be constant and equal to r. ht )ηt . In turn. ht )
(4.7) holding with equality and:
Mart´ Uribe ın
λt = β(ct . iterate the ﬁrst-order condition (4. ht ) but instead Uh (ct . ht )(1 + rt )Et λt+1
(4.10) forward to obtain: ηt = −Et
∞ j=1 θt+j θt+1
U (ct+j . ht )ηt . ht+1 ) −Uh (ct . ht )−βc (ct . ht+1 ) + Et ηt+1 β(ct+1 .
We assume free capital mobility. ht ) − ηt βc (ct . ht )Et λt+1 At+1 Fk (kt+1 . the marginal disutility
of labor is not simply Uh (ct . ht ) + ηt βh (ct .9) (4. Intuitively. ht ) but rather by Uc (ct . Equating the domestic and world interest rates.13)
. except for the fact that the marginal utility of consumption is not given simply by Uc (ct . The second term in this expression reﬂects the fact that an increase in current consumption lowers the discount factor (βc < 0). ht+1 ) + 1 − δ + Φ (kt+2 − kt+1 ) (4.8)
λt = Uc (ct . a unit decline in the discount factor reduces utility in period t by ηt . ht ) − βh (ct .

t ≥ 0. and k0 and the exogenous process { t }.
t+1
∼ N IID(0.9) and (4. t (4. h) =
−1
β(c.
−ψ1
The assumed functional forms for the period utility function and the discount factor imply that the marginal rate of substitution between consumption and leisure depends only on labor. ht . it . which in equilibrium equals the real wage rate. λt .
(4. The above expres-
. ct . yt . In eﬀect.4)-(4.15)
The right-hand side of this expression is the marginal product of labor. Chapter 4 The law of motion of the productivity shock is given by:
41
ln At+1 = ρ ln At +
t+1 . who uses the following functional forms for preferences and technology: c − ω −1 hω 1−γ
1−γ
U (c. combining equations (4. ht ). σ 2 ).14).11) yields hω−1 = At Fh (kt .Lectures in Open Economy Macroeconomics. h) = k α h1−α Φ(x) = φ 2 x . kt+1 . ηt . The left-hand side is the marginal rate of substitution of leisure for consumption. h) = 1 + c − ω −1 hω F (k. given the initial conditions A0 .14)
A competitive equilibrium is a set of processes {dt . 2 φ > 0. d−1 .
We parameterize the model following Mendoza (1991). rt } satisfying (4.

455 ψ1 . This parameter determines the stationarity of the model and the speed of convergence to the steady state. The parameter values are shown on table 4. The value assigned to ψ1 is set so as to match the average Canadian trade-balance-to-GDP ratio. equilibrium condition (4. Mendoza calibrates the model to the Canadian economy. The time unit is meant to be a year.42
Mart´ Uribe ın
Table 4. use equation (4. We also follow Mendoza (1991) in assigning values to the structural parameters of the model.1: Calibration of the Small Open RBC Economy γ 2 ω 1.
It follows from this expression that the steady-state capital-labor ratio is independent of the parameter ψ1 . To see how in steady state this ratio is linked to the value of ψ1 . All parameter values are standard in the real-business-cycle literature.15) implies that the steady-state value of hours is also independent of ψ1 and given by k h
α
1 ω−1
h = (1 − α)
.12) in steady state to get k = h α r+δ
1 1−α
.028 r 0.42 σ 0. Given the capital-labor ratio.0129
sion thus states that the labor supply depends only upon the wage rate and in particular that it is independent of the level of wealth.1.11 α .04 δ 0.32 φ 0.
. It is of interest to review the calibration of the parameter ψ1 deﬁning the elasticity of the discount factor with respect to the composite c − hω /ω.1 ρ 0.

Now note that in the steady state the trade balance. rt . and ω. investment (i = δk). we can ﬁnd directly the steady-state values of capital.14) where the vector xt ≡ {dt−1 . h. which uses the speciﬁc functional form assumed for the discount factor. Clearly. λt . Chapter 4
43
Given the steady-state values of hours and the capital-labor ratio. independently of ψ1 . and output (y = k α h1−α ).
Approximating Equilibrium Dynamics
We look for solutions to the equilibrium conditions (4. is given by y − c − i.4)-(4.Lectures in Open Economy Macroeconomics. we have that the transversality condition limj→∞ Et dt+j /(1 + r)j = 0 is always satisﬁed. Because in any such solution the stock of debt is bounded. yt .8)-(4. tb.8) imply the following steady-state condition relating the trade balance to ψ1 : [1 + y − i − tb − hω /ω]−ψ1 (1 + r) = 1.4)-(4. At } ﬂuctuates in a small neighborhood around its nonstochastic steady-state level.6) and (4. it follows that this expression can be solved for ψ1 given tb/y.
Recalling that y. ht . kt . r. δ. the larger is the trade-balance-to-output ratio. The above expression can be solved for the trade balance-to-output ratio to obtain: (1 + r)1/ψ1 + tb i =1− − y y y
hω ω
−1
.14) of ten equations in ten variables given by the elements of the vector xt . The system
. We thus focus on bounded solutions to the system (4. given α. ηt . ct . it . nd i are independent of ψ1 . This expression and equilibrium condition (4. the larger is ψ1 .

The resulting linear system can be readily solved using well-established techniques. The technique we will employ here consists in applying a ﬁrst-order Taylor expansion (i. This is the case.
where A and B are square matrices conformable with xt . Appendix A displays the linearized equilibrium conditions of the RBC model studied
. We therefore must resort to an approximate solution. where w denotes the steady-state value of wt . linearizing) the system of equations around the nonstochastic steady state. This is the case.. we introduce a convenient variable transformation. with output or investment. Some other variables are more naturally expressed in levels. Closed form solutions for this type of system are not typically available. Before linearizing the equilibrium conditions.e. ˆ The linearized version of the equilibrium system can then be written as
Axt+1 = Bxt .44 can be written as Et f (xt+1 . we deﬁne wt ≡ wt − w.
Mart´ Uribe ın
This expression describes a system of nonlinear stochastic diﬀerence equations. with net interest rates or variables that can take negative values. It is useful to express some variables in terms of percent deviations from their steady-state value. for instance. such as the trade balance. say wt . Note that ˆ for small deviations of wt from w it is the case that wt ≈ (wt − w)/w. There are a number of techniques that have been devised to solve dynamic systems like the one we are studying. For any such variable. xt ) = 0. we deﬁne wt ≡ log(wt /w). For this type of variable. for instance.

the elements of A and B. we impose a terminal condition requiring that at any point in time the system be expected to converge to the nonstochastic steady state. and yt . The vector xt contains 10 variables. ct . namely. That appendix also shows how to compute second moments and impulse response functions. Formally. State variables are variables whose values in any period t ≥ 0 are either predetermined (i..
Appendix B shows in some detail how to solve linear stochastic systems like the one describing the dynamics of our linearized equilibrium conditions. that is. Co-states are endogenous variables whose values are not predetermined in period t. The linearized system has three known initial conditions k0 . and At . d−1 . the terminal condition takes the form
j→∞
lim |Et xt+j | = 0.
. determined before ˆ t) or determined in t but in an exogenous fashion. To determine the initial value of the remaining seven variables. λt .e. kt and ˆ ˆ dt−1 are endogenous state variables and At is an exogenous state variable. All the coeﬃcients of the linear system. ηt . kt . ˆ t . In our model. are known functions of the deep structural parameters of the model to which we assigned values when we calibrated the model. 3 are ˆ state variables.Lectures in Open Economy Macroeconomics. Chapter 4
45
here. Of these 10 variables. The remaining 7 elements of xt that is. dt−1 . are ˆ ˆ h ˆ ˆ co-state variables. and A0 . rt . it .

59 0.46
Mart´ Uribe ın
Table 4.8 2 1. For instance.xt−1 0.94 0. the model overestimates the correlations of hours and consumption with output. Standard deviations are measured in percentage points.3 ρxt .9
Model ρxt .GDPt 1 0.61 0.1 2. it correctly implies a volatility ranking featuring investment above output and output above consumption.026
Note.31 0.54 0. For we picked values for the parameters σ .5 2.2
The Model’s Performance
Table 4.3 9.66 1 -0. the volatility of investment.33 0.66 ρxt .61 0.xt−1 0.5 1.2 displays some unconditional second moments of interest implied by our model. On the downside. φ.07 0. Also in line with the data is the model’s prediction of a countercyclical trade balance-to-output ratio. It should not come as a surprise that the model does very well in replicating the volatility of output. This prediction is due to the assumed functional form for the period utility index. Note in particular that the implied correlation between hours and output is exactly unity. In eﬀect.61 0.7 0.8 2. and the serial correlation of output. and ρ so as to match these three moments.15).8 -0. equating the marginal product of
.1 1.1 2.012 0.2: Empirical and Theoretical Second Moments Variable σx t y c i h
tb y ca y
Canadian Data ρxt . Empirical moments are taken from Mendoza (1991).7 0.5 9.13 σx t 3.
4.64 0. But the model performs relatively well along other dimensions. equilibrium condition (4.GDPt 1 0.

The model predicts an expansion in output.5 1 0. investment. The log-linearized version of this condition t is ω ht = yt . displays the impulse response functions of a number of variables of interest to a technology shock of size 1 in period 0.5 1 Output
0
2
4 Hours
6
8
10
0
2
4 6 8 Trade Balance / GDP
10
0
2
4 6 8 Current Account / GDP
10
labor to the marginal rate of substitution between consumption and leisure.5 0 1.5 0 1 0.
. which implies that ht and yt are perfectly correlated.1.5 0 −0. consumption. resulting in a deterioration of the trade balance.5 0 10 5 0 −5 1 0.5 −1 0 2 4 6 8 10 0 2 4 6 Investment 8 10 0.5 0 −0. Figure 4.Lectures in Open Economy Macroeconomics. Chapter 4
47
Figure 4.5 −1 0 2 4 6 8 10 1. The increase in domestic absorption is larger than the increase in output.5 1 0. can be written as hω = (1−α)yt . and hours.1: Responses to a Positive Technology Shock
Consumption 2 1.

48
Mart´ Uribe ın
Figure 4. Second. To illustrate this conclusion.2: Response of the Trade-Balance-To-Output Ratio to a Positive Technology Shock
0.2
−0.2 displays the impulse response function of the trade balance-to-GDP ratio to a technology shock of unit size in period 0 under three alternative parameter
. capital adjustment costs must not be too stringent.2
0
−0.8
0
1
2
3
4
5
6
7
8
9
4. In particular.4
0.2. two conditions must be met for the model to generate a deterioration in the external accounts in response to a mean-reverting improvement in total factor productivity. ﬁgure 4.1
The Role of Capital Adjustment Costs
In previous chapters. First. we deduced that the negative response of the trade balance to a positive technology shock was not a general implication of the neoclassical model.6
−0.6 benchmark high φ low ρ 0.4
−0. the productivity shock must be suﬃciently persistent.

1. As a result. by equation (4.3
Alternative Ways to Induce Stationarity
In the RBC model analyzed thus far households have endogenous discount factors. because the productivity shock is expected to die out quickly. The solid line reproduces the benchmark case from ﬁgure 4. We will refer to that model as the ‘internal discount factor model. In addition. At the same time. The crossed line depicts the case of high capital adjustment costs.084.’
. the response of aggregate domestic demand is weaker. As a result. Chapter 4
49
speciﬁcations.Lectures in Open Economy Macroeconomics. Here the parameter φ equals 0. a value three times as large as in the benchmark case. high adjustment costs discourage ﬁrms from increasing investment spending by as much as in the benchmark economy. The broken line depicts an economy where the persistence of the productivity shock is half as large as in the benchmark economy (ρ = 0. the initial responses of output and hours are identical in both economies (recall that. results in an improvement in the trade balance when productivity shocks are not very persistent. the temporariness of the shock induces households to save most of the increase in income to smooth consumption over time. the response of investment is relatively weak. because the size of the productivity shock is the same as in the benchmark economy. In this environment. the expansion in aggregate domestic absorption is modest.
4. The combination of a weak response in domestic absorption and an unchanged response in output.15). In this case. ht depends only on kt and At ).21). leading to an improvement in the trade balance-to-output ratio.

Our analysis follows closely Schmitt-Groh´ and Uribe (2003). Had we assumed a constant discount factor. Two problems emerge when the linear approximation possesses a unit root. In this section. one can no longer claim that the linear system behaves like the original nonlinear system— which is ultimately the focus of interest—when the underlying shocks have suﬃciently small supports. which are the most common descriptive statistics of the business cycle. but expands their analysis by including a e model with an internal interest-rate premium. it is impossible to compute unconditional second moments. etc. serial correlations. such as standard deviations. First.
4.3.50
Mart´ Uribe ın
or IDF model. The inclusion of an endogenous discount factor responds to the need to obtain stationary dynamics up to ﬁrst order. suppose that the discount factor depends not upon the agent’s own consumption and eﬀort.1
External Discount Factor (EDF)
Consider an alternative formulation of the endogenous discount factor model where domestic agents do not internalize the fact that their discount factor depends on their own levels of consumption and eﬀort.. correlations with output. when the variables of interest contain random walk elements. Alternatively. we analyze and compare alternative ways of inducing stationarity in small open economy models. the log-linearized equilibrium dynamics would have contained a random walk component. Second. but rather on the average per capita levels of these
.

Formally. (4. The ﬁrst-order conditions of the household’s maximization problem are (4.7).7).14). ht )(1 + rt )Et λt+1 c ˜
(4. d−1 . (4.22) (4.17)-(4. (4. ˜ ˜ kt+1 . individual and average per capita levels of consumption and eﬀort are identical. ht . which ˜ the individual household takes as given. Chapter 4 variables.13).21)
A competitive equilibrium is a set of processes {dt .
(4. ht .
. preferences are described by (4. ct . it . rt . That is.2).Lectures in Open Economy Macroeconomics. (4.4)-(4.17)
λt = Uc (ct .20) In equilibrium. λt . ht ) = λt At Fh (kt . ht )Et λt+1 At+1 Fk (kt+1 . (4. given A0 . ht ) −Uh (ct . yt .19)
λt [1+Φ (kt+1 −kt )] = β(˜t .22) all holding with equality.18) (4. and k0 and the stochastic process { t }. (4.16) holding with equality and: λt = β(˜t .2). and
51
θt+1 = β(˜t .16)
˜ where ct and ht denote average per capital consumption and hours. At } satisfying (4. ct = ct ˜ and ˜ ht = ht . (4.1).4)-(4. ht )
(4. ct . ht )θt c ˜
t ≥ 0. ht+1 ) + 1 − δ + Φ (kt+2 − kt+1 ) c ˜ (4.

equation (4.1.1). The economy with a noninternalized discount factor features an externality.52
Mart´ Uribe ın
Note that the equilibrium conditions include one Euler equation less.23)
where r denotes the world interest rate and p(·) is a country-speciﬁc interest rate premium.
It is remarkable that the degree of computational complexity is reversed when the computational technique consists in iterating a Bellman equation over a discretized state space. Formally.
4. which we denote by dt . stationarity is induced by assuming that the interest rate faced by domestic agents. than the standard endogenous-discountfactor model of subsection 4. θt = β t . A similar comment applies to the computation of equilibrium in a model with an interest-rate premium that depends on the aggregate level of extern! al debt to be discussed in the next subsection. Unlike in the previous model. ηt .2
External Debt-Elastic Interest Rate (EDEIR)
Under an external debt-elastic interest rate. rt . which complicates signiﬁcantly the task of computing the equilibrium by value function iterations. (4. Speciﬁcally. preferences are assumed to display a constant subjective rate of discount. These fewer elements facilitate the computation of the equilibrium dynamics using perturbation methods.
1
.1 We evaluate the model using the same functional forms and parameter values as in the IDF model. and one variable less. Preferences are given by equation (4.3.10). where β ∈ (0. The function p(·) is assumed to be strictly increasing. is increasing in ˜ the aggregate level of foreign debt. rt is given by ˜ rt = r + p(dt ). 1) is a constant parameter.

We adopt the same forms for the functions U . We calibrate the parameters γ. given (4. δ. ht .
(4. ht ) = λt . rt . (4. that is.26)
λt [1+Φ (kt+1 −kt )] = βEt λt+1 At+1 Fk (kt+1 . ¯ where ψ2 and d are constant parameters. and k0 . F .24) (4. ht ) = λt At Fh (kt . kt+1 . ct . and (4. (4. A0 . We use the following functional form for the risk premium:
¯
p(d) = ψ2 ed−d − 1 .25) (4.7). and Φ as in the IDF of subsection 4. λt }∞ t=0 satisfying (4.27) Because agents are assumed to be identical. in equilibrium aggregate per capita debt equals individual debt. and σ using the values shown in table 4. −Uh (ct . ˜ dt = dt . yt .4)-(4. φ.4)-(4. α. Chapter 4
53
The representative agent’s ﬁrst-order conditions are (4. We set the subjective discount factor equal to the world
. dt+1 . ω.1.Lectures in Open Economy Macroeconomics. ρ.7) holding with equality and λt = β(1 + rt )Et λt+1 Uc (ct . d−1 .23)-(4. r. ht ).28) all holding with equality.14).28)
˜ A competitive equilibrium is a set of processes {dt . ht+1 ) + 1 − δ + Φ (kt+2 − kt+1 ) .1. it .

Formally. that is. We set d so that the steady-state level of foreign debt equals the one implied by Model 1.3.29)
. The resulting values of β. dt . d. To see this. the model is identical to the model featuring an external debt-elastic interest rate. note that in steady state.96 0. In all other aspects.23) and (4. 1+r
¯ The parameter d equals the steady-state level of foreign debt.3: Model 2: Calibration of Parameters Not Shared With Model 1 ¯ d β ψ2 0. we set the parameter ψ2 so as to ensure that this model and Model 1 generate the same volatility in the current-account¯ to-GDP ratio.3
Internal Debt-Elastic Interest Rate (IDEIR)
The model with an internal debt-elastic interest rate assumes that the interest rate faced by domestic agents is increasing in the individual debt position.7442 0.24) together with the assumed form of the interest-rate premium imply that 1 = β 1 + r + ψ2 ed−d − 1
¯
.
4. and ψ2 are given in Table 4. in the IDEIR model the interest rate is given by rt = r + p(dt ).54
Mart´ Uribe ın
Table 4. β= 1 .3.000742
interest rate. the equilibrium conditions (4. (4. Finally. If follows that in the steady state the interest rate premium is ¯ nil. The fact that β(1 + r) = 1 then implies
¯ that d = d.

30)
This expression features the derivative of the premium with respect to debt because households internalize the fact that as they increase their debt positions. A competitive equilibrium is a set of processes {dt . (4.25)-(4. yields d = 0. We assume the same functional forms and parameter values as in the model with an external interest-rate premium. the steady-state version of equation (4.7442.30) imposes the following restriction on d. λt }∞ t=0 satisfying (4. given (4. and k0 . r denotes the world interest rate and p(·) is a householdspeciﬁc interest-rate premium.
¯ which. (4. However. which now takes the form
λt = β[1 + r + p(dt ) + p (dt )dt ]Et λt+1 .
(4. rt . so does the interest rate they face in ﬁnancial markets. given d = 0.7). as before.
¯
(1 + d)ed−d = 1. it .30).27).29). Chapter 4
55
where. all holding with equality. kt+1 .Lectures in Open Economy Macroeconomics. and (4. In the household’s problem the only optimality condition that changes relative to the case with an external premium is the Euler equation for debt accumulation. given by ∂[ρ(dt )dt ]/∂dt . ht . yt . We note that in the model analyzed in this subsection the steady-state level of debt is no longer equal ¯ to d. the marginal country premium. An
. ¯ The fact that the steady-state debt is lower than d implies that the country premium is negative in the steady state.4045212. is nil in the steady state. Recalling that β(1 + r) = 1. ct .4)-(4.14). A0 . d−1 .

Preferences and technology are as in the EDEIR model of section 4. 2
dt = (1 + rt−1 )dt−1 − yt + ct + it + Φ(kt+1 − kt ) +
(4.3. equation (4. In contrast to what is assumed in the EDEIR model.7). (4.2. but the marginal premium is positive and equal ¯ to ψ2 d.56
Mart´ Uribe ın
¯ alternative calibration strategy is to impose d = d.13) holds. The value of this increase in consumption in terms of utility is given by the left-hand side of the above
. (4. that is. here the interest rate at which domestic households can borrow from the rest of the world is constant and equal to the world interest. The sequential budget constraint of the household is given by ψ3 ¯ (dt − d)2 .5)-(4. and adjust β to ensure that equation (4. The ﬁrst-order conditions associated with the household’s maximization problem are (4. In this case.31) holding with equality and ¯ λt [1 − ψ3 (dt − d)] = β(1 + rt )Et λt+1 (4.30) holds in steady state.25)-(4.31)
¯ where ψ3 and d are constant parameters deﬁning the portfolio adjustment cost function.
4.27).32)
This optimality condition states that if the household chooses to borrow an additional unit.4
Portfolio Adjustment Costs (PAC)
In this model. the country premium vanishes in the steady state.3. then current consumption increases by one unit minus the ¯ marginal portfolio adjustment cost ψ3 (dt − d). stationarity is induced by assuming that agents face convex costs of holding assets in quantities diﬀerent from some long-run level.

1 generate the same volatility in the current-account-to-GDP ratio. and the interest rate faced by domestic households (equations (4. Preferences and technology are parameterized as in the EDEIR model. This parameter value is almost identical to that assigned to ψ2 in the EDEIR model. and σ take the values displayed in table 4. the Euler equations associated with the optimal choice of foreign bonds (equations (4.27). Chapter 4
57
equation. λt }∞ t=0 satisfying (4. and k0 . r. d−1 . This is because the log-linearized versions of models 2 and 3 are almost identical. rt . we assign the value 0. The parameters γ. At the optimum.13) and (4. The value of this repayment in terms of today’s utility is given by the right-hand side. A competitive equilibrium is a set of processes {dt .7). As in model 2. Finally.32) imply that the parameter ¯ ¯ ¯ d determines the steady-state level of foreign debt (d = d). ω. This assumption and equation (4.5)-(4. and EDEIR (see table 4. (4. ρ. δ.32)).1. kt+1 . φ. We calibrate d so that the steady-state level of foreign debt equals the one implied by models IDF. ct .4) and (4. ht . and (4.32) all holding with equality. which ensures that this model and the IDF model of section 4. The log-linearized versions of the resource constraints are the same in both models. EDF.13). Next period.14). the subjective discount factor is assumed to satisfy β(1+r) = 1.3). (4.! The log-linear approximation to the domestic interest rate is given by 1 + rt = ψ2 d(1 + r)−1 dt in the EDEIR and by 1 + rt = 0 in the PAC
.24) and (4.31).Lectures in Open Economy Macroeconomics. it . (4. given (4. A0 .23)). the marginal beneﬁt of a unit debt increase must equal its marginal cost. Indeed. α.31)).25)-(4. the models share all equilibrium conditions but the resource constraint (equations (4.00074 to ψ3 . the household must repay the additional unit of debt plus interest. yt .

(4.58
Mart´ Uribe ın
model.34)
. In those models agents have access to a single ﬁnancial asset that pays a risk-free real rate of return.5
Complete Asset Markets (CAM)
All model economies considered thus far feature incomplete asset markets.3.33)
where rt+1 is a stochastic discount factor such that the period-t price of a random payment bt+1 in period t + 1 is given by Et rt+1 bt+1 .
(4. it follows that 1/[Et rt+1 ] denotes the risk-free real interest rate in period t. Households are assumed to be subject to a no-Ponzi-game constraint of the form
j→∞
lim Et qt+j bt+j ≥ 0. Preferences and technology are as in model 2. The period-by-period budget constraint of the household is given by
Et rt+1 bt+1 = bt + yt − ct − it − Φ(kt+1 − kt ). the log-linearized versions of the Euler equation for debt are λt = ψ2 d(1 + r)−1 dt + Et λt+1 in model 2 and λt = ψ3 ddt + Et λt+1 in Model 3. In the model studied in this subsection. It follows that for small values of ψ2 and ψ3 satisfying ψ2 = (1 + r)ψ3 the EDEIR and PAC models imply similar dynamics. In turn. agents have access to a complete array of state-contingent claims. This assumption per se induces stationarity in the equilibrium dynamics.
4. Note that because Et rt+1 is the price in period t of an asset that pays 1 unit of good in every state of period t + 1.

35).Lectures in Open Economy Macroeconomics. t t+1
(4. one ﬁrst-order condition of the foreign household is an equation similar to (4. . The variable qt represents the stochastic discount factor between periods 0 and t. Consequently. (4.25)-(4. we have
λ∗ rt+1 = βλ∗ . and (4.36)
Note that we are assuming that domestic and foreign households share the same subjective discount factor.35)
A diﬀerence between this expression and the Euler equations that arise in the models with incomplete asset markets studied in previous sections is that under complete markets in each period t there is one ﬁrst-order condition for each possible state in period t + 1. (4. (4. We have that qt satisﬁes q t = r 1 r2 . with q0 ≡ 1. Combining the domestic and foreign Euler
.6).33). whereas under incomplete markets the above Euler equation holds only in expectations.27). (4.5). such that the period-0 price of a random payment bt in period t is given by E0 qt bt .34) holding with equality and λt rt+1 = βλt+1 . . The ﬁrst-order conditions associated with the household’s maximization problem are (4. r t . Chapter 4
59
at all dates and under all contingencies.
In the rest of the world. Letting starred letters denote foreign variables or functions. agents have access to the same array of ﬁnancial assets as in the domestic economy.

5). λt }∞ t=0 satisfying (4. (4. The parameter ψ4 is set so as to ensure that the steadystate level of consumption is the same in this model as in the IDF. φ. it . α. λt λ∗ t
Mart´ Uribe ın
This expression holds at all dates and under all contingencies.35) and (4. where ψ4 ≡ ξλ∗ is a constant parameter. β.60 equations—equations (4. ω. F . t
where ξ is a constant parameter determining diﬀerences in wealth across countries. where λ∗ is a parameter. This means that λ∗ must be taken as an exogenous variable. EDEIR. The functions U . and PAC models. and k0 .
λt = ξλ∗ . and σ take the values displayed in tables 4.36)—yields λ∗ λt+1 = t+1 .6). (4. The above equilibrium condition then becomes λ t = ψ4 . The parameters γ. We assume that the domestic economy is small. Formally. δ. given (4. This means that the domestic marginal utility of consumption is proportional to its foreign counterpart. A0 . and Φ are parameterized as in the previous models. (4. kt+1 .14). yt . we assume that λ∗ is constant t and equal to λ∗ .27).1 and 4.25)-(4.37). ρ. and (4.37)
. Because we are interested only t in the eﬀects of domestic productivity shocks. A competitive equilibrium is a set of processes {ct . ht . EDF.3.

d−1 .3. (c) agents face no frictions in adjusting the size of their portfolios. ct . The appendix shows the log-linear version of the IDF model of subsection 4. We calibrate the model using the parameter values displayed in tables 4. and therefore does not have well deﬁned unconditional second moments. yt . CAM. A competitive equilibrium in the nonstationary model is a set of processes {dt . EDF. in this section we describe a version of the small open economy model that displays no stationarity. and (d) markets are incomplete in the sense that domestic households have only access to a single risk-free international bond.6
The Nonstationary Case (NC)
For comparison with the models considered thus far. The Matlab computer code used to compute the unconditional second moments and
2 Model 5 is nonstationary. and k0 .14).3. IDEIR. kt+1 .24)t=0 (4. it . (b) the interest rate at which domestic agents borrow from the rest of the world is constant (and equal to the subjective discount factor).
4.13). we compute the equilibrium dynamics by solving a loglinear approximation to the set of equilibrium conditions.7).Lectures in Open Economy Macroeconomics.
. λt }∞ satisfying (4. and (4. This speciﬁcation of the model induces a random walk component in the equilibrium marginal utility of consumption and the net foreign asset position. In this model (a) the discount factor is constant.4)-(4.3. ht .1 and 4. A0 . rt .27) all holding with equality. and NC models.7
Quantitative Results
Table 4.4 displays a number of unconditional second moments of interest implied by the IDF.2 For all models. EDEIR. PAC. Chapter 4
61
4.1. given (4. (4.

In response to a positive technology shock. its role in determining the cyclicality of the trade balance is smaller. For all variables but consumption and the trade-balance-to-GDP ratio. whereas the models featuring incomplete asset markets imply that it is negative.
. the only small but noticeable diﬀerence is given by the responses of consumption and the trade-balanceto-GDP ratio in the complete markets model. consumption increases less when markets are complete than when markets are incomplete. which as expected predicts less volatile consumption. The main result of this section is that regardless of how stationarity is induced. the model’s predictions regarding second moments are virtually identical.3 demonstrates that all of the models being compared imply virtually identical impulse response functions to a technology shock. EDEIR.4. This in turn. the complete markets case. the CAM model predicts that the correlation between output and the trade balance is positive.62
Mart´ Uribe ın
impulse response functions for all models presented in this section is available at www. Each panel shows the impulse response of a particular variable in the six models. EDF. the impulse response functions are so similar that to the naked eye the graph appears to show just a single line.econ. Figure 4. The low volatility of consumption in the complete markets model introduces an additional diﬀerence between the predictions of this model and the IDF. As a result. Because consumption is smoother in the CAM model. This result is evident from table 4. and PAC models. The only noticeable diﬀerence arises in the CAM model.duke. Again. leads to a smaller decline in the trade balance in the period in which the technology shock occurs. IDEIR.edu/∼uribe.

61 0.5 0.61 0.5 1.9 9.51 0.1 2.1 1.43 0.5 0.069 0.1 1.66 1 -0.8 1.05
PAC 3.013 0.32 0.07 0. CAM = Complete Asset Markets.069 0.67 1 -0.068 0. ct−1 ) corr(it .1 2. yt ) corr(it . Chapter 4
63
Table 4.012 0. cat−1 ) y y
EDF 3.66 1 0.4: Implied Second Moments IDF Volatilities: std(yt ) std(ct ) std(it ) std(ht ) std( tbtt ) y std( catt ) y corr(yt .94 0.041
EDEIR 3.1 2.3 0.62 0.62 0.32 0.61 0.62 0.5 1.
.Lectures in Open Economy Macroeconomics.026
Serial Correlations: 0. EDF = External Discount Factor. yt ) corr( tbtt .62 0. ht−1 )
t−1 corr( tbtt . yt−1 ) corr(ct .7 9 2. tbt−1 ) y y t−1 corr( catt .33 0.8 1.7 0.31 0.13
Note.39
Correlations with Output: corr(ct .1 2.5 9 2. Standard deviations are measured in percent per year.67 1 -0.66 1 -0.6 1.1 2.1 2.61 0.1 1.62 0. yt ) y 1 0.025
IDEIR 3.043 0.07 0.78 0.4 0. IDF = Internal Discount Factor.84 0.85 0.3 0.5 0.1 2. EDEIR = External Debt-Elastic Interest Rate. yt ) y corr( catt . yt ) corr(ht .6
3.61 0.1 1.61 0.044 0.32 0.1 2.051
CAM 3.3 9.89 0. NC = Nonstationary Case. it−1 ) corr(ht . IDEIR = Internal Debt-Elastic Interest Rate.76 0.07 0.62 0.78 0.94 0.68 1 -0.7 9 2.7 0.1 1. PAC = Portfolio Adjustment Costs.036 0.5 0.3 9.61 0.1 1.1 1.5 0.

5 −1 0 2 4 6 8 10 0 2 4 6 Investment 8 10 0.5 0 10 5 0 −5 1 0.5 1 0. Circles: Model without stationarity inducing elements.
.5 1 Consumption
0
2
4 Hours
6
8
10
0
2
4 6 8 Trade Balance / GDP
10
0
2
4 6 8 Current Account / GDP
10
Note.5 0 −0.3: Impulse Response to a Unit Technology Shock in Models 1 .64
Mart´ Uribe ın
Figure 4.5 −1 0 2 4 6 8 10 1.5 0 1.5 1 0. Dash-dotted line: Portfolio adjustment cost model. Squares: Endogenous discount factor model without internalization.5 0 −0. Solid line: Endogenous discount factor model. Dashed line: Debt-elastic interest rate model. Dotted line: complete asset markets model.5 0 1 0.5
Output 2 1.

In the log-linearization we are using the particular forms assumed for the
.4
Appendix A: Log-Linearization
Let xt ≡ log(xt /¯) denote the log-deviation of xt from its steady-state value.
βch
≡ hβch /βc .Lectures in Open Economy Macroeconomics. Chapter 4
65
4. x Then.
βc cc
≡ cβc /β.
βh
≡ hβh /β. si = i/y. stb ≡ tb/y. sc ≡ c/y.
ch
βcc
≡ cβcc /βc .
c
≡
= cUcc /Uc .
= hUch /Uc . where cUc /U . taking a ﬁrst-order log-linearization of the model we get: r rt−1 + stb (1 + r)dt−1 − r[yt − sc ct − si it ] 1+r yt = At + αkt + (1 − α)ht kt+1 = (1 − δ)kt + δ it λt = λt = (1 − β) c (1 − β) c − β [
βc
stb dt = stb
r rt + 1+r
cc ct
βc ct
+
βh ht
+ Et λt+1
βc
+
ch ht ]
−
β (1 − β)
c
−β
[ηt +
βc βc ct
βcc ct
+
βch ht ]
ηt = (1 − β)[ c Et ct+1 + (1 − β) h (1 − β) h + β [
βh hc ct + hh ht ]+
h Et ht+1 ]
+ β[Et ηt+1 + [ηt +
βh
+
βh ht ]
β (1 − β)
βh h
+β
βhc ct + βhh ht ]
= λt +At +αkt −αht
λt + φk kt+1 − φk kt =
βc ct
+
βh ht
+ Et λt+1 + β(β −1 + δ − 1)[Et At+1
+(1 − α)Et ht+1 − (1 − α)kt+1 + βφkEt kt+2 − βφk kt+1
rt = 0 ˆ ˆ At = ρAt−1 + t .

.66
Mart´ Uribe ın
production function and the capital adjustment cost function.

In addition. we introduced one particular strategy.38)
where Et denotes the mathematical expectations operator conditional on information available at time t. Here we explain in detail how to solve the resulting system of linear stochastic diﬀerence equations. the complete set of equi-
. xt ) = 0. The vector xt denotes predetermined (or state) variables and the vector yt denotes nonpredetermined (or control) variables. (Beyond the initial condition. Reduced forms of this sort are common in Macroeconomics. yt .5
Appendix B: Solving Dynamic General Equilibrium Models
The equilibrium conditions of the simple real business cycle model we studied in the previous chapter takes the form of a nonlinear stochastic vector diﬀerence equation. A problem that one must face is that. The initial value of the state vector x0 is an initial condition for the economy. In the previous chapter.
(4. it is impossible to solve such systems. xt+1 . The equilibrium conditions of a wide variety of dynamic stochastic general equilibrium models can be written in the form of a nonlinear stochastic vector diﬀerence equation
Et f (yt+1 . consisting in linearizing the equilibrium conditions around the nonstochastic steady state. Chapter 4
67
4. we show how to use the solution to compute second moments and impulse response functions.Lectures in Open Economy Macroeconomics. in general. But fortunately one can obtain good approximations to the true solution in relatively easy ways.

3 The
is assumed to have a bounded support and to be independently
and identically distributed.40)
The vector xt of predetermined variables is of size nx × 1 and the vector yt of nonpredetermined variables is of size ny × 1. like a no-Ponzi game constraint. We deﬁne n = nx + ny . with mean zero and variance/covariance matrix ˜ I.68
Mart´ Uribe ın
librium conditions also includes a terminal condition.38) is of the form: yt = g (xt ) ˆ and ˆ xt+1 = h(xt ) + ησ
t+1 . The function f then maps Rny × Rny × Rnx × Rnx into Rn .
(4. we t assume that x2 follows the exogenous stochastic process given by t ˜ x2 = h(x2 .) The state vector xt can be partitioned as xt = [x1 . The eigenvalues of the Jacobian of the function h with respect to its ﬁrst argument evaluated at the non-stochastic steady state are assumed to lie within the unit circle. We omit such a constraint here because we focus on approximating stationary solutions. The solution to models belonging to the class given in equation (4. Speciﬁcally.39)
(4. σ) + η σ ˜ t+1 t where both the vector x2 and the innovation t vector
t
t+1 .
3 It is straightforward to accommodate the case in which the size of the innovations vector t is diﬀerent from that of x2 . t
. x2 ] .
t
are of order n × 1. The vector x1 consists of endogenous predetermined state t t t variables and the vector x2 of exogenous state variables.

Chapter 4 The matrix η is of order nx × n and is given by ∅ η = . that is. x ¯
Taking a Taylor series approximation of the functions g and h around the point (x. η ˜
69
ˆ The shape of the functions h and g will in general depend on the amount ˆ of uncertainty in the economy. a perturbation methods ﬁnds a local approximation of the functions g and h. σ) = (¯.41)
(4.
(4. The key idea of perturbation methods is to interpret the solution to the model as a function of the state vector xt and of the parameter σ scaling the amount of uncertainty in the economy. σ ). By a local approximation. x ¯
.
yt = g(xt . σ) + ησ
t+1 .Lectures in Open Economy Macroeconomics. we mean an approximation that is valid in the neighborhood of a particular point (¯.
Given this interpretation. σ) and xt+1 = h(xt .42)
where the function g maps Rnx × R+ into Rny and the function h maps Rnx × R+ into Rnx . σ ) we have (for the moment to keep the notation simple.

σ ). g(x. σ )(σ − σ )2 + . σ) = 0
∀x.
Fxk σj (x. σ )(x − x) + gσ (¯. σ )(σ − σ ) x ¯ x ¯ ¯ x ¯ ¯ 1 + gxx (¯. σ) ≡ Et f (g(h(x. substitute the proposed solution given by equations (4. . . k. j. σ )(σ − σ )2 + . Formally. σ )(x − x) + hσ (¯. x ¯ To identify these derivatives.44)
. We use a prime to indicate variables dated in period t + 1. . x ¯ ¯ 2 h(x. σ) = h(¯. σ ) + hx (¯. x) (4.42) into equation (4. . σ) + ησ . σ) + ησ . σ )(x − x)(σ − σ ) x ¯ ¯ ¯ 1 + hσσ (¯. x ¯ ¯ 2 The unknowns of an nth order expansion are the n-th order derivatives of the functions g and h evaluated at the point (¯. Because F (x. σ )(x − x)2 + gxσ (¯. σ )(x − x)(σ − σ ) x ¯ ¯ x ¯ ¯ ¯ 2 1 + gσσ (¯. and deﬁne
F (x. σ )(σ − σ ) x ¯ x ¯ ¯ x ¯ ¯ 1 + hxx (¯.38). . σ) = g(¯. σ.
Here we are dropping time subscripts. it must be the case that the derivatives of any order of F must also be equal to zero. σ) must be equal to zero for any possible values of x and σ. σ ) + gx (¯. h(x. σ).41) and (4.70 let’s assume that nx=ny=1)
Mart´ Uribe ın
g(x.
(4.43) = 0. σ). σ )(x − x)2 x ¯ ¯ 2 +hxσ (¯.

0) = y x ¯ and h(¯. y . a particularly convenient point to approximate the functions g and h around is the non-stochastic steady state. g(¯. x ¯ The remaining unknown coeﬃcients of the ﬁrst-order approximation to g
. x. 0) + hx (¯. The reason why the steady state is a particularly convenient point is that in most cases it is possible to solve for the steady state. With the steady state values in hand. We deﬁne the non-stochastic steady state as vectors (¯. We are looking for approximations to g and h around the point (x. 0)(x − x) + hσ (¯. 0) = x. 0) and x = h(¯. 0)σ x x ¯ x As explained earlier. 0)σ x x ¯ x
h(x. one can then ﬁnd the derivatives of the function F . y ) such x ¯ that f (¯. σ) = (¯. note that if σ = 0. 0) + gx (¯. ¯ x ¯ x then Et f = f . As will become clear below. σ) denotes the derivative of F with respect to x taken k times and with respect to σ taken j times. x) = 0. 0) of the form x
g(x. To see this. σ) = g(¯. 0). σ) = h(¯. xt = x ¯ and σ = 0. y ¯ ¯ ¯ It is clear that y = g(¯.Lectures in Open Economy Macroeconomics. 0)(x − x) + gσ (¯. Chapter 4
71
where Fxk σj (x.

0) = 0. 0) = 0.44). σ). by equation (4. 0) = 0
To ﬁnd those derivatives let’s repeat equation (4. x) = 0.43)
F (x. σ) ≡ Et f (g(h(x. it must be the case that: Fσ (¯. σ) + ησ . h(x. 0) = Et fy [gx (hσ + η ) + gσ ] + fy gσ + fx (hσ + η ) x = fy [gx hσ + gσ ] + fy gσ + fx hσ
This is a system of n equations. Then imposing
Fσ (¯. x and x Fx (¯. x
one can identify gσ and hσ : hσ fy + f y =0 gσ
fy gx + fx
.
Taking derivative with respect to the scalar σ we ﬁnd:
Fσ (¯. g(x. σ) + ησ . σ).72
Mart´ Uribe ın
and h are identiﬁed by using the fact that.

the policy function is independent of the variance-covariance matrix of
t. we have that hσ = 0. Thus. one need not correct the constant term of the approximation to the policy function for the size of the variance of the shocks. In this sense. This question is at the heart of the recent literature on optimal ﬁscal and monetary stabilization policy.
This is an important limitation of
ﬁrst-order perturbation techniques. Thus. These two expressions represent an important theoretical result. any two policies that give rise to the same steady state yield. up to ﬁrst-order the mean of the rate of return of all all assets must be same. that is. ﬁrst-order approximation techniques cannot be used to study risk premia. we can say that in a ﬁrst-order approximation the certainty ¯ equivalence principle holds. Another important question that can in general not be addressed with ﬁrst-order perturbation techniques is how uncertainty aﬀects welfare.Lectures in Open Economy Macroeconomics. up to ﬁrst order.
. and gσ = 0. Chapter 4
73
This equation is linear and homogeneous in gσ and hσ . if a unique solution exists. Because in a ﬁrst-order accurate solution the unconditional expectation of a variable is equal to the non-stochastic steady state. They show that in general. This result implies that in a ﬁrst-order approximation the expected values of xt and yt are equal to their non-stochastic steady-state values x and ¯ y . For example. Because in many economic applications we are interested in ﬁnding the eﬀect of uncertainty on the economy.

hx xt = xt+1 . In this case. x. then postmultiplying the above system equation ??) by ˆ ¯ xt we obtain: ˆ
I I A ˆ ˆ h x xt = B xt gx gx Consider for the moment. x) = (¯. y .
Mart´ Uribe ın
To ﬁnd gx and hx diﬀerentiate (4. x .43) with respect to x to obtain the following system
Fx (¯.
Let xt ≡ xt − x.74 up to ﬁrst-order the same level of welfare. y. x) are y ¯ ¯ ¯ known. a perfect foresight equilibrium. Imposing
Fx (¯. The above expression represents a system of n × nx quadratic equations in the n × nx unknowns given by the elements of gx and hx . 0) = fy gx hx + fy gx + fx hx + fx x
Note that the derivatives of f evaluated at (y . ˆ ˆ
I I A ˆ ˆ xt+1 = B xt gx gx
. Note that both A and B are known. 0) = 0 x
the above expression can be written as:
[fx
I fy ] hx = −[fx gx
I fy ] gx
Let A = [fx
fy ] and B = −[fx
fy ].

we use the generalized Schur decomposition of the matrices A and B. st = . b. A matrix z is orthonormal if z z = zz = I. z.4 The generalized Schur decomposition of A and B is given by upper triangular matrices a and b and orthonormal matrices q and z satisfying:5 qAz = a and qBz = b. Let st ≡ z [I. Chapter 4 We are interested in solutions in which
75
t→∞
lim |ˆt | < ∞ x
We will use this limiting conditions to ﬁnd the matrix gx . In particular.
5 4
s1 t
. 2 0 a22 0 b22 z21 z22 st
More formal descriptions of the method can be found in Sims (1996) and Klein (2000). x Then we have that ast+1 = bst Now partition a. To solve the above system. gx ]ˆt . we will use the Schur decomposition method. b = . and st as a11 a12 b11 b12 z11 z12 a= . z = . Recall that a matrix a is said to be upper triangular if elements aij = 0 for i > j.Lectures in Open Economy Macroeconomics.

without loss of generality.6 Thus. t+1 t 22 Assume. and s2 is of t order ny × 1. and that the number of ratios greater than one is equal to the number of state variables. t+1 t
or b−1 a22 s2 = s2 . x
Because this condition has to hold for any value of the state vector. (c) The eigenvalues of an upper triangular matrix are the elements of its main diagonal. (a) The inverse of a nonsingular upper triangular matrix is upper triangular. (b) the product of two upper triangular matrices is upper triangular. the eigenvalues of b−1 a22 are all less than unity in modulus. that the ratios abs(aii /bii ) are decreasing in i. By construction. Suppose further that the number of ratios less than unity is exactly equal to the number of control variables. z12 is of order nx × ny . it ˆ follows that it must be the case that
z12 + z22 gx = 0. by t t+j the deﬁnition of s2 . xt . 22 the requirement limj→∞ |s2 | < ∞ is satisﬁed only if s2 = 0. Then we have that a22 s2 = b22 s2 . Namely.76
Mart´ Uribe ın
where a22 and b22 are of order ny × ny . nx . this restriction implies that t
(z12 + z22 gx )ˆt = 0.
6
. In turn.
Here we are applying a number of properties of upper triangular matrices. ny .

x
so that
−1 −1
hx = [z11 − z21 z22
z12 ]−1 a11 −1 b11 [z11 − z21 z22
z12 ]. Chapter 4 Solving this expression for gx yields
−1
77
gx = −z22
z12 .
The fact that s2 = 0 also implies that t a11 s1 = b11 s1 . x t Replacing gx . we get
−1 −1
xt+1 = [z11 − z21 z22 ˆ
z12 ]−1 a11 −1 b11 [z11 − z21 z22
z12 ]ˆt .Lectures in Open Economy Macroeconomics. t+1 t
or s1 = a11 −1 b11 s1 t+1 t Now s1 = (z11 + z21 gx)ˆt . x
Combining this expression with the equation describing the evolution of st shown two lines above. we have s1 = [z11 − z21 z22 t
−1
z12 ]ˆt .
.

Now use element (1.78
Mart´ Uribe ın
We can simplify this expression for hx by using the following restrictions: z11 z11 + z21 z21 z11 z12 + z21 z22 I =zz= z12 z11 + z22 z21 z12 z12 + z22 z22 to write:7
−1 hx = z11 a−1 b11 z11 . 1) of z z to write z21 z21 = I − z11 z11 . 1) of z z to get z12 z11 = −1 −1 −1 −z22 z21 . 11
4. use element (2. In this case there is a unique local equilibrium. which is less than the number of control variables. nx . Then the requirement that we wish to study equilibria in which limj→∞ Et |ˆt+j | < ∞ will only yield m restrictions.
. Let’s ﬁrst consider the case that the number of eigenvalues of D with modulus greater than unity is equal to m < ny. −1 Use this expression to eliminate z22 z12 From the square bracket in the expression for hx . we get [z11 + (I − z11 z11 )z11 −1 ].6
Local Existence and Uniqueness of Equilibrium
In the above discussion. It follows that one can choose arbitrary initial values for ny − m elements of y0 and the resulting ﬁrst order solution will still be
7 To obtain this simple expression for hx . But not for every economy this is the case. ny . Then this square bracket becomes [z11 + z21 z21 z11 −1 ]. and that the number of eigenvalues of D with modulus greater than one is equal to the number of state variables. Premultiply by z22 and post multiply by z11 to get z22 z12 = −z21 z11 −1 . Using this equation to eliminate z21 z21 from the expression −1 in square brackets. rather than x ny restrictions. which is simply z11 . we assumed that the number of eigenvalues of D with modulus less than unity is exactly equal to the number of control variables.

Lectures in Open Economy Macroeconomics. if the number of eigenvalues of D with modulus greater than unity is greater than the number of control variables. because x0 is a vector of predetermined or exogenous variables and therefore its elements can take arbitrary values.
(4. we say no local equilibrium exists. In this case the equilibrium is indeterminate. Let again m denote the number of eigenvalues of D greater than unity in modulus and assume that m > ny . Then in order to ensure that lim Et |ˆt+j | < ∞ we must set m elements of [x0 y0 ] equal to zero. which is given by
xt+1 = hx xt + ση ˆ ˆ
t+1 .45)
Covariance Matrix of xt Let Σ x ≡ E x t xt ˆˆ denote the unconditional variance/covariance matrix of xt and let ˆ Σ ≡ σ 2 ηη . then no local equilibrium exists. On the other hand.
4. But this can never be the case. ny .
. In this case. Chapter 4
79
expected to converge back to the steady state.7
Second Moments
Start with the equilibrium law of motion of the deviation of the state vector with respect to its steady-state value. x This implies that m − ny elements of x0 must be functions of the remaining nx − (m − ny ) elements.

and the symbol ⊗ denotes the Kronecker product. B. where the vec operator transforms a matrix into a vector by stacking its columns.
Mart´ Uribe ın
Method 1
One way to obtain Σx is to make use of the following useful result. and C be matrices whose dimensions are such that the product ABC exists.
. Thus if the vec operator is applied to both sides of
Σx = hx Σx hx + Σ .80 Then we have that Σ x = h x Σx hx + Σ . We will describe two numerical methods to compute Σx . Then vec(ABC) = (C ⊗ A) · vec(B). Let A.
where F = hx ⊗ hx .
the result is
vec(Σx ) = vec(hx Σx hx ) + vec(Σ ) = F vec(Σx ) + vec(Σ ).

0 = hx Σ
.t+1
. The eigenvalues of F are products of the eigenvalues of the matrix hx .t Σ . x x
Method 2 The following iterative procedure. This implies that (I − F) is nonsingular and we can indeed solve for Σx .t hx.0
.t hx.Lectures in Open Economy Macroeconomics.
Σx.t + Σ Σx. Because all eigenvalues of the matrix hx have by construction modulus less than one.0 = I hx. called doubling algorithm. One possible drawback of this method is that one has to invert a matrix that has dimension n2 × n2 .t Σx.t + Σ hx.t+1 = hx.t
= hx. may be faster than the one described above in cases in which the number of state variables (nx ) is large.t hx.t+1 = hx. Chapter 4 Solving the above expression for vec(Σx ) we obtain vec(Σx ) = (I − F)−1 vec(Σ )
81
provided that the inverse of (I −F) exists.t
=Σ
.t Σ
. it follows that all eigenvalues of F are less than one in modulus.

Consider for instance the covariance matrix E xt xt−j for j > 0. which we can write as: yt = gx xt . Then ¯ ¯ ˆ ˆ
E yt yt = Egx xt xt gx ˆˆ ˆ ˆ = gx [E xt xt ]gx ˆ ˆ = gx Σx gx
and.
. more generally. xt has been computed. or control vector yt is given by yt = y + gx (xt − x). consider the variance covariance matrix of linear combinations of the state vector xt . x
for j ≥ 0.82 Other second moments
Mart´ Uribe ın
Once the covariance matrix of the state vector. it is easy to ﬁnd other second moments of interest.
E yt yt−j = gx [E xt xt−j ]gx ˆˆ ˆ ˆ = gx hj Σx gx . Let µt = ση t . ˆˆ
j−1
E xt xt−j = ˆˆ =
E[hj xt−j xˆ
+
hk µt−k ]ˆt−j x x
k=0 j hx E xt−j xt−j ˆ ˆ
= h j Σx x
Similarly. the co-state. For instance.

x ˆ ˆ x x x
t ≥ 0. y x
4. letting x denote the innovation to the state vector in period 0. Using the law of motion Et xt+1 = hx xt for the ˆ ˆ state vector.edu/~uribe/2nd_order. say zt in period t + j to an impulse in period t is deﬁned as:
IR(zt+j ) ≡ Et zt+j − Et−1 zt+j
The impulse response function traces the expected behavior of the system from period t on given information available in period t.duke.9
Matlab Code For Linear Perturbation Methods
Stephanie Schmitt-Groh´ and I have written a suite of programs that are e posted on the courses webpage: www. relative to what was expected at time t − 1.m computes the matrices gx and hx using the Schur de-
. The program gx_hx. that is. Chapter 4
83
4.8
Impulse Response Functions
The impulse response to a variable. and applying the law of iterated expectations we get that the impulse response of the state vector in period t is given by
IR(ˆt ) ≡ E0 xt − E−1 xt = ht [x0 − E−1 x0 ] = ht [ησ 0 ] = ht x.
The response of the vector of controls yt is given by ˆ IR(ˆt ) = gx ht x.htm.Lectures in Open Economy Macroeconomics.econ. x = ησ 0 .

xt ) = 0.10
Higher Order Approximations
In this chapter.. we focused on a ﬁrst-order approximation to the solution of a nonlinear system of stochastic diﬀerence equations of the form Et f (xt+1 . obtaining the coeﬃcients of the ith order terms of the appro! ximate solution given all lower-order coeﬃcients involves solving a linear system of equations. since the coeﬃcients of the linear terms are those of the ﬁrst order approximation. Schmitt-Groh´ and Uribe (2004) describe in detail how to obtain a e second-order approximation to the solution of the nonlinear system and provide MATLAB code that implements the approximation. Namely.
.m computes second moments. obtaining a higherorder approximation to the solution of the non-linear system is a sequential procedure. But higher order approximations are relatively easy to obtain.m computes impulse response functions. Speciﬁcally. for j > 1 and i < j. The program ir. Indeed. there is a sense in which higher order approximations are simpler than the ﬁrst order approximation.84
Mart´ Uribe ın
composition method. the coeﬃcients of the ith term of the jth-order approximation are given by the coeﬃcients of the ith term of the ith order approximation. More importantly. then obtaining the second order approximation requires only to compute the coeﬃcients of the quadratic terms. The program mom.
4. So if the ﬁrst-order approximation to the solution is available.

and p as in SGU. Compute the impulse response functions of output. ρ. and correlation with output of output. α. Modify the model by assuming that agents internalize the dependence of the interest rate premium on the level of debt. ω. Compute the unconditional standard deviation. and ψ2 using tables 1 and 2 in SGU. Compare these statistics to those associated with model 2 in SGU (reported in their table 3). 4. σ . 1. Derive the model’s equilibrium conditions. ¯ Calibrate the parameters γ. investment. consumption. φ.11. e model 2). 3. the trade balance-to-output ratio. suppose that the function p(·) depends upon the individual debt position. Compare these im-
. 2. d. dt . and the current account-to-output ratio implied by the model. hours. Use the same forms for the functions U . 2003. and the current account-to-output ratio implied by the model.Lectures in Open Economy Macroeconomics.1
Exercise
An RBC Small Open Economy with an internal debtelastic interest-rate premium
Consider RBC open economy model with a debt elastic interest rate premium studied in Schmitt-Groh´ and Uribe (SGU) (JIE. Φ. hours. F . Chapter 4
85
4. δ. r.11
4. Calculate the model’s nonstochastic steady state and compare it to that of model 2 in SGU. serial correlation. investment. the trade balance-to-output ratio. β. Speciﬁcally. section 3. dt . consumption. rather ˜ than on the aggregate per capita level of debt.

Hint: You might ﬁnd it convenient to use as a basis the matlab code associated with the SGU paper.htm
.duke.edu/~uribe/closing.econ. located at www.86
Mart´ Uribe ın pulse responses to those associated with model 2 in SGU (shown in their ﬁgure 1).

the available theoretical explanations fall into two categories: One is that emerging market economies are subject to more volatile shocks than are developed countries.Chapter 5
The Terms of Trade
Three key stylized facts documented in chapter 1 are: (1) that emerging market economies are about twice as volatile as developed economies. and (3) that the trade-balance-to-output ratio is signiﬁcantly more countercyclical in emerging markets than it is in developed countries. The second category of explanations argues that in emerging countries government policy tends to amplify business-cycle ﬂuctuations whereas in developed countries public policy tends to mitigate aggregate instability. Broadly. Explaining this striking contrast between emerging and industrialized economies is at the top of the research agenda in small-open-economy macroeconomics. This and the following two chapters provide a progress report on the identiﬁcation and quantiﬁcation of exogenous sources of busi87
. (2) that private consumption spending is more volatile than output in emerging countries. but less volatile than output in developed countries.

agricultural products. Letting Ptx and Ptm denote indices of world prices of exports and imports for a particular country. the real exchange rate (rer) is deﬁned as the relative price of consumption in terms of importable goods. the terms of trade have the potential to be an important source of business cycles in developing countries. Then the real exchange rate is given by Ptc /Ptm .88
Mart´ Uribe ın
ness cycles in small open economies. the terms of trade can be regarded as an exogenous source of aggregate ﬂuctuations. or oil. At the same time. A number of empirical regularities emerge from the table:
. such as metals. In the table. the terms of trade for that country are given by tott ≡ Ptx /Ptm . Because primary commodities display large ﬂuctuations over time.1 displays summary statistics relating the terms of trade to output.2
Empirical Regularities
Table 5. The present chapter concentrates on terms-of-trade shocks. and the real exchange rate in the postwar era.
5.1
Deﬁning the Terms of Trade
The terms of trade are deﬁned as the relative price of exports in terms of imports. the components of aggregate demand. Typically.
5. Speciﬁcally. let Ptc denote a domestic CPI index. It follows that for many small countries. emerging countries specialize in exports of a few primary commodities. emerging countries are normally small players in the world markets for the goods they export or import.

Indonesia. The group of oil-exporting countries is formed by Mexico. tables 1 and 3-6. The group of developed countries is formed by the US. Italy. y) ρ(tot.67 0.45 0.77 0. India. The group of developing countries is formed by Argentina.39 0. and tb denote. c) ρ(tot. Saudi Arabia. i.24 0. rer) Developed Countries 4.38 0. Philippines.0 0. Germany.70 0. The terms of trade are measured as the ratio of export to import unit values with 1900=100. respectively.40 0.
. Venezuela. Cameroon. Note: tot.30 0. All other variables are measured per capita at constant import prices.28 0.47 0.33 0.70 Developing Countries 10. Algeria. consumption.52 0. and Japan. c.74 0. Peru.50 1. France. tott−1 ) σ(tot)/σ(y) ρ(tot. y.19 0. investment. and Nigeria. The sample is 1955 to 1990 at annual frequency.0 0. Chapter 5
89
Table 5. the terms of trade. Korea. Mexico. Brazil.Lectures in Open Economy Macroeconomics. Taiwan. UK. i) ρ(tot. Chile. output.78 0. Venezuela. Canada.40 0.42
Source: Mendoza (1995). and the trade balance.1: The Terms of Trade and Business Cycles Summary Statistic σ(tot) ρ(tott .34 0. All variables are expressed in percent deviations from a HP trend constructed using a smoothing parameter of 100. and Thailand.07 Oil Exporting Countries 18. tb) ρ(tot.

5. The terms of trade are positively correlated with the real exchange rate. The terms of trade are half as volatile as output in developed countries.1 is mute on the importance of terms
of trade shocks in explaining movements in aggregate activity. More than half a century ago. 6. 75 percent as volatile as output in developing countries.
5. Harberger (1950) and Laursen
. 3. The terms of trade display positive but small serial correlation. The terms of trade are twice as volatile in emerging countries as in developed countries. The terms of trade are procyclical. 2. we attempt to answer this question by combining the empirical information contained in table 5.2. and 150 percent as volatile as output in oil-exporting countries. The information provided in table 5. They are twice as procyclical in developed countries as in developing countries. and they are almost twice as volatile in oil-exporting countries as in developing countries. 4.90
Mart´ Uribe ın 1. The correlation between the terms of trade and the trade balance is positive but small. This correlation is high for developed countries but almost nil for less developed countries. Later in this chapter.1 with the theoretical predictions of a fully speciﬁed dynamic general equilibrium model of the open economy.1
TOT-TB Correlation: Two Early Explanations
The eﬀects of terms-of-trade shocks on the trade balance is an old subject of investigation.

it denotes private investment. Consider the following behavioral equations deﬁning the dynamics of each component of aggregate demand. This view remained more or less unchallenged until the early 1980s. using a dynamic optimizing model of the current account. gt denotes public consumption. concluded that the eﬀect of terms of trade shocks on the trade balance depends crucially on the perceived persistence of the terms of trade. and mt denotes imports. This conclusion became known as the Harberger-LaursenMetzler (HLM) eﬀect.
where yt denotes output. This view became known as the Obstfeld-Razin-Svensson (ORS) eﬀect. Public consumption and
. xt denotes exports.e. the conclusion that rising terms of trade should be associated with an improving trade balance.
The Harberger-Laursen-Metzler Eﬀect A simple way to obtain a positive relation between the terms of trade and the trade balance in the context of a Keynesian model is by starting with the national accounting identity
y t = ct + gt + it + xt − m t . In their model a positive relation between terms of trade and the trade balance (i. when Obstfeld (1982) and Svensson and Razin (1983). within the context of a keynesian model. Chapter 5
91
and Metzler (1950) formalized. ct denotes private consumption.. Let us look at the HLM and ORS eﬀects in some more detail.Lectures in Open Economy Macroeconomics. the HLM eﬀect) weakens as the terms of trade become more persistent and may even be overturned if the terms of trade are of a permanent nature.

In the jargon of the 1950s. where g and ¯ are parameters. respectively. respectively. ¯
with α ∈ (0. 1). we will assume that these two varibles are constant over time and given by
gt = g ¯
and it = ¯ i. For simplicity. whereas the term c + g + ¯ is referred to as the autonomous component of ¯ ¯ i domestic absorption. 1) and c > 0 are parameters. with µ ∈ (0. Output as well as all components of aggregate demand are expressed in terms of import goods. is given by xt = tott qt . the value of exports in terms of importables. xt .
. Imports are assumed to be ¯ proportional to output. The quantity of goods exported in period t is denoted by qt . Consumption is assumed to be ¯ i an increasing linear function of output
ct = c + αyt .92
Mart´ Uribe ın
private investment are assumed to be independent of output. Thus. the parameters α and µ are referred to as the marginal propensities to consume and import. mt = µyt .

this theory implies that an improvement in the terms of trade (an increase in tott ) gives rise to an expansion in the trade surplus. we can write 1−α µ(¯ + g + ¯ c ¯ i) ¯ tott q − . as a larger fraction of income is used to buy foreign goods. and mt from the national income identity. It is worth noting that in the context of this model. µ. xt . This positive relation between the terms of trade and the trade balance is stronger the larger is the volume of exports. The reason why µ increases the TOT multiplier is that a higher value of µ weakens the endogenous expansion in aggregate demand to an exogenous increase in exports. and the smaller is the marginal propensity to consume α. gt . a larger value of α reduces the TOT multiplier because it exacerbates the endogenous response of aggregate demand to a TOT shock through private consumption. q . ¯ where q is a positive parameter.Lectures in Open Economy Macroeconomics. Using the behavioral equations to eliminate ¯ ct . Similarly. qt . it . 1+µ−α
Letting tbt ≡ xt − mt denote the trade balance. and solving for output yields yt = ¯ c + g + ¯ + tott q ¯ ¯ i . the smaller is the marginal propensity ¯ to import. Chapter 5
93
where tott denotes the terms of trade. and the quantity of goods exported. the sign of the eﬀect
. 1+µ−α 1+µ−α
tbt =
Clearly. The terms of trade are assumed to evolve exogenously. is assumed to be constant and vigen by qt = q .

yt = 1 for all t. The economy is small in world product markets.
The Obstfeld-Razin-Svensson Eﬀect The ORS eﬀect is cast within the dynamic optimizing theoretical framework that diﬀers fundamentally from the reduced-form Keynesian model we used to derive the HLM eﬀect. with tott taking the place of yt . The model is therefore identical to the stochastic-endowment economy studied in chapter 2. Consider the small. This is the main contrast with the Obstfeld-Razin-Svensson eﬀect.5). Suppose that the good the household consumes is diﬀerent from the good it is endowed with. therefore exports the totality of its endowment and imports the totality of its consumption. The resource constraint is then given by
dt = (1 + r)dt−1 + ct − tott . or the terms of trade.
The borrowing constraint given in (2. This is an economy inhabited by an inﬁnitely lived representative household with preferences described by the intertemporal utility function given in (2. Assume for simplicity that the endowment of exportable goods is constant and normalized to unity. Let tott denote the relative world price of exported goods in terms of imported goods.3) prevents the household from engaging in Ponzi games.94
Mart´ Uribe ın
of a TOT shock on the trade balance is independent of whether the terms of trade shocks are permanent or temporary in nature. endowment economy studied in chapter 2. The household. open. We can then use the results derived in chapter 2 to draw the
. so it takes the evolution of tott as exogenous.

The
. A simple way to do this is to modify the RBC model of chapter 4 by assuming again that households do not consume the good they produce.1 plots the serial correlation of the terms of trade against the correlation of the trade balance with the terms of trade for 30 countries. an improvement in the terms of trade can lead to a deterioration in the current account driven by investment expenditures. In this case. The 30 observations were taken from Mendoza (1995). Africa. Is the ORS eﬀect borne out in the data? If so. so agents can aﬀord assuming higher current debts without sacriﬁcing future expenditures. Figure 5. the productivity shock At can be interpreted as a terms-of-trade shock. In this case. Agents save in order to ensure higher future consumption. including the G7 countries and 23 selected developing countries from Latin America.Lectures in Open Economy Macroeconomics. table 1. shown with circles. the value of income is expected to grow over time. but as the serial correlation of the terms of trade shock increases. and the Middle East. displays no pattern. An increase in the terms of trade produces an improvement in the trade balance if the terms of trade shock is transitory. an improvement in the terms of trade induces a trade balance deﬁcit. The cloud of points. we should observe that countries experiencing more persistent terms-of-trade shocks should display lower correlations between the terms of trade and the trade balance than countries facing less persistent terms of trade shocks. When terms of trade are nonstationary. This conclusion can be extended to a model with endogenous labor supply and capital accumulation. Chapter 5
95
following conclusion: if the terms of trade are stationary then an increase in the terms of trade produces an improvement in the current account. East Asia.

3 0. Venezuela). Canada. T OT ) = 0. Saudi Arabia.
. and Egypt). Brazil. and Tunisia). Philippines.23 + 0. Note: Each point corresponds to one country. and Japan). 6 countries from Latin America.1
0. Nigeria. The dashed-dotted line is the OLS ﬁt after eliminating the G7 countries. United Kingdom. and 8 countries from Africa (Algeria.2
0
−0. (Argentina. Saudi Arabia. Peru. Kenya.28+0.5
0.8
0. Morocco. T OT ) = 0. Sudan. 6 countries from Asia (Taiwan.03ρ(T OT ). The solid line is the OLS ﬁt and is given by corr(T B. Italy.2 0. Germany.4 TB−TOT Correlation
0.6
Argentina
0. T OT ) = 0. The dashed line is the OLS ﬁt after eliminating Argentina from the sample and is given by corr(T B.2
−0.4 TOT Serial Correlation
0.12ρ(T OT ). The sample includes the G-7 countries (United States.1
0
0. France. Mexico. Zaire.6 −0.96
Mart´ Uribe ın
Figure 5.4
−0.7
Source: Mendoza (1995).6
0. The TOT serial correlation and the TB-TOT correlation are computed over the period 1955-1990.14ρ(T OT ). and Argentina from the sample and is given by corr(T B. Cameroon.35−0. India. Indonesia.1: TOT Persistence and TB-TOT Correlations
0. Chile. table 1. 3 countries from the Middle East (Israel. and Thailand). Korea.

.1 The corresponding ﬁtted relationship is shown with a broken line on ﬁgure 5. Eliminating Argentina from the sample one obtains a positive OLS slope of 0. The sign of the slope is indeed in line with the ORS eﬀect: As the terms of trade shocks become more persistent. Chapter 5
97
OLS ﬁt of the 30 points. Argentina is the only country whose elimination from the sample results in a positive slope. Does this conclusion suggest that the empirical evidence presented here is against the Obstfeld-Razin-Svensson eﬀect? Not necessarily.03 and is shown with a dash-dotted line in ﬁgure 5. that the negative slope in the OLS regression is driven by a single observation.1. Countries in this group would include all of the G7 nations.Lectures in Open Economy Macroeconomics. The ﬁtted line using this reduced sample has a negligible slope equal to 0. Argentina. they should be expected to induce a smaller response in the trade balance. displays a small negative slope of -0. gives us a better idea of what the relation between the TB-TOT correlation and the TOT persistence looks for small emerging countries that take their terms of trade exogenously. It is apparent in the graph. For these countries. the only country in the sample with a negative serial correlation of the terms of trade. however. a major oil exporter. A number of countries in ﬁgure 5. the largest economies in the world.1.12. We conclude that the observed relationship between the TB-TOT correlation and the persistence of TOT is close to nil. the terms of trade are not likely to be exogenous.1 are likely to be large players in the world markets for the goods and services they import and/or export. and Saudi Arabia. Eliminating these 8 countries (as well as the outlier Argentina) from the sample. The ORS
1 Indeed.14. shown with a solid line.

In this case. At the same time. high world interest rates are associated with contractions in aggregate demand and improvements in the trade balance in emerging countries. Moreover. with deteriorated terms of trade for the emerging countries producing those commodities. In turn. High world interest rates may be associated with depressed economic activity in developed and emerging economies alike. A case in point is given by world-interest-rate shocks. for that matter—is identiﬁcation. the terms of trade and the trade balance are moving at the same time. An important step in the process of isolating terms-of-trade shocks—or any kind of shock. Not controlling for these shocks may result in erroneously attributing part of their eﬀect on the trade balance to the terms of trade. Suppose further that this other country or set of countries generates a substantial fraction of the demand for exports or the supply of imports of the country in question. Data analysis based purely on statistical methods will in general not result in a successful iden-
. judging the empirical validity of the ORS eﬀect only on the grounds of raw correlations would be misplaced. The raw data is in principle driven by a multitude of shocks. of which the terms of trade is just one. as a result. some of these shocks may directly aﬀect both the trade balance and the terms of trade. As another example. suppose that domestic technology shocks are correlated with technology shocks in another country or set of countries. but attributing all of the movement in the trade balance to changes in the terms of trade would be clearly misleading.98
Mart´ Uribe ın
eﬀect requires isolating the eﬀect of TOT shocks on the trade balance. and. Under this scenario. low levels of economic activity in the developed world are likely to be associated with a weak demand for primary commodities.

Economic theory must be used be at center stage in the identiﬁcation process. The household block of the model is identical to that of the standard RBC model studied in chapter 4.3
Terms-of-Trade Shocks in an RBC Model
Consider expanding the real-business-cycle model of chapter 4 to allow for terms-of-trade shocks. which follows Mendoza (1995).Lectures in Open Economy Macroeconomics. Chapter 5
99
tiﬁcation of technology shocks. The main diﬀerence with the model of chapter 4 is that the model studied here features three sectors: a sector producing importable goods. an exportable good is either an exported good or a good that is sold domestically but is highly substitutable with a good that is exported. In doing this. An importable good is either an imported good or a good that is produced domestically but is highly substitutable with a good that is imported.3.
5. we follow the work of Mendoza (1995). and a sector producing nontradable goods. The following exercise. Similarly.
5. A nontradable good is a good that is neither exportable nor importable. The economy is populated by a large number of identical house-
. represents an early step in the task of identifying the effects of terms-of-trade shocks on economic activity in emerging economies. a sector producing exportable goods.1
Households
This block of the model is identical to that of the RBC model studied in chapter 4.

which they rent at the rate ut .2)
where the function β is assumed to take the form
β(c. ht denotes labor eﬀort. h) = [1 + c(1 − h)ω ]−β . and U is a period utility function taking the form [c (1 − h)ω ]1−γ .
(5. ht ). kt . ht ). The stock of capital evolves according to the following law of motion:
kt+1 = (1 − δ)kt + it − Φ(kt+1 − kt ).
We established in chapter 4 that the endogeneity of the discount factor serves the purpose of rendering the deterministic steady state independent of the country’s initial net foreign asset position.1)
where ct denotes consumption.
Households oﬀer labor services for a wage wt and own the stock of capital.100 holds with preferences described by the utility function
∞
Mart´ Uribe ın
E0
t=0
θt U (ct .3)
.
(5.
(5. h) =
The variable θt /θt−1 is a time-varying discount factor and is assumed to evolve according to the following familiar law of motion:
θt+1 = θt β(ct . 1−γ
U (c.

the ﬁrst-order conditions of the household’s maximization problem
. Households are assumed to be able to borrow or lend freely in international ﬁnancial markets by buying or issuing risk-free bonds denominated in units of importable goods and paying the constant interest rate r ∗ .2) and (5. the period budget constraint of the household can be written as
dt = (1 + r ∗ )dt−1 + pc ct + it − wt ht − ut kt .Lectures in Open Economy Macroeconomics. 1] denotes the capital depreciation rate. The function Φ introduces capital adjustment costs. t which serve the role of numeraire. and is assumed to satisfy Φ(0) = Φ (0) = 0 and Φ > 0. The parameter δ ∈ [0. t
(5. the steady-state level of capital is not aﬀected by the presence of adjustment costs. Households are subject to a no-Ponzigame constraint of the form Et dt+j ≤ 0.4)
The relative prices pc . and ut are expressed in terms of importable goods. Chapter 5
101
where it denotes gross investment. As discussed in chapter 4.5).5)
The household seeks to maximize the utility function (5. Under these assumptions. capital adjustment costs help curb the volatility of investment in small open economy models like the one studied here. Letting dt denote the debt position assumed by the household in period t and pc t denote the price of the consumption good. which is assumed to be an importable good.4). wt . j→∞ (1 + r ∗ )j lim
(5. Letting θt ηt and θt λt denote the Lagrange multipliers on (5.2)(5.1) subject to (5.

9) ηt = −Et U (ct+1 . These ﬁrms operate a CES production function that takes tradable consumption goods. ht+1 ) + Et ηt+1 β(ct+1 .2). ht ) − ηt βc (ct . The ﬁrstorder conditions associated with this proﬁt-maximization problem are (5.10)
5.8)
λt [1 + Φ (kt+1 − kt )] = β(ct . (5.5) holding with equality.4). t t ct = [χ(cT )−µ + (1 − χ)(cN )−µ ]−1/µ . t t t t t where pT and pN denote. and nontradable consumption goods. ht ) = λt pc t
(5. ct . They choose output and inputs to maximize proﬁts. ht )Et λt+1 ut+1 + 1 − δ + Φ (kt+2 − kt+1 ) (5. and
Mart´ Uribe ın
Uc (ct . (5. cN . t t
(5.11)
. ht ) + ηt βh (ct . ht ) = λt wt λt = β(ct .11)
with µ > −1. which are given by pc ct − pT cT − pN cN . cT .3.2
Production of Consumption Goods
The consumption good. is produced by domestic ﬁrms. ht+1 ) (5.102 are (5. respectively.6)
−Uh (ct .7) (5. as inputs. Firms operate in perfectly competitive product and input markets. ht )(1 + rt )Et λt+1
(5. the relative prices of tradable and t t nontradable consumption goods in terms of importable goods. Formally.

are produced using importable t consumption goods. cX . Chapter 5 and cN t = cT t ct = cT t 1−χ χ 1 χ
103
1 1+µ
pT t N pt pT t pc t
1 1+µ
.13)
It is clear from the ﬁrst of these eﬃciency conditions that the elasticity of substitution between tradable and nontradable goods is given by 1/(1 + µ). then the share of tradables in total consumption. which are given by
pT cT − pX cX − cM .14)
where α ∈ (0. t t t increases as the relative price of tradables in terms of consumption.Lectures in Open Economy Macroeconomics. Firms are competitive and aim at maximizing proﬁts. one observes that if the elasticity of substitution between tradables and nontradables is less than unity (or µ > 0). pT /pc . 1) is a parameter.
(5. t t t
(5.
1 1+µ
(5.3.12)
1 1+µ
. t t increases.
cT = (cX )α (cM )1−α . From the second optimality condition. cM . given by pT cT /(pc ct ).3
Production of Tradable Consumption Goods
Tradable consumption goods. denoted cT . Formally.
5. t t t t t
. and exportable consumption goods. via a Cobbt t Douglas production function.

17)
M M yt = AM (kt )αM . the three production technologies are given by
X X yt = AX (kt )αX .14) and: pX cX t t =α pT cT t t cM t = 1 − α. Exportable. The optimality t conditions associated with this problem are (5. This implication is a consequence of the assumption of Cobb-Douglas technology in the production of tradable consumption. whereas nontradable goods are produced using labor services only.4
Production of Importable. t
(5.18)
.3. respectively. T cT pt t
(5. Formally.16)
These optimality conditions state that the shares of consumption of exportables and importables in total consumption expenditure in tradable goods are constant and equal to α and 1 − α.15)
(5. we have that the relative price of importables in terms of the numeraire is always unity (pM = 1). and Nontradable Goods
Exportable and importable goods are produced with capital as the only input. or the terms of trade.104
Mart´ Uribe ın
where pX denotes the relative price of exportable goods in terms of imt portable goods.
5. t
(5. Note that because the importable good plays the role of numeraire.

. Chapter 5 and
N yt = AN (hN )αN . and the vari-
able hN denotes labor services employed in the nontradable sector.21)
and wt hN t = αN . t t
105
(5. yt denotes output of imN portable goods. M yt
(5. and yt denotes output of nontradable goods. X pX yt t M ut kt = αM .Lectures in Open Economy Macroeconomics. and as a consequence of the assumption of Cobb-Douglas technologies.
i The variable kt denotes the capital stock in sector i = X. which are given by
X M N X M pX yt + yt + pN yt − wt hN − ut (kt + kt ). M. M . Firms t demand input quantities to maximize proﬁts.20)
(5.19)
X M where yt denotes output of exportable goods. N .22)
According to these expressions. t t t
The optimality conditions associated with this problem are
X ut kt = αX . input shares are constant. N pN yt t (5. The factors Ai t
denote exogenous and stochastic technology shocks in sectors i = X.

in equilibrium the evolution of the net foreign debt position of the economy is given by
X M dt = (1 + r ∗ )dt−1 − pX (yt − cX ) − yt + cM + it . and nontradable). t t t
(5. the markets for capital. That is.
.23) (5.106
Mart´ Uribe ın
5.25)
Also. and nontradables must clear.5
Market Clearing
In equilibrium. t
(5. and the terms of trade. We assume that all shocks follow autoregressive processes of order one.
(5. labor.3.
X M kt = kt + kt . exportable. Mendoza (1995) imposes four restrictions on the joint distribution of the exogenous shocks: (1) all four shocks share the same persistence. (4) Innovations to productivity shocks and terms-of-trade shocks are allowed to be correlated. (2) The sectorial productivity shocks are assumed to be perfectly correlated. t and
N cN = yt .3.26)
5.6
Driving Forces
There are four sources of uncertainty in this economy: One productivity shock in each of the three sectors (importable. (3) The technology shocks aﬀecting the production of importables and exportables are assumed to be identical.24)
ht = hN .

7
Competitive Equilibrium
A stationary competitive equilibrium is a set of stationary processes {ct . T t . ht . one matching key macroeconomic relations in developed countries.
5.26).8
Calibration
Mendoza (1995) presents two calibrations of the model. dt . cT . pT . pc . t
X M N X M cX .
= ψT
p t
T + νt . yt . cN . Chapter 5 These assumptions give rise to the following laws of motion:
p t. In calibrating the driving forces of the developed-country version of the model. hN . pX }∞ .3. kt . cM . yt .
We are now ready to deﬁne a competitive equilibrium. the parameters ρ and σ
p
are set to match the average serial cor-
. ηt .6)-(5.3. given the initial conditions t=0 k0 and d−1 and the exogenous processes {AX . t t t t t t t
λt }∞ satisfying equations (5. t N t
= ψN
T t . AN . σν T ). kt . ut .Lectures in Open Economy Macroeconomics.3) and (5. E( p . it . p t
107
ln pX = ρ ln pX + t t−1
∼ N (0. kt . wt . AM . and the other matching key macroeconomic relations in developing countries. νt ) = 0. yt .
T 2 T νt ∼ N (0. σ 2p ). N t . t t t t t=0
5.
ln AX = ρ ln AX + t t−1 ln AM = ρ ln AM + t t−1 ln AN = ρ ln AN + t t−1
T t
T t . pN .

2 displays the parameter values implied by the above restrictions. Mendoza (1995) sets the correlation between the productivity shock in the exportable sector and the terms of trade at 0. and σ
p
= 0.165. This completes the calibration of the parameters deﬁning exogenous driving forces in the developed-country model.014.473.
Using estimates of productivity shocks in ﬁve industrialized countries by Stockman and Tesar (1995).
Based on correlations between Solow residuals and terms of trade in ﬁve developed countries.047
1 − ρ2 . Mendoza (1995) sets the volatility of productivity shocks in the importable and exportable sectors at 0.019
1 − ρ2 .
2 2 ψT σ 2p + σν T
Table 5.165. This implies that ψT σ
p
= 0.
and ψN
2 2 ψT σ 2p + σν T = 0.019 and the volatility of the productivity shock in the nontraded sector at 0.014
1 − ρ2 . This implies that
2 2 ψT σ 2p + σν T = 0.
.108
Mart´ Uribe ın
relation and standard deviation of the terms of trade for the group of G7 countries. Using the information presented in table 1 of Mendoza (1995) yields ρ = 0.

041 0.156 0.5 0.61 -0.34 0.47 0.028 1.1 0.74 0.032 0.3 2.57 0.04 0.011 0.009
.56 0.15 0.74 0.067 0.028 2.22 0.Lectures in Open Economy Macroeconomics.1 0.70 0.27 0.009 Developing Country 0.017 0.49 0.04 0.2: Calibration Parameter σp σν T ρ ψT ψN r∗ αX αM αN δ φ γ µ α ω β Developed Country 0.41 -0.35 0. Chapter 5
109
Table 5.08 0.79 0.

and ψT σ
p
= −0. Mendoza sets the standard deviation of productivity shocks in the traded sectors at 0.46.04 and their correlation with the terms of trade at -0.46 to match the observed average standard deviation of GDP and the correlation of GDP with TOT in developing countries.2.04
1 − ρ2 . This implies:
ρ = 0. the
parameter ρ and σp are picked to match the average serial correlation and standard deviation of the terms of trade for the group of developing countries reported in table 1 of Mendoza (1995).
2 2 ψT σ 2p + σν T
The implied parameter values are shown in table 5.414.
and σ
p
= 0.74 as in the developed-country model.110
Mart´ Uribe ın In calibrating the driving forces of the developing-country model.
Mendoza (1995) assumes that the standard deviation of productivity shocks in the traded sector are larger than in the nontraded sector by the same proportion as in developed countries. This means that the parameter ψN takes the value 0. This yields the restrictions:
2 2 ψT σ 2p + σν T = 0. This completes the calibration of the exogenous driving forces for the developing-country version
.12
1 − ρ2 .

38 . Table 5.32 1.34 .57 1 1 .3: Data and Model Predictions Variable
σx σT OT
G7 TOT Data 1 Model 1 TB Data 1.32 .18 .60 .42 .70 .80 .47 .38 .9
Model Performance
Table 5.27 1.14 .51 .30 .74 .62 Model 3.17 -.78 .32 .18 -.47 1.39 .57 Source: Mendoza
DCs 1 1 1.5 GDP Data 1.
2
See Mendoza.96 .70 .99 .33 .58 . 1995 for more details.26 .28 . Chapter 5
111
Table 5.67 .39 . The following list highlights a number of empirical regularities and comments on the model’s ability to capture them.95
ρx.86 1.85 .52 .23 .66 .19 .37 .01 I Data 1.89 .78 .
of the model.2
5.11 1 1 .47 .3.69 .69 Model .49 .GDP G7 DCs .32 -.Lectures in Open Economy Macroeconomics.60 .72 .75 .44 Model .94
ρxt .78 .12 .90 Model 2.49 .59 Model 1.25 .2 also displays the values assigned to the remaining parameters of the model.T OT G7 DCs 1 1 .71
ρx.78 .03 .41 .84 .43 .3 presents a number of data summary statistics from developed (G7) and developing countries (DCs) and their theoretical counterparts.44 .86 C Data 1.81 .08 .25 .79 (1995).62 .25
1.0 RER Data 1.45 1 1 .11 .41 .82 .68 .
.49 .xt−1 G7 DCs .

recall that the parameter ρ is set to pin down the serial correlation of the terms of trade in developing and G7 countries. particularly for developing countries. the real exchange rate (RER) is measured as the ratio of the domestic CPI to an exchange-rate-adjusted. the terms of trade are procyclical. In the model. The model appears to overestimate the countercyclicality of the trade balance. In the data. This fact is matched by construction. The trade balance is countercyclical in DCs but procyclical in G7 countries. the RER is deﬁned as the relative price of consumption in terms of importables and denoted by pc . 2. In the model. 4. although much less so in developing countries than in G7 countries. The model captures this fact relatively well. but underestimates the TB-TOT correlation. 3. trade-weighted average of foreign CPIs. 5. The terms of trade are positively correlated with the trade balance. the trade balance is countercyclical in both. The model fails to capture this fact. The model captures this empirical regularity. 6. The terms of trade are less volatile than GDP. for other authors estimate negative TB-GDP correlations for developed countries.112
Mart´ Uribe ın
1. developed and developing countries. In the t
. The failure of the model to capture the procyclicality of the trade balance in developed countries should be taken with caution. The observed terms of trade are somewhat persistent. In the data.

7.45 for both developing and G7 countries).75 for both types of country. When the same experiment is performed in the context of the developingcountry version of the model.3. The RER is somewhat persistent (with a serial correlation of less than 0. although it overestimates somewhat the RER-GDP correlation. Therefore. In the context of the model of this section. The model captures this fact. the RER is highly persistent. the volatility of output deviations from trend measured at import prices falls from 4. Mendoza (1995) ﬁnds that when the volatility of all productivity shocks is set equal to zero in the developed-country version of the model.Lectures in Open Economy Macroeconomics. Chapter 5
113
data. one must set all productivity shocks at their deterministic steady-state values.
5. This is accomplished by setting σν T = ψT = 0. In the model. in the model the terms of trade explain about 88 percent of the volatility of output. The reason for this
.6 percent. with an autocorrelation above 0. the RER is procyclical. the terms of trade explain about 66 percent of output movements. shutting oﬀ the variance of the productivity shocks results in an increase in the volatility of output.1 percent to 3. When output is measured in terms of domestic prices. one can run the counterfactual experiment of computing equilibrium dynamics after shutting oﬀ all sources of uncertainty other than the terms of trade themselves.10
How Important Are the Terms of Trade?
To assess the contribution of the terms of trade to explaining business cycles in developed and developing countries.

114

Mart´ Uribe ın

increase in output volatility is that in the benchmark calibration the terms of trade are negatively correlated with productivity shocks—note in table 5.2 that ψT < 0 for developing countries. Taking this result literally would lead to the illogical conclusion that terms of trade explain more than 100 percent of output ﬂuctuations in the developing-country model. What is wrong? One diﬃculty with the way we have measured the contribution of the terms of trade is that it is not based on a variance decomposition of output. A more satisfactory way to assess the importance of terms-of-trade and productivity shocks would be to deﬁne the terms of trade shock as

p t

and

T the productivity shock as νt . One justiﬁcation for this classiﬁcation is that p t

aﬀects both the terms of trade and sectoral total factor productivities,

T while νt aﬀects sectoral total factor productivities but not the terms of

trade. Under this deﬁnition of shocks, the model with only terms of trade shocks results when σν T is set equal to zero. Note that the parameter ψT must not be set to zero. An advantage of this approach is that, because the variance of output can be decomposed into a nonnegative fraction explained by

p t T and a nonnegative part explained by νt , the contribution of the terms

of trade will always be a nonnegative number no larger than 100 percent. Exercise 5.1 Using the model presented in this section, compute a variance decomposition of output. What fraction of output is explained by terms-oftrade shocks in the developed- and developing-country versions of the model?

Chapter 6

**Interest-Rate Shocks
**

Business cycles in emerging market economies are correlated with the interest rate that these countries face in international ﬁnancial markets. This observation is illustrated in ﬁgure 6.1, which depicts detrended output and the country interest rate for seven developing economies between 1994 and 2001. Periods of low interest rates are typically associated with economic expansions and times of high interest rates are often characterized by depressed levels of aggregate activity.1 Data like those shown in ﬁgure 6.1 have motivated researches to ask what fraction of observed business cycle ﬂuctuations in emerging markets is due to movements in country interest rate. This question is complicated by the fact that the country interest rate is unlikely to be completely exogenous to the country’s domestic conditions.2 To clarify ideas, let Rt denote the gross

The estimated correlations (p-values) are: Argentina -0.67 (0.00), Brazil -0.51 (0.00), Ecuador -0.80 (0.00), Mexico -0.58 (0.00), Peru -0.37 (0.12), the Philippines -0.02 (0.95), South Africa -0.07 (0.71). 2 There is a large literature arguing that domestic variables aﬀect the interest rate at which emerging markets borrow externally. See, for example, Edwards (1984), Cline

1

115

116

Mart´ Uribe ın

**Figure 6.1: Country Interest Rates and Output in Seven Emerging Countries
**

Argentina 0.15 0.1 0.05 0 −0.05 −0.1 94 95 96 97 98 99 00 01 94 95 96 97 98 Mexico 0.15 0.3 0.1 0.2 0.05 0.1 0 0 94 95 96 97 98 Peru 0.1 0.05 0 −0.05 94 95 96 97 98 99 00 01 0 0.1 99 00 01 −0.05 94 95 96 97 98 99 00 01 99 00 01 0.1 0.05 0 0.15 Brazil

Ecuador 0.4

Philippines

0.05

94

95

96

97

98

99

00

01

South Africa 0.06 0.04 0.02 0 94 95 96 97 98 99 00 01

Output

Country Interest Rate

Note: Output is seasonally adjusted and detrended using a log-linear trend. Country interest rates are real yields on dollar-denominated bonds of emerging countries issued in international ﬁnancial markets. Data source: output, IFS; interest rates, EMBI+. Source: Uribe and Yue (2006).

may induce foreign lenders to lower spreads on believes that the country’s ability to repay its debts has improved.
Here. Because the interest-rate premium is country speciﬁc.S. Chapter 6
117
interest rate at which the country borrows in international markets. The researcher. In response to this output increase. borrow and lend from one another. An increase in output. i. or the interest rate at which developed countries. Assume. We can’t say the same.. foreign lenders reduce the country spread in response to expansions in aggregate activity. and St denotes the gross country interest-rate spread. it is reasonable to assume that the world interest rate Rt . consider the following example. that Rt is countercyclical. however. If the country in question is a small player in international ﬁnancial markets. Suppose now that a domestic productivity shock induces an expansion in output. The researcher. like the U. a Colombian spread. Interpreting the country interest rate as an exogenous variable when in reality it has an endogenous component is likely to result in an overstatement of the importance of interest rates in explaining business cycles. for instance. furthermore.
is completely exogenous to the emerging country’s domestic conditions. and Cline and Barnes (1997).. etc. who believes Rt is exogenous. This interest rate can be expressed as Rt = Rt St . about the country spread St . or the
us country interest rate. erroneously attributes
(1995). as many emerging
us economies are. in the data we ﬁnd an Argentine spread. or country interest-rate premium.e.Lectures in Open Economy Macroeconomics. the interest rate falls. its contribution to generating business cycles is nil. wrongly assumes that the interest rate is purely exogenous. Thus. however. Suppose that the interest rate Rt is purely endogenous. To see why.
. Rus denotes the world interest rate.

We measure Rt as the
3-month gross Treasury bill rate divided by the average gross US inﬂation
. and Rt denotes the gross real (emerging) country interest rate.1
An Empirical Model
Our empirical model takes the form of a ﬁrst-order VAR system: yt ˆ yt−1 ˆ ˆt−1 ı tbyt−1 ˆ us Rt−1 ˆ Rt−1
y t i t tby t rus t r t
A
ˆt ı tbyt = B ˆ us Rt ˆt R
+
(6. it denotes real gross domestic
us investment. The right conclusion. tbyt denotes the trade balance to output ratio. the identiﬁcation process must combine statistical methods and economic theory. Rt denotes the
gross real US interest rate. It follows that in order to quantify the macroeconomic eﬀects of interest rate shocks. Necessarily. is that all of the movement in output is due to the productivity shock. A hat on yt and it denotes log deviations from a log-linear
us us trend.1)
where yt denotes real gross domestic output. The particular combination adopted in this chapter draws heavily from Uribe and Yue (2006). the ﬁrst step is to identify the exogenous components of country spreads and world interest rate shocks. of course.118
Mart´ Uribe ın
part of the increase in output to the decline in Rt .
6. A hat on Rt and Rt denotes simply the log.

A4i = B4i =
0.e. conceivably. Uribe and Yue (2006) adopt this restriction primarily because it is reasonable to assume that disturbances in a particular (small) emerging country will not aﬀect the real interest rate of a large country like the United States.
3
r. Chapter 6
119
over the past four quarters. This identiﬁcation strategy is a natural one. t
and
tby t )
aﬀect ﬁnancial
markets contemporaneously.. Because Rt and Rt appear at the bottom of the system. di! scuss an alternative identiﬁcation strategy consisting in placing ﬁnancial variables ‘above’ real variables ﬁrst in the VAR system. it seems reasonable to assume that ﬁnancial markets are able to react quickly to news about the state of the business cycle. Output. Morgan’s EMBI+ stripped spread and the US real interest rate.
. 4 Uribe and Yue (2006). this
identiﬁcation strategy presupposes that innovations in world interest rates (
rus ) t
and innovations in country interest rates ( r ) percolate into domestic t
real variables with a one-period lag.Lectures in Open Economy Macroeconomics. for. P. decisions such as employment and spending on durable consumption goods and investment goods take time to plan and implement. At the same time. is that the world
us interest rate Rt follows a simple univariate AR(1) process (i.4 An additional restriction imposed on the VAR system. the identiﬁcation scheme implies that real domestic shocks ( y . and the trade balance are seasonally adjusted. Uribe and Yue (2006) impose the restriction that the matrix A be lower triangular with unit diagus onal elements. Also. t
i. To identify the shocks in the empirical model.3 We measure Rt as the sum of J. for all i = 4). t
can equivalently be interpreted as a
Using a more forward looking measure of inﬂation expectations to compute the US real interest rate does not signiﬁcantly alter our main results. investment. The country-interest-rate shock.

the estimated residual of the newly deﬁned bottom equation. t
are identical to those associated
r t
Therefore. t
Moreover.
6. it is obvious that the impulse response
s t
functions of yt .1) the ˆ ˆ ˆ ˆ us country interest rate Rt using the deﬁnition of country spread.
us Clearly. t
is identical to
r. how and by how much do country spreads move in response to innovations in emerging-country fundamentals? Fourth. To see this. throughout the paper we indistinctly refer to
as a
country interest rate shock or as a country spread shock.120
Mart´ Uribe ın
country spread shock.1). because Rt appears as a regressor in the bottom equation of the
VAR system. St ≡ Rt −Rt . how do US-interest-rate shocks and country-spread shocks aﬀect real domestic variables such as output. how important are USinterest-rate shocks and country-spread shocks in accounting for movements in country spreads? We answer these questions with the help of impulse response functions and variance decompositions. call it
s. ˆt .2 displays with solid lines the impulse response function implied
. investment.2
Impulse Response Functions
Figure 6. and the trade balance? Second. Uribe and Yue use it to address a number of questions central to disentangle the eﬀects of country-spread shocks and world-interest-rate shocks on aggregate activity in emerging markets: First. how important are US-interest-rate shocks and country-spread shocks in explaining movements in aggregate activity in emerging countries? Fifth. After estimating the VAR system (6. and tbyt associated with ˆ ı with
r. consider substituting in equation (6. how do country spreads respond to innovations in US interest rates? Third.

4 −0. The responses of the Trade Balance-to-GDP ratio. (2) The responses of Output and Investment are expressed in percent deviations from their respective log-linear trends.Lectures in Open Economy Macroeconomics. the country interest rate. the US interest rate.15 −0.2 −0. and the country spread are expressed in percentage points.5
10
15
20
Country Interest Rate 1 0.25 −0. and broken lines depict two-standard-deviation error bands.2 0 5
Country Spread
10
15
20
Notes: (1) Solid lines depict point estimates of impulse responses.2 0 5 10 15 20 1 0.4 0.2: Impulse Response To Country-Spread Shock
Output 0 −0.5 0.6 0.05 −0.2 −0.2 5 10 15 20 Investment
Trade Balance−to−GDP Ratio 1 0.1 −0.4 0.4 0.8 0.6 0. The two-standard-error bands are computed using the delta method.3 0.6 −0.2 0.1 0 5 10 15 20 −1 5 0
World Interest Rate
−0.3 5 10 15 20 0 −0.
.8 0.8 −1 −1. Chapter 6
121
Figure 6.

the country spread itself increases and then quickly falls toward its steady-state level. t
Broken lines depict two-standard-deviation bands. because of our maintained assumption that external ﬁnancial shocks take one quarter to aﬀect production and absorption. The point estimates of the impulse response functions of output.5 In response to an
unanticipated country-spread shock. aggregate activity and gross domestic investment contract. t
The eﬀects of US interest-rate shocks on domestic variables and coun-
try spreads are measured with signiﬁcant uncertainty.
r. while net exports improve.1) to a unit innovation in the country spread shock. The half life of the country spread response is about one year. They are unchanged in the period of impact. investment.1) to a one percentage point increase in the US interest rate shock. However.122
Mart´ Uribe ın
by the VAR system (6. The adverse spread shock produces a larger contraction in aggregate domestic absorption than! in aggregate output. as indicated by the width of the 2-standard-deviation error bands. That is. This is reﬂected in the fact that the trade balance improves in the two periods following the shock. investment. Output. In the two periods following the country-spread shock. the quantitative eﬀects of an innovation in the US interest rate are much more pronounced than those caused by a country-spread disturbance of equal magnitude.
. and the trade balance-to-output ratio respond as one would expect. and the trade balance. Figure 6. and subsequently recover gradually until they reach their pre-shock level. are qualitatively similar to those associated with an innovation in the country spread. For
5
These bands are computed using the delta method.
rus . output and investment fall. however.3 displays the response of the variables included in the VAR system (6.

The responses of the Trade Balance-to-GDP ratio.5 2 1.5 0 −0. Chapter 6
123
Figure 6. and broken lines depict two-standard-deviation error bands. and the US interest rate are expressed in percentage points.5 5 10 15 20 2.5 2 1. (2) The responses of Output and Investment are expressed in percent deviations from their respective log-linear trends.2 0 0 5 10 15 20 5 1 0.5 0.5 1 0.5 0.6 1 0.5 1 0.8
World Interest Rate
10
15
20
Country Interest Rate 3 2.Lectures in Open Economy Macroeconomics.
.5 −2 −3 −1 −4 −5 −1.5 −1 5
Country Spread
10
15
20
Notes: (1) Solid lines depict point estimates of impulse responses.5 0 −0. the country interest rate.4 0.3: Impulse Response To A US-Interest-Rate Shock
Output 1 0 0 −1 −0.5 5 10 15 20 −6 5 10 15 20 Investment
Trade Balance−to−GDP Ratio 2 1.

by construction. However.124
Mart´ Uribe ın
instance. the country spread initially falls. As a result. The negative impact eﬀect is in line with the ﬁndings of Eichengreen and Mody (1998) and Kamin and Kleist (1999). the terms-oftrade. however. The model is vague about the precise nature of output shocks. country spreads increase signiﬁcantly in response to innovations in the US interest rate but with a short delay. In eﬀect. that because the models estimated by these authors are static in nature. in the period of impact the country interest rate increases but by less than the jump in the US interest rate. the country spread recovers quickly and after a couple of quarters it is more than one percentage point above its pre-shock level.
impinge upon the variables of
our empirical model. they are unable to capture the rich dynamic relation linking these two variables. Thus. Figure 6. the trough in the output response is twice as large under a USinterest-rate shock than under a country-spread shock. It is remarkable that the impulse response function of the country spread to a US-interest-rate shock displays a delayed overshooting. We note. The overshooting of country spreads is responsible for the much larger response of domestic variables to an innovat! ion in the US interest rate than to an innovation in the country spread of equal magnitude. They can reﬂect variations in total factor productivity. etc. The response of investment is about three times as large as that
. The response of output. and the trade balance is very much in line with the impulse response to a positive productivity shock implied by the small open economy RBC model (see ﬁgure 4.1).
y t. We now ask how innovations in output.4 depicts the impulse response function to a one-percent increase in the output shock. investment.

and broken lines depict two-standard-deviation error bands.3 −0.8 0.2 0 5 10 15 20 0.6 −0.5
World Interest Rate
−0.4 −0.2 −0.4: Impulse Response To An Output Shock
Output 1 3 0.6 −0. (2) The responses of Output and Investment are expressed in percent deviations from their respective log-linear trends.8 5
Country Spread
10
15
20
Notes: (1) Solid lines depict point estimates of impulse response functions.4 −0.Lectures in Open Economy Macroeconomics.2 −0. the country interest rate.5 5 10 15 20 −1 5 0 0.8 5 10 15 20 0 −0. The responses of the Trade Balance-to-GDP ratio. Chapter 6
125
Figure 6.5 Investment
Trade Balance−to−GDP Ratio 1 0 −0.4 −0.5 0 5 10 15 20 2.4 1 0.5 2 1.
.1 −0.6 0.5
10
15
20
Country Interest Rate 0 −0. and the US interest rate are expressed in percentage points.2 −0.

1) at diﬀerent horizons. t
Broken lines depict the
fraction of the variance of the forecasting error explained by US-interestrate shocks (
rus ). The countercyclical behavior of the country spread in response to output shocks suggests that country interest rates behave in ways that exacerbates the business-cycle eﬀects of output shocks.3
Variance Decompositions
Figure 6.4 percent and after two quarters starts to improve.126
Mart´ Uribe ın
of output. the increase in output produces a signiﬁcant reduction in the country spread of about 0.
6. t
Because
rus t
and
r t
are orthogonal disturbances.5 displays the variance decomposition of the variables contained in the VAR system (6. converging gradually to its steady-state level.6 percent. the trade balance deteriorates signiﬁcantly by about 0. At the same time. More interestingly. the
vertical diﬀerence between the solid line and the broken line represents the variance of the forecasting error explained by country-spread shocks at dif-
. Solid lines show the fraction of the variance of the forecasting error explained jointly by US-interest-rate shocks and country-spread shocks (
rus t
and
r ). The half life of the country spread response is about ﬁve quarters.

Lectures in Open Economy Macroeconomics, Chapter 6

127

**Figure 6.5: Variance Decomposition at Diﬀerent Horizons
**

Output 0.3 0.25 0.25 0.2 0.15 0.1 0.05 0 0.2 0.15 0.1 0.05 5 10 15 quarters 20 0 5 10 15 quarters World Interest Rate 2 0.4 1.5 0.3 1 20 Investment

Trade Balances−to−GDP Ratio

0.2

0.1

0.5

0

5

10 15 quarters Country Interest Rate

20

0

5

10 15 quarters Country Spread

20

0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 5 10 15 quarters 20

0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 5 10 15 quarters 20

rus

rus

+

r

Note: Solid lines depict the fraction of the variance of the kquarter-ahead forecasting error explained jointly by rus and r t t at diﬀerent horizons. Broken lines depict the fraction of the variance of the forecasting error explained by rus at diﬀerent t horizons.

128

Mart´ Uribe ın

ferent horizons.6,7 Note that as the forecasting horizon approaches inﬁnity, the decomposition of the variance of the forecasting error coincides with the decomposition of the unconditional variance of the series in question. For the purpose of the present discussion, we associate business-cycle ﬂuctuations with the variance of the forecasting error at a horizon of about ﬁve years. Researchers typically deﬁne business cycles as movements in time series of frequencies ranging from 6 quarters to 32 quarters (Stock and Watson, 1999). Our choice of horizon falls in the middle of this window. According to our estimate of the VAR system given in equation (6.1), innovations in the US interest rate,

rus , t

explain about 20 percent of move-

ments in aggregate activity in emerging countries at business cycle frequency. At the same time, country-spread shocks,

r, t

account for about 12 percent of

6 ˆ us ˆ These forecasting errors are computed as follows. Let xt ≡ [ˆt ˆt tbyt Rt Rt ] be the y ı vector of variables included in the VAR system and t ≡ [ y i tby !rus r ] the vector of t t t t t disturbances of the VAR system. Then, one can write the MA(∞) representation of xt as xt = ∞ Cj t−j , where Cj ≡ (A−1 B)j A−1 . The error in forecasting xt+h at time t j=0 for h > 0, that is, xt+h − Et xt+h , is given by h Cj t+h−j . The variance/covariance j=0 h matrix of this h-step-ahead forecasting error is given by Σx,h ≡ j=0 Cj Σ Cj , where Σ is the (diagonal) variance/covariance matrix of t . Thus, the variance of the h-stepahead forecasting error of xt is simply the vector containing the diagonal elements of Σx,h . In turn, the variance of the error of the h-step-ahead forecasting error of xt due to rus ,h a particular shock, say rus , is given by the diagonal elements of the matrix Σx, ≡ t !h (Cj Λ4 )Σ (Cj Λ4 ) , where Λ4 is a 5×5 matrix with all elements equal to zero except j=0 element (4,4), which takes the value one. Then, the broken lines in ﬁgure 6.5 are given by rus ,h the element-by-element ratio of the diagonal elements of Σx, to the diagonal elements x,h of the matrix Σ for diﬀerent values of h. The diﬀerence between the solid lines and the broken lines (i.e., the fraction of the variance of the forecasting error due to r ) is t computed in a similar fashion but using the matrix Λ5 . 7 We observe that the estimates of y , i , tby , and r (i.e., the sample residuals of t t t t the ﬁrst, second, third, and ﬁfth equations of the VAR system) are orthogonal to each us other. But because yt , ˆt , and tbyt are excluded from the Rt equation, we have that ˆ i rus the estimates of t will in general not be orthogonal to the! estimates of y , i , or tby . t t t However, under our maintained speciﬁcation assumption that the US real interest rate does not systematically respond to the state of the business cycle in emerging countries, this lack of orthogonality should disappear as the sample size increases.

Lectures in Open Economy Macroeconomics, Chapter 6

129

aggregate ﬂuctuations in these countries. Thus, around one third of business cycles in emerging economies is explained by disturbances in external ﬁnancial variables. These disturbances play an even stronger role in explaining movements in international transactions. In eﬀect, US-interest-rate shocks and country-spread shocks are responsible for about 43 percent of movements in the trade balance-to-output ratio in the countries included in our panel. Variations in country spreads are largely explained by innovations in US interest rates and innovations in country-spreads themselves. Jointly, these two sources of uncertainty account for about 85 percent of ﬂuctuations in country spreads. Most of this fraction, about 60 percentage points, is attributed to country-spread shocks. This last result concurs with Eichengreen and Mody (1998), who interpret this ﬁnding as suggesting that arbitrary revisions in investors sentiments play a signiﬁcant role in explaining the behavior of country spreads. The impulse response functions shown in ﬁgure 6.4 establish empirically that country spreads respond signiﬁcantly and systematically to domestic macroeconomic variables. At the same time, the variance decomposition performed in this section indicates that domestic variables are responsible for about 15 percent of the variance of country spreads at business-cycle frequency. A natural question raised by these ﬁndings is whether the feedback from endogenous domestic variables to country spreads exacerbates domestic volatility. Here we make a ﬁrst step at answering this question. ˆ Speciﬁcally, we modify the Rt equation of the VAR system by setting to zero the coeﬃcients on yt−i , ˆt−i , and tbyt−i for i = 0, 1. We then compute ˆ i

Dev.
. the volatility of the country interest rate falls by about one third.1 shows that the presence of feedback from domestic variables to country spreads signiﬁcantly increases domestic volatility.1: Aggregate Volatility With and Without Feedback of Spreads from Domestic Variables Model Variable Feedback No Feedback Std.4955 4. The eﬀect of feedback on the cyclical behavior of the country spread itself is even stronger.9260 4. For one should not expect that in response to changes in the coeﬃcients deﬁning the spread process all other coeﬃcients of the VAR system will remain unaltered. This more satisfactory approach necessarily involves the use of a theoretical model economy where private decisions change in response to alterations in the country-spread process. Of course.0674 ˆ ı 14.1060 11. Dev. As such.6450 3. ˆt . when feedback is negated. tbyt and Rt in the modiﬁed VAR system ˆ i at business-cycle frequency (20 quarters).3846 3. In particular. Table 6.1 serve solely as a way to motivate a more adequate approach to the question they aim to address. Std. when we shut oﬀ the endogenous feedback. y ˆ 3.130
Mart´ Uribe ın
Table 6.5198 tby R 6. In eﬀect. the results of table 6. We compare these volatilities to those emerging ! from the original VAR model.7696
ˆ the implied volatility of yt . this counterfactual exercise is subject to Lucas’ (1976) celebrated critique. We follow this route next. the volatility of output falls by 16 percent and the volatility of investment and the trade balance-to-GDP ratio fall by about 20 percent.

Lectures in Open Economy Macroeconomics, Chapter 6

131

6.4

A Theoretical Model

The process of identifying country-spread shocks and US-interest-rate shocks involves a number of restrictions on the matrices deﬁning the VAR system (6.1). To assess the plausibility of these restrictions, it is necessary to use the predictions of some theory of the business cycle as a metric. If the estimated shocks imply similar business cycle ﬂuctuations in the empirical as in theoretical models, we conclude that according to the proposed theory, the identiﬁed shocks are plausible. Accordingly, we will assess the plausibility of our estimated shocks in four steps: First, we develop a standard model of the business cycle in small open economies. Second, we estimate the deep structural parameters of the model. Third, we feed into the model the estimated version of the fourth and ﬁfth equations of the VAR system (6.1), describing the stochastic laws of motion of the US interest rate and the country spread. Finally, we compare estimated impulse responses (i.e., those shown in ﬁgures 6.2 and 6.3) with those implied by the proposed theoretical framework. The basis of the theoretical model presented here is the standard neoclassical growth model of the small open economy (e.g., Mendoza, 1991). We depart from the canonical version of the small-open-economy RBC model along four dimensions. First, as in the empirical model, we assume that in each period, production and absorption decisions are made prior to the realization of that period’s world-interest-rate shock and country-spread shock. Thus, innovations in the world interest rate or the country spread are assumed to have allocative eﬀects with a one-period lag. Second, preferences are as-

132

Mart´ Uribe ın

sumed to feature external habit formation, or catching up with the Joneses as in Abel (1990). This feature improves the predictions of the standard model by preventing an excessive contraction in private non-business absorption in response to external ﬁnancial shocks. Habit formation has been shown to help explain asset prices and business ﬂuctuations in both developed economies (e.g., B! oldrin, Christiano, and Fisher, 2001) and emerging countries (e.g., Uribe, 2002). Third, ﬁrms are assumed to be subject to a working-capital constraint. This constraint introduces a direct supply side eﬀect of changes in the cost of borrowing in international ﬁnancial markets, and allows the model to predict a more realistic response of domestic output to external ﬁnancial shocks. Fourth, the process of capital accumulation is assumed to be subject to gestation lags and convex adjustment costs. In combination, these two frictions prevent excessive investment volatility, induce persistence, and allow for the observed nonmonotonic (hump-shaped) response of investment in response to a variety of shocks (see Uribe, 1997).

6.4.1

Households

Consider a small open economy populated by a large number of inﬁnitely lived households with preferences described by the following utility function

∞

E0

t=0

β t U (ct − µ˜t−1 , ht ), c

(6.2)

where ct denotes consumption in period t, ct denotes the cross-sectional ˜ average level of consumption in period t, and ht denotes the fraction of time devoted to work in period t. Households take as given the process

Lectures in Open Economy Macroeconomics, Chapter 6

133

for ct . The single-period utility index U is assumed to be increasing in its ˜ ﬁrst argument, decreasing in its second argument, concave, and smooth. The parameter β ∈ (0, 1) denotes a subjective discount factor, and the parameter µ measures the intensity of external habit formation. Households have access to two types of asset, physical capital and an internationally traded bond. The capital stock is assumed to be owned entirely by domestic residents. Households have three sources of income: wages, capital rents, and interest income bond holdings. Each period, households allocate their wealth to purchases of consumption goods, purchases of investment goods, and purchases of ﬁnancial assets. The household’s periodby-period budget constraint is given by

dt = Rt−1 dt−1 + Ψ(dt ) + ct + it − wt ht − ut kt ,

(6.3)

where dt denotes the household’s debt position in period t, Rt denotes the gross interest rate faced by domestic residents in ﬁnancial markets, wt denotes the wage rate, ut denotes the rental rate of capital, kt denotes the stock of physical capital, and it denotes gross domestic investment. We assume that households face costs of adjusting their foreign asset position. We introduce these adjustment costs with the sole purpose of eliminating the familiar unit root built in the dynamics of standard formulations of the small open economy model. The debt-adjustment cost function Ψ(·) is assumed ¯ ¯ ¯ to be convex and to satisfy Ψ(d) = Ψ (d) = 0, for some d > 0. Earlier in chapter 4, we compared a number of standard alternative ways to induce stationarity in the small open economy framework, including the one used

that are increasing and convex in the volume of intermediation. banks face operational costs. Suppose there is a continuum of banks of measure one that behave competitively. which is precisely the shadow interest rate faced by domestic agents in the centralized problem (see the Euler condition (6.
(6. which are d d given by Rt [dt − Ψ(dt )] − Rt dt . It follows from the ﬁrst-order condition associated with this problem that the interest rate charged to domestic residents d R is given by Rt = 1−Ψ t(dt ) . The problem of domestic banks is then to choose the volume dt so as to maximize proﬁts.134
Mart´ Uribe ın
here. In addition.
8
.8 The process of capital accumulation displays adjustment costs in the form of gestation lags and convex costs as in Uribe (1997).10) below). This digression will be of use later in the paper when we analyze the ﬁrm’s problem.6)
where δ ∈ (0. and conclude that they all produce virtually identical implications for business ﬂuctuations! . Then investment in period t is given by 1 4
3
it =
sit . 1. Producing one unit of capital good requires investing 1/4 units of goods for four consecutive periods. 3.4)
In turn. 1) denotes the rate of depreciation of physical capital. Suppose that ﬁnancial transactions between domestic and foreign residents require ﬁnancial intermediation by domestic institutions (banks).5)
The stock of capital obeys the following law of motion: s3t kt
kt+1 = (1 − δ)kt + kt Φ
.
(6. Let sit denote the number of investment projects started in t − i for i = 0. Ψ(dt ). Bank proﬁts are assumed to be distributed to domestic households in a lump-sum fashion. The
The debt adjustment cost can be decentralized as follows. the evolution of sit is given by
si+1t+1 = sit . 2.
i=0
(6. They capture funds from foreign investors at the country rate Rt d and lend to domestic agents at the rate Rt . taking as given Rt and Rt . dt .

where λt . The Lagrangian associated with the household’s optimization problem can be written as:
∞
L = E0
t=0
β
t
U (ct − µ˜t−1 . k0 . As discussed in chapters 3 and 4.6). These last two assumptions ensure the absence of adjustment costs in the steady state and that the steady-state level of investment is independent of Φ. and a borrowing constraint of the form dt+j+1
j s=0 Rt+s
j→∞
lim Et
≤0
(6. R−1 d−1 . dt+1 }∞ so as to t=0 maximize the utility function (6. λt νit . The introduction of capital adjustment costs is commonplace in models of the small open economy. as deﬁned by the function Φ. and λt qt are the Lagrange multipliers associated with con-
. which is assumed to be strictly increasing. adjustment costs are a convenient and plausible way to avoid excessive investment volatility in response to changes in the interest rate faced by the country in international markets. and sit for c t=0 i = 0. the laws of motion of total investment. ut }∞ as well as c0 .4)-(6. ht+1 .2) subject to the budget constraint (6. Rt . and to satisfy Φ(δ) = δ and Φ (δ) = 1.t+1 . Chapter 6
135
process of capital accumulation is assumed to be subject to adjustment costs. and the capital stock given by equations (6. s0. wt . Households choose contingent plans {ct+1 . investment projects. 1. The household takes as given the processes {˜t−1 .Lectures in Open Economy Macroeconomics. h0 . 2. concave.3).7)
that prevents the possibility of Ponzi schemes. 3. ht ) + λt c s3t kt
1 dt − Rt−1 dt−1 − Ψ(dt ) + wt ht + ut kt − 4
2
3
sit − ct
i=0
+
λt qt (1 − δ)kt + kt Φ
− kt+1 + λt
i=0
νit (sit − si+1t+1 ) .

ht+1 ) c c Et [wt+1 λt+1 ] = −Uh (ct+1 − µ˜t .10) (6. ht+1 .3). the variables ct+1 .14)
s3t+1 kt+1 s3t+1 kt+1
−
+ λt+1 ut+1 . (6.4)-(6.8) states that in period t households choose consumption and leisure for period t + 1 in such as way as to equate the marginal utility of consumption in period t + 1 to the expected marginal utility of wealth in that period. because of our assumed information structure.15)
It is important to recall that.9) deﬁnes the household’s labor supply schedule. Et λt+1 . ht+1 ) λt 1 − Ψ (dt ) = βRt Et λt+1 1 Et λt+1 ν0t+1 = Et λt+1 4 βEt λt+1 ν1t+1 = βEt λt+1 ν2t+1 = βEt λt+1 qt+1 Φ λt qt = βEt λt+1 qt+1 1 − δ + Φ β Et λt+1 + λt ν0t 4 β Et λt+1 + λt ν1t 4 = β Et λt+1 + λt ν2t 4 s3t+1 Φ kt+1 s3t+1 kt+1 (6. respectively. Equation (6.11) (6. bec cause current consumption cannot react to unanticipated changes in wealth.6). Equation (6. and (6.8) (6. and s0t+1 all reside in the information set of period t. ht )).12) (6.5). Note that in general the marginal utility of wealth will diﬀer from the marginal utility of consumption (λt = Uc (ct − µ˜t−1 . (6.7) all holding with equality and Et λt+1 = Uc (ct+1 − µ˜t .3). (6.136
Mart´ Uribe ın
straints (6.9) (6.13) (6. by equating
. The optimality conditions associated with the household’s problem are (6.

(6.11)-(6.14) links the cost of producing a unit of capital to the shadow price of installed capital. Equation (6. the relevant rate of return on this type of asset is not simply the market rate Rt but rather the shadow rate of return Rt /[1 − Ψ (dt )]. net of depreciation and adjustment costs. ht ). Finally. Note that.Lectures in Open Economy Macroeconomics. Equation (6. to the discounted value of renting the unit of capital for one period and then selling it. because of the presence of frictions to adjust bond holdings. above its steady-state level d.15) is a pricing condition for physical capital. we have that Ψ (dt ) > 0 so that the shadow rate of return is higher than the market rate of return. It equates the revenue from selling one unit of capital today.
yt = F (kt . qt . or Tobin’s Q.13) show how to price investment projects at diﬀerent stages of completion. say. qt .2
Firms
Output is produced by means of a production function that takes labor services and physical capital as inputs. Intuitively. Chapter 6
137
the marginal disutility of eﬀort in period t + 1 to the expected utility value of the wage rate in that period.10) is an asset pricing relation equating the intertemporal marginal rate of substitution in ! consumption to the rate of return on ﬁnancial assets. The price of an investment project in its ith quarter of gestation equals the price of a project in the i-1 quarter of gestation plus 1/4 units of goods. ut+1 + qt+1 . when the household’s debt position ¯ is. equation (6.4. providing further incentives for households to save.16)
.
6. thereby reducing their debt positions. Equations (6.

d The interest rate Rt will in general diﬀer from the country interest rate
Rt —the interest rate that domestic banks face in international ﬁnancial markets—because of the presence of debt-adjustment costs. The production process is subject to a working-capital constraint that requires ﬁrms to hold non-interest-bearing assets to ﬁnance a fraction of the wage bill each period. t t−1
d where πt denotes distributed proﬁts in period t. evolves according to the t following expression
d df = Rt−1 df − F (kt . we can rewrite the above expression as t d Rt − 1 d Rt
at = at−1 − F (kt . increasing in both arguments.
. as shown in footnote 8. Firms hire labor and capital services from perfectly competitive markets. denoted by df . Then. the working-capital constraint takes the form
κt ≥ ηwt ht .
where κt denotes the amount of working capital held by the representative ﬁrm in period t. and Rt ≡
Rt 1−Ψ (dt )
is the
interest rate faced by nonﬁnancial domestic agents.138
Mart´ Uribe ın
where the function F is assumed to be homogeneous of degree one. and concave. Deﬁne the ﬁrm’s total net liabilities at the end of period t as at =
d Rt df − κt . ht ) + wt ht + ut kt + πt − κt−1 + κt . Formally. ht ) + wt ht + ut kt + πt + Rt
κt . The debt position of the ﬁrm.
η ≥ 0.

17) to eliminate πt from the ﬁrm’s objective function the ﬁrm’s problem can be stated as choosing processes for at . and kt so as to maximize
∞
E0
t=0
βt
λt λ0
at − at−1 + F (kt . ht ) + wt ht 1 + η d Rt
d Rt − 1 d Rt
+ u t kt + π t . Using constraint (6.Lectures in Open Economy Macroeconomics. That is.
The ﬁrm’s objective is to maximize the present discounted value of the stream of proﬁts distributed to its owners. ht . the domestic residents.17)
It is clear from this expression that the assumed working-capital constraint
d d increases the unit labor cost by a fraction η(Rt − 1)/Rt .
. This implies that the working-capital constraint will always bind. So we can use the working-capital constraint holding with equality to eliminate κt from the above expression to get at = at−1 − F (kt . λ0
We use the household’s marginal utility of wealth as the stochastic discount factor because households own domestic ﬁrms. which is increasing d in the interest rate Rt .
subject to a no-Ponzi-game borrowing constraint of the form at+j j d s=0 Rt+s
j→∞
lim Et
≤ 0. Chapter 6
139
We will limit attention to the case in which the interest rate is positive at all times. for otherwise the ﬁrm would incur in unnecessary ﬁnancial costs. ht ) − wt ht 1 + η d Rt
d Rt − 1 d Rt
− u t kt . which would be suboptimal.
∞
max E0
t=0
βt
λt πt .
(6.

the no-Ponzi-game constraint holding with equality. under a cashd in-advance timing the wedge takes the form 1 + η(Rt − 1). We also note that the above three equations together with the assumption that the production technology is homogeneous of degree one imply that proﬁts are zero at all times (πt = 0 ∀ t).10).9 We
also observe that any process at satisfying equation (6.140
Mart´ Uribe ın
The ﬁrst-order conditions associated with this problem are (6. This distortion is
d larger the larger the opportunity cost of holding working capital.19)
It is clear from the ﬁrst of these two eﬃciency conditions that the workingcapital constraint distorts the labor market by introducing a wedge between the marginal product of labor and the real wage rate. ht ) − wt ht 1 + η
− u t kt
In this case.
9
. η.
The precise form taken by this wedge depends on the particular timing assumed in modeling the use of working capital. For instance. Then. ht ) = ut . ht ) = wt 1 + η
(6. as well as the amount of debt intermediated by local banks. dt represents the country’s net debt position.17).
(6. Alternative assumptions give rise to diﬀerent speciﬁcations of the wedge. (6. or the higher the intensity of the working capital constraint. an optimal plan consists in holding no liabilities at all times (at = 0 for all t ≥ 0). (Rt − d 1)/Rt . We assume that ﬁrms start out with no liabilities. with distributed proﬁts given by
d Rt − 1 d Rt
πt = F (kt .18)
Fk (kt . Here we adopt the shopping-time timing.17) and the ﬁrm’s no-Ponzi-game constraint is optimal. and
d Rt − 1 d Rt
Fh (kt .

In formal terms.79ˆt + 0.Lectures in Open Economy Macroeconomics.).11ˆt − 0. Chapter 6
141
6. the variable tbyt stands for the trade balance-
.29tbyt − 0. It turns out that
us up to ﬁrst order.63Rt−1 + 0. we can generate the corresponding theoretical MA(∞) representation and compare it to its empirical counterpart.61ˆt−1 + 0.
where
is an i.i. So using the economic model.d. investment. As indicated earlier. etc. we produced empirical estimates of the coeﬃcients associated with
r t
and
rus t
in the MA(∞) representation
of the endogenous variables of interest (output. This is because we empirically identiﬁed not only the distribution of the two shocks we wish to study. We therefore close our model by introducing the law of motion of the country interest rate Rt .3
Driving Forces
One advantage of our method to assess the plausibility of the identiﬁed USinterest-rate shocks and country-spread shocks is that one need not feed into the model shocks other than those whose eﬀects one is interested in studying. This process is the estimate of the bottom equation of the VAR system (6. one only needs to know the laws of motion of Rt and Rt
to construct the coeﬃcients of the theoretical MA(∞) representation.50Rt + 0. but also their contribution to business cycles in emerging economies.20) ı + 0.4.19tbyt−1 +
r r t.35Rt−1 − 0.1) and is given by ˆ ˆ ˆ us ˆ us Rt = 0.031.12ˆt−1 y y ı (6. disturbance with mean zero and standard deviation
0.

¯ because of our maintained assumption that Ψ (d) = 0. qt .
(6.21)
Because the process for the country interest rate deﬁned by equation (6. dt .4
Equilibrium. However.22)
where 0.20)
us involves the world interest rate Rt .142 to-GDP ratio and is given by:10 yt − ct − it − Ψ(dt ) .007. λt . 3. 2 satisfying
In an economy like the one described by our theoretical model. tbyt . ht+1 . yt .4.d. individual consumption equals average consumption across households. and Parameter Values
In equilibrium all households consume identical quantities. which is assumed to be an exogenous
random variable. we must also include this variable’s law of motion as part of the set of equations deﬁning the equilibrium behavior of the theoretical
us model. innovation with mean zero and standard deviation
6. or
c t = ct . ˜
t ≥ −1. the national income and product accounts would measure private consumption as ct + Ψ(dt ) and not simply as ct . 2. sit+1 for ˜
d i = 0. kt+1 . we estimate Rt as follows an AR(1) process and obtain
ˆ us ˆ us Rt = 0. Rt . Rt . Functional Forms.i. yt
Mart´ Uribe ın
tbyt =
(6. where the debtadjustment cost Ψ(dt ) are incurred by households.
is an i.
10
. it . 1. Thus. Accordingly.
(6. wt .23)
An equilibrium is a set of processes ct+1 .83Rt−1 +
rus t
rus t . ct+1 . 1. and νit for i = 0. it follows that both measures of private consumption are identical up to ﬁrst order. ut .

we set γ = 2. and α = . Chapter 6
143
conditions (6. Following Mendoza (1991). and Evans (2001) and consists in choosing values
. Eichenbaum. h0 . capital adjustment costs. ψ. c−1 .32.21). and debt adjustment costs. This value is consistent with an average US interest rate of about 4 percent and an average country premium of 7 percent. There is no readily available estimates for these parameters for emerging economies. c − µ˜ − ω −1 hω c 1−γ
1−γ
U (c − µ˜. given c0 . (6. Our estimation procedure follows Christiano. We therefore proceed to estimate them. and µ.455.
F (k.18)-(6. 2 ψ ¯ (d − d)2 . i−1 .3)-(6. 2 φ > 0. We adopt the following standard functional forms for preferences. a standard value in business-cycle studies. the processes for the exogenous innovations
rus t
and
r. There remain four parameters to assign values to. φ.22) describing the evolution of the world interest
rate. ω = 1. t
and equation (6. the time unit is meant to be one quarter.16). y−1 . We set the steadystate real interest rate faced by the small economy in international ﬁnancial markets at 11 percent per year. and (6. all holding with equality.23). h) = k α h1−α . both of which are in line with actual data. We set the depreciation rate at 10 percent per year. Φ(x) = x − φ (x − δ)2 .Lectures in Open Economy Macroeconomics.
Ψ(d) =
In calibrating the model. h) = c
−1
. technology. η. i0 .

The implied debt adjustment costs are small. η. φ. a 10 percent increase in dt over its steady-state value d maintained over one year has a resource cost of 4 × 10−6 percent of annual GDP. Thus. φ = 72. η = 1. capital adjustment costs appear as more signiﬁcant. φ. η. investment. and the country interest rate). Then our estimate of (ψ. the set of estimated impulse responses used in the estimation procedure are originated by a single shock.φ. is that here the estimation procedure requires ﬁtting impulse responses to multiple sources of uncertainty (i.
11
. φ. For ¯ example. µ) is given by
argmax{ψ.2.144
Mart´ Uribe ın
for the four parameters so as to minimize the distance between the estimated impulse response functions shown in ﬁgure 6. country-interest-rate shocks and world-interest-rate shocks. η.2. we are setting 4 parameter values to match 192 points.e.00042. µ)] Σ−1e [IRe − IRm (ψ. and µ = 0.8.η. whereas in Christiano et al. Speciﬁcally. On the other hand. let IRe denote the 192×1 vector of estimated impulse response functions and IRm (ψ. µ)]. the trade balance.2 and the corresponding impulse responses implied by the model. This matrix penalizes those elements of the estimated impulse response functions associated with large error intervals. µ) the corresponding vector of impulse responses implied by the theoretical model. η.µ} [IRe − IRm (ψ. to 2 shocks (the US-interest-rate shock and the country-spread shock). The resulting parameter estimates are ψ = 0.11 In our exercise we consider the ﬁrst 24 quarters of the impuls! e response functions of 4 variables (output. For
A key diﬀerence between the exercise presented here and that in Christiano et al. which is a function of the four parameters we seek to estimate.. φ. IR where ΣIRe is a 192×192 diagonal matrix containing the variance of the impulse response function along the diagonal.

a 10 percent increase in investment for one year produces an increase in the capital stock of 0.12 The left col12 The Matlab code used to produce theoretical impulse response functions is available on line at http://www. In the absence of capital adjustment costs.econ. Chapter 6
145
Table 6. The estimated value of η implies that ﬁrms maintain a level of working capital equivalent to about 3.8 0.
6. Constantinides.32 72.html. Table 6.77% Description Subjective discount factor Inverse of intertemporal elasticity of substitution Habit formation parameter 1/(ω − 1) = Labor supply elasticity capital elasticity of output Capital adjustment cost parameter Debt adjustment cost parameter Depreciation rate (quarterly) Fraction of wage bill subject to working-capital constraint Steady-state real country interest rate (quarterly)
instance. 1990).6 depicts impulse response functions of output.. the estimated degree of habit formation is modest compared to the values typically used to explain asset-price regularities in closed economies (e.5
Theoretical and Estimated Impulse Responses
Figure 6.2 2. the capital stock increases by 0.204 1.455 0.025 1. starting in a steady-state situation.88 percent.00042 0.edu/∼uribe/uribe yue jie/uribe yue jie. and the country interest rate.
.g. the trade balance-to-GDP ratio.6 months of wage payments.2: Parameter Values Symbol β γ µ ω α φ ψ δ η R Value 0. investment.2 gathers all parameter values. Finally.973 2 0.96 percent.Lectures in Open Economy Macroeconomics.duke.

5 −1 −1.1 0 0 5 10 15 20 0 0
Response of TB/GDP to εr
5
10
15
20
Response of Country Interest Rate to εrus 3 2 1 0 0 5 10 15 20 1 0.2 0 0
Response of Country Interest Rate to εr
5
10
15
20
Estimated IR
-x-x.5 0 5 10 15 20 −0.
.4 1.6 0. and the second column displays impulse responses to a country-spread shock ( r ).6: Theoretical and Estimated Impulse Response Functions
Response of Output to εrus 0 0 −0.2 −0.2 0.Model IR
2-std error bands around Estimated IR
Note: The ﬁrst column displays impulse responses to a US interest rate shock ( rus ).4 0.1 −0.146
Mart´ Uribe ın
Figure 6.5
5
10
15
20
Response of TB/GDP to εrus 2 0.5 0.8 0.5 0.3 1 0.3 0 5 10 15 20 Response of Output to εr
Response of Investment to εrus 0 −2 −4 −1 −6 0 5 10 15 20 0 0
Response of Investment to εr
−0.

The model replicates three key qualitative features of the estimated impulse response functions: First. ˆt−1 . A natural ˆ ˆ ı ı question is to what extent the endogeneity of country spreads contributes
. ˆt . Chapter 6 umn shows impulse responses to a US-interest-rate shock (
rus ). the trade balance improves in response to either shock. the exogenous country-spread shock
r t
(the sentiment component). yt−1 . This information is reproduced from ﬁgures 6. the country spread St = Rt − Rt moves in response to four types of variable: its own lagged value St−1 (the autoregressive component). and broken lines depict the associated two-standard-error bands. yt . Fourth. tbyt . tbyt−1 . Third. the country interest rate displays a monotonic response to a country-spread shock. t
147 and the
right column shows impulse responses to a country-spread shock ( r ). current
us us and past US interest rates Rt and Rt−1 ).1) is indeed successful in isolating country-spread shocks and US-interest-rate shocks from the data.6
The Endogeneity of Country Spreads
According to the estimated process for the country interest rate given in ˆ ˆ ˆ us equation (6. Second.20). Solid t lines display empirical impulse response functions. and current and past values of a
ˆ ˆ set of domestic endogenous variables. Crossed lines depict theoretical impulse response functions. the country interest rate displays a hump-shaped response to an innovation in the US interest rate. We therefore conclude that the scheme used to identify the parameters of the VAR system (6.
6. output and investment contract in response to either a US-interest-rate shock or a country-spread shock.2 and 6.3.Lectures in Open Economy Macroeconomics.

it implies that. St−1 = Rt−1 − Rt−1 . First. t
.148 to exacerbating aggregate ﬂuctuations in emerging countries. This parametrization has two properties of interest.79ˆt + 0. given the past value ˆ ˆ ˆ us of the country spread. The reason is that doing so would alter the estimated process of the country spread shock
r. The process for the US interest rate is assumed to be unchanged (i. we assume that the process for the country interest rate is given by ˆ ˆ ˆ us ˆ us Rt = 0.29tbyt − 0. We note that in conducting this and the next counterfactual exercises we do not reestimate the VAR system.
Mart´ Uribe ın
We address this question by means of two counterfactual exercises.e. St . the current country spread. given by equation (6. To this end.19tbyt−1 +
r t..12ˆt−1 y y ı (6.11ˆt − 0.63Rt−1 − 0. we calculate the volatility of endogenous macroeconomic variables due to US-interest-rate shocks in a world where the country spread does not directly depend on the US interest rate.
This process diﬀers from the one shown in equation (6.22)). does not directly depend upon current or past values of the US interest rate.24) ı + 0. which is the negative of the coeﬃcient on the lagged country interest rate.63Rt−1 + Rt − 0. Speciﬁcally.61ˆt−1 + 0. The ﬁrst exercise aims at gauging the degree to which country spreads amplify the eﬀects of world-interest-rate shocks. the above speciﬁcation of the country-interest-rate process preserves the dynamics of the model in response to country-spread shocks.20) only in that the coeﬃcient on the contemporaneous US interest rate is unity and the coefﬁcient on the lagged US interest rate equals -0. Second.63.

when Rt follows
. Namely.819 0.e.20) or the one given in equation (6.420 0.623 4.469 0. due to r Baseline No yt ˆ Baseline No yt ˆ us ˆ. changes in the endogenous and the sentiment components of the country spread process. Dev. Chapter 6
149
Table 6. The result is shown in table 6.784 0.983 Note: The variable S denotes the country spread and is deﬁned as S = R/Rus . or tby us ˆ.. under the process for Rt given in equation (6.429 4.Lectures in Open Economy Macroeconomics.515 0.245 0. ˆ The precise question we wish to answer is: what process for Rt induces higher volatility in macroeconomic variables in response to US-interest-rate shocks.24) and in each case compute a variance decomposition of output and other endogenous variables of interest.24)? To answer this question. the one given in equation (6.429 4.3.866 1.622 2.580 1. due to rus Std.429 3. We ﬁnd that when the country spread is assumed not to respond directly to variations in the US interest rate (i. Dev.983 S 2.3: Endogeneity of Country Spreads and Aggregate Instability Std.885 0..547 1.347 1. or tby Variable Model No R ı Model No R ı y ˆ 1. A hat on a variable denotes log-deviation from its non-stochastic steady-state value.547 1.175 tby 1.24)) the standard deviation of output and the trade balance-to-output ratio explained by US-interest-rate shocks is about two thirds smaller than in the baseline scenario (i.446 R 3.640 4.663 0.639 ˆ ı 2. we feed the theoretical model ﬁrst with equation (6.509 1.
This would amount to introducing two changes at the same time.429 3.110 0.e.819 0.663 0.319 0.20) and then with equation (6.

this particular exercise should be conducted in an environment with a richer battery of shocks capable of explaining a larger fraction of observed business cycles than that accounted by rus and r alone.35Rt−1 +
r t.25)
Table 6. we use our theoretical model to compute the volatility of endogenous domestic variables in an environment where country spreads do not respond to domestic variables.20) with the process ˆ ˆ ˆ us ˆ us Rt = 0.50Rt + 0. A second counterfactual experiment we wish to conduct aims to assess the macroeconomic consequences of the fact that country spreads move in response to changes in domestic variables. To this end. Speciﬁcally.
Ideally. we replace the process for Rt given in equation (6.3 displays the outcome of this exercise. and the trade balance-to-output ratio explained jointly by US-interest-rate shocks and country-spread shocks ( and
r) t rus t
falls by about one fourth when the feedback from endogenous do-
mestic variables to country spreads is shut oﬀ.63Rt−1 + 0.13 We conclude that the fact that country spreads respond to the state of domestic business conditions signiﬁcantly exacerbates aggregate instability in emerging countries.150
Mart´ Uribe ın
the process given in equation (6. We ﬁnd that the equilibrium volatility of output.20)).
(6. t t
13
. such as output and the external accounts. This indicates that the aggregate eﬀects of US-interest-rate shocks are strongly ampliﬁed by the dependence of country spreads on US interest rates. investment.

we do observe a signiﬁcant amount of borrowing and lending across nations. one reason why residents of a given country honor their debts with other residents of the same country is because creditors are protected by a government able and willing to apply force against delinquent debtors. If agents have no incentives to pay their international debts. At the international level the situation is quite diﬀerent. then lenders should have no reason to lend internationally to begin with. It follows that international borrowers must have reasons to repay their debts other than 151
. A key distinction between international and domestic debts is that the latter are enforceable. Countries typically have in place domestic judicial systems capable of punishing defaulters. For there is no such a thing as a supernational authority with the capacity to enforce ﬁnancial contracts between residents of diﬀerent countries. Defaulting on international ﬁnancial contracts appears to have no legal consequences.Chapter 7
Sovereign Debt
Why do countries pay their international debts? This is a fundamental question in open-economy macroeconomics. Thus. Yet.

Economic sanctions may take many forms. we will present some stylized facts about international lending and default that will guide us in evaluating the existing theories. A reputational motive to pay international debts arises when creditor countries have the ability to exclude from international ﬁnancial markets countries with a reputation of being defaulters. as it precludes use of the current account to smooth out consumption in response to aggregate domestic income shocks. import tariﬀs and quotas.1 displays average debt-to-GNP ratios over the period 1970-2000 for a number of emerging countries that defaulted upon or restructured their external debt at least once between 1824 and 1999.
7. Before plunging into theoretical models of country debt. Intuitively. the smaller the incentives for debtor countries to default. As a result.152 pure legal enforcement. trade embargoes. etc. The table also
.1
Empirical Regularities
Table 7. such as seizures of debtor country’s assets located abroad. the stronger is the ability of creditor countries to impose economic sanctions. countries may choose to repay their debts simply to preserve their reputation and thereby maintain access to international ﬁnancing. Being isolated from international ﬁnancial markets is costly. This chapter investigates whether the existing theories of sovereign debt are capable of explaining the observed levels of sovereign debt.
Mart´ Uribe ın
Two main reasons are typically oﬀered for why countries honer their international debts: economic sanctions and reputation. however.

Egypt 2002. Philippines. Chapter 7
153
Table 7.
. Colombia.1 Average Country Spread 1756 845 186 649 442 593 464 663 1021 638
Country Argentina Brazil Chile Colombia Egypt Mexico Philippines Turkey Venezuela Average
Notes: The sample includes only emerging countries with at least one external-debt default or restructuring episode between 1824 and 1999. Turkey 1978. 1999. Morgan. Chile.P. Brazil.6 38. Debt-to-GNP ratios are averages over the period 1970-2000.0 46.1: Debt-to-GNP Ratios and Country Premiums Among Defaulters Average Debt-to-GNP Ratio 37.7 58. and expressed in basis points.1 30.3 58. Egypt 1984.4 33. Source: Own calculations based on Reinhart.5 41.2 31.6 21. with varying starting dates as follows: Argentina 1993. and Turkey.4 50. Mexico. Venezuela 1982 and 1995. Colombia did not register an external default or restructuring episode between 1970 and 2002.3 44. and Savastano (2003). Debt-to-GNP ratios at the beginning of a default episodes are averages over the following default dates in the interval 1970-2002: Argentina 1982 and 2001. Chile 1972 and 1983.2 55.0 46.6 70.7 112. and are averages through 2002. 1997. produced by J.Lectures in Open Economy Macroeconomics. 1992. Brazil 1983. Philippines 1983. and Venezuela. tables 3 and 6. Mexico 1982. Country spreads are measured by EMBI country spreads. Rogoﬀ.1 63.1 Debt-to-GNP Ratio at Year of Default 54.7 70.

it follows that at the beginning of a default
. the debt-to-GNP ratio at the onset of a default or restructuring episode was on average 14 percentage points above normal times. Table 7. they estimate ∂ log country spread = 1. the interest rate at which these 9 countries borrowed in the international ﬁnancial market was on average about 6 percentage points above the interest rate at which developed countries borrow from one another. about whether the higher debt-to-GDP ratios observed at the brink of default episodes obey to a contraction in aggregate activity or to a faster-than-average accumulation of debt in periods immediately preceding default. if.3 (see their table 3. as documented in table 7.154
Mart´ Uribe ın
displays average debt-to-GNP ratios at the beginning of default or restructuring episodes. however. and the average country spread during normal times is about 640 basis points. estimate a semielasticity of the spread with respect to the debt-to-GNP ratio of 1. The information provided in the table is silent. for the countries considered in the sample.1 also shows the country premium paid by the 9 emerging countries listed over a period starting on average in 1996 and ending in 2002. the debt-to-GNP ratio is 14 percentage points higher at the beginning of a debt default or restructuring episode than during normal times. The data suggest that at the time of default debt-to-GNP ratios are signiﬁcantly above average. There is evidence that country spreads are higher the higher the debt-to-GNP ratio.3. That is. column 4). ∂debt-to-GNP ratio Thus. Akitoby and Stratmann (2006). In eﬀect.1. During this period.

After a debt crisis.Lectures in Open Economy Macroeconomics.16 percent. That is. Table 7. one would conclude that default causes countries to be in ﬁnancial autarky for about a decade. Country spreads are known to respond systematically to other variables that may take diﬀerent values during normal and default times. This increase in spreads might seem small for a country that is at the brink of default. countries defaulted on average once every 33 years.2 displays empirical probabilities of default for 9 emerging countries over the period 1824-1999. Table 7.3 × 640 × 0. But the connection between state of default and ﬁnancial autarky should not be taken too far. Chapter 7
155
episode the country premium increases on average by 1. countries are in state of default for about 11 years on average. Akitoby and Stratmann (2006) and others have documented that spreads increase signiﬁcantly with the rate of inﬂation and decrease signiﬁcantly with the foreign reserve-to-GDP ratio. however. If one assumes that while in state of default countries have little access to fresh funds from international markets. that this increase in the country premium is only the part of the total increase in country spreads that is attributable to changes in the debt-to-GNP ratio. doesn’t necessarily preclude the possibility of obtaining new loans (possibly from
.14 = 116 basis points. or 1. For being in state of default with a set of lenders. the probability of default is about 3 prevent per year. On average. these two factors contribute to higher country spreads in periods immediately preceding a default episode. Because around default periods inﬂation tends to be high and foreign reserves tend to be low. We note.2 also reports the average number of years countries are in state of default or restructuring after a default or restructuring episode. For instance.

051 0. the average probability is conditional on at least one default in the sample period.040 0. table 1.006 0.023 0. Source: Own calculations based on Reinhart.034 0. and Savastano (2003).2: Probability of Default and Length of Default State 1824-1999 Probability of Default per year 0.030 Years in State of Default per Default Episode 11 6 14 10 11 6 32 5 7 11
Country Argentina Brazil Chile Colombia Egypt Mexico Philippines Turkey Venezuela Average
Note: The sample includes only emerging countries with at least one external-debt default or restructuring episode between 1824 and 1999.
. Rogoﬀ.156
Mart´ Uribe ın
Table 7.011 0. Therefore.040 0.017 0.046 0.

Chapter 7 other lenders) or issuing new external debt. the cumulative output loss associated with default and restructuring episodes in the 1980s was of about 4 percent over four yeas. Rose estimates an equation of the form
M
Tijt = β0 +
m=0
φijtm Rijt−m + βij Xijt +
ijt .
157
7. then maintaining access to international trade could represent a reason why countries tend to honor their international ﬁnancial obligations. The Paris Club is an informal association of creditor-country ﬁnance ministers and central bankers that meets to negotiate bilateral debt rescheduling agreements with debtor-
.
where Tijt is a measure of average bilateral trade between countries i and j in period t.Lectures in Open Economy Macroeconomics. Rose identiﬁes default with dates in which a country enters a debt restructuring deal with the Paris Club. Sturzenegger (2003) ﬁnds that after controlling for a number of factors that explain economic growth.2
The Cost of Default
Default and debt restructuring episodes are typically accompanied by signiﬁcant declines in aggregate activity. The question of
whether default disrupts international trade is of interest because if for some reason trade between two countries is signiﬁcantly diminished as a result of one country defaulting on its ﬁnancial debts with other cuntries. Rose (2005) investigates this issue empirically. Default episodes are also associated with disruptions in international trade.

respectively. sharing of a common language. or 2. area. The inception of IMF programs is associated with an accumulated contraction in trade of about 10 percent over three years. distance. a common language. population. The sample contains 283 Paris-Club debt-restructuring deals. or membership in a regional free trade agreement trade more. 1. Rose estimates the parameter
.158
Mart´ Uribe ın
country governments. Default. as measured by the debt restructuring variable Rijt has a significant and negative eﬀect on bilateral trade. The data set used for the estimation of the model covers all bilateral trades between 217 countries between 1948 and 1997 at an annual frequency. Specifically. membership to the same free trade agreement. and most of the colonial eﬀects are large and positive. Countries that share a common currency. countries that are more distant geographically trade less. Rose ﬁnds sensible estimates of the parameters pertaining to the gravity model. if neither. or both countries i and j engaged in an IMF program at time t. a common border. one. that takes the values 0. The variable Xijt is a vector of regressors including (current and possibly lagged) characteristics of the country pair ij at time t such as output. Landlocked countries and islands trade less. Rose’s empirical model belongs to the family of gravity models. whereas high-income country pairs trade more. etc. The regressor Rijt proxies default and is a binary variable equal to unity if countries i and j renegotiated debt in period t in the context of the Paris Club and zero otherwise. sharing of land borders. country pair-speciﬁc dummies. The main focus of Rose’s work is the estimation of the coeﬃcients φijm . The vector Xijt also includes current and lagged values of a variable IM Fijt .

Lectures in Open Economy Macroeconomics. Thus. Consider a small open economy populated by a large number of identical individuals. to be about 15 years.
. The central assumption behind this theory is that somehow foreign lenders coordinate to exclude from international ﬁnancial markets countries with a reputation for being defaulters. Preferences are described by the utility function
∞
E0
t=0
β t u(ct ). M . countries pay their international debts to preserve their reputation and in that way maintain access to external ﬁnancing. As a result. Chapter 7
159
φijm to be on average about 0. the cumulative eﬀect of default on trade is about one year worth of trade in the long run. Our version of the Eaton-Gersowitz model follows Arellano (2005). countries that default are punished by having to operate in autarky. This means that entering in a debt restructuring agreement with a member of the Paris Club leads to a decline in bilateral trade of about 7 percent per year for about 15 years. Rose concludes that one reason why countries pay back their international ﬁnancial obligations is fear of trade disruptions in the case of default. Thus. Based on this ﬁnding.
7.07 and the lag length.3
A Reputational Model of Sovereign Debt
Eaton and Gersowitz (1981) pioneered a literature that explains sovereign debt with a reputational argument. Autarky is costly because it implies that countries cannot share systematic income risks with the rest of the world.

Households that are in good standing and choose not to default face the following budget constraint:
c + d = y + q(d )d .. If the household is in good ﬁnancial standing.d. We drop the time subscript in expressions where all variables are dated in the current period.
where d denotes the household’s debt due in the current period. This endowment is assumed to be exogenous. If it chooses to default.i. it can choose to default on its debt obligations or to honor its debt. d denotes the debt acquired in the current period and due in the next period. and
. then it immediately acquires a bad ﬁnancial status. the household can be either in good ﬁnancial standing or in bad ﬁnancial standing If the household is in bad ﬁnancial standing. and β ∈ (0. then it is prevented from borrowing or lending in ﬁnancial markets. Each period t ≥ 0.160
Mart´ Uribe ın
where ct denotes consumption in period t. Formally. stochastic. the representative household is endowed with yt units of consumption goods. with a distribution featuring a bounded support Y = [y. and i. u is a period utility function assumed to be strictly increasing and strictly concave. then it maintains its good ﬁnancial standing until the beginning of the next period. When in autarky the household is forced to consume its endowment. ¯ At the beginning of each period. 1) is a parameter denoting the subjective discount factor. If it chooses to honor its debt. consumption of a household in bad ﬁnancial standing is given by c = y. y ].

nature of the endowment. not defaulting) is denoted by v c (d. then bond prices would depend on the level of current endowment. The household enters in bad standing when it defaults on its ﬁnancial obligations.i. y denotes next period’s endowment. y) and is given by
v c (d. d .Lectures in Open Economy Macroeconomics. d. This means that once the household falls into bad standing. y) = max u(y + q(d )d − d) + βEv g (d . This is because of the assumed i.
d
. If instead we had assumed that y were serially correlated. y ) . but not on the level of debt acquired in the previous period and due in the current period. it remains in that status forever. Notice also that q(·) is independent of the current level of output.
We assume that ‘bad ﬁnancial standing’ is an absorbent state.
For an agent in good standing.. Chapter 7
161
q(d ) denotes the market price of the household’s debt. This is because the default decision in the next period depends on the amount of debt due then. the value function associated with continuing to participate in capital markets (i. Note that the price of debt depends on the amount of debt acquired in the current period and due next period.d. The value function associated with bad ﬁnancial standing is denoted v b (y) and is given by
v b (y) = u(y) + βEv b (y ).e. which implies that its current value conveys no information about future expected endowment levels.
Here.

households choose to default when servicing the debt entails a cost in terms of forgone current consumption that is larger than the inconvenience of living in ﬁnancial autarky forever. we demonstrate that this intuition is in fact correct.
Because it is never in the agent’s interest to default when its asset position is nonnegative (or d ≤ 0). Formally. We do so in steps. and is given by
v g (d. It is then reasonable to conjecture that default is more likely the larger the level of debt and the lower the current endowment. y) = max{v b (y). it follows that D(d) is empty for all d ≤ 0.
Mart´ Uribe ın
where v g (d. y) denotes the value function associated with being in good ﬁnancial standing. the default set is deﬁned by
D(d) = {y ∈ Y : v b (y) > v c (d. In this economy. In what follows. ¯ The parameter d > 0 is a debt limit that prevents agents from engaging in Ponzi games. y)}. v c (d. We denote the default set by D(d).
The Default Set The default set contains all endowment levels at which a household chooses to default given a particular level of debt. y)}.162 subject to ¯ d ≤ d. The following proposition shows that at debt levels for which the default
.

v g (d. y ) ≥ v b (y ). This proposition states that if the household has a level of debt that puts it in risk of default.2 If y2 ∈ D(d) and y1 < y2 . it must devote part of its current endowment to servicing the debt.
. then D(d) = ∅.1 If D(d) = ∅. q(d)d − d ≥ 0 and because. by assumption. Speciﬁcally. We now establish that in this economy households tend to default in bad times. by deﬁnition. Chapter 7 set is not empty.
Proposition 7.Lectures in Open Economy Macroeconomics. The second inequality holds because. In other worlds. y) ≡ max u(y + q(d )d − d) + βEv g (d . y )
¯ d <d
≥ u(y + q d d − d) + βEv g (d.
163
v c (d. we show that if a household with a certain level of debt and income chooses to default then it will also choose to default at the same level of debt and lower income. Then. It follows that if q(d)d − d ≥ 0 ¯ for some d ≤ d. ¯ Proposition 7. then q(d )d − d < 0 for all d ≤ d. The ﬁrst inequality follows from the fact that d was picked arbitrarily.
for all y. the economy must run trade surpluses. y ) ≥ u(y) + βEv b (y ) ≡ v b (y). then if it is to continue to participate in the ﬁnancial market. if the default set is not empty then it is indeed an interval with lower bound given by the lowest endowment set level y. ¯ Proof: Suppose that q d d − d ≥ 0 for some d ≤ d. then y1 ∈ D(d).

y) ≡ ∂v c (d. then v b (y1 ) > v c (d. Let
b c vy (y) ≡ ∂v b (y)/∂y and vy (d. v b (y) − v c (d. It follows that y ∗ (d) is
given either by y or (implicitly) by v b (y ∗ (d)) = v c (d. [y. where y ∗ (d) is increasing in d if y ∗ (d) < y . y ∗ (d)]. y) > 0. y ∗ (d)). by strict concavity
b c of u. Diﬀerentiating ¯ this expression yields
c vd (d. Consider any y ∈ Y such that y ∈ D(d).1. y) is a decreasing function of y for all y ∈ D(d). every y ∈ D(d) satisﬁes v b (y) − v c (d.164
Mart´ Uribe ın
Proof: Suppose D(d) = ∅. By proposition 7. We have shown that vy (y ∗ (d))−vy (d. y ∗ (d)) <
. Equivalently. Proposition 7. y2 ) for y2 ∈ D(d).
b c we showed that vy (y) − vy (d. b c vy (y) = u (y) and vy (d. if y2 ∈ D(d). y) < 0 for all y ∈ D(d). b c dd vy (y ∗ (d)) − vy (d.
for all y ∈ D(d). y ∗ (d)) dy ∗ (d) . y) ≡ ∂v c (d. that u (y + q(d )d − d) > u (y). we
¯ have that q(d )d − d < 0 for all d ≤ d. This implies. That is.3 If D(d) = ∅. By deﬁnition. By the envelope theorem. the larger the probability of default. It follows that vy (y) − vy (d. We have shown that the default set is an interval with a lower bound given by the lowest endowment y. y) < 0. Put diﬀerently. y) = u (y + q(d )d − d). y)/∂y. the higher the debt. y)/∂d. We now show that the default set D(d) is a larger interval the larger the stock of debt. y1 ) for y1 < y2 . then D(d) is an interval. then y1 ∈ D(d) for any y1 < y2 . ¯
Proof: We already proved that the default set D(d) is an interval. y ∗ (d)) c b c where vd (d. This means that if v b (y2 ) > v c (d. At the same time.

the larger the stock of debt. It follows that the expected rate of return on the debt must equal r ∗ . equating the expected rate of return on the domestic debt to the risk-free world interest rate. one obtains Prob {y ≥ y ∗ (d )} . dd as stated in the proposition. default is more likely the lower the level of output. If the country DOES default. the higher the probability of default. These two results are in line with the stylized facts presented earlier in this chapter. Summarizing. Chapter 7
165
c 0. table7. Let the world interest rate be constant and equal to r ∗ > 0. y) and applying the envelope theorem. If the country does not default. In eﬀect. Second. We then conclude
that dy ∗ (d) > 0.
Default Risk and the Country Premium We now consider the behavior of the country interest-rate premium in this economy. We assume that foreign lenders are risk neutral and perfectly competitive. y ∗ (d)) = −u (y ∗ (d) + q(d )d − d) < 0. Using the deﬁnition of vd (d. given the stock of debt. it folc lows that vd (d.Lectures in Open Economy Macroeconomics. Therefore. foreign lenders receive nothing. q(d )
1 + r∗ =
.1 shows that at the time of default countries tend to display above-average debt-to-GNP ratios. we have obtained two important results: First. foreign lenders receive 1/q(d ) units of goods per unit lent.

is nondecreasing in the stock of debt.4
Saving and the Breakdown of Reputational Lending
A key assumption of the reputational model of sovereign debt is that when a country defaults foreign lenders coordinate to exclude this country from the possibility to borrow or lend in international ﬁnancial markets. given by the diﬀerence between 1/q(d ) and 1+r ∗ .
7.3. dd 1 + r∗ The inequality follows because by deﬁnition F ≥ 0 and because. 1 + r∗
q(d ) =
Taking derivative with respect to next period’s debt yields dq(d ) −F (y ∗ (d ))y ∗ (d ) = ≤ 0. It follows that the country spread. At a ﬁrst glance. given by 1/q(d ) − 1 − r ∗ is nondecreasing in the stock of debt. by proposition7. Letting F (y) denote the cumulative density function of the endowment shock. it might seem that what is important is that defaulters be precluded from borrowing in international ﬁnancial markets. y ∗ (d ) ≥ 0.166
Mart´ Uribe ın
The numerator on the right side of this expression is the probability that the country will not default next period. We summarize this result in the following proposition: Proposition 7. we can write 1 − F (y ∗ (d )) . Bullow and
.4 The country spread.

no foreign investor will be willing to lend to the country in period T − 2. To illustrate this insight in a simple setting. then no foreign investor would want to lend to this country in period T − 1. Thus.Assume that default is punished with perpetual exclusion from borrowing in international ﬁnancial markets. consider a deterministic economy. a
. that the prohibiting defaulters to lend (or save) to foreign agents is crucial for the reputational model to work. if the country is excluded from borrowing starting in period T − 1. but that lending in these markets is allowed after default. dT −2 ≤ 0. however.Lectures in Open Economy Macroeconomics. dT −1 ≤ 0. To see this. no lending at all could be supported on reputational grounds alone. This assumption and the fact that the economy operates under perfect foresight imply that any reputational equilibrium featuring positive debt at at least one date must be characterized by no default. It follows from this result that in an equilibrium with positive external
1 For an example of a deterministic model with sovereign debt supported by reputation. we arrive at the conclusion that default in period T implies no debt at any time. since default would occur for sure one period later. Continuing with this logic. then it will have no incentives to honor any debts outstanding in that period. Suppose that a reputational equilibrium supports a path for external debt given by {dt }∞ . As a result. see Eaton and Fern´ndez (1995). That is. dt ≤ 0 for all t ≥ 0. In turn. notice that if the country defaults at some date T > 0.1 . If delinquent countries were not allowed to borrow but could run current account surpluses. That is. Chapter 7
167
Rogoﬀ (1989) have shown. where dt denotes the level of external debt t=0 assumed in period t and due in period t + 1.

where tbT +1 is the trade balance prevailing in period T + 1 under the original debt sequence {dt }. That is. That is. The reason is that the country could default on this debt in T + 1—and therefore be excluded from borrowing internationally forever thereafter—and still be able to run trade balances no larger than the ones ˜ that would have obtained in the absence of default. because the probability of default is nil. which implies that ˜ dT +1 = dT +1 − (1 + r ∗ )dT .
for t ≥ 0. Let dT be the maximum level of external debt in this equilibrium sequence. Does it pay for the country to honor this debt? The answer is no. we have that dT +1 < 0. According to the economy’s no-default resource constraint. where tbt denotes the trade balance in period t.1)
˜ Because by assumption dT ≥ dT +1 and r ∗ > 0. the country premium is nil. we have that −tbT +1 = dT +1 − (1 + r ∗ )dT . That is. the country can generate the no-default level of trade balance in
. Let the debt position acquired in the period of default be ˜ dT +1 = −tbT +1 .
(7. let dt for t > T denote the post-default path of external debt. To see this. dT ≥ dt for all t ≥ −1.168
Mart´ Uribe ın
debt the interest rate must equal the world interest rate r ∗ > 0. The evolution of the equilibrium level of debt is then given by
dt = (1 + r ∗ )dt−1 − tbt .

we derived the breakdown of the reputation model under saving using a model without uncertainty. Let the external debt position in period T + 2 be ˜ ˜ dT +2 = (1 + r ∗ )dT +1 − tbT +2 . It follows that it pays for the country to default immediately after reaching the largest debt level dT . It follows that no external debt can be supported in equilibrium on reputational grounds. we obtain ˜ dT +2 = dT +2 − (1 + r ∗ )2 dT < 0. Using (7. tbT +2 is the trade balance prevailing in period T + 2 under the original debt sequence {dt }. again. can be supported by a debt path dT +j satisfying ˜ dT +j = dT +j − (1 + r ∗ )j dT < 0. one obtains that the no-default sequence of trade bal˜ ances.
for all j ≥ 1. 1989). Continuing in this way.
.Lectures in Open Economy Macroeconomics. But we showed that default in this perfect foresight economy implies zero debt at all times. tbT +j for j > 0. But the result also holds in a stochastic environment (see Bullow and Rogoﬀ. The inequality follows because by assumption dT +2 ≤ dT and r ∗ > 0. Chapter 7
169
period T +1 without having to borrow internationally. For simplicity.
where.1) and the fact that tbT +2 = (1 + r ∗ )dT +1 − dT +2 .

170
Mart´ Uribe ın
.

171
. 149-166. Mark.” NBER working paper No.” American Economic Review (91). Akitoby.” working paper. 2001. “Estimation of Dynamic Models With Error Components. Arellano. M. Bond. 277-297. 1981. 38-42. and Hsiao. B.References
Abel. 1990. C. C.” Review of Economic Studies (58).” IMF working paper wp/06/16. January 2006. and T. “Some Tests of Speciﬁcation for Panel Data: Monte Carlo Evidence and An Application to Employment Equations. August 2004. Arellano. and S. 598-606. Andrew. University of Minnesota. Stratmann “Fiscal Policy and Financial Markets. “Asset Pricing Lessons for Modeling Business Cycles.” Journal of the American Statistical Association (76). and Jonas Fisher. pp.. Anderson. Lawrence J. December 2005.” American Economic Review (80). Michele. H. Christiano. Aguiar. 10734. 1991. Boldrin. “Default Risk and Income Fluctuations in Emerging Markets. pp. and Gita Gopinath “Emerging Market Business Cycles: The Cycle is the Trend. “Asset Prices Under Habit Formation and Catching Up With The Joneses. T.

Sturzenegger “Default Episodes in the 1980s and 1990s: What have We Learned?. and M. Lawrence J. “Debt with Potential Repudiation: Theoretical
.172
Mart´ Uribe ın
Bullow. Cline. Eaton. Washington.. J. Correia. 98.” Econometrica (55). Gersowitz. Institute for International Finance/ 1995. “Does Saving Anticipate Declining Labor Income? An Alternative Test of the Permanent Income Hypothesis. 1995. DC. 519-543. pages 2031-2077. Grossman and K. DC... Eaton. “Habit Persistence: A Resolution of the Equity Premium Puzzle. pp. a Rogoﬀ editors. Christiano.” European Economic Review (39). November 1987. and K. P.” Research paper 97-1. and Charles Evans “Nominal Rigidities and the Dynamic Eﬀects of a Shock to Monetary Policy. William R. Constantinides. George M. July 2001. 8403. and Kevin S. Handbook of International Economics III. John Y. 1249-1273.” manuscript. chapter 39.” Journal of Political Economy (1990). Rogoﬀ. The Netherlands: Elsevier Science.” in G. Campbell.” American Economic Review (79). Amsterdam. J. 10891113. and R. International Debt Reexamined. Martin Eichenbaum. 1995. and Sergio Rebelo. 43-50. J. Universidad Torcuato Di Tella. Jo˜o C. “Sovereign Debt. November 24. “Sovereign Debt: Is to Forgive to Forget?. 2003. Fern´ndez. Cline.” NBER working paper No. Institute for International Finance. William R. and F. “Business Cycles in a a Small Open Economy. Washington. March 1989.. 1997. Isabel..Neves. Barnes “Spreads and Risk in Emerging Market Lending. Chuhan.

Barry and Ashoka Mody “What Explains Changing Spreads on Emerging-Market Debt: Fundamentals or Market Sentiment?. Income. 1981. Hall. and John B. Herschel I. and Ann L. “Dynamics of Open Economy Business Cycle Models: Understanding the Role of the Discount Factor.. 1984. 726-734. “Currency Depreciation. and Reputation. and the Balance of Trade. Kim.” American Economic Review (78).” American Economic Review (74). Edwards.” NBER Working Paper No. Robert. 1978. pp. Steven and Karsten von Kleist “The Evolution and Determinants of Emerging Market Credit Spreads in the 1990s. February 1998. Sunghyun Henry and Ayhan Kose. “Stochastic Implications of the Life Cycle-Permanent Income Hypothesis: Theory and Evidence. 47-60. Sebastian. Klein. Paul. 971-987. December 1988. 263-90.” Review of Economic Studies (47). Harberger. 1999. Kamin. 1088-1097. Repudiation. Owen.” BIS Working Paper No. 9-15. Chapter 7
173
and Empirical Analysis.” Economic Letters (65). Judson.Lectures in Open Economy Macroeconomics. “Using the Generalized Schur Form to Solve a Multivariate
. 6408. 68.” Journal of Political Economy (86). Grossman. May 1999. pp. 1950. “Estimating Dynamic Panel Data Models A Guide For Macroeconomists. 2001. “LDC Foreign Borrowing and Default Risk: An Empirical Investigation. “Sovereign Debt as a Contingent Claim: Excusable Dafault. Eichengreen. 289309. Ruth A.” Journal of Political Economy (58). Van Huyck. Arnold C.” Macroeconomic Dynamics (7).

Pablo A. the Real Exchange Rate. 311-25. “Devaluation Risk and the Businessın Cycle Implications of Exchange-Rate Management.” International Economic Review (36). and Economic Fluctuations. November 1950. Finn E. Svend and Lloyd A. Mendoza.” Carnegie-Rochester Conference Series on Public Policy (53).” CarnegieRochester Conference Series on Public Policy (1). 19-46.” Review of Economics and Statistics (32). 1405-1423.” Economic Review (Federal Reserve Bank of Dallas). Rogers. Zarazaga.” manuscript. Credit and Banking (11). Jr. and John H.” Journal of Economic Dynamics and Control (24). Lucas. 101-137. James M. Enrique.E... 1976. Nason. and Fabrizio Perri “Business Cycles in Emerging Markets:The Role of Interest Rates. February 1995. 2000. “Real Business cycles in a small-open economy. “Is the Business Cycle of Argentina Diﬀerent?. Kydland. January 2006.” Journal of Money. Enrique and Mart´ Uribe. 281-299.174
Mart´ Uribe ın Linear Rational Expectations Model. 1979. 159-187. December 2000. Mendoza. Neumeyer.. 1991. Enrique. New York University. “The Terms of Trade.
Krugman.. J. “The Present-Value Model of the Current Account Has Been Rejected: Round Up the Usual Suspects. Metzler. M. 1997. Laursen. and Carlos E.” American Economic Review (81). 21-36. R. “A Model of Balance of Payments Crises. No-
. “Flexible Exchange Rates and the Theory of Employment. 797-818. “Econometric Policy Evaluation: A Critique. 239-96. Mendoza.” Journal of International Economics (68). Paul R.

Chapter 7 vember 2001. Schmitt-Groh´. 482-509. 72-81. Stephanie and Mart´ Uribe. and Banking (33). Obstfeld. October 2003.” Cahiers Economiques et Monetaires. “Solving Dynamic General e ın
.” Quarterly Journal of Economics (97). Stephanie and Mart´ Uribe. Stephanie. “The Logic of Currency Crises.” Journal of Monetary Economics (26). Reinhart.
175
Obstfeld. 2003. “Stabilization Policy and the e ın Costs of Dollarization. and Miguel Savastano. 251-270. 163-185 . Andrew K. May 1982. 1-74. Obstfeld. 1990. “The international transmission of economic ﬂuce tuations: Eﬀects of U.” Journal of Development Economics (77(1)). 257-287. 45-75.” Journal of International Economics (61). business cycles on the Canadian economy. 189-213. Stephanie and Mart´ Uribe.. Schmitt-Groh´. 189-206. Bank of France (43). Schmitt-Groh´. Kenneth Rogoﬀ.” Journal of Money. Obstfeld.” Journal of International Economics (44). Maurice. Maurice. 1994. “Debt Intolerance. Schmitt-Groh´.. “Aggregate Spending and the Terms of Trade: Is There a Laursen-Metzler Eﬀect?.S. April 1998. “Closing Small Open Economy e ın Models. impatience.Lectures in Open Economy Macroeconomics. May 2001. Rose. “Rational and Self-Fulﬁlling Balance of Payments Crises. 2005. “Intertemporal dependence. Carmen M. and dynamics.” American Economic Review (76). 1986. Maurice. Credit. “One Reason Countries Pay Their Debts: Renegotiation and International Trade. pp. Maurice.” Brookings Papers On Economic Activity (2003(1)).

The Netherlands: Elsevier Science. 3-64. 1995. Stockman. University of Pennsylvania. James H. Stock.. 197-221. Mart´ “The Price-Consumption Puzzle of Currency Pegs. Mart´ and Z. and Linda L.” American Economic Review (85). “Tastes and Technology in a TwoCountry Model of the Business Cycle: Explaining International Comovements. 1994. 1997. “Adjustment of a Small Open Economy to External Shocks. 533-569. and Assaf Razin. Christopher “Solving Linear Rational Expectations Models. and Mark W.” Dissertation. Abdelhak S. January 2004. Uribe.” Journal of International Economics (forthcoming). 2006. 1983.” in John B. Watson. Uribe. 28).” Journal of Monetary Economics (39). February.. of Monetary Economics (49). Sims. Uribe. Mart´ “Exchange rate based inﬂation stabilization: the initial real ın.” Journal of Political Economy (91). Vivian Yue. Amsterdam.” Journal ın.
. available on line at the authors’ websites. Taylor and Michael Woodford. Svensson. eds. “Business Cycle Fluctuations in US Macroeconomic Time Series. Yale University/ 1996. Lars E. eﬀects of credible plans. April 2002. pp.176
Mart´ Uribe ın Equilibrium Models Using a Second-Order Approximation to the Policy Function.
Senhadji. 168-85. “The Terms of Trade and the Current Account: The Harberger-Laursen-Metzler Eﬀect. 1999.” Journal of Economic Dynamics and Control (vol. 97-125.O. Tesar.” mimeo. Handbook of Macroeconomics. 755-775. “Country Spreads and Emerging Counın tries: Who Drives Whom?. p. Alan C.

The University of Edinburgh Press Edinburgh. 485-504. (Ed.). “Time Preference. Chapter 7
177
Uzawa.” in Wolfe.N. Value.Lectures in Open Economy Macroeconomics. J. Capital and Growth: Papers in Honor of Sir John Hicks. 1968. pp.. H.
. the Consumption Function and Optimum Asset Holdings.