This action might not be possible to undo. Are you sure you want to continue?

Mart´ Uribe2 ın This draft, December 31, 2010 Newer versions maintained at http://www.columbia.edu/~mu2166/lecture_notes.html

I would like to thank Javier Garc´ ıa-Cicco, Felix Hammermann, and Stephanie Schmitt-Groh´ for comments and suggestions. Comments welcome. e 2 Columbia University and NBER. E-mail: martin.uribe@columbia.edu.

1

ii

Contents

1 A First Look at the Data 1.1 Duration and Amplitude of Business Cycles in Emerging and Developed Countries . . . . . . . . . . . . . . . . . . . . . . . 2 An 2.1 2.2 2.3 2.4 Endowment Economy The Model Economy . . . . . . . . . . . . . The Economy’s Response to Output Shocks Nonstationary Income Shocks . . . . . . . Testing the Model . . . . . . . . . . . . . . 1 5 9 9 16 19 24 29 30 33 38 40 42 44 47 48 57 60 62 63 65 67 69 71 74

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

3 An Economy with Capital 3.1 The Basic Framework . . . . . . . . . . . . . . . . . . 3.1.1 A Permanent Productivity Shock . . . . . . . . 3.1.2 Adjustment to Temporary Productivity shocks 3.2 Capital Adjustment Costs . . . . . . . . . . . . . . . . 3.2.1 Dynamics of the Capital Stock . . . . . . . . . 3.2.2 A Permanent Technology Shock . . . . . . . . . 4 The 4.1 4.2 4.3 4.4

. . . . . .

. . . . . .

. . . . . .

. . . . . .

Small-Open-Economy Real-Business-Cycle Model The Model . . . . . . . . . . . . . . . . . . . . . . . . . . The Model’s Performance . . . . . . . . . . . . . . . . . . The Roles of Persistence and Capital Adjustment Costs . Alternative Ways to Induce Stationarity . . . . . . . . . . 4.4.1 External Discount Factor (EDF) . . . . . . . . . . 4.4.2 External Debt-Elastic Interest Rate (EDEIR) . . 4.4.3 Internal Debt-Elastic Interest Rate (IDEIR) . . . 4.4.4 Portfolio Adjustment Costs (PAC) . . . . . . . . . 4.4.5 Complete Asset Markets (CAM) . . . . . . . . . . 4.4.6 The Nonstationary Case (NC) . . . . . . . . . . . iii

. . . . . . . . . .

. . . . . . . . . .

iv

CONTENTS 4.4.7 Quantitative Results . . . . . . . . . . . . . . . . . . . 4.4.8 The Perpetual-Youth Model . . . . . . . . . . . . . . . Emerging-Country Business Cycles Through the Lens of the SOE RBC Model . . . . . . . . . . . . . . . . . . . . . . . . Appendix A: Log-Linearization . . . . . . . . . . . . . . . . . Appendix B: Solving Dynamic General Equilibrium Models . Local Existence and Uniqueness of Equilibrium . . . . . . . . Second Moments . . . . . . . . . . . . . . . . . . . . . . . . . Impulse Response Functions . . . . . . . . . . . . . . . . . . . Matlab Code For Linear Perturbation Methods . . . . . . . . Higher Order Approximations . . . . . . . . . . . . . . . . . . Exercises . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75 76 91 100 101 112 113 117 117 118 119

4.5 4.6 4.7 4.8 4.9 4.10 4.11 4.12 4.13

5 Interest-Rate Shocks 129 5.1 An Empirical Model . . . . . . . . . . . . . . . . . . . . . . . 132 5.2 Impulse Response Functions . . . . . . . . . . . . . . . . . . . 134 5.3 Variance Decompositions . . . . . . . . . . . . . . . . . . . . . 140 5.4 A Theoretical Model . . . . . . . . . . . . . . . . . . . . . . . 145 5.4.1 Households . . . . . . . . . . . . . . . . . . . . . . . . 146 5.4.2 Firms . . . . . . . . . . . . . . . . . . . . . . . . . . . 151 5.4.3 Driving Forces . . . . . . . . . . . . . . . . . . . . . . 155 5.4.4 Equilibrium, Functional Forms, and Parameter Values 156 5.5 Theoretical and Estimated Impulse Responses . . . . . . . . . 159 5.6 The Endogeneity of Country Spreads . . . . . . . . . . . . . . 161 6 The Terms of Trade 165 6.1 Deﬁning the Terms of Trade . . . . . . . . . . . . . . . . . . . 166 6.2 Empirical Regularities . . . . . . . . . . . . . . . . . . . . . . 166 6.2.1 TOT-TB Correlation: Two Early Explanations . . . . 168 6.3 Terms-of-Trade Shocks in an RBC Model . . . . . . . . . . . 177 6.3.1 Households . . . . . . . . . . . . . . . . . . . . . . . . 177 6.3.2 Production of Consumption Goods . . . . . . . . . . . 180 6.3.3 Production of Tradable Consumption Goods . . . . . 181 6.3.4 Production of Importable, Exportable, and Nontradable Goods . . . . . . . . . . . . . . . . . . . . . . . . 182 6.3.5 Market Clearing . . . . . . . . . . . . . . . . . . . . . 184 6.3.6 Driving Forces . . . . . . . . . . . . . . . . . . . . . . 184 6.3.7 Competitive Equilibrium . . . . . . . . . . . . . . . . 185 6.3.8 Calibration . . . . . . . . . . . . . . . . . . . . . . . . 185 6.3.9 Model Performance . . . . . . . . . . . . . . . . . . . . 189

CONTENTS 6.3.10 How Important Are the Terms of Trade?

v . . . . . . . 191 193 . 194 . 197 . 199 . 200 . 206 . 207 . 227 . 243 . 248 249 251 255 261 262 264 278 286

7 Overborrowing 7.1 Imperfect Policy Credibility . . . . . . . . . . . . . . 7.1.1 The Government . . . . . . . . . . . . . . . . . 7.1.2 Equilibrium Under A Permanent Trade Reform 7.1.3 Equilibrium Under A Temporary Tariﬀ Reform 7.2 Financial Externalities . . . . . . . . . . . . . . . . . . 7.2.1 The No Overborrowing Result . . . . . . . . . 7.2.2 The Case of Overborrowing . . . . . . . . . . . 7.2.3 The Case of Underborrowing . . . . . . . . . . 7.2.4 Discussion . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

8 Sovereign Debt 8.1 Empirical Regularities . . . . . . . . . . . . . . . . . . . . 8.2 The Cost of Default . . . . . . . . . . . . . . . . . . . . . 8.3 Default Incentives With State-Contingent Contracts . . . 8.3.1 Optimal Debt Contract With Commitment . . . . 8.3.2 Optimal Debt Contract Without Commitment . . 8.4 Default Incentives With Non-State-Contingent Contracts 8.5 Saving and the Breakdown of Reputational Lending . . .

. . . . . . .

. . . . . . .

vi

CONTENTS

The volatility of detrended output is 4.Chapter 1 A First Look at the Data The characterization of business cycles in small open economies requres. Periods of 1 .8 in Canada. The group of developed economies is typically deﬁned by countries with high income per capita.1 illustrates this contrast by displaying key business-cycle properties in Argentina and Canada.6 in Argentina and only 2. Examples of developed small open economies are Canada and Belgium. and the group of emerging economies is composed of middle income countries. the distinction between developed and emerging economies. Table 1. 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. critically. and examples of small open emerging economies are Argentina and Malaysia. 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.

4 2. xt−1 ) 0.70 0.13 0.76 0.9 4.61 corr(xt . GDPt ) 1 1 0.70 0.7 corr(xt .3 1.2 Mart´ Uribe ın Table 1. . Standard deviations are measured in percentage points from trend.84 -0.5 13.8 2.1 2.6 2. Kydland and Zarazaga (1997).66 0.0 3.48 0. For Argentina.31 0.94 0.80 0.37 Source: Mendoza (1991).59 0.64 -0. data on hours and productivity are limited to the manufacturing sector.0 1.54 0.96 0.8 5.3 9.79 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.

hours. Also.2 displays average business cycle facts in developed and emerging economies. A third diﬀerence between Canadian and Argentine business cycles is that in Argentina consumption appears to be more volatile than output at businesscycle frequencies. Two additional diﬀerences between the business cycle in Argentina and Canada are that in Argentina the correlation of the domestic components of aggregate demand (consumption and investment) with GDP are twice as high as in Canada. The diﬀerences between the business cycles of Argentina and Canada hold more generally between emerging and developed countries.Lecture in Open Economy Macroeconomics. In particular. and that in Argentina hours and productivity are less correlated with GDP than in Canada. Chapter 1 3 economic boom (contraction) are characterized by relatively larger trade deﬁcits (surpluses) in emerging countries than in developed countries. emerging countries are signiﬁcantly more volatile and display a much more countercyclical trade-balance share than developed countries. For all countries. whereas the reverse is the case in Canada. 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. these variables move in tandem with output. The table displays average 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). investment. One dimension along which business cycles in Argentina and Canada are similar is the procyclicality of consumption. In both countries. The data is detrended using a band-pass ﬁlter that leaves out all frequencies above 32 quarters and below 6 quarters. Table 1. and productivity. consumption is .

Denmark.4 Mart´ Uribe ın Table 1. New Zealand.32 3. Finland.26 0. Source: Aguiar and Gopinath (2004).87 Developed Countries 1. Emerging countries: Argentina.04 0. Spain. Canada. Mexico.71 -0. and Turkey. .69 0. Brazil.2: Business Cycles: Emerging Vs.96 2. Thailand. Austria.09 -0. Switzerland. Data are detrended using a band-pass ﬁlter including frequencies between 6 and 32 quarters with 12 leads and lags. Sweden. Belgium. Philippines.y Emerging Countries 2.9 0. Norway.94 3.75 Note: Average values of moments for 13 small emerging countries and 13 small developed countries.86 0. Portugal. Slovak Republic. Korea.02 1.95 0. Netherlands. Developed Countries: Australia.87 0.y ρi.42 0. South Africa.09 0.58 0.y ρc. Ecuador.74 0. Malaysia. Israel. Peru. Developed Economies Moment σy σ∆y ρy ρ∆y σc /σy σi /σy σtb/y ρtb/y.23 1.

We now decompose business cycles into contractions and expansions and estimate for each of these phases of the cycle its duration and magnitude. consisting in identifying peaks and troughs in the logarithm of real quarterly GDP. for j = −2. The duration of a contraction is the period of time that goes from a peak to the next trough. They deﬁne a peak as an output observation that is larger than the two immediately preceeding and succeeding observations. The amplitude of a contraction is the percentage fall in output between a peak and the next trough.Lecture in Open Economy Macroeconomics. a peak takes place when yt > yt+j . Formally. 1. for j = −2. The duration of a cycle is the period of time that goes from one peak to the next. Chapter 1 5 more volatile than output in emerging countries but less volatile than output in developed countries. or a level of yt satistying yt < yt+j .1 Duration and Amplitude of Business Cycles in Emerging and Developed Countries We have documented that emerging countries display signiﬁcantly more output volatility than developed countries. And the duration of an expansion is the period of time that goes from a trough to the next peak. . 1. letting yt denote the logarithm of real GDP. Calder´n and Fuentes (2010) adopt o a classical approach to characterizing business cycles in emerging and developed countries. a trough is deﬁned as an output observation that is lower than its two immediately preceeding and succeeding observations. −1. Similarly. −1. 2. 2. The amplitude of an expansion is the percentage increase in output between a trough and the next peak. 1.

which results from adding the average durations of con- . Japan.6 Mart´ Uribe ın Table 1. However. The table shows that contractions are on average quite short (about 11 months) and of about the same duration in the emerging and developed countries.and Venezuela.6 23.2 versus 2. Germany. Chile.2 20. Costa Rica. Table 1.2 21. Uruguay.3 displays the average duration and amplitude of business cycles in two groups of countries. The coutries included in the Latin America group are: Argentina. o Note: The data is quarterly real GDP from 1980:1 to 2006:4. Finally. Spain. However. contractions are much more pronounced in emerging countries than in developed countries (6. The countries included in the OECD group are Australia. and the United States. Peru. Mexico. Brazil. Sweden. Ecuador. Italy. France. Portugal. New Zealand. United Kingdom. Bolivia.2 percent of GDP).2 Source: Calder´n and Fuentes (2010).3 2.3: Duration and Amplitude of Business Cycles in Emerging and Developed Economies Group of Countries Latin America OECD Duration Contraction Expansion 3. Colombia. emerging countries are more cyclical developed economies in the sense that in the former set of countries complete cycles are shorter (20 quarters versus 27 quarters. Canada. expansions are much longer than contractions. and are shorter in emerging countries than in developed countries (16 versus 23 quarters).5 16.0 3. We will identify the former group with emerging countries and the latter with developed countries. 12 from Latin America and 12 from the pertaining to the OECD. Paraguay. At the same time.8 Amplitude Contraction Expansion 6. their amplitude is about the same in both groups of countries (about20 percent of GDP).

Chapter 1 7 tractions and expansions). The general pattern that emerges is of emerging countries being more volatile. .Lecture in Open Economy Macroeconomics. as they display more cycles per unit of time and more pronounced contractions on average.

8 Mart´ Uribe ın .

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 ). The model developed in this chapter is simple enough to allow for a full characterization of its equilibrium dynamics using pen and paper. 9 (2. 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 1.1) . 2.Chapter 2 An Endowment Economy The purpose of this chapter is to build a canonical dynamic.

10 Mart´ Uribe ın where ct denotes consumption and U denotes the single-period utility function. dt − dt−1 . has two sources. households receive an exogenous and stochastic endowment and have the ability to borrow or lend in a risk-free real bond that pays a constant interest rate. (1 + r)j j→∞ lim Et (2. Each period.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 in the long run. The above constraint states that the change in the level of debt. The evolution of the debt position of the representative household is given by dt = (1 + r)dt−1 + ct − yt . This is because if the allocation {ct . and yt is an exogenous and stochastic endowment of goods. Households are subject to the following borrowing constraint that prevents them from engaging in Ponzi games: dt+j ≤ 0. ct − yt . dt }∞ satisﬁes the no-Ponzi-game constraint with strict inequality. which is assumed to be strictly increasing and strictly concave. (2. interest services on previously acquired debt. r denotes the interest rate. The optimal allocation of consumption and debt will always feature this constraint holding with strict equality. then t=0 . rdt−1 . and excess expenditure over income. This endowment process represents the sole source of uncertainty in this economy. assumed to be constant.2) where dt denotes the debt position assumed in period t.

the cost and beneﬁt of postponing consumption must equal each other in the margin. (2. We now derive an intertemporal resource constraint by combining the household’s sequential budget constraint (2.4) The interpretation of this expression is simple. dt }∞ that also satisﬁes the not=0 Ponzi-game constraint and that satisﬁes ct ≥ ct . its period-t utility falls by U (ct ).3). 1 + r. and the following Euler condition: U (ct ) = β(1 + r)Et U (ct+1 ).2) and (2.1) subject to (2. This alternative allocation is clearly strictly preferred to the original one because the single period utility function is strictly increasing.2). The optimality conditions associated with this problem are (2. The household chooses processes for ct and dt for t ≥ 0. Begin by expressing the sequential budget constraint in period t as (1 + r)dt−1 = yt − ct + dt .Lecture in Open Economy Macroeconomics. yielding β(1 + r)Et U (ct+1 ) utils. If the household sacriﬁces one unit of consumption in period t and invests it in ﬁnancial assets. Chapter 2 11 one can choose an alternative allocation {ct . In period t + 1 the household receives the unit of goods invested plus interests. so as to maximize (2.3) holding with equality. At the optimal allocation.3) holding with equality—also known as the transversality condition. with ct > ct for at least one date t ≥ 0. Get rid of dt by combining this expression with itself evaluated one period . (2. for all t ≥ 0.2) and the no-Ponzi-scheme constraint (2.

3) holding with equality) to get the following intertemporal resource constraint: ∞ (1 + r)dt−1 = Et j=0 yt+j − ct+j . if dt−1 (1 + r) < 0. 1+r 1+r Mart´ Uribe ın (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 + . According to this intertemporal resource constraint. if dt−1 (1 + r) > 0. However. yt+j could in principle be larger than ct+j for all j ≥ 0. then it must run a trade surplus in at least one period. 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. First we require that the subjective and pecuniary rates of . yt+j must exceed ct+j for at least one j ≥ 0. 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. i. (1 + r)j Intuitively. if the economy starts as a net external debtor. At this point.e. that is.12 forward. i. then it could in principle run a perpetual trade deﬁcit. if the economy starts as a net creditor of the rest of the world. we make two additional assumptions that greatly facilitates the analysis. This yields: yt+1 − ct+1 dt+1 + ...e. that is.

(2. be equal to each other. ¯ 2 (2. our model becomes essentially Hall’s (1978) permanent income model of consumption. Finally. that is. using again the Euler equation (2. Take expectations conditional on information available at time t and use the law of iterated expectations to obtain Et ct+1 = Et ct+2 . the Euler condition (2.1 This particular functional form makes it possible to obtain ¯ a closed-form solution of the model. Repeating this procedure j times. Chapter 2 discount. we can write ct = Et ct+2 .6) one period to obtain ct+1 = Et+1 ct+2 . Second.Lecture in Open Economy Macroeconomics. Under these assumptions. we can deduce that ct = Et ct+j . β and 1/(1 + r). Now lead the Euler equation (2. we assume that the period utility index is quadratic and given by 1 U (c) = − (c − c)2 . This assumption eliminates long-run growth in consumption when the economy features no stochastic shocks. .6) which says that consumption follows a random walk: At each point in time.4) collapses to ct = Et ct+1 . 13 β(1 + r) = 1.6) to replace Et ct+1 by ct . Use this result to get rid of expected future consumption in the above intertemporal 1 After imposing this assumption. households expect to maintain a constant level of consumption over time.5) with c < c.

and to debt ser- vice.7).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. the j-period-ahead forecast of output in period t is given by Et yt+j = ρj yt . The larger is ρ. Using this expression to eliminate expectations of future income from equation (2.i. we assume that the endowment process follows an AR(1) process of the form yt = ρyt−1 + t . ct .d. 1) deﬁnes the serial correlation of the endowment process. To be able to fully characterize the equilibrium in this economy. Given this autoregressive structure of the endowment. rdt−1 .14 Mart´ Uribe ın resource constraint to obtain (after slightly rearranging terms) r Et 1+r ∞ rdt−1 + ct = j=0 yt+j . the more persistent is the endowment process. we obtain r = yt 1+r = ∞ rdt−1 + ct j=0 ρ 1+r j r yt . innovation and the parameter ρ ∈ [0. where t denotes an i. (1 + r)j (2. 1+r−ρ .

Chapter 2 Solving for ct . In the present model. we have that a unit increase in the endowment leads to a less-than-unit increase in consumption. it follows . The current account is deﬁned as the sum of the trade balance and net investment income on the country’s net foreign asset position.e. we have that tbt ≡ yt − ct . letting cat denote the current account in period t.8) Because ρ is less than unity. 1+r−ρ 15 ct = (2.. The trade balance is given by the diﬀerence between exports and imports of goods and services. Formally. Two key variables in open economy macroeconomics are the trade balance and the current account. the trade balance is given by the diﬀerence between output and consumption. i.Lecture in Open Economy Macroeconomics. we obtain r yt − rdt−1 . We can then write the equilibrium levels of the trade balance and the current account as: 1−ρ yt 1+r−ρ tbt = rdt−1 + and cat = 1−ρ yt . Because the current account equals the change in the country’s net foreign asset position. cat = −(dt − dt−1 ). Formally. letting tbt denote the trade balance in period t. we have that cat ≡ −rdt−1 + tbt . 1+r−ρ Note that the current account inherits the stochastic process of the underlying endowment shock.

so that endowment shocks are positively serially correlated. 2. 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. the endowment shock is assumed to be purely transitory.8). the current account plays the role of a shock absorber. Recall that we are assuming that 0 < ρ < 1. an increase in the endowment leads to a permanent deterioration in the trade balance. Households borrow to ﬁnance negative income . external debt follows a random walk and is therefore nonstationary. households smooth consumption by eating a tiny part of the current windfall and leaving the rest for future consumption. According to equation (2. Most of the endowment increase—a fraction 1/(1+r)—is saved. Thus. A temporary increase in the endowment produces a gradual but permanent decline in the stock of foreign liabilities. 1+r−ρ dt = dt−1 − According to this expression. ρ = 0. The longrun behavior of the trade balance is governed by the dynamics of external debt. Two polar cases are of interest. In this case. Because income is expected to fall quickly to its long-run level. The intuition for this result is clear.16 Mart´ Uribe ın that the equilibrium evolution of the stock of external debt is given by 1−ρ yt .2 The Economy’s Response to Output Shocks Consider the response of our model economy to an unanticipated increase in output. In the ﬁrst case.

an increase in income today is not accompanied by the expectation of a future decline in income. the current account is procyclical. is illustrated in ﬁgure 2.1. In the extreme case of purely transitory shocks. For. as documented in chapter 1. This prediction represents a serious problem for this model. the current account is nil and the stock of debt remains constant over time. In response to the positive endowment shock. That is. households expect the higher level of income to persist over time. consumption experiences a once-and-for-all increase. these two variables converge gradually to their respective long-run levels. the current account is countercyclical in small open economies. As a result. it improves during expansions and deteriorates during contractions. the more volatile is the current account. when endowment shocks are permanent. This expansion in domestic absorption is smaller than the initial increase in income. households allocate all innovations in their endowments to current consumption. Intuitively. ρ → 1. 1). In this case. households are able to sustain an expendituren path consisting in consuming the totality of the current income shock. More importantly. The other polar case emerges when shocks are highly persistent. especially in developing countries. the trade balance and the current account improve. Note that the trade balance converges to .Lecture in Open Economy Macroeconomics. Chapter 2 17 shocks and save in response to positive shocks. and. as a result. Rather. which is close to the volatility of the endowment shock itself for small values of r. It follows that the more temporary are endowment shocks. After the initial increase. As a result. The intermediate case of a gradually trend-reverting endowment process. the standard deviation of the current account is given by σy /(1 + r). which takes place when ρ ∈ (0.

18 Mart´ Uribe ın Endowment Consumption c t y t 0 t 0 t Trade Balance and Current Account External Debt 0 ca 0 t t t tbt dt Figure 2.1: Response to a Positive Endowment Shock .

Thus. Fixing this problem is at the heart of what follows in this and the next two chapters.Lecture in Open Economy Macroeconomics. is an i. shock with mean zero and variance σ 2 . a positive endowment shock produces a shortrun improvement in the trade balance but a long-run deterioration in this variable.’ A central failure of the model is the prediction of a procyclical current account. the level of income is . whose service requiers a smaller trade surplus. Chapter 2 19 a level lower than the pre-shock level.i. adjust to permanent shocks. let ∆yt ≡ yt − yt−1 denote the change in endowment between periods t − 1 and t. in this model. 2. Summarizing. Suppose that ∆yt evolves according to the following autoregressive process: ∆yt = ρ∆yt−1 + t . Speciﬁcally. external borrowing is conducted under the principle: ‘ﬁnance temporary shocks. rather than its level.d. According to this process. and ρ ∈ [0. 1) where t is a constant parameter. displays mean reversion. This is because in the long-run the economy settles at a lower level of external debt. which captures the essential elements of what has become known as the intertemporal approach to the current account.3 Nonstationary Income Shocks Suppose now that the rate of change of output.

Faced with such an income proﬁle. and the Euler equa- . Figure 2.2: Stationary Versus Nonstationary Endowment Shocks nonstationary.2) and the no-Ponzi-game constraint (2.2 provides a graphical expression of this intuition.5) subject to the sequential resource 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. thereby producing a countercyclical tendency in the current account. the no-Ponzi-game constraint holding with equality. consumption-smoothing households have an incentive to borrow against future income. The following model formalizes this story.20 Mart´ Uribe ın Stationary Endowment Nonstationary Endowment y c t t ct y t 0 t 0 t Figure 2. The ﬁrst-order conditions associated with this problem are the sequential budget constraint. in the sense that a positive output shock ( t > 0) produces a permanent future expected increase in the level of output.3). As before. This is the basic intuition why allowing for a nonstationary output process can help explain the observed countercyclicality of the trade balance and the current account at business-cycle frequencies.

Chapter 2 21 tion (2. the current account deteriorates in response to a positive innovation in output. (1 + r)j Rearranging. Using this result in the above expression. the above expression is valid regardless of whether output follows a stationary or a nonstationary process.6). (1 + r)j Using this expression and recalling that the current account is deﬁned as cat = yt − ct − rdt−1 .7). we obtain ∞ cat = −Et j=1 ∆yt+j . Now using the assumed autorregressive structure for the change in the endowment. we have not used the assumed stochastic properties of output. This implication is an important improve- . we can write the current account as: cat = −ρ ∆yt .Lecture in Open Economy Macroeconomics. (1 + r)j This expression states that the current account equals the present discounted value of future expected income decreases. Note that in deriving this expression. which we reproduce here for convenience r ct = −rdt−1 + Et 1+r ∞ j=0 yt+j . Using these optimality conditions yields the expression for consumption given in equation (2. 1+r−ρ According to this formula. we can write r Et 1+r ∞ cat = yt − j=0 yt+j . Therefore. we have that Et ∆yt+j = ρj ∆yt .

whereas in chapter 1 we reported statistics pertaining to the growth rates of consumption and output (i. respectively. it is necessary that future income be expected to be higher than current income. Are implied changes in consumption more or less volatile than changes in output? This question is important because. ∆ct and ∆yt ).2 We start with the deﬁnition of the current account cat = yt − ct − rdt−1 . letting σ∆c and σ∆y denote the standard deviations of ∆ct ≡ ct − ct−1 and ∆yt . The intuition behind this result is clear. Strictly speaking. Formally. We note that the countercyclicality of the current account in the model with nonstationary shocks depends crucially on output changes being positively serially correlated.22 Mart´ Uribe ın ment relative to the model with stationary shocks. because here we are analyzing changes in the levels of consumption and output (i. which happens only if ∆yt is positively serially correlated..e. the current account increases in response to a positive endowment shock.. as we saw in chapter 1. developing countries are characterized by consumption growth being more volatile than output growth. the current account ceases to be countercyclical. or ρ > 0. ∆ct /ct and ∆yt /yt ). we wish to ﬁnd 2 2 out conditions under which σ∆c can be higher than σ∆y in equilibrium. For an unexpected increase in income to induce an increase in consumption larger than the increase in income itself.e. when ρ is zero or negative. In eﬀect. Recall that when the endowment level is stationary. 2 . this exercise is not comparable to the data displayed in chapter 1.

we can write the standard deviation of consumption changes as σ∆c 1+r = σ∆y 1+r−ρ 1 − ρ2 . that when the level of income is a random walk. we obtain 23 cat − cat−1 = ∆yt − ∆ct − r(dt−1 − dt−2 ). namely. Then.9) This expression implies that the standard deviation of consumpiton changes is given by σ∆c = 1+r σ. consumption and income move hand in hand. 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 right-hand side of this expression equals unity at ρ = 0. This result conﬁrms the one obtained earlier in this chapter. Noting that dt−1 − dt−2 = −cat−1 and solving for ∆ct . Chapter 2 Taking diﬀerences.Lecture in Open Economy Macroeconomics. 1) for which the volatility of consumption changes is . 1+r−ρ (2. It follows that there are values of ρ in the interval (0. 1+r−ρ 2 The AR(1) speciﬁcaiton of ∆yt implies that σ∆y (1 − ρ2 ) = σ 2 . so their changes are equally volatile. The right hand side of the above expression is increasing in ρ at ρ = 0.

. whereas.d. is given by yt − ct − rdt−1 we can write equation (2.4 Testing the Model Hall (1978) was the ﬁrst to explore the econometric implication of the simple model developed in this chapter. ∆ct follows an i. Hall’s work motivated a large empirical literature devoted to testing the empirical relevance of the model described above. the variance of ∆yt becomes inﬁnitely large as changes in income become a random walk. This is because as ρ → 1. in particular. denoted cat . Hall tested the prediction that consumption must follow a random walk. deduced and tested a number of theoretical restrictions on the equilibrium behavior of national savings. In the context of the open economy. Noting that the current account in period t. Here we review these restrictions and their empirical validity. process with ﬁnite variance for all values of ρ ∈ [0. We start by deriving a representation of the current account that involves expected future changes in income. Campbell’s restrictions are readily expressed in terms of the current account.24 Mart´ Uribe ın indeed higher than that of income changes. This property ceases to hold as ∆yt becomes highly persistent. Speciﬁcally.i. Campbell (1987). 2. 1).7) as ∞ −(1 + r)cat = −yt + rEt j=1 (1 + r)−j yt+j .9) shows. as expression (2.

. it is simple to show that ∞ ∞ 25 −yt + rEt j=1 (1 + r)−j yt+j = (1 + r) j=1 (1 + r)−j Et ∆yt+j .10) Intuitively. (2. 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. Similarly. In this case the country saves for a rainy day.Lecture in Open Economy Macroeconomics. Combining the above two expression we can write the current account as ∞ cat = − j=1 (1 + r)−j Et ∆yt+j . cat Consider estimating a VAR system including xt : xt = Dxt−1 + t . Chapter 2 Deﬁning ∆xt+1 = xt+1 − xt . we have that the forecast of xt+j given Ht is given by Et [xt+j |Ht ] = D j xt . Then. Let Ht denote the information contained in the vector xt . Consider now an empirical representation of the time series ∆yt and cat . the country chooses to build its net foreign asset position (runs a current account surplus) when income is expected to decline in the future. from the above VAR system. Deﬁne ∆yt xt = .

Nason and Rogers (2006) perform an econometric test of this restriction. cat Let F ≡ − 1 0 [I − D/(1 + r)]−1 D/(1 + r). they calibrate r at 3.04. The Wald statistic associated with null hypothesis that F = [0 1] is 16. Their data strongly rejects the above cross-equation restriction of the model. The regression coeﬃcients of the right-hand side regression is F . In computing F .10) onto the vector xt . which reﬂects the discrepancy of F from [0 1].1. Now consider running a regression of the left and right hand side of equation (2. They estimate the VAR system using Canadian data on the current account and GDP net of investment and government spending. So the model implies the following restriction on the vector F : F = [0 1]. we obtain that the regression coefﬁcient for the left-hand side regression is the vector [0 1]. The estimation sample is 1963:Q1 to 1997:Q4. Since xt includes cat as one element. The VAR system that Nason and Rogers estimate includes 4 lags. with an asymptotic p-value of 0. This p-value means that if the null hypothesis was true.26 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) . would take a value of 16.7 percent per year. then the Wald statistic.1 or higher only 4 out of 100 times. Consider now an additional testable cross-equation restriction on the .

Speciﬁcally. a variation of the model economy where the bliss point is a random variable. (2. where c is still a constant. ¯ ¯ shock with mean zero. the orthogonality condition stating that cat+1 − (1 + r)cat − ∆yt+1 be orthogonal to variables dated t or earlier. and µt is an i.d.10) it follows that 27 Et cat+1 − (1 + r)cat − Et ∆yt+1 = 0. This projection yields the orthogonality restriction [0 1][D −(1+r)I]−[1 0]D = [0 0].11) According to this expression. equation (2.and right-hand sides of this expression on the information set Ht .5) by c + µt . one can consider projecting the above expression onto ∆yt−1 and cat−1 . 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.Lecture in Open Economy Macroeconomics. 3 . if one runs a regression of this variable on current and past values of xt . 4 In particular.3 This restriction is not valid in a more general version of the model featuring private demand shocks. Chapter 2 theoretical model. because in general µt is correlated with cat . for instance.11) becomes Et cat+1 − (1 + r)cat − Et ∆yt+1 = µt . Clearly.4 Both of the orthogonality conditions discussed here are strongly rejected by the data. all coeﬃcients should be equal to zero. This yields the orthogonality condition [0 1][D − (1 + r)I]D − [1 0]D2 = [0 0]. From equation (2. In particular. Nevertheless. the variable cat+1 − (1 + r)cat − ∆yt+1 is unpredictable in period t. Consider. Nason and Rogers (2006) ﬁnd that a test of the hypothesis that all coeﬃcients are zero in a regression of Consider projecting the left. will not hold. In this environment. replace c ¯ in equation (2.i.

. The p-value associated with a regression featuring as regressors past values of xt is 0.28 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.

Here. which. which was maintained throughout the previous chapter. we show that allowing for capital accumulation can help resolve this problem. in turn. 29 . is an endogenous variable. an important result derived in the previous chapter is that the simple endowment economy model driven by stationary endowment shocks fails to predict the countercyclicality of the trade balance. For output is perhaps the main variable any theory of the business cycle should aim to explain. the assumption that output is an exogenously given stochastic process.Chapter 3 An Economy with Capital In this chapter. First. The purpose of introducing physical capital in the model is twofold. we introduce capital accumulation in the simple small open economy of chapter 2. In this chapter we provide a partial remedy to this problem by assuming that output is produced with physical capital. is unsatisfactory if the goal is to understand observed business cycles. Second.

β ∈ (0. kt+1 = kt + it .3) (3. we assume that the capital stock does not depreciate. For the sake of simplicity. we relax this . The variable θt denotes an exogenous nonstochastic productivity factor. yt = θt F (kt ). strictly concave.2) where bt denotes real bonds bought in period t yielding the constant interest rate r > 0.30 Mart´ Uribe ın 3.1 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 ). Households seek to maximize this utility function subject to the following three constraints: bt = (1 + r)bt−1 + yt − ct − it . and to satisfy the Inada conditions. and twice continuously diﬀerentiable. kt denotes the stock of physical capital. t=0 (3. 1) denotes the subjective discount factor. Later in these notes. yt denotes output in period t. assumed to be strictly increasing. j→∞ (1 + r)j lim (3. strictly concave. The function F describes the production technology and is assumed to be strictly increasing. and it denotes investment. and U denotes the period utility function.1) where ct denotes consumption. and bt+j ≥ 0.

Lecture in Open Economy Macroeconomics. λt = βλt+1 [1 + θt+1 F (kt+1 )]. This assumption together with the ﬁrst two of the above optimality conditions implies that consumption is constant over time. (3. we assume that β(1 + r) = 1.4) . λt = β(1 + r)λt+1 . and the transversality condition bt = 0. t→∞ (1 + r)t lim As in the endowment-economy model of chapter 2. Chapter 3 assumptions of a no depreciation and of deterministic productivity. to avoid inessential long-run dynamics. 31 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 ]} . The ﬁrst-order conditions corresponding to this problem are U (ct ) = λt . ∀t ≥ 0. ct+1 = ct . bt = (1 + r)bt−1 + θt F (kt ) − ct − kt+1 + kt .

but also the present discounted value of the diﬀerences between output and investment. b−1 . (1 + r)j (3. kt+1 = κ(θt+1 . κ2 < 0. kt+1 .32 Mart´ Uribe ın Using this expression. we follow the same steps as in the derivation of its counter- . κ1 > 0. given the initial stock of physical capital. the optimality conditions can be reduced to the following two expressions: r = θt+1 F (kt+1 ) (3.5) and ct = rbt−1 + r 1+r ∞ j=0 θt+j F (kt+j ) − kt+j+1 + kt+j . ∞ θt+j F (kt+j )−kt+j+1 +kt+j . θt+1 . Formally. It follows from this equilibrium condition that next period’s level of physical capital.6) for t ≥ 0. which includes not only initial ﬁnancial wealth. and a decreasing function of the opportunity cost of holding physical capital.6). A perfect foresight equilibrium is a set of sequences {ct .5) states that households invest in physical capital until the marginal product of capital equals the rate of return on foreign bonds. is an increasing function of the future expected level productivity. and the deterministic sequence of productivity shocks {θt }∞ . k0 .6). j=0 (1+r)j To obtain equilibrium condition (3. t=0 Equilibrium condition (3. r. r). the initial net external asset position.4)-(3. b−1 . kt+1 }∞ t=0 satisfying (3. Equilibrium condition (3.6) says that consumption equals the interest ﬂow on a broad deﬁnition of wealth. bt .

The intertemporal resource constraint (3. Suppose further that before period 0 the economy was in a steady state in which bond holdings and the capital stock were constant ¯ ¯ b and equal to ¯ and k ≡ κ(θ.7). θt = θ for t ≤ −1 and θt = θ for t ≥ 0. we have that investment is constant and equal to zero. we have that k0 = k.6) and evaluating that . that is. respectively. from k to a level k ≡ κ(θ . the technology factor θt was con¯ stant and equal to θ. Plugging this path for the capital stock into the intertemporal resource constraint (3. Finally.Lecture in Open Economy Macroeconomics. when households expected θ0 to ¯ ¯ be equal to θ.1 A Permanent Productivity Shock Suppose that up until period -1 inclusive. ¯ i0 = k − k > 0. the trade balance is ¯ ¯ constant and large enough to serve the external debt. where y ≡ θF (k) denotes the steady¯ b ¯ state level of output. tb ≡ −r¯ b.1. In period 0. Chapter 3 part for the endowment economy. Because k0 was chosen in period −1. Suppose now that in period 0 there is a permanent. unexpected increase ¯ ¯ ¯ in the technology factor from θ to θ > θ. r) > k. and it = 0. kt = k for t ≥ 1. That is. investment experiences a one-time increase that raises the level of capital available for production in ¯ ¯ period 1. r). ca = 0.6) then implies that consumption ¯ ¯ is constant and equal to c = r¯ + y. Because the current account equals the change in the net international asset position (cat = bt+1 − bt ). Thus. 33 3. for t ≥ 1. Because in this steady state the capital stock is constant and because the depreciation rate of capital is assumed to be zero. equation (2. k1 . we have that in the steady state the current account equals zero.

we have that ¯ c > r¯ + θ F (k) > r¯ + θF (k) = c. Therefore. t ¯ ¯ by c ≡ θF (k) + r¯ To show that the proposed allocation is feasible.5) to replace r for θ F (k ) in the expression within curly brackets to obtain 1 ¯ ¯ θ F (k ) − F (k) − θ F (k ) k − k 1+r ¯ c = r¯ + θ F (k) + b .1 1 I thank Alberto Felettigh for providing this proof. we have that ¯ F (k )−F (k) ¯ k −k > F (k ). Clearly. b b ¯ ¯ ¯ This establishes that consumption experiences a once-and-for-all increase in period 0. 1+r 1+r (3. given because θ > θ. In previous versions of this book. let us plug the ¯ b.34 equation at t = 0 we get Mart´ Uribe ın c0 = r¯ + b r 1 ¯ ¯ θ F (k) − k + k + θ F (k ). That is.7) This expression can be used to show that in response to the permanent technology shock. consumption and investment paths cs and is into the sequential budget constraint (3. This means that the expression within curly brackets is strictly possitive.2) to t t . ct = c > c for all t ≥ 0. rewrite the above expression as ¯ ¯ c = r¯ + θ F (k) + b 1 ¯ ¯ θ F (k ) − F (k) − r k − k 1+r . Use equation (3. Because F is assumed to be strictly concave. t t ¯ the consumption path cs is strictly preferred to the pre-shock path. consumption experiences a permanent increase. I used to oﬀer the following alternative demostration: Consider the following suboptimal ¯ b paths for consumption and investment: cs = θ F (k) + r¯ and is = 0 for all tt ≥ 0. To see this.

amounts to ¯ assuming that output is given by the exogenous amount θt F (k). tb.8) vanishes. ¯ −r¯ − (k − k) < −r¯ Recalling also that the trade balance prevailing b b.2 In the present economy.8). trade balance deteriorates on impact. the optimal consumption path is constant starting in period 0.7) to write ¯ tb0 = tb − 1 ¯ ¯ [θ F (k ) − θ F (k) + (k − k)]. Obviously. is to set k = k on the right-hand side of equation (3. ¯ ¯ before the shock. . in which a permanent endowment shock leaves the trade balance unchanged. limt→∞ ¯ b t so the proposed suboptimal allocation satisﬁes the no-Ponzi-game condition (3. in period zero households expect future output.3). Note that in this case. it is of use to decompose the change in the trade balance into an eﬀect due to the investment surge and an eﬀect stemming from changes in future expected output. implying that tb0 = tb. obtain the sequence of asset positions bs = ¯ for all t ≥ 0. To better grasp the intuition behind this result. a permanent positive technology shock causes the trade balance to deteriorate in period zero for two reasons: First. This is equivalent to assuming that households cannot invest in physical capital. 2 A quick way to refresh the eﬀect of a permanent output shock in the endowment ¯ economy. the expression within square brackets on the right-hand side of equation (3. use equation (3. which. Chapter 3 35 ¯ The trade balance in period 0 is given by tb0 = θ F (k)−c −i0 . This result together with the fact that. We have established the existence of a feasible consumption path that is strictly preferred to the pre-shock consumption allocation.8) The expression within square brackets is unambiguously positive. once-and-for-all increase in period 0. This result is signiﬁcantly diﬀerent from the one obtained in the endowment economy.4). To this end.Lecture in Open Economy Macroeconomics. 1+r (3. It follows that the optimal consumption path must also be strictly preferred to the pre-shock consumption path. from equilibrium condition (3. The above expression therefore implies that the trade balance falls in period 0. essentially. equals −r¯ we have that tb0 < tb. Recalling ¯ ¯ that i0 = k − k > 0 and that c > θ F (k) + r¯ we deduce that tb0 < b. θ F (k ). to be higher than b/(1+r)t = 0. implies that consumption must experience a permanent. This means that the b.

This second source of trade deﬁcit appears in the second term of the expression within square brackets in equation (3. capital accumulation. note that ca0 = r¯ + tb0 and that b ¯ is predetermined in period 0. the current account returns b permanently to zero. write tb1 = y1 −c1 − ¯ i1 and note that between periods 0 and 1 output increases (from θ F (k) to ¯ θ F (k )). We then showed that the modiﬁed model produces very diﬀerent predictions regarding the initial behavior of . Evaluating the intertemporal budget constraint (3. the positive productivity shock generates an investment surge in period 0.8). We started with the simple small open economy of chapter 2 and introduced a single modiﬁcation. consumption is unchanged. Let’s take stock of the results obtained thus far. To see this. equals zero for periods t ≥ 1. it is useful to characterize the behavior of net foreign asset holdings from period 1 onward. This means that the current account.6) at any period t ≥ 1. This eﬀect is reﬂected in the ﬁrst term of the expression within square brackets in equation (3. we readily ﬁnd that bt = (c − θ F (k )) for t ≥ 0.) The trade balance remains constant after period 1. In period 1. and investment falls (from k − k > 0 to k − k = 0.36 Mart´ Uribe ın ¯ current output. namely.8). consumption smoothing households borrow from future income to consume in the present. Second. To show this. Finally. the identity b0 = ¯ + ca0 and the fact that the current account deteriorates in b period 0 imply that net foreign asset holdings fall in period 0. As a result. To establish this. θ F (k). as already explained. The trade balance improves in period 1. The current account deteriorates in period 0 by the same magnitude as the trade balance. Thtis expression states that net foreign asset holdings are constant from t = 0 onward. given by bt − bt−1 .

households expect the productivity of capital to be high in the future. dampening the increase in domestic absorption in the date the shock occurs. whereas in the endowment economy a permanent output shock leaves the trade balance unchanged. investment falls to zero in period 1 and remains nil thereafter. investment spending is spread over a number of periods.Lecture in Open Economy Macroeconomics. Chapter 3 37 the trade balance in response to a positive shock. To this end. Consequently. In turn. Note that in response to the increase in future expected productivity. Indeed. To highlight the importance of these two factors in generating a deterioration of the trade balance in response to a positive productivity shock. and thus have an incentive to raise the future stock of physical capital. the entire adjustment in investment occurs in period zero. In the present economy. In the presence of costs of adjusting the stock of capital. they expand investment in the present period. This channel is closed in the endowment economy of the previous chapter. What is behind the novel predictions of the present model? One factor that contributes to causing an initial deterioration of the trade balance is the combination of a demand for goods for investment purposes and a persistent productivity shock. this increase in the demand for goods tends to deteriorate the trade balance. In our model economy. . A second 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. in the next two sections we analyze separately an economy with purely temporary productivity shocks and an economy with capital adjustment costs. the positive productivity shock in period 0 is expected to last over time. a permanent increase in productivity causes the trade balance to initially deteriorate.

This is intuitive. That is ¯ ¯ y0 = y−1 + (θ − θ)F (k). The productivity of capital unexpectedly increases in period zero. everybody expects θt = θ for all t > 0. This output increase . That is. But k0 is ﬁxed in period zero. from θF (k) to θ F (k). ¯ where k is the level of capital inherited in period 0. As a result. households would like to have more capital in that period. however. Investment in period zero can only increase the future stock of capital.1. The positive productivity shock in period zero does produce an increase ¯ ¯ ¯ in output in that period.5) implies that the capital stock. Suppose also that in period -1 people assigned ¯ a zero probability to the event that θ0 would be diﬀerent from θ. it is worth analyzing the eﬀect of a purely temporary shock. θ0 = θ > θ. namely.38 Mart´ Uribe ın 3. and therefore also invest¯ ment. because its productivity is expected to go back down to its historic ¯ level θ right after period 0. ¯ ¯ where y−1 ≡ θF (k) is the pre-shock level of output. a zero probability event happens. suppose that up until period -1 inclusive the productivity factor θt ¯ was constant and equal to θ. Specifically. But agents have no incentives to have a higher capital stock in the future. Furthermore. equation (3. In this case. That is. In period ¯ 0. suppose that everybody correctly expects the productivity shock to ¯ be purely temporary. are unaﬀected by the productivity shock.2 Adjustment to Temporary Productivity shocks To stress the importance of persistence in productivity movements in inducing a deterioration of the trade balance in response to a positive output shock. kt = k for all t ≥ 0.

6) we have that r ¯ ¯ (θ − θ)F (k). 1+r tb0 − tb−1 = (y0 − y−1 ) − (c0 − c−1 ) − (i0 − i−1 ) = This expression shows that the trade balance improves on impact. consumers save most of the purely temporary increase in income in order to smooth consumption over time. . households invest the entire increase in output in the international ﬁnancial market and increase consumption by the interest ﬂow associated with that ﬁnancial investment. The reason for this counterfactual response is simple: Firms have no incentive to invest. Combining the above two expressions and recalling that investment is unaﬀected by the temporary shock.Lecture in Open Economy Macroeconomics. we get that the trade balance in period 0 is given by 1 ¯ ¯ (θ − θ)F (k) > 0. At the same time. Chapter 3 induces higher consumption: Using equation (3. we can derive the following principle: Principle I: The more persistent are productivity shocks. As a consequence. as the increase in the productivity of capital is short lived. Comparing the results obtained under the two polar cases of permanent and purely temporary productivity shocks. the more likely is the trade balance to experience an initial deterioration. Basically. domestic absorption increases but by less than the increase in output. 1+r 39 c0 = c−1 + ¯ ¯ where c−1 ≡ rb−1 + θF (θ) is the pre-shock level of consumption.

except that now changes in the stock of capital come at a cost. 2 kt bt = (1 + r)bt−1 + θt F (kt ) − ct − it − Here. . in the context of a model featuring a more ﬂexible notion of persistence. 1998. among many e others). This means that capital adjustment costs are nil in the steady state. capital adjustment costs are given by i2 /(2kt ) and are a strictly cont vex function of investment. Capital adjustment costs—in a variety of forms—are a regular feature of small open economy models because they help dampen the volatility of investment over the business cycle (see. 3. so that the law of motion of the capital stock is given by kt+1 = kt + it . Note further that the slope of the adjustment-cost function.g. Mendoza. given by it /kt . and Schmitt-Groh´. 1991.. also vanishes in the steady state. As will be cler shortly. e. Suppose that the sequential budget constraint is of the form 1 i2 t . we assume that physical capital does not depreciate.2 Capital Adjustment Costs Consider now an economy identical to the one analyzed thus far.40 Mart´ Uribe ın We will analyze this principle in more detail later. Note that the level of this function vanishes at the steady-state value of investment. it = 0. As in the economy without adjustment costs. this feature implies that in the steady state the relative price of capital goods in terms of consumption goods is unity.

Chapter 3 41 Households seek to maximize the utility function given in (3. (1 + r)j In deriving these optimality conditions. ct = rbt−1 + and ct+1 = ct .Lecture in Open Economy Macroeconomics. 2 kt+1 = kt + it . kt 1+ (3.9) 1 (1 + r)qt = θt+1 F (kt+1 ) + (it+1 /kt+1 )2 + qt+1 .10) j=0 θt+j F (kt+j ) − it+j − 1 (i2 /kt+j ) 2 t+j . We continue to assume that β(1 + r) = 1. r 1+r ∞ (3.1) subject to the above two restrictions and the no-Ponzi-game constraint (3. The Lagrange multiplier qt represents the shadow relative price of capital . The variables λt and qt denote Lagrange multipliers on the sequential budget constraint and the law of motion of the capital stock. we combined the sequential budget constraint with the transversality condition to obtain a single intertemporal budget constraint and made use of the fact that the Lagrange multiplier λt is constant for t ≥ 0.3). 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 . respectively. The ﬁrst-order conditions associated with the household problem are: it = qt .

As qt increases.42 Mart´ Uribe ın in terms of consumption goods. on the left-hand side. instead of using qt units of goods to buy one unit of capital. an extra unit of capital reduces tomorrow’s adjustment costs by (it+1 /kt+1 )2 /2. 1 + it /kt . In turn. This is the left-hand side of equation (3. the increase in investment raises the marginal adjustment cost.10). Alternatively. This unit yields θt+1 F (kt+1 ) units of output next period. the agent can engage in a ﬁnancial investment by purchasing qt units of bonds in period t which yields a gross return of (1+r)qt in period t+1.11) . it /kt . nonlinear diﬀerence equations in kt and qt : kt+1 = qt kt (3. In addition. Equation (3.10). agents have incentives to devote more resources to the production of physical capital.9) equates the marginal cost of producing a unit of capital. qt . Consider ﬁrst the rate of return of investing in physical capital. the unit of capital can be sold next period at the price qt+1 . we obtain the following two ﬁrst-order. and is known as Tobin’s q.1 Dynamics of the Capital Stock Eliminating it from the optimality conditions. 3. At the optimum both strategies must yield the same return. The sum of these three sources of income form the right hand side of (3. Optimality condition (3. Adding one unit of capital to the existing stock costs qt .10) compares the rate of return on bonds (left-hand side) to the rate of return on physical capital (right-hand side). which tends to restore the equality between the marginal cost and marginal revenue of capital goods. Finally. on the right-hand side.2. to the marginal revenue of selling a unit of capital.

13) Above the locus KK . 1+r (3. Chapter 3 43 Figure 3. the capital stock grows over time.Lecture in Open Economy Macroeconomics. The locus QQ corresponds to the . q = 1.12) The perfect-foresight solution to these equations is depicted in ﬁgure 3. and below the locus KK the capital stock declines over time.11). That is. The horizontal line KK corresponds to the pairs (kt . qt ) for which kt+1 = kt in equation (3. (3.1: The Dynamics of the Capital Stock q Q S a • 1 K • K′ S′ Q′ k0 k* k qt = θt+1 F (qt kt ) + (qt+1 − 1)2 /2 + qt+1 .1.

The price of capital. The locus SS represents the converging saddle path. 1. If the initial capital stock is diﬀerent from its long-run level. . 3.2 A Permanent Technology Shock Suppose now that in period 0 the technology factor θt increases permanently ¯ ¯ from θ to θ > θ. is not aﬀected in the long run. rq = θF (qk) + (q − 1)2 /2. qt ) for which qt+1 = qt in equation (3. The system (3. the locus QQ is downward sloping.14)shows that the locus QQ shifts up and to the right. the long-run level of capital experiences a permanent increase.12) is saddle-path stable. It follows that in response to a permanent increase in productivity. For qt near unity.11)-(3. which we denote by k ∗ .44 pairs (kt . Equation (3. Above and to the right of QQ . equations (3. and the steady-state value of Tobin’s q.2. adjustment costs vanish in the steady state. This is not surprising. It is clear from equation (3. This is the same value obtained in the economy without adjustment costs. as noted earlier. both q and k converge monotonically to their steady states along the saddle path.13) and (3.12). q decreases over time. That is. because. The value of k ∗ is implicitly determined by the expression r = θF (k ∗ ). qt .14) determine the steady-state value of the capital stock.14) Jointly. Mart´ Uribe ın (3. q increases over time and below and to the left of QQ .13) that the locus KK is not aﬀected by the productivity shock. on the other hand.

Then the new steady-state values of k and q are given by k ∗ and 1. In the frictionless environment. The price of installed capital. Chapter 3 45 Consider now the transition to the new steady state. the capital stock does not move. Along this transition. Suppose that the steady-state value of capital prior to the innovation in productivity is k0 in ﬁgure 3. the capital stock increases monotonically towards its new steady-state k ∗ .Lecture in Open Economy Macroeconomics. After the initial impact. This increase in the price of installed capital induces an increase in investment.1. investment experiences a one-time jump equal to k ∗ −k0 in period zero. and because investment tends to be less responsive to productivity shocks in the presence of adjustment costs. In eﬀect. it follows that the trade balance is less likely to deteriorate in response to a positive innovation in productivity in the environment with frictions. The equilibrium dynamics of investment in the presence of adjustment costs are quite diﬀerent from those arising in the absence thereof. In particular. the initial increase in investment is smaller. as the capital stock adjusts gradually to its long-run level.3 The diﬀerent behavior of investment with and without adjustment costs has consequences for the equilibrium dynamics of the trade balance. qt decreases toward 1. point a in the ﬁgure. 3 . Under capital adjustment costs. capital and investment display a mute response. In the period of the shock. because investment is part of domestic absorption. which in turn makes capital grow over time. jumps to the new saddle path. The following principle therefore emerges: Principle II: The more pronounced are capital adjustment costs. qt . the less It is straightforward to see that the response of the model to a purely temporary productivity shock is identical to that of the model without adjustment costs.

46 Mart´ Uribe ın likely is the trade balance to experience an initial deterioration in response to a positive productivity shock. In light of principles I and II derived in this chapter it is natural to ask what the model would predict for the behavior of the trade balance in response to productivity shocks when one introduces realistic degrees of capital adjustment costs and persistence in the productivity-shock process. . We address this issue in the next chapter.

First. The resulting theoretical framework is known as the Small-Open-Economy Real-Business-Cycle model. we arrived at the conclusion that a model driven by productivity shocks can explain the observed countercyclicality of the current account.Chapter 4 The Small-Open-Economy Real-Business-Cycle Model In the previous two chapters. uncertainty in the technology shock process. and capital depreciation. succinctly. or. capital adjustment costs must not be too strong. 47 . productivity shocks must be suﬃciently persistent. the SOE-RBC model. We also established that two features of the model are important for making this prediction possible. Second. we extend the model of the previous chapter by allowing for three features that add realism to the model’s implied dynamics: endogenous labor supply and demand. In this chapter.

(4. This property is desirable for a purely technical reason.4) where dt denotes the household’s debt position at the end of period t. This preference speciﬁcation was conceived by Uzawa (1968) and introduced in the small-open-economy literature by Mendoza (1991).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 . it makes it possible to rely on linear approximations to characterize equilibrium dynamics. In particular. (4. ht )θt t ≥ 0. 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. Namely. βh > 0. 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 ). under these preferences the steady state is independent of the initial net foreign asset position of the economy.3) where βc < 0. yt denotes domestic output.2) (4. rt denotes the interest rate at which domestic residents can borrow in period t.48 Mart´ Uribe ın 4. ht ). (4. ct denotes consumption. it denotes gross in- . θt+1 = β(ct .1) θ0 = 1.

Households choose processes {ct . (4. The function Φ(·) is meant to capture capital adjustment costs and is assumed to satisfy Φ(0) = Φ (0) = 0.6) to eliminate. This shock represents the single source of aggregate ﬂuctuations in the present model. small open economy models typically include capital adjustment costs to avoid excessive investment volatility in response to variations in the productivity of domestic capital or in the foreign interest rate. (4. As pointed out earlier. respectively. ht ). dt .6) and a no-Ponzi constraint of the form j→∞ lim Et dt+j j s=0 (1 + rs ) ≤ 0. yt and it .2)-(4. yt = At F (kt . it . and kt denotes physical capital.1) subject to (4. kt+1 . ht . Chapter 4 49 vestment. 1) denotes the rate of depreciation of physical capital.6) where δ ∈ (0. The stock of capital evolves according to kt+1 = it + (1 − δ)kt .7) Use equations (4.Lecture in Open Economy Macroeconomics. The restrictions imposed on Φ ensure that in the steady state adjustment costs are nil and the relative price of capital goods in terms of consumption goods. Output is produced by means of a linearly homogeneous production function that takes capital and labor services as inputs. yt . is unity. from the .5) and (4. (4. or Tobin’s q. θt+1 }∞ so as to maxt=0 imize the utility function (4.5) where At is an exogenous and stochastic productivity shock.

In turn. ht ) − ηt βc (ct .9) (4. ht ) + ct + kt+1 − (1 − δ)kt + Φ(kt+1 − kt ). ht+1 ) + Et ηt+1 β(ct+1 . a unit decline in the discount factor reduces utility in period t by ηt . ht+1 ) + 1 − δ + Φ (kt+2 − kt+1 ) (4.11) forward to obtain ηt = −Et ∞ j=1 θt+j θt+1 U (ct+j .8) Uc (ct . ht ) = λt − Uh (ct . iterate the ﬁrst-order condition (4. ht )ηt .12) These ﬁrst-order conditions are fairly standard. Let θt λt denote the Lagrange multiplier associated with this sequential budget constraint and θt ηt be the Lagrange multiplier associated with (4.11) λt [1+Φ (kt+1 −kt )] = β(ct .4). ht )Et λt+1 At+1 Fk (kt+1 . This operation yields Mart´ Uribe ın dt = (1 + rt−1 )dt−1 − At F (kt . ht+1 ) (4. the ﬁrst-order conditions associated with the household’s maximization problem are (4. ηt equals the expected present discounted value of utility from period t + 1 onward.7) holding with equality and: λt = β(ct . Similarly. ht ) but rather by Uc (ct . ht ) ηt = −Et U (ct+1 . the marginal . ht )(1 + rt )Et λt+1 (4. ht )−βc (ct . Then.50 sequential budget constraint (4.3)-(4. ht ) = λt At Fh (kt .10) (4. To see this. Intuitively.3). ht ) + ηt βh (ct . The second term in this expression reﬂects the fact that an increase in current consumption lowers the discount factor (βc < 0). except for the fact that the marginal utility of consumption is not given simply by Uc (ct . ht+j ).

who uses the following functional forms for preferences and technology: c − ω −1 hω U (c. yt . A competitive equilibrium is a set of processes {dt . 1) measures the serial correlation of the technology shock and the innovation distributed N (0.4)-(4. ht )ηt . λt . d−1 .d. and k0 and the exogenous process { t }.Lecture in Open Economy Macroeconomics. (4. white noise −1 β(c. ht )−βh (ct . given the initial conditions A0 . At } satisfying (4.i. The world interest rate is assumed to be constant and equal to r. We assume free capital mobility. ht ) but instead Uh (ct . h) = 1−γ 1−γ t is assumed to be an i.14). Chapter 4 51 disutility of labor is not simply Uh (ct . ηt . h) = k α h1−α Φ(x) = φ 2 x . kt+1 . 2 φ > 0. ht .14) where the parameter ρ ∈ (−1. it . We parameterize the model following Mendoza (1991). −ψ1 . rt . h) = 1 + c − ω −1 hω F (k. σ 2 ). yields rt = r. (4. Equating the domestic and world interest rates. ct .13) The law of motion of the productivity shock is given by the following ﬁrstorder autoregressive process: ln At+1 = ρ ln At + t+1 .

α.0129 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. is independent of the level of consumption. as it indicates the real wage at which ﬁrms are willing to hire an extra unit of labor.455 implies. In eﬀect.04 δ 0. The value of ω of 1. t (4. coincides with the demand for labor. The left-hand side is the marginal rate of substitution of leisure for consumption. The values assigned to the parameters γ.42 σ 0.9) and (4.11 α 0. and r are quite standard in the real-business-cycle literature. It represents the supply of labor.15). We also follow Mendoza (1991) in assigning values to the structural parameters of the model. the labor supply depends only upon the wage rate and.028 r 0.455 ψ1 0 . δ. a relatively high labor supply elasticity .1. The time unit is meant to be a year. ω.15) The right-hand side of this expression is the marginal product of labor.1: Calibration of the Small Open RBC Economy γ 2 ω 1. It follows that under the assumed functional forms for preferences. which. The parameter values are shown in table 4. because it indicates the real wage at which households are willing to supply an extra unit of labor to the market.1 ρ 0.10) yields hω−1 = At Fh (kt .32 φ 0. Mendoza calibrates the model to the Canadian economy. by equation (4. in particular.52 Mart´ Uribe ın Table 4. in in a decentralized setting. combining equations (4.. ht ).

The calibrated value of δ implies that capital goods depreciate at a rate of 10 percent per year. This parameter determines the stationarity of the model and the speed of convergence of the net external debt position to the steady state. equilibrium condition (4.2. To see how in steady state this ratio is linked to the value of ψ1 . The value assigned to ψ1 is picked to match the average Canadian tradebalance-to-GDP ratio of about 2 percent. use equation (4. and output (y = k α h1−α ). Use the resource constriant . The calibrated value of r of 4 percent is in line with the average real rate of return of broad measures of the stock market in developed countries over the postwar period. Given the steady-state values of hours and the capital-labor ratio. we can ﬁnd directly the steady-state values of capital. It is of interest to explain in more detail the calibration of the parameter ψ1 deﬁning the elasticity of the discount factor with respect to the composite c − hω /ω. Chapter 4 53 of 1/(ω − 1) = 2. Given the capital-labor ratio. independently of ψ1 .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 . It follows from this expression that the steady-state capital-labor ratio is independent of the parameter ψ1 . investment (i = δk). The assumed value of α implies a share of labor income in GDP of 68 percent.Lecture in Open Economy Macroeconomics.

and set the parameter β to match the observed trade- . it follows that this expression can be solved for ψ1 given tb/y. where c and t ¯ h denote the steady-state values of ct and ht . matching the observed average trade-balance-to-output ratio might require a value of ψ1 that does aﬀect the behavior of the model at business-cycle frequency. h. For this reason. With this speciﬁcation.54 Mart´ Uribe ın c = y − i − tb to eliminate c from the steady-state version of equilibrium condition (4. This expression uses the speciﬁc functional form assumed for the discount factor and can be solved for the trade balance-to-output ratio to obtain: (1 + r)1/ψ1 + i tb =1− − y y y hω ω −1 . the larger is ψ1 . and β > 0 is a parameter. nd i are all independent of ψ1 . one may want to set ψ1 at a small level so as to ensure stationarity without aﬀecting the predictions of the model at business-cycle frequency. On the one hand. δ. the parameter ψ1 governs both the steady-state trade-balance-to-output ratio and the stationarity of the equilibrium dynamics. Clearly. r. the larger is the assumed steady-state trade-balance-to-output ratio. On the other hand. This dual role may create a conﬂict. Recalling that y. just ¯ to ensure stationarity.8) to obtain the following relation between the steady-state level of the trade balance and the parameter ψ1 : [1+y−i−tb−hω /ω]−ψ1 (1+r) = 1. α. and ω. it might be useful to consider a two-parameter speciﬁcations of the discount ¯ factor. Note that in our assumed speciﬁcation of the endogenous discount factor. one can ﬁx the parameter ψ1 at a small value. such as β(ct . ht ) = β[1 + (ct − c) − ω −1 (hω − hω )]−ψ1 .

ηt . The resulting linear system can be readily solved using well-established techniques. we have that the transversality condition limj→∞ Et dt+j /(1 + r)j = 0 is always satisﬁed. kt . and ρ are calibrated to match the standard deviation of investment.. the parameters φ. linearizing) the system of equations around the nonstochastic steady state.4)-(4. We therefore must resort to an approximate solution. σ . We can therefore focus on bounded solutions to the system (4.e. it . Approximating Equilibrium Dynamics We look for solutions to the equilibrium conditions (4. The system can be written as Et f (xt+1 .8)-(4.1.4)-(4. In any such solution the stock of debt is bounded. rt . xt ) = 0. yt . The technique we will employ here consists in applying a ﬁrst-order Taylor expansion (i. This expression describes a system of nonlinear stochastic diﬀerence equations.14) of ten equations in ten variables given by the elements of the vector xt . There are a number of techniques that have been devised to solve dynamic systems like the one we are studying. λt . Chapter 4 balance-to-output ratio. we introduce a convenient .Lecture in Open Economy Macroeconomics. At } ﬂuctuates in a small neighborhood around its nonstochastic steady-state level.6) and (4. ht . ct . 55 Finally.14) where the vector xt ≡ {dt−1 . the standard deviation of output. Before linearizing the equilibrium conditions. and the serial correlation of output in Canada as shown in table 1. Closed form solutions to this type of system are not typically available.

dt−1 . Appendix A displays the linearized equilibrium conditions of the RBC model studied here. which we denote generically by wt . that . are ˆ ˆ ˆ ˆ ˆ co-state variables. State variables are variables whose values in any period t ≥ 0 are either predetermined (i. It is useful to express some variables in terms of percent deviations from their steady-state values. All the coeﬃcients of the linear system. wt × 100 is approximately equal to the percent deviation of wt from its steady-state level. Of these 10 variables.e. rt . for instance. Note ˆ that for small deviations of wt from w it is the case that wt ≈ (wt − w)/w. we deﬁne its log deviation from steady state. where w denotes the steady-state value of wt . kt and ˆ ˆ dt−1 are endogenous state variables. That is.56 Mart´ Uribe ın variable transformation. it . with net interest rates or variables that can take negative values. For any such variable. ˆ The linearized version of the equilibrium system can then be written as Axt+1 = Bxt . that is. 3 are state ˆ variables. λt . determined before ˆ t) or determined in t but in an exogenous fashion. and At is an exogenous state variable. namely. kt . we deﬁne wt ≡ wt − w.. such as the trade balance. ηt . ct . Some other variables are more naturally expressed in levels. This is the case. For this type of variable. ht . The vector xt contains 10 variables. In our model. and yt . and At . Co-states are endogenous variables whose values are not predetermined in period t. wt ≡ log(wt /w). where A and B are square matrices conformable with xt . The remaining 7 elements of xt .

and the serial correlation of output. Chapter 4 57 is. and ρ to match these three moments. To determine the initial value of the remaining seven variables. Matlab code to compute second moments and impulse response functions implied by the model studied here is available online at http://www. That appendix also shows how to compute second moments and impulse response functions. the elements of A and B. the volatility of investment. we impose a terminal condition requiring that at any point in time the system be expected to converge to the nonstochastic steady state. But the model performs relatively well along other dimensions.edu/~mu2166/closing. φ. and A0 . d−1 . For we picked values for the parameters σ .Lecture in Open Economy Macroeconomics.2 The Model’s Performance Table 4. For instance. are known functions of the deep structural parameters of the model to which we assigned values when we calibrated the model.columbia.htm. Formally.2 displays some unconditional second moments of interest implied by our model. it correctly implies a volatility . The linearized system has three known initial conditions k0 . Appendix B shows in some detail how to solve linear stochastic systems like the one describing the dynamics of our linearized equilibrium conditions. 4. the terminal condition takes the form j→∞ lim |Et xt+j | = 0. It should not come as a surprise that the model does very well at replicating the volatility of output.

2: Empirical and Theoretical Second Moments Variable σx t y c i h tb y ca y Canadian Data ρxt .61 0. This prediction is due to the assumed functional form for the period utility index.58 Mart´ Uribe ın Table 4.5 9.3 9. which equates the marginal product of labor to the marginal .61 0. ranking featuring investment volatility above output volatility and output volatility above consumption volatility.94 0. were set independently of the cyclical properties of the trade balance.5 2.7 0.15).3 ρxt . the model predicts too little countercyclicality in the trade balance and overestimates the correlations of both hours and consumption with output. On the downside.07 0.026 Note.1 1.8 -0.xt−1 0.012 0. which.66 1 -0.7 0.33 0. as we established in the previous chapter. Standard deviations are measured in percentage points.1 2.8 2.61 0.64 0.GDPt 1 0. Also in line with the data is the model’s prediction of a countercyclical trade balance-to-output ratio.13 σx t 3. equilibrium condition (4.66 ρxt .8 2 1. are the key determinants of the cyclicality of the trade-balance-to-output ratio.xt−1 0.5 1.1 2. Note in particular that the implied correlation between hours and output is exactly unity. In eﬀect.9 Model ρxt . This prediction is of interest because the parameters φ and ρ governing the degree of capital adjustment costs and the persistence of the productivity shock. Empirical moments are taken from Mendoza (1991).31 0.54 0.59 0.GDPt 1 0.

Interestingly. and hours.. The log-linearized version of this condition is ω ht = yt .5 0 0 2 4 6 8 10 0 2 4 6 8 10 Trade Balance / GDP 1 0. investment. This stark prediction is a consequence of the particular functional form assumed for the period utility function. consumption.5 1 0.5 1 Consumption 0 2 4 Hours 6 8 10 Investment 10 5 0 −5 1. displays the impulse response functions of a number of variables of interest to a technology shock of size 1 in period 0. can be written as hω = (1 − α)yt .1. the increase in domestic absorption (i.5 0 1.5 0 −0.5 0 0 2 4 6 8 10 0.5 −1 0 2 4 6 8 10 1 0.5 −1 0 2 Current Account / GDP 4 6 8 10 rate of substitution between consumption and leisure. In response to this innovation. the model predicts an expansion in output.e.5 0 −0. Chapter 4 59 Figure 4.1: Responses to a One-Percent Productivity Shock Output 2 1.5 1 0. t which implies that ht and yt are perfectly correlated.Lecture in Open Economy Macroeconomics. the increase in ct +it ) that takes place following the expansionary technology . Figure 4.

60

Mart´ Uribe ın

shock is larger than the increase in output, resulting in a deterioration of the trade balance.

4.3

The Roles of Persistence and Capital Adjustment Costs

In the previous chapter, we deduced that the negative response of the trade balance to a positive technology shock was not a general implication of the neoclassical model. In particular, Principles I and II of the previous chapter state that 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. First, capital adjustment costs must not be too stringent. Second, the productivity shock must be suﬃciently persistent. To illustrate this conclusion, ﬁgure 4.2 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 speciﬁcations. The solid line reproduces the benchmark case from ﬁgure 4.1. The broken line depicts an economy where the persistence of the productivity shock is half as large as in the benchmark economy (ρ = 0.21). In this case, because the productivity shock is expected to die out quickly, the response of investment is relatively weak. In addition, the temporariness of the shock induces households to save most of the increase in income to smooth consumption over time. As a result, the expansion in aggregate domestic absorption is modest. At the same time, because the size of the productivity shock is the same as in the benchmark economy, the initial responses of output and hours are identical

Lecture in Open Economy Macroeconomics, Chapter 4

61

**Figure 4.2: Response of the Trade-Balance-To-Output Ratio to a Positive Technology Shock
**

0.6 benchmark high φ low ρ 0.4

0.2

0

−0.2

−0.4

−0.6

−0.8

0

1

2

3

4

5

6

7

8

9

62

Mart´ Uribe ın

in both economies (recall that, by equation (4.15), ht depends only on kt and At , and that kt is predetermined in period t). The combination of a weak response in domestic absorption and an initial response in output that is independent of the value of ρ, results in an improvement in the trade balance when productivity shocks are not too persistent. The crossed line depicts the case of high capital adjustment costs. Here the parameter φ equals 0.084, a value three times as large as in the benchmark case. In this environment, high adjustment costs discourage ﬁrms from increasing investment spending by as much as in the benchmark economy. As a result, the response of aggregate domestic demand is weaker, leading to an improvement in the trade balance-to-output ratio.

4.4

Alternative Ways to Induce Stationarity

The small open economy RBC model analyzed thus far features households with endogenous subjective discount factors. We will refer to that model as the ‘internal discount factor model,’ or IDF model. As mentioned earlier in this chapter, the inclusion of an endogenous discount factor responds to the need to obtain stationary dynamics up to ﬁrst order. Had we assumed a constant discount factor, the log-linearized equilibrium dynamics would have contained a random walk component. Two problems emerge when the linear approximation possesses a unit root. First, 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. Second, when the variables of interest contain

Lecture in Open Economy Macroeconomics, Chapter 4

63

random walk elements, it is impossible to compute unconditional second moments, such as standard deviations, serial correlations, correlations with output, etc., which are the most common descriptive statistics of the business cycle. In this section, we analyze and compare alternative ways of inducing stationarity in small open economy models. Our analysis follows closely Schmitt-Groh´ and Uribe (2003), but expands their analysis by including e two additional approaches to inducing stationarity: a model with an internal interest-rate premium, and a model with perpetually-young consumers.

4.4.1

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. Speciﬁcally, suppose that the discount factor depends not upon the agent’s own consumption and eﬀort, but rather on the average per capita levels of these variables. Formally, preferences are described by (4.1), (4.2), and

θt+1 = β(˜t , ht )θt c ˜

t ≥ 0,

(4.16)

where ct and ˜ t denote average per capita consumption and hours, which ˜ h the individual household takes as given. The ﬁrst-order conditions associated with the household’s maximization

64

Mart´ Uribe ın

problem are (4.2), (4.4)-(4.7), (4.16) holding with equality, and:

λt = β(˜t , ht )(1 + rt )Et λt+1 c ˜

(4.17)

λt = Uc (ct , ht ) − Uh (ct , ht ) = λt At Fh (kt , ht )

(4.18) (4.19)

λt [1+Φ (kt+1 −kt )] = β(˜t , ht )Et λt+1 At+1 Fk (kt+1 , ht+1 ) + 1 − δ + Φ (kt+2 − kt+1 ) c ˜ (4.20) In equilibrium, individual and average per capita levels of consumption and eﬀort are identical. That is, ct = ct ˜ and ht = ˜ t . h (4.22) (4.21)

A competitive equilibrium is a set of processes {dt , ct , ht , ct , ht , yt , it , ˜ ˜ kt+1 , λt , rt , At } satisfying (4.4)-(4.7), (4.13), (4.14), (4.17)-(4.22) all holding with equality, given A0 , d−1 , and k0 and the stochastic process { t }. Note that the equilibrium conditions include one fewer Euler equation (equation (4.11)), and one fewer variable, ηt , than the internal discount-factor model of subsection 4.1. The smaller size of the external-discount-factor model slightly speeds up the computation of equilibrium dynamics using perturbation techniques. Remarkably, the reverse is true if the model is solved using value-function iterations over a discretized state space. The reason is that in the value-function formulation, the aggregate variables ct ˜

Lecture in Open Economy Macroeconomics, Chapter 4

65

˜ and ht add two state variables. And each of these state variables comes at a cost in terms of computational speed.1 We evaluate the model using the same functional forms and parameter values as in the IDF model.

4.4.2

External Debt-Elastic Interest Rate (EDEIR)

Under an external debt-elastic interest rate, stationarity is induced by assuming that the interest rate faced by domestic agents, rt , is increasing in ˜ the aggregate level of foreign debt, which we denote by dt . Speciﬁcally, rt is given by ˜ rt = r + p(dt ), (4.23)

where r denotes the world interest rate and p(·) is a country-speciﬁc interest rate premium. The function p(·) is assumed to be strictly increasing. Preferences are given by equation (4.1). Unlike in the IDF or EDF model,s however, preferences are assumed to display a constant subjective rate of discount. Formally, we assume that

θt = β t ,

where β ∈ (0, 1) is a constant parameter. The representative agent’s ﬁrst-order conditions are (4.4)-(4.7) holding with equality and λt = β(1 + rt )Et λt+1

1

(4.24)

A similar comment applies to the computation of equilibrium in a model with an interest-rate premium that depends on the aggregate per capita level of external al debt to be discussed in the next subsection.

˜ given A0 .1. and k0 .26) λt [1+Φ (kt+1 −kt )] = βEt λt+1 At+1 Fk (kt+1 . and the process { t }. We set the subjective discount factor equal to the world interest rate. rt .14). φ. Mart´ Uribe ın (4. and (4. ht+1 ) + 1 − δ + Φ (kt+2 − kt+1 ) . ht . ρ. λt .25) (4. dt . We calibrate the parameters γ. d−1 . it .66 Uc (ct . ω. equilibrium conditions (4. r. and σ using the values shown in table 4. (4. At } satisfying (4. ht ) = λt . ht ). ct .23)-(4. kt+1 . δ. that is. note that in steady state. this. 1+r To see ¯ ¯ The parameter d equals the steady-state level of foreign debt. We adopt the same forms for the functions U .23) and (4.28) all holding with equality.27) Because agents are assumed to be identical. in equilibrium aggregate per capita debt equals individual debt.1. − Uh (ct .24) . ht ) = λt At Fh (kt . (4.4)-(4. α. d−1 = d−1 . (4.28) ˜ A competitive equilibrium is a set of processes {dt . ¯ where ψ2 and d are constant parameters. F . and Φ as in the IDF model of subsection 4. We use the following functional form for the risk premium: p(d) = ψ2 ed−d − 1 . β= 1 . ˜ dt = dt .7). that is. yt .

(4. In all other aspects. 4. as before.1.96 ¯ d 0. It follows that in the steady state the interest rate premium is nil. Formally. This means that in deciding its optimal expenditure and savings plan. Note that the argument of the interest-rate premium function is the household’s own net debt position. and p(·) is a householdspeciﬁc interest-rate premium. Finally.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.4. The fact that β(1 + r) = 1 then implies that ¯ d = d.3.Lecture in Open Economy Macroeconomics. d. dt . the model is identical to the EDEIR model. the household .29) where. in the IDEIR model the interest rate is given by rt = r + p(dt ).3: Calibration of the Model with an External Debt Elastic Interest Rate β 0.1 generate the same volatility of the current-account-to-GDP ¯ ratio.000742 together with the assumed form of the interest-rate premium imply that 1 = β 1 + r + ψ2 ed−d − 1 ¯ . and ψ2 are given in Table 4. r denotes the world interest rate. The resulting values of β.7442 ψ2 0. Chapter 4 67 Table 4. ¯ We set d so that the steady-state level of foreign debt equals the one implied by the calibrated version of the IDF model of section 4. we set the parameter ψ2 to ensure that the model analyzed here and the IDF model of section 4.

30) imposes the following restriction on d. We set d = 0. (4. ct . which now takes the form λt = β[1 + r + p(dt ) + p (dt )dt ]Et λt+1 .14).7).30) This expression features the derivative of the premium with respect to debt because households internalize the fact that as their net debt increases. The implied steady-state level of debt is then given by ¯ d = 0. At } satisfying (4. ¯ (1 + d)ed−d = 1. yt . λt .27). The fact that the steady-state debt is lower than d implies .2. and the exogenous process { t }. kt+1 . recall that β(1 + r) = 1 and note that the steady-state version of equation (4. and (4.4045212. except in the special case in which ¯ ¯ d = 0. (4. ¯ which does not admit the solution d = d.4)-(4.7442.30). rt .2. it . (4. We note that in the model analyzed here ¯ the steady-state level of debt is no longer equal to d.4. so does the interest rate they face in ﬁnancial markets. (4. and k0 .4.29). given A0 . A competitive equilibrium is a set of processes {dt .25)-(4. ht .68 Mart´ Uribe ın will take into account the fact that a change in his debt position alters the marginal cost of funds. To see this. which is the value imposed in the EDEIR model of section 4. We assume the same functional forms and parameter values as in the EDEIR model of section 4. all holding with equality. d−1 . The only optimality condition that changes relative to the EDEIR model is the Euler equation for debt accumulation.

4. and adjust β to ensure that equation (4. However.27). That is.32) . An alternative calibration ¯ strategy is to impose d = d. equation (4.4. In this case. the country premium vanishes in the ¯ steady state. Chapter 4 69 that the country premium is negative in the steady state. but the marginal premium is positive and equal to ψ2 d. Preferences and technology are as in the EDEIR model of section 4. in contrast to what is assumed in the EDEIR model. the marginal country premium. The sequential budget constraint of the household is given by ψ3 ¯ (dt − d)2 . given by ∂[ρ(dt )dt ]/∂dt .4.13) is assumed to hold. as it is in the EDEIR economy of section 4.2. 4.31) ¯ where ψ3 and d are constant parameters deﬁning the portfolio adjustment cost function.30) holds in steady state. The ﬁrst-order conditions associated with the household’s maximization problem are (4.25)-(4. stationarity is induced by assuming that agents face convex costs of holding assets in quantities diﬀerent from some long-run level.7).31) holding with equality and ¯ λt [1 − ψ3 (dt − d)] = β(1 + rt )Et λt+1 (4. 2 dt = (1 + rt−1 )dt−1 − yt + ct + it + Φ(kt+1 − kt ) + (4.5)-(4. However. (4. is nil in the steady state.Lecture in Open Economy Macroeconomics.2.4 Portfolio Adjustment Costs (PAC) In this model. (4. here the interest rate at which domestic households can borrow from the rest of the world is assumed to be constant and equal to the world interest rate.

(4.2. ct .31). The parameters γ. ω. (4.4.4) and (4. d−1 . and k0 . Preferences and technology are parameterized as in the EDEIR model of section 4. λt .13). the marginal beneﬁt of a unit debt increase must equal its marginal cost. At } satisfying (4.25)-(4.70 Mart´ Uribe ın This optimality condition states that if the household chooses to borrow an additional unit. the Euler equations associated with the optimal choice of foreign .1 generate the same volatility in the current-account-to-GDP ratio. yt . As in the EDEIR model of section 4.32) imply that the parameter d determines the steady-state level ¯ ¯ of foreign debt (d = d). Finally. we assign the value 0.32) all holding with equality. This is because the log-linearized versions of models 2 and 3 are almost identical.00074 to ψ3 . α. which ensures that this model and the IDF model of section 4. and σ take the values displayed in table 4. (4. and EDEIR (see table 4.31)). rt . the household must repay the additional unit of debt plus interest. We calibrate d so that the steady-state level of foreign debt equals the one implied by models IDF. δ.2. The value of this repayment in terms of today’s utility is given by the right-hand side of the above optimality condition. A competitive equilibrium is a set of processes {dt . φ. and (4. At the optimum.14). the models share all equilibrium conditions but the resource constraint (compare equations (4. ht .7). given A0 .3). The value of this increase in consumption in terms of utility is given by the left-hand side of the above equation. This assumption and ¯ equation (4. and the exogenous process { t }. kt+1 .4. This parameter value is almost identical to that assigned to ψ2 in the EDEIR model.27). (4. Next period. it . EDF. the subjective discount factor is assumed to satisfy β(1 + r) = 1. ρ.5)-(4.1. then current consumption increases by one unit minus the ¯ marginal portfolio adjustment cost ψ3 (dt − d). Indeed. r.

4. The log-linear approximation to the domestic interest rate is given by 1 + rt = ψ2 d(1 + r)−1 dt in the EDEIR model and by 1 + rt = 0 in the PAC model. As we will see.t+1 .32)).t+1 bt+1 = bt + yt − ct − it − Φ(kt+1 − kt ). 4.5 Complete Asset Markets (CAM) All model economies considered thus far feature incomplete asset markets. Using these results. and the interest rate faced by domestic households (compare equations (4. (4. In the model studied in this subsection.2 Note that 2 To clarify the nature of the stochastic discount factor rt. the log-linearized versions of the resource constraints are the same in both models.33) where rt.2.13) and (4.23)). the log-linearized versions of the Euler equation for debt can be written as λt = ψ2 d(1 + r)−1 dt + Et λt+1 in the EDEIR model and as λt = ψ3 ddt + Et λt+1 in the PAC model. Chapter 4 71 bonds (compare equations (4.Lecture in Open Economy Macroeconomics.24) and (4. satisfying ψ2 = (1 + r)ψ3 .4. agents have access to a complete array of state-contingent claims. deﬁne the current state . It follows that for small values of ψ2 and ψ3 .t+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.t+1 bt+1 . In those models. agents have access to a single ﬁnancial asset that pays a non-state-contingent rate of return. The period-by-period budget constraint of the household is given by Et rt. the EDEIR and PAC models imply similar dynamics. However. this assumption per se induces stationarity in the equilibrium dynamics. Preferences and technology are as in the EDEIR model of section 4.

34) at all dates and under all contingencies.t+j bt+j is the periodt price of a stochastic payment bt+j in period t + j.5). Now |S ) let rt.t+1 ] denotes the risk-free real interest rate in period t. Let p(S t+1 |S t ) denote the price of a contingent claim that pays one unit of consumption in a particular state S t+1 following the current state S t . But this expression is simply the conditional expectation S t+1 |S t rt. (4. (4.t+j represents the stochastic discount factor such that Et rt.34) holding with equality and λt rt. we can write the price of the portfolio as t+1 |S t ).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. Households are assumed to be subject to a no-Ponzi-game constraint of the form j→∞ lim Et rt. let bt+1 ≡ b(S t+1 |S t ).t+1 bt+1 . given information available at the current state S t . Then the current price of a portfolio composed of b(S t+1 |S t ) units of contingent claims paying in states S t+1 following S t is given by S t+1 |S t p(S t+1 |S t )b(S t+1 |S t ). it follows that 1/[Et rt.t+1 is the price in period t of an asset that pays 1 unit of good in every state of period t + 1.t+1 bt+1 π(S Et rt. and (4. (4.6).27). (4. (4.33).t+1 = βλt+1 . The variable rt. Now let π(S t+1 |S t ) denote the probability of occurrence of state S t+1 . The ﬁrst-order conditions associated with the household’s maximization problem are (4. Also. Multiplying and dividing the expression inside the summation sign by π(S t+1 |S t ) p(S t+1 t we can write the price of the portfolio as S t+1 |S t π(S t+1 |S t ) π(S t+1|S t) b(S t+1 |S t ).72 Mart´ Uribe ın because Et rt. whereas under incomplete markets of nature as S t .t+1 rt+1.25)-(4.t+2 · · · rt+j−1.t+j bt+j ≥ 0.t+j ≡ rt. . Then.t+1 ≡ p(S t+1 |S t )/π(S t+1 |S t ) be the price of a contingent claim that pays in state S t+1 |S t scaled by the inverse of the probability of occurrence of the state in which the claim pays.

36)—yields λ∗ λt+1 = t+1 . This means that the domestic marginal utility of consumption is indeed proportional to its foreign counterpart.36) where λ∗ denotes the foreign household’s marginal utility of wealth. λ∗ rt. t Because we are interested only in the eﬀects of domestic productivity shocks. we assume that λ∗ is constant and equal to λ∗ . Formally. where λ∗ is a parat .t+1 = βλ∗ . one ﬁrst-order condition of the foreign household is an equation similar to (4.35). Combining the domestic and foreign Euler equations— equations (4. 73 Foreign households are assumed to have access to the same array of ﬁnancial assets as do domestic households. Chapter 4 the above Euler equation holds only in expectations.35) and (4. λt λ∗ t This expression holds at all dates and under all contingencies. Note t that we are assuming that domestic and foreign households share the same subjective discount factor. t t+1 (4.Lecture in Open Economy Macroeconomics. λt = ξλ∗ . This means that λ∗ is exogenously determined. We assume that the domestic economy is small. Consequently. t where ξ is a constant parameter determining diﬀerences in wealth across countries.

(4. and σ take the values displayed in tables 4. δ. At } satisfying (4. and k0 .27).6 The Nonstationary Case (NC) For comparison with the models considered thus far. the deterministic steady state of the model depends on the assumed initial level net foreign debt. (b) the interest rate at which domestic agents borrow from the rest of the world is constant (and equal to the subjective discount factor). λt .37) A competitive equilibrium is a set of processes {ct . φ.3. up to ﬁrst order. kt+1 . Under this speciﬁcation.4. where ψ4 ≡ ξλ∗ is a constant parameter. Also. β.25)-(4. and (d) markets are incomplete. The parameters γ.37). (4. The above equilibrium condition then becomes Mart´ Uribe ın λ t = ψ4 . given A0 . (4. the equilibrium dynamics contain a random wal component in variables such .1 and 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. in the sense that domestic households have only access to a single risk-free international bond. we consider a version of the small open economy model that displays no stationarity. F . it .74 meter.5). and Φ are parameterized as in the previous models. α. In this model (a) the discount factor is constant. The functions U . and PAC models.6). ω. (4. 4. and the exogenous process { t }. ρ. and (4. EDF. yt . (c) agents face no frictions in adjusting the size of their portfolios.14). ht . EDEIR.

EDF. given d−1 .4.7). EDF. the trade balance.4. the model’s predictions regarding second moments are virtually identical. (4. λt . .3. EDEIR.13). ct . and therefore does not have well deﬁned unconditional second moments. it . This result is evident from table 4.4)-(4. we compute the equilibrium dynamics by solving a log-linear approximation to the set of equilibrium conditions. PAC. and CAM models.1 and 4.4 displays a number of unconditional second moments of interest implied by the IDF. ht . 3 The NC model is nonstationary up to ﬁrst order. Chapter 4 as consumption. and (4. as might be expected. EDEIR. which. predicts less volatile consumption. (4.1.columbia.7 Quantitative Results Table 4. The main result of this section is that regardless of how stationarity is induced.14). The low volatility of consumption in the complete markets model introduces an additional diﬀerence between the predictions of this model and those of the IDF.27) all holding with equality. the complete markets case. The appendix shows the log-linear version of the IDF model of subsection 4. yt .Lecture in Open Economy Macroeconomics.3 For all models. and net external debt. The Matlab computer code used to compute the unconditional second moments and impulse response functions for all models presented in this section is available at www. and the exogenous process { t }. The only noticeable diﬀerence arises in the CAM model. and k0 . 75 A competitive equilibrium in the nonstationary model is a set of processes {dt . We calibrate the model using the parameter values displayed in tables 4. IDEIR. At } satisfying (4.edu/~mu2166. rt . 4. kt+1 .24)-(4.

leads to a smaller decline in the trade balance in the period in which the technology shock occurs.8 The Perpetual-Youth Model In this subsection. Each panel shows the impulse response of a particular variable in the six models.3 demonstrates that all of the models being compared imply virtually identical impulse response functions to a technology shock. Again. small-openeconomy version of the perpetual-youth model due to Blanchard (1985). consumption increases less when markets are complete than when markets are incomplete. its role in determining the cyclicality of the trade balance is smaller. This in turn. the impulse response functions are so similar that to the naked eye the graph appears to show just a single line. Figure 4. 4. whereas the models featuring incomplete asset markets all imply that it is negative. For all variables. We treat this case separately because exact aggregation in the presence of aggregate uncertainty requires the adoption of a particular functional form for the period utility index. It is a discrete-time. and PAC models: Because consumption is smoother in the CAM model. the only small but noticeable diﬀerence is given by the responses of consumption and the trade-balance-to-GDP ratio in the complete markets model.76 Mart´ Uribe ın IDEIR. In response to a positive technology shock. As . As a result.4. Cardia (1991) represents an early adoption of the perpetual-youth model in the context of a small open economy. I present an additional way to induce stationarity in the small open economy RBC model. the CAM model predicts that the correlation between output and the trade balance is positive.

66 1 -0.31 0.61 0. Standard deviations are measured in percent per year. NC = Nonstationary Case.043 0.041 EDEIR 3. yt ) y corr( catt . cat−1 ) y y EDF 3. EDEIR = External Debt-Elastic Interest Rate. CAM = Complete Asset Markets.5 1.3 9.012 0.78 0.7 0.61 0. yt−1 ) corr(ct .94 0.89 0.67 1 -0.1 2. EDF = External Discount Factor. IDEIR = Internal Debt-Elastic Interest Rate.5 0. yt ) y 1 0.62 0.1 1.013 0.1 2.7 9 2.6 1.051 CAM 3.62 0.07 0. .6 3. it−1 ) corr(ht .07 0.5 0.61 0.62 0.5 1.61 0.62 0. PAC = Portfolio Adjustment Costs.1 2. yt ) corr(it .07 0.61 0.036 0.1 1.1 2.05 PAC 3.9 9.1 1.7 0.044 0.068 0.5 0. tbt−1 ) y y t−1 corr( catt .1 1.76 0.3 0.8 1.78 0.069 0.5 0.025 IDEIR 3.66 1 0.94 0.1 2.32 0.62 0.Lecture in Open Economy Macroeconomics.67 1 -0.62 0.1 2.33 0.51 0.026 Serial Correlations: 0. IDF = Internal Discount Factor. yt ) corr(ht . yt ) corr( tbtt .8 1.1 1.13 Note.3 0.32 0.5 0.61 0.66 1 -0.3 9.7 9 2. Chapter 4 77 Table 4.39 Correlations with Output: corr(ct .4 0.1 2.1 2.32 0.61 0.1 1.68 1 -0.5 9 2.1 1.43 0.069 0.84 0. ht−1 ) t−1 corr( tbtt . ct−1 ) corr(it .85 0.4: Implied Second Moments IDF Volatilities: std(yt ) std(ct ) std(it ) std(ht ) std( tbtt ) y std( catt ) y corr(yt .

5 1 0.5 0 10 5 0 −5 1 0.5 1 0.5 0 1. Solid line: Endogenous discount factor model. Squares: Endogenous discount factor model without internalization.78 Mart´ Uribe ın Figure 4.5 −1 0 2 4 6 8 10 1. Circles: Model without stationarity inducing elements.5 −1 0 2 4 6 8 10 0 2 4 6 Investment 8 10 0. Dash-dotted line: Portfolio adjustment cost model. .5 0 −0.5 0 1 0.5 0 −0.3: Impulse Response to a Unit Technology Shock in Models 1 .5 Output 2 1.5 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. Dashed line: Debt-elastic interest rate model. Dotted line: complete asset markets model.

applying the operator (1 − θ) −∞ t−s s=t θ on both sides of the expression ds. Speciﬁcally. to keep the size of the population constant over time.t−1 + µt . Chapter 4 79 we will see. Here. where dt denotes the aggregate debt position in period t. and µt is a white noise common to all agents.t−1 + µt —yields dt = θdt−1 + µt . In performing the aggregation. Assume that those agents who die leave their outstanding debts unpaid and that newborns inherit no debts.t = ds. We now depart from the representative-agent assumption by introducing a constant and ageindependent probability of death at the individual level. the aggregate dynamics of the model are very much in line with those obtained from the other models studied earlier in this section. 1) of dying. recall that dt. however. Adding the left.t−1 = 0.t = ds. ds. we assume that 1 − θ agents are born each period. representative-agent economy of chapter 2.and right-hand sides of the law of motion for debt over all agents alive in period t—i. individual agents face a probability 1 − θ ∈ (0. In addition. This was exactly the equilibrium evolution of debt we obtained in the quadraticpreference. because agents are born free of debts.t denotes the net debt position in period t of an agent born in period s. The Basic Intuition The basic intuition behind why the assumption of ﬁnite lives by itself helps to eliminate the unit root in the aggregate net foreign asset position can be seen from the following simple example.Lecture in Open Economy Macroeconomics.e. . assume that the population is constant over time and normalized to unity. Each period.. Consider an economy in which debt holdings of individual agents follow a pure random walk of the form ds.

we embed this basic stationarity result in the smallopen-economy real-business-cycle model. 1) represents a subjective discount factor. Following ¯ the preference speciﬁcation used in all of the models studied in this chapter. ω (4. the risk-free inter- . As will become clear shortly.t .t denote consumption and hours worked in period t by an agent born in period s. But we depart from the preference speciﬁcations used earlier in this chapter by assuming a quadratic period utility index. the resulting law of motion for the aggregate level of debt is mean reverting at the survival rate θ.t − x)2 ¯ t=0 where cs. The key diﬀerence with the representative agent model is that here each period a fraction 1 − θ of the stock of debt simply disappears. Agents can borrow from foreign lenders by means of a bond paying a constant real interest rate.t − hω s.t and hs.80 Mart´ Uribe ın Clearly. we assume that agents derive utility from a quasi-diﬀerence between consumption and leisure. The parameter β ∈ (0.t = cs. In what follows. Households Each agent maximizes the utility function 1 − E0 2 with xs. this assumption is essential to achieve aggregation in the presence of aggregate uncertainty.38) ∞ (βθ)t (xs. and x is a parameter denoting a satiation point. in the world ﬁnancial market.

t − πt − wt hs. they have access to an actuarily fair insurance market.40) The ﬁrst-order conditions associated with the agent’s maximization prob- .t+j ≤ 0 (4. Foreign agents are assumed to lend to a large number of domestic agents so that the fraction of unpaid loans due to death is deterministic.t−1 + cs. It follows that the gross interest rate on the domestic agent’s asset position (whether this position is positive or negative) is given by (1 + r)/θ. we assume that agents do not trade shares and that the shares of the dead are passed to the newborn in an egalitarian fashion. To facilitate aggregation. Domestic agents can also lend at the rate r. Chapter 4 81 est rate is given by r. proﬁts received from the ownership of stock shares and the real wage rate.Lecture in Open Economy Macroeconomics. To compensate foreign lenders for these losses.t . Agents are assumed to be subject to the following no-Madoﬀgame constraint j→∞ lim Et θ 1+r j ds. the insurance contract in this market speciﬁes that lenders receive a constant premium over the interest rate r while alive and must leave their assets to the insurance company in case of death. The debts of deceased domestic agents are assumed to go unpaid. Thus. In addition.39) where πt and wt denote. (4. domestic agents therefore pay a constant premium over the world interest rate. share holdings are identical across agents. The agent’s sequential budget constraint can be written as 1+r θ ds. respectively.t = ds.

t − x) To facilitate notation.t − x) = λs.39).82 lem are (4. ¯ hω−1 = wt .44) Use equations (4.38) and (5. (4. t (4. we introduce the following auxiliary variable: 1 ω yt ≡ πt + 1 − ¯ wt ht − x. and Mart´ Uribe ın − (xs.t and write hω−1 = wt . s.t and λs.43) Note that hs.39).t+1 (4.45) which is the same for all generations s because both proﬁts and hours worked are independent of the age of the cohort. This yields. it is independent of the agent’s birth date). (4. Then the sequential budget con- .40) holding with equality.t = ds.e.t from the sequential budget constraint (4.t is independent of s (i.λs.t .42) (4.t = β(1 + r)Et .9) to eliminate cs.41) (4..38).t + x + (xs.t−1 − πt − 1 − ¯ ¯ wt hs. 1+r θ 1 ω ds. This means that we can drop the subscript s from hs. (4.

t−1 = Et j=0 θ 1+r j [yt+j − (xs.t − x) β(1 + r)2 − θ Solve for xs.t = ¯ ds.t−1 = Et j=0 θ 1+r j yt+j − β(1 + r)2 ¯ (xs. to obtain 1+r θ ∞ ds.43) to replace Et xs.t − x) Now iterate this expression forward and use the transversality condition (i.t+j yields 1+r θ ∞ ds.e.t+j − x)] ¯ Using equations (4. equation (4.t−1 − yt + (xs.t−1 ). y βθ(1 + r) xs.t = x + ¯ (4.47) We now aggregate individual variables by summing over generations born ..Lecture in Open Economy Macroeconomics.t to obtain β(1 + r)2 − θ (˜t − ds.40) holding with equality). and can be expressed recursively as θ θ ˜ yt + Et yt+1 1+r 1+r yt = ˜ (4. Speciﬁcally.46) where yt denotes a weighted average of current and future expected values ˜ of yt . yt is deﬁned as ˜ θ yt ≡ ˜ Et 1+r ∞ j=0 θ 1+r j yt+j .41) and (4. Chapter 4 straint becomes: 1+r θ 83 ds.

keep in mind that dt.49) We now turn to the economic units producing consumption and capital goods. y βθ(1 + r) xt = x + ¯ (4. respectively. agents inherit no debts at birth. Now multiply (4.46) by (1 − θ)θ t−s and then sum for s = t to s = −∞ to obtain the following expression for the aggregate version of equation (4. and.t−1 = 0.84 Mart´ Uribe ın at time s ≤ t.t and −∞ dt ≡ (1 − θ) s=t θ t−s ds.39) to obtain − (xt − x) = λt ¯ and dt = (1 + r)dt−1 − yt + xt − x ¯ (4. (1 − θ)θ people born in t − 1. That is. Notice that at time t there are alive 1 − θ people born in t. . (1 − θ)θ s people born in period t − s. in general.41) and the budget constraint(4. Let −∞ xt ≡ (1 − θ) s=t θ t−s xs.48) In performing this step.t denote the aggregate levels of xs.t .46): β(1 + r)2 − θ (˜t − θdt−1 ).t and bs. Finally.50) (4. aggregate the ﬁrst-order condition (4.

πt .51) and At Fh (kt . and pay dividends.54) . The ﬁrst-order conditions associated with the ﬁrm’s proﬁt-maximization problem are At Fk (kt . Formally. which are given by: At F (kt . Chapter 4 Firms Producing Consumption Goods 85 We assume the existence of competitive ﬁrms that hire capital and labor services to produce consumption goods.Lecture in Open Economy Macroeconomics. rent out capital. (4. these ﬁrms maximize proﬁts. which we reproduce here for convenience kt+1 = (1 − δ)kt + it (4. where the productivity factor At and the production function F are as described earlier in this section.52) Firms Producing Capital Goods We assume that there exist ﬁrms that buy consumption goods to transform them into investment goods. dividends in period t are given by πt = ut kt − it − Φ(kt+1 − kt ) (4. ht ) − wt ht − ut kt . ht ) = wt .53) The evolution of capital follows the law of motion given earlier in this chapter. ht ) = ut (4.

at any given time.’ For the representative household does not exist. For there is no market for such claims. we will assume that the ﬁrm uses the discount factor β j λt+j /λt to discount quantities of goods delivered in a particular state of period t + j into period t. respectively. the owners of the ﬁrm change over time. current investment decisions of the ﬁrm have consequences not only for current proﬁts but also for future proﬁts.That is. Note also that we use as the subjective discount factor . Note that this discount factor uses the average marginal utility of consumption of agents alive in period t and t+j.86 Mart´ Uribe ın Because investment goods take one period to become productive capital and because of the presence of adjustment costs. given by λt and λt+j . It follows that the optimization problem of the capitalproducing ﬁrm is inherently dynamic. it cannot use the price of state-contingent claims to discount future proﬁts. With this in mind. the ﬁrm operates in a ﬁnancial environment characterized by incomplete asset markets. For this reason. This issue does not have a clear answer for two reasons: ﬁrst. Recall that the shares of the dead are distributed in equal parts among the newborn. The ﬁrm must maximize some present discounted value of current and future expected proﬁts. the ﬁrm converts goods into utils by multiplying the amount of goods by the average marginal utility of consumption of the shareholders alive at that time. A problem that emerges at this point is what discount factor should the ﬁrm use. One must therefore introduce assumptions regarding the ﬁrm’s discounting behavior. Second. These assumptions will in general not be inocuous with respect to the dynamics of capital accumulation. It follows that the ﬁrm cannot use as its discount factor the intertemporal marginal rate of substitution of the ‘representative household.

44). (4. yt . ¯ . t 1 ω yt ≡ πt + 1 − yt = ˜ xt = x + ¯ ¯ wt ht − x.47)-(4.55) Equilibrium Equations (4.55) form a system of eleven equations in eleven unknowns: xt .Lecture in Open Economy Macroeconomics.45). yt . wt . πt . y βθ(1 + r) −(xt − x) = λt . λt . θ θ ˜ yt + Et yt+1 . ht . (4. dt . ut . This is because the number of shareholders is constant over time (and equal to unity). unlike the size of a cohort born at a particular date. we reproduce ˜ the system of equilibrium conditions for convenience: hω−1 = wt . it . Chapter 4 87 the parameter β and not βθ. Below. kt . The Lagrangean associated with the capital-producing optimization problem is then given by ∞ L = Et j=0 βj λt+j [ut+j kt+j − kt+j+1 + (1 − δ)kt+j − Φ(kt+j+1 − kt+j )] λt The ﬁrst-order condition with respect to kt+1 is λt [1 + Φ (kt+1 − kt )] = βEt λt+1 ut+1 + 1 − δ + Φ (kt+2 − kt+1 ) (4. which declines at the mortality rate 1 − θ. 1+r 1+r β(1 + r)2 − θ (˜t − θdt−1 ).

although in this case individual levels of debt display a trend in the deterministic equilibrium.) The forcing term yt − (1 + r − θ)/θ˜t represents the diﬀerence y between current and permanent income. 1) of the agents die and are replaced by newborns holding no ﬁnancial assets. kt+1 = (1 − δ)kt + it . (This intuition also goes through when β(1 + r) = 1. on average. As a result. the current aggregate level of debt is only a fraction θ of the previous period’s level of debt.88 dt = (1 + r)dt−1 − yt + xt − x. ¯ At Fk (kt . the evolution of per capita external debt is given by dt = θdt−1 + (1 + r − θ)/θ˜t − yt . Mart´ Uribe ın λt [1 + Φ (kt+1 − kt )] = βEt λt+1 ut+1 + 1 − δ + Φ (kt+2 − kt+1 ) . πt = ut kt − it − Φ(kt+1 − kt ). The reason why the aggregate level of external debt is trend reverting in equilibrium is the fact that each period a fraction 1 − θ ∈ (0. ht ) = ut . This expression shows that the stock of debt per capital y does not follows a random walk as was the case in the representative-agent economy with constant subjective discount factor and world interest rate. 1). ht ) = wt . It is of interest to consider the special case in which β(1 + r) = 1. The mean reverting property of aggregate external debt obtains in spite of the fact that individual debt positions follow a random walk. At Fh (kt . the current account . the (autoregressive) coeﬃcient on past external debt is θ ∈ (0. In equilibrium. In eﬀect. In this case.

as in the calibration of the models discussed earlier in this section. 2 ρ = 0. θ = 0. in the sense that β(1 + r) < 1. the model generates the same ranking of volatilities and quite . Chapter 4 absorbs these income deviations.5 displays unconditional second moments implied by the model and ﬁgure 4.1.04. x = 0. That is. The model dynamics are quite similar to those obtained under the alternative ways of inducing stationarity discussed in this section. φ = 0. and ¯ β = 0.51. Note that the calibration makes agents relatively impatient. the aggregate level of debt is given by 1 r − 1 d= β(1+r)2 −θ rβ(1+r)(1+r−θ) 2 − β(1+r) −θ rβ(1+r) y In the special case in which β(1 + r) equals unity. wish to hold constant debt levels over time. In particular. the steady-state aggregate stock of debt is nil. 89 In the deterministic steady state.42.028. and a positive steady-state trade-balance-to-output ratio of about 2 percent.9933. Implied Business Cycles As in the previous subsections. h) = k α h1−α and Φ(x) = φ x2 . ω = 1. in the steady state both the aggregate and the individual levels of debt are zero in this case.9596. Table 4. we use the functional forms F (k. The reason for imposing tis parameterization is that it allows for a positive aggregate level of external debt in the steady state. This is because in this case agents. all of whom are born with no debts.32. r = 0.Lecture in Open Economy Macroeconomics. And we calibrate the model as follows: δ = 0.4550. α = 0.4 depicts impulse responses to a one-percent increase in total factor productivity.

5 0 0 2 4 6 8 10 0 2 4 6 8 10 Trade Balance / GDP 1 0.4: Perpetual Youth Model: Impulse Response to a One-Percent Productivity Shock Output 2 1.099 0. Standard deviations are measured in percentage points.9 2.065 0.5 0 0 2 4 6 8 10 0.2 1.63 0.5 1 0.8 8.69 1 -0.GDPt 1 0.63 0.90 Mart´ Uribe ın Table 4.5 0 −0.xt−1 0.5 −1 0 2 4 6 8 10 1 0.46 0.5 0 −0.059 Note.5 1 0.4 ρxt . Figure 4.8 0.6 1.5 −1 0 2 Current Account / GDP 4 6 8 10 .5: Perpetual Youth Model: Implied Second Moments Variable Output Consumption Investment Hours Trade-Balance-To-Output Ratio Current-Account-To-Output Ratio σx t 3.5 1 Consumption 0 2 4 Hours 6 8 10 Investment 10 5 0 −5 1.2 2.3 ρxt .89 0.5 0 1.

consumption. in response to a positive innovation in total factor productivity. Speciﬁcally. In section 2. and hours and a deterioration in the trade balance and the current account. In addition. Aguiar and Gopinath (2007) pursue this strategy and argue that the SOE-RBC model is an adequate framework for understanding aggregate ﬂuctuations in developing countrires provided it is augmented to allow for nonstationary productivity shocks.2). we suggested that a possible solution to this problem is to introduce nonstationary income shocks. investment. Contrary to this prediction. the perpetual-youth model implies expansions in output. they consider a small open economy populated by a large number of identical households seeking to maximize the utility function ∞ max E0 t=0 βt [Ctγ (1 − ht )1−γ ]1−σ − 1 1−σ . emerging countries are characterized by consumption being more volatile than output (see table1.Lecture in Open Economy Macroeconomics. Chapter 4 91 similar patterns of autocorrelations and correlations with output.5 Emerging-Country Business Cycles Through the Lens of the SOE RBC Model Can the SOE RBC model explain business cycles in emerging economies? One problem with the SOE-RBC model studied thus far is that it predicts consumption to be smoother than output (see table 4.3.2). as in the other models discussed in this section. 4.

as in section 4. is assumed to be debt elastic: ˜ ¯ Rt = R∗ + ψ eDt+1 /Xt −d − 1 . The optimality conditions associated with the household’s problem are 1 − γ Ct = (1 − α)At Xt γ 1 − ht Ct γ(1−σ)−1 Kt X t ht α . and to some no-Ponzi-game constraint of the type presented earlier in this chapter. ˜ where Dt denotes the cross-sectional average level of external debt per capita in period t. which. .2. Kt denotes the stock of physical capital. because all households are identical.4. Λt = βRt Et Λt+1 . and Rt denotes the gross interest rate charged by the rest of the world. Dt denotes net external debt. In the above expressions.92 subject to Dt+1 φ = Dt + Ct + Kt+1 − (1 − δ)Kt + Rt 2 Kt+1 −g Kt 2 Mart´ Uribe ın Kt − At Ktα (Xt ht )1−α . This economy is driven by a stationary productivity shock At and a nonstationary productivity shock Xt . (1 − ht )(1−γ)(1−σ) = Λt . Ct denotes consumption. In equilibrium. ht denotes hours worked. we have that ˜ D t = Dt .

The main diﬀerence between the present model and the one studied earlier in this chapter is the introduction of the nonstationary productivity shock Xt . This last characteristic is reﬂected . Assume that Xt and At are mutually independent random variables with laws of motion given by ln At = ρa ln At−1 + σa a t and ln(gt /g) = ρa ln(gt−1 /g) + σg g . where Λt denotes the Lagrange multiplier associated with the sequential budget constraint of the household. The variables and g t are assumed to be exogenous. Note that the productivity factor Xt is nonstationary in the sense that it displays both secular growth. mutually independent white noises distributed N (0. t where gt ≡ Xt Xt−1 a t denotes the gross growth rate of Xt .Lecture in Open Economy Macroeconomics. 1). Chapter 4 and Kt+1 −g Kt Kt+1 Xt+1 ht+1 φ 2 α−1 93 Λt 1 + φ = βEt Λt+1 1 − δ + αAt+1 +φ Kt+2 Kt+1 Kt+2 −g Kt+1 − Kt+2 −g Kt+1 2 . and a random walk cmponent. The parameter g > 1 denotes the gross growth rate of productivity in a nonstochastic equilibrium path. at an average rate g.

and the stock of external debt. let ct ≡ Ct /Xt−1 . including consumption. dt ≡ Dt /Xt−1 . Fortunately. which in this case coincides with the Solow residual. Then we have that under the present technology speciﬁcation SRt = At Xt1−α . the Solow residual inherits the nonstationarity of Xt . however. Speciﬁcally. Let SRt denote total factor productivity. because At is a stationary random variable independent of Xt . And this properties will be transmitted in equilibrium to other variables of the model. Rt = R∗ + ψ edt+1 −d − 1 . and λt ≡ Xt−1 1+(σ−1)γ Λt . Then. the marginal utility of wealth. the capital stock. kt ≡ Kt /Xt−1 . Because non of thise variables exhibits a deterministic steady sate. investment. 1 − γ ct = (1 − α)At gt γ 1 − ht ct γ(1−σ)−1 kt gt ht α (1 − ht )(1−γ)(1−σ) = λt . one can write the system of equilibrium conditions in stationary form as gt dt+1 φ = dt + ct + gt kt+1 − (1 − δ)kt + Rt 2 gt kt+1 −g kt ¯ 2 α kt − At kt (ht )1−α . Clearly. there exists a simple stationary transformation of the variables of the model whose equilibirum behavior is described by a system of equations very similar to the one that governs the joint determination of the original variables.94 Mart´ Uribe ın in the fact that an innovation in gt has a permenent eﬀect on the level of Xt . it is impossible to linearize the model around such point.

the capital stock. Second. as shown in table 4. investment. ρa . in equilibrium. Because the number of estimated parameters (six) is smaller than the number of mo- . mean reverting. In other words. the shares of consumption. and net external debt all share the same stochastic trend Xt . it follows that in this model consumption.6. Recalling that the variable transformations involve dividing by the productivity factor Xt .Lecture in Open Economy Macroeconomics.8 and the observed average growth rate of GDP. possesses two properties. The econometric estimation consists in picking values for σa . and external debt in GDP are all stationary variables. up to ﬁrst order. and g to match the empirical second moments displayed in table 4. the rational expectations dynamics are. This property of the model is known as the balancegrowth property. All other parameters of the model are calibrate using standard values in business-cycle analysis. σg . investment. together with the laws of motion of gt and At . Chapter 4 λt = βRt gt γ(1−σ)−1 95 Et λt+1 λt 1 + φ gt kt+1 −g kt = βgt γ(1−σ)−1 Et λt+1 1 − δ + αAt+1 gt+1 kt+2 −g kt+1 − φ 2 kt+1 gt+1 ht+1 α−1 +φ gt+1 kt+2 kt+1 gt+1 kt+2 −g kt+1 2 This system. Aguiar and Gopinath (2007) econometrically estimate the parameters deﬁning the laws of motion of the two productivity shocks as well as the parameter governing the strength of capital adjustment costs. φ. it has a deterministic steady state that is independent of initial conditions. ρg . The data are from Mexico and the sample is 1980:Q1 to 2003:Q1. output. capital. First.

05 Table 4. consider the growth rate of the Solow residual.10 ψ 0. To this end.0066 φ 1.95 g 1. g .7: Estimated Parameters σg 0.0213 σa 0. Because the two terms on the right-hand side of this expression are mutually independent.96 Mart´ Uribe ın Table 4.37 ments matched (eleven).98 γ 0. the estimation procedure uses a weighting matrix following the GMM technique. Before studying how the model ﬁts the data.6: Calibrated Parameters β 0.36 b 0. we can ask what fraction of the variance of ∆ ln SRt is explaind by gt . it is of interest to calculate the importance of the nonstationary component of productivity in driving movements in the Solow residual implied by the econometrically estimated parameters. the variance of gt is given by σg /(1 − ρ2 ). It is straightforward to deduce that the variance of ∆ ln At is given by 2 2 2σa /(1 + ρa ).68 σ 2 δ 0.00 ρa 0.0053 ρg 0. which can be written as ∆ ln SRt = ∆ ln At + (1 − α)gt .001 α 0. At the same time.

02132 /(1 − 0.02132 /(1 − 0.002 ) 2 × 0. How should we interpret these results? There are three aspects of the .32)2 × 0. we have that var((1 − α)gt ) var(∆ ln SRt ) 2 (1 − α)2 σg /(1 − ρ2 ) g 2 2 2σa /(1 + ρa ) + (1 − α)2 σg /(1 − ρ2 ) g 97 = = (4.56) (1 − 0.8793. This is an indication that the data favor the hypothesis that nonstationary technology shocks play a signiﬁcant role in moving total factor productivity at business-cycle frequency in emerging economies.95) + (1 − 0. As explained in detail in section 2. the estimated parameters imply that the nonstationary component of productivity explains 88 percent of the variance of the growth rate of the Solow residual. as in most other emerging countries. That is. Table 4. These ﬁndings suggest that nonstationary productivity shocks to be more relevant in emerging economies than in developed ones.8 displays ten empirical and theoretical second moments.32)2 × 0.Lecture in Open Economy Macroeconomics.002 ) = 0. the presence of nonstationary productivity shocks plays a key role in making this prediction possible.3. The estimated model does a very good job at matching all of the moments shown in the table. Of particular relevance is the ability of the model to match the fact that in Mexico. Aguiar and Gopinath (2007) also perform this exercise using Canadian data and ﬁnd that the nonstationary component explains only 40 percent of movements in total factor productivity. consumption is more volatile than output.00532 /(1 + 0. Chapter 4 Therefore.

80 Note.82 0. econometric estimation that deserve special comments. The second diﬃculty with the econometric strategy pursued in this section is that it allows room only for productivity shocks. But . 1980:Q1 to 2003:Q1. This would not be a big problem if no other candidate shocks had been identiﬁed as potentially important in driving business cycles in emerging economies.52 1.13 1. Growth rates are unﬁltered.82 0.50 0. is relatively short.18 -0. Variables in levels were HP ﬁltered using a parameter of 1.600. One is that the data sample.8: Model Fit Statistic σ(y) σ(∆y) σ(c)/σ(y) σ(i)/σ(y) σ(nx)/σ(y) ρ(y) ρ(∆y) ρ(y. Short samples can lead to misleading results.10 3.42 1.26 4.83 0. and conclude that it is not possible to determine reliably whether the nonstationary component is more important in Mexico or in Canada (see their ﬁgure 2).91 Model 2.40 1.27 -0. i) Data 2. It is well known that the only reliable way to disentangle the stationary and nonstationary components of a time series is to use long samples.91 0.15 0. c) ρ(y. or unitroot.80 0.83 0. component in total factor productivity. nx) ρ(y.95 0. In fact. Aguiar and Gopinath (2007) analyze direct evidence on Solow residuals for Mexico and Canada over the period 1980-2000. Source: Aguiar and Gopinath (2007).75 0. Recall that the main purpose of the estimation procedure is to identify the random-walk.98 Mart´ Uribe ın Table 4.

Lecture in Open Economy Macroeconomics, Chapter 4

99

this is not the case. For example, a growing number of studies show that world interest-rate shocks and country-spread shocks play an important role in driving business cycles in emerging countries. Omitting these and other relevant shocks in the econometric estimation may induce a bias in favor of the shocks that are included. Finally, the present model limits attention to a frictionless neoclassical framework. This might also be an oversimpliﬁcation. For instance, a large body of worked points at ﬁnancial frictions, including default risk, balance-sheet eﬀects, etc., as important ampliﬁers of business cycles in emerging economies. Omitting these sources of friction might cause a spurious increase in the estimated variance and persistence of the exogenous driving processes. In the next chapter, we attempt to address these concerns by ﬁrst studying the importance of interest-rate and country-spread shocks as drivers of business cycles in developing countries. Then, we estimate a business cycle model with ﬁnancial frictions in which stationary and nonstationary productivity shocks compete with interest-rate and country-spread shocks in explaining business cycles. As an attempt to obtain a reliable measure of the nonstationary component of productivity, we will estimate the model using long data samples.

100

Mart´ Uribe ın

4.6

Appendix A: Log-Linearization

**The log-linear version of the system (4.4)-(4.6) and (4.8)-(4.14) is given by 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 [ηt +

βc βc ct βcc ct

+

ch ht ]

−

β (1 − β)

βc c

−β

+

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 , where cUc /U ,
**

βc cc

≡ cβc /β,

βh

≡ hβh /β,

ch

βcc

≡ cβcc /βc ,

βch

≡ hβch /βc ,

c

≡

= cUcc /Uc ,

= hUch /Uc , stb ≡ tb/y, sc ≡ c/y, si = i/y.

In the log-linearization we are using the particular forms assumed for the production function and the capital adjustment cost function.

Lecture in Open Economy Macroeconomics, Chapter 4

101

4.7

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. Reduced forms of this sort are common in Macroeconomics. A problem that one must face is that, in general, it is impossible to solve such systems. But fortunately one can obtain good approximations to the true solution in relatively easy ways. In the previous chapter, we introduced one particular strategy, consisting in linearizing the equilibrium conditions around the nonstochastic steady state. Here we explain in detail how to solve the resulting system of linear stochastic diﬀerence equations. In addition, we show how to use the solution to compute second moments and impulse response functions. 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 , yt , xt+1 , xt ) = 0,

(4.57)

where Et denotes the mathematical expectations operator conditional on information available at time t. The vector xt denotes predetermined (or state) variables and the vector yt denotes nonpredetermined (or control) variables. The initial value of the state vector x0 is an initial condition for the economy. (Beyond the initial condition, the complete set of equi-

102

Mart´ Uribe ın

librium conditions also includes a terminal condition, like a no-Ponzi game constraint. We omit such a constraint here because we focus on approximating stationary solutions.) The state vector xt can be partitioned as xt = [x1 ; x2 ] . The vector x1 consists of endogenous predetermined state t t t variables and the vector x2 of exogenous state variables. Speciﬁcally, we t assume that x2 follows the exogenous stochastic process given by t ˜ x2 = h(x2 , σ) + η σ ˜ t+1 t where both the vector x2 and the innovation t vector

t

t+1 ,

t

are of order n × 1.4 The

is assumed to have a bounded support and to be independently

and identically distributed, with mean zero and variance/covariance matrix ˜ I. 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. The solution to models belonging to the class given in equation (4.57) is of the form: yt = g (xt ) ˆ and ˆ xt+1 = h(xt ) + ησ

t+1 .

(4.58)

(4.59)

The vector xt of predetermined variables is of size nx × 1 and the vector yt of nonpredetermined variables is of size ny × 1. We deﬁne n = nx + ny . The function f then maps Rny × Rny × Rnx × Rnx into Rn .

4 It is straightforward to accommodate the case in which the size of the innovations vector t is diﬀerent from that of x2 . t

Lecture in Open Economy Macroeconomics, Chapter 4 The matrix η is of order nx × n and is given by ∅ η = . η ˜

103

ˆ The shape of the functions h and g will in general depend on the amount ˆ of uncertainty in the economy. 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, that is,

yt = g(xt , σ) and xt+1 = h(xt , σ) + ησ

t+1 ,

(4.60)

(4.61)

where the function g maps Rnx × R+ into Rny and the function h maps Rnx × R+ into Rnx .

Given this interpretation, a perturbation methods ﬁnds a local approximation of the functions g and h. By a local approximation, we mean an approximation that is valid in the neighborhood of a particular point (¯, σ ). x ¯

Taking a Taylor series approximation of the functions g and h around the point (x, σ) = (¯, σ ) we have (for the moment to keep the notation simple, x ¯

104 let’s assume that nx=ny=1)

Mart´ Uribe ın

g(x, σ) = g(¯, σ ) + gx (¯, σ )(x − x) + gσ (¯, σ )(σ − σ ) x ¯ x ¯ ¯ x ¯ ¯ 1 + gxx (¯, σ )(x − x)2 + gxσ (¯, σ )(x − x)(σ − σ ) x ¯ ¯ x ¯ ¯ ¯ 2 1 + gσσ (¯, σ )(σ − σ )2 + . . . x ¯ ¯ 2 h(x, σ) = h(¯, σ ) + hx (¯, σ )(x − x) + hσ (¯, σ )(σ − σ ) x ¯ x ¯ ¯ x ¯ ¯ 1 + hxx (¯, σ )(x − x)2 x ¯ ¯ 2 +hxσ (¯, σ )(x − x)(σ − σ ) x ¯ ¯ ¯ 1 + hσσ (¯, σ )(σ − σ )2 + . . . , 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 (¯, σ ). x ¯ To identify these derivatives, substitute the proposed solution given by equations (4.60) and (4.61) into equation (4.57), and deﬁne

F (x, σ) ≡ Et f (g(h(x, σ) + ησ , σ), g(x, σ), h(x, σ) + ησ , x) (4.62) = 0.

Here we are dropping time subscripts. We use a prime to indicate variables dated in period t + 1. Because F (x, σ) must be equal to zero for any possible values of x and σ, it must be the case that the derivatives of any order of F must also be equal to zero. Formally,

Fxk σj (x, σ) = 0

∀x, σ, j, k,

(4.63)

x. y . note that if σ = 0. σ) = (¯. σ) denotes the derivative of F with respect to x taken k times and with respect to σ taken j times. 0). 0) and x = h(¯. one can then ﬁnd the derivatives of the function F . x) = 0. y ) such x ¯ that f (¯. g(¯. 0) + gx (¯. 0) + hx (¯. σ) = g(¯. xt = x ¯ and σ = 0. With the steady state values in hand. 0) = y x ¯ and h(¯. 0)(x − x) + hσ (¯. As will become clear below. Chapter 4 105 where Fxk σj (x. x ¯ The remaining unknown coeﬃcients of the ﬁrst-order approximation to g . 0)(x − x) + gσ (¯. We deﬁne the non-stochastic steady state as vectors (¯. σ) = h(¯. To see this. 0)σ x x ¯ x As explained earlier.Lecture in Open Economy Macroeconomics. a particularly convenient point to approximate the functions g and h around is the non-stochastic steady state. y ¯ ¯ ¯ It is clear that y = g(¯. 0) of the form x g(x. We are looking for approximations to g and h around the point (x. 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. ¯ x ¯ x then Et f = f . 0)σ x x ¯ x h(x. 0) = x.

Taking derivative with respect to the scalar σ we ﬁnd: Fσ (¯. g(x. σ). 0) = 0 To ﬁnd those derivatives let’s repeat equation (4. 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.106 Mart´ Uribe ın and h are identiﬁed by using the fact that. σ) + ησ . it must be the case that: Fσ (¯. σ) + ησ .63). 0) = 0. h(x.62) F (x. Then imposing Fσ (¯. by equation (4. σ). x and x Fx (¯. x one can identify gσ and hσ : hσ fy + f y =0 gσ fy gx + fx . σ) ≡ Et f (g(h(x. x) = 0. 0) = 0.

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. In this sense. Chapter 4 107 This equation is linear and homogeneous in gσ and hσ . This is an important limitation of ﬁrst-order perturbation techniques. This question is at the heart of the recent literature on optimal ﬁscal and monetary stabilization policy. Because in many economic applications we are interested in ﬁnding the eﬀect of uncertainty on the economy. and gσ = 0. For example. the policy function is independent of the variance-covariance matrix of t. They show that in general. one need not correct the constant term of the approximation to the policy function for the size of the variance of the shocks. we have that hσ = 0. These two expressions represent an important theoretical result. that is.Lecture in Open Economy Macroeconomics. any two policies that give rise to the same steady state yield. Because in a ﬁrst-order accurate solution the unconditional expectation of a variable is equal to the non-stochastic steady state. Thus. ﬁrst-order approximation techniques cannot be used to study risk premia. up to ﬁrst order. if a unique solution exists. . 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 . up to ﬁrst-order the mean of the rate of return of all all assets must be same. Thus.

x) = (¯. x. ˆ ˆ I I A ˆ ˆ xt+1 = B xt gx gx . In this case.108 up to ﬁrst-order the same level of welfare. y . x . y. The above expression represents a system of n × nx quadratic equations in the n × nx unknowns given by the elements of gx and hx . Mart´ Uribe ın To ﬁnd gx and hx diﬀerentiate (4. x) are y ¯ ¯ ¯ known. hx xt = xt+1 .62) with respect to x to obtain the following system Fx (¯. a perfect foresight equilibrium. Imposing Fx (¯. Let xt ≡ xt − x. 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 ]. 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. 0) = fy gx hx + fy gx + fx hx + fx x Note that the derivatives of f evaluated at (y .

Recall that a matrix a is said to be upper triangular if elements aij = 0 for i > j. z. Chapter 4 We are interested in solutions in which 109 t→∞ lim |ˆt | < ∞ x We will use this limiting conditions to ﬁnd the matrix gx . we use the generalized Schur decomposition of the matrices A and B. A matrix z is orthonormal if z z = zz = I.5 The generalized Schur decomposition of A and B is given by upper triangular matrices a and b and orthonormal matrices q and z satisfying:6 qAz = a and qBz = b. b = . z = . gx ]ˆt . 6 5 s1 t . st = . In particular. b. To solve the above system. Let st ≡ z [I. x Then we have that ast+1 = bst Now partition a. we will use the Schur decomposition method. 2 0 a22 0 b22 z21 z22 st More formal descriptions of the method can be found in Sims (1996) and Klein (2000). and st as a11 a12 b11 b12 z11 z12 a= .Lecture in Open Economy Macroeconomics.

ny . (a) The inverse of a nonsingular upper triangular matrix is upper triangular. Then we have that a22 s2 = b22 s2 . Suppose further that the number of ratios less than unity is exactly equal to the number of control variables. Namely. 7 . xt . By construction. 22 the requirement limj→∞ |s2 | < ∞ is satisﬁed only if s2 = 0. the eigenvalues of b−1 a22 are all less than unity in modulus. without loss of generality. nx . x Because this condition has to hold for any value of the state vector. z12 is of order nx × ny . t+1 t 22 Assume. t+1 t or b−1 a22 s2 = s2 . it ˆ follows that it must be the case that z12 + z22 gx = 0.110 Mart´ Uribe ın where a22 and b22 are of order ny × ny . that the ratios abs(aii /bii ) are decreasing in i. In turn. (c) The eigenvalues of an upper triangular matrix are the elements of its main diagonal. and that the number of ratios greater than one is equal to the number of state variables. and s2 is of t order ny × 1. Here we are applying a number of properties of upper triangular matrices. this restriction implies that t (z12 + z22 gx )ˆt = 0. (b) the product of two upper triangular matrices is upper triangular.7 Thus. by t t+j the deﬁnition of s2 .

x so that −1 −1 hx = [z11 − z21 z22 z12 ]−1 a11 −1 b11 [z11 − z21 z22 z12 ]. 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 . Chapter 4 Solving this expression for gx yields −1 111 gx = −z22 z12 . x t Replacing gx . we get −1 −1 xt+1 = [z11 − z21 z22 ˆ z12 ]−1 a11 −1 b11 [z11 − z21 z22 z12 ]ˆt . The fact that s2 = 0 also implies that t a11 s1 = b11 s1 .Lecture in Open Economy Macroeconomics.

. It follows that one can choose arbitrary initial values for ny − m elements of y0 and the resulting ﬁrst order solution will still be 8 To obtain this simple expression for hx . Now use element (1. which is simply z11 . 1) of z z to get z12 z11 = −1 −1 −1 −z22 z21 . Premultiply by z22 and post multiply by z11 to get z22 z12 = −z21 z11 −1 . which is less than the number of control variables. and that the number of eigenvalues of D with modulus greater than one is equal to the number of state variables. 1) of z z to write z21 z21 = I − z11 z11 . Let’s ﬁrst consider the case that the number of eigenvalues of D with modulus greater than unity is equal to m < ny. ny . But not for every economy this is the case. In this case there is a unique local equilibrium.8 Local Existence and Uniqueness of Equilibrium In the above discussion. Then this square bracket becomes [z11 + z21 z21 z11 −1 ]. 11 4.112 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:8 −1 hx = z11 a−1 b11 z11 . Then the requirement that we wish to study equilibria in which limj→∞ Et |ˆt+j | < ∞ will only yield m restrictions. rather than x ny restrictions. nx . −1 Use this expression to eliminate z22 z12 From the square bracket in the expression for hx . use element (2. Using this equation to eliminate z21 z21 from the expression −1 in square brackets. we assumed that the number of eigenvalues of D with modulus less than unity is exactly equal to the number of control variables. we get [z11 + (I − z11 z11 )z11 −1 ].

In this case the equilibrium is indeterminate. x This implies that m − ny elements of x0 must be functions of the remaining nx − (m − ny ) elements. (4. But this can never be the case. Chapter 4 113 expected to converge back to the steady state. we say no local equilibrium exists.64) 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. if the number of eigenvalues of D with modulus greater than unity is greater than the number of control variables. 4. because x0 is a vector of predetermined or exogenous variables and therefore its elements can take arbitrary values. On the other hand. Then in order to ensure that lim Et |ˆt+j | < ∞ we must set m elements of [x0 y0 ] equal to zero.9 Second Moments Start with the equilibrium law of motion of the deviation of the state vector with respect to its steady-state value. Let again m denote the number of eigenvalues of D greater than unity in modulus and assume that m > ny . ny . . which is given by xt+1 = hx xt + ση ˆ ˆ t+1 .Lecture in Open Economy Macroeconomics. In this case.

. Then vec(ABC) = (C ⊗ A) · vec(B). We will describe two numerical methods to compute Σx . 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. where F = hx ⊗ hx .114 Then we have that Σ x = h x Σx hx + Σ . Let A. where the vec operator transforms a matrix into a vector by stacking its columns. Thus if the vec operator is applied to both sides of Σx = hx Σx hx + Σ . and the symbol ⊗ denotes the Kronecker product. B. the result is vec(Σx ) = vec(hx Σx hx ) + vec(Σ ) = F vec(Σx ) + vec(Σ ).

The eigenvalues of F are products of the eigenvalues of the matrix hx .t Σ .t Σx. it follows that all eigenvalues of F are less than one in modulus.t+1 = hx.t = hx. One possible drawback of this method is that one has to invert a matrix that has dimension n2 × n2 .t+1 = hx.t hx.0 .t + Σ Σx.t =Σ .t hx. x x Method 2 The following iterative procedure.t + Σ hx.t+1 .Lecture in Open Economy Macroeconomics.0 = I hx. Σx. Because all eigenvalues of the matrix hx have by construction modulus less than one.0 = hx Σ . This implies that (I − F) is nonsingular and we can indeed solve for Σx . called doubling algorithm.t Σ . may be faster than the one described above in cases in which the number of state variables (nx ) is large.t hx. Chapter 4 Solving the above expression for vec(Σx ) we obtain vec(Σx ) = (I − F)−1 vec(Σ ) 115 provided that the inverse of (I −F) exists.

Let µt = ση t . the co-state. x for j ≥ 0. which we can write as: yt = gx xt . consider the variance covariance matrix of linear combinations of the state vector xt . it is easy to ﬁnd other second moments of interest. Then ¯ ¯ ˆ ˆ E yt yt = Egx xt xt gx ˆˆ ˆ ˆ = gx [E xt xt ]gx ˆ ˆ = gx Σx gx and.116 Other second moments Mart´ Uribe ın Once the covariance matrix of the state vector. or control vector yt is given by yt = y + gx (xt − x). E yt yt−j = gx [E xt xt−j ]gx ˆˆ ˆ ˆ = gx hj Σx gx . 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. For instance. . xt has been computed. more generally. Consider for instance the covariance matrix E xt xt−j for j > 0.

The program gx_hx. y x 4.edu/~mu2166/2nd_order. that is.columbia. 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. Chapter 4 117 4.Lecture in Open Economy Macroeconomics. htm.m computes the matrices gx and hx using the Schur .10 Impulse Response Functions The impulse response to a variable.11 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. Using the law of motion Et xt+1 = hx xt for the ˆ ˆ state vector. 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. x ˆ ˆ x x x t ≥ 0. The response of the vector of controls yt is given by ˆ IR(ˆt ) = gx ht x. relative to what was expected at time t − 1. letting x denote the innovation to the state vector in period 0. x = ησ 0 .

More importantly. there is a sense in which higher order approximations are simpler than the ﬁrst order approximation.. .118 Mart´ Uribe ın decomposition method. The program mom. then obtaining the second order approximation requires only to compute the coeﬃcients of the quadratic terms. 4. The program ir. Speciﬁcally.m computes impulse response functions. for j > 1 and i < j. Namely. 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 . So if the ﬁrst-order approximation to the solution is available. 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. 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.12 Higher Order Approximations In this chapter. Indeed. But higher order approximations are relatively easy to obtain. since the coeﬃcients of the linear terms are those of the ﬁrst order approximation. obtaining a higherorder approximation to the solution of the non-linear system is a sequential procedure. xt ) = 0.m computes second moments. obtaining the coeﬃcients of the ith order terms of the approximate solution given all lower-order coeﬃcients involves solving a linear system of equations.

where AT and AN denote. Assume that At is a composite of tradable and nontradable goods produced via the following aggregator function: 1/(1−1/µ) At = η(AT )1−1/µ + (1 − η)(AN )1−1/µ t t . 1−σ where Ct denotes consumption and ht denotes hours worked. respectively.13 Exercises 1. The Eﬀect of Nonstationary Productivity Shocks on the Trade Balance in the Preence of Nontraded Goods The purpose of this assignment is to explore whether nonstationary productivity shocks are useful to understand business cycles in emerging markets when the one-sector SOE RBC model studied in class is augmented to include a nontraded sector. where It denotes gross domestic investment.Lecture in Open Economy Macroeconomics. Let At denote domestic absorption. Consider an economy populated by a large number of identical households with preferences described by the utility function ∞ E0 t=0 βt [Ctγ (1 − ht )1−γ ]1−σ − 1 . domestic absorptions of tradt t able and nontradable goods. Output of tradable and nontradable . That is. Chapter 4 119 4. At satisﬁes the identity At = Ct + It .

Market clearing in the nontraded sector requires that domestic absorption equals production: YtN = AN . stochastic productivity shock and Xt represents an exogenous.120 Mart´ Uribe ın goods. nonstationary productivity shock. stationary. The variable zt represents an exogenous. YtT and YtN . are produced under constant returns to scale using capital and labor as inputs: YtT = zt (KtT )1−αT (Xt hT )αT t and YtN = zt (KtN )1−αN (Xt hN )αN . capital and labor services emt ployed in sector i. stochastic. t . N . respectively. for i = T. respectively. The capital stocks evolve according to the following laws of motion: φ − 2 T Kt+1 −g KtT 2 T Kt+1 = (1 − δ)KtT + ItT KtT and N Kt+1 = (1 − δ)KtN + ItN φ − 2 N Kt+1 −g KtN 2 KtN . where Iti denotes investment in sector i = T. N and satisﬁes It = ItT + ItN . t where Kti and hi denote. denoted.

t . t t Assume that the country has access to a single. Chapter 4 Also. one-period. t Rt with Rt = R∗ + ψ eDt+1 /Xt −d − 1 . Finally. ˜ where Dt denotes average external debt per capita in period t. we have that in equilibrium the following condition must hold: ˜ Dt = Dt . market clearing in the labor market requires that 121 h t = hT + hN . Because households are homogeneous. t And the nonstationary productivity shock evolves according to ln(gt /g) = ρa ln(gt−1 /g) + σg g .Lecture in Open Economy Macroeconomics. Dt . is given by Dt+1 = Dt + AT − YtT . The evolution of the household’s net foreign debt position. internationally traded bond that pays the debt-elastic interest rate Rt when held between periods t and t + 1. assume that the stationary productivity shock zt follows an AR(1) process of the form ˜ ¯ ln zt = ρz ln zt−1 + σz z .

001 αT 0.0066 φ 1.37 µ 0.95 g 1. σg > 0. Use the following table to calibrate the model: β 0.36 d 0.80 σ 2 δ 0.40 αN 0.65 in the steady state. Compute the steady state of the model. Count the number of variables and equations. deﬁned as pN YtN /(pN YtN + YtT ) equals 0. σz . (a) Write down the complete set of equilibrium conditions.122 where gt ≡ ρz . and distributed N (0. Report the t t .00 ρz 0. 1). (b) Let pN and pt be the (shadow) relative prices of the nontradable t good and the composite good in terms of tradable goods. Are pN and pt stationary variables? Why? t t (c) Write down the complete set of equilibrium conditions in stationary form along a balanced-growth path.0213 σz 0. In addressing the numerical portions of the following questions. write your answers to 4 decimal places. Xt−1 and g t are independent white noises ρg 0. ρg ∈ (−1.0053 z t Mart´ Uribe ın Xt .05 σg 0. respectively.98 γ 0. (d) Derive a restriction on R∗ that guarantees that the steady-state value of dt+1 ≡ Dt+1 /Xt along the balanced-growth path equals d.10 ψ 0. Write down expressions for pN and pt in terms of AT and t t AN . 1).44 Guidelines for calibrating the parameter η are given below. Set η to ensure that the share of nontraded output in total output.

(e) Report the numerical values of the eigenvalues of the matrix hx deﬁning the linearized equilibrium law of motion of the state vector. evaluate the ability of the present traded-nontraded model to explain key stylized facts in emerging countries.Lecture in Open Economy Macroeconomics. respectively. (f) Deﬁne GDP as Yt = (YtT + pt YtN )/pt . . Compute the unconditional standard deviations of the growth rates of consumption and output. (g) Compute other second moments of your choice and discuss whether they are in line with empirical regularities in emerging economies. (i) Based on your answers to the previous three questions. deﬁned. (h) Explain how you would reparameterize the model to make it coincide with the one-sector model studied in Aguiar and Gopinath (2007). deﬁned as tbyt ≡ T Bt /Yt . Answer the previous two questions under your proposed parameterization. Deﬁne the trade balance as the variable T Bt that satisﬁes T Bt = Yt − Ct − It . Chapter 4 123 numerical value of η and the steady-sate values of all endogenous variables of the model. as ∆Ct ≡ log(Ct /Ct−1 ) and ∆Yt ≡ log(Yt /Yt−1 ). Compute the correlation between ∆Yt and the trade-balance-to-output ratio. Provide intuition making sure to emphasize the diﬀerences you can identify between the dynamics implied by the one-good SOE RBC model and the present two-good SOE RBC model.

Again. and the current-accountto-output ratio implied by the IDF model driven by interest-rate shocks.04. Shut down the productivity shock by setting σ = 0.1. hours. investment. (a) Consider the IDF economy studied in chapter 4. was derived in the context of a model driven by technology shocks. Inducing Stationarity and Interest-Rate Shocks Mart´ Uribe ın One result derived in chapter 4 of the lecture notes is that the business cycle implied by the SOE RBC model is not aﬀected by the method used to induce stationarity. Set r = 0. compute the statistics considered in table 4. ¯ ¯ 2 where µt ∼ N (0. consumption. set σ = 0. the trade-balance-to-output ratio. Using this version of the IDF model.012. This exercise aims to establish whether that result is robust to assuming that business cycles are driven by world-interest-rate shocks.124 2. and σµ = 0. (b) Now consider the EDEIR model. Replace the ˜ formulation rt = r + p(dt ) with ∗ ˜ rt = rt + p(dt ) .8. ¯ Calibrate all other parameters of the model at the values given in table 4. Replace the interest-rate speciﬁcation rt = r with the process rt − r = ξ(rt−1 − r ) + µt . Provide intuition for these results. ξ = 0.4 and make a table. however. This result. σµ ). Make a ﬁgure showing impulse responses of output.

Assume that the external shock . Does the stationarity-inducing mechanism make any diﬀerence for the business-cycles implied by the SOE model driven by interest-rate shocks? 3. Calibrate ¯ all other parameters of this model mimicking the calibration of the EDEIR model studied in the lecture notes. ¯ ¯ 125 Calibrate r. Use the resulting calibrated model to compute unconditional second moments and impulse responses. ξ. To facilitate comparison. Business Cycles in a Small Open Economy with Complete Asset Markets and External Shocks Consider the small open economy model with complete asset markets (CAM) studied in chapter 4 of the lecture notes. (c) Compare the predictions of the IDF and EDEIR models driven by interest rate socks. place the information generated here in the same table and ﬁgure produced in the previous question. and σµ as in the previous question. Provide intuition for your results. which can be interpreted as an external shock. where xt is an exogenous and stochastic random variable. Replace the equilibrium condition λt = ψ4 with the expression λt = x t . Suppose that the productivity factor At is constant and normalized to 1. Chapter 4 and ∗ ∗ rt − r = ξ(rt−1 − r ) + µt .Lecture in Open Economy Macroeconomics.

it . ct .9 and σ = 0. and . ht . serial correlation. ht . Answer questions 2.a and 2.2 of the lecture notes. and tbt /yt . Calibrate all other parameters of the model following the calibration of the CAM model of chaper 4 of the lecture notes. (c) Now replace the values of ρ and σ given above with values such that the volatility and serial correlation of output implied by the model are the same as those reported for the Canadian economy in table 4. set the steady state value of xt in such a way that the steady-state level of consumption equals the level of steady-state consumption in the version of the CAM modelstudied in the lecture notes.02.b using these new parameter values. it . where xt ≡ ln(xt /x) and x denotes the non-stochastic steady-state ˆ level of xt . evaluate the ability of external shocks (as deﬁned here) to explain business cycles t) of yt . Let ρ = 0. and correlation with output of yt . (b) Produce a ﬁgure with 5 plots depicting the impulse responses to an external shock (a unit innovation in tbt /yt . ct . Finally.126 follows a process of the form Mart´ Uribe ın xt = ρˆt−1 + t . (d) Based on your answer to the previous question. σ 2 ). ˆ x t ∼ N (0. (a) Produce a table displaying the unconditional standard deviation.

Chapter 4 in Canada.Lecture in Open Economy Macroeconomics. 127 .

128 Mart´ Uribe ın .

Ecuador -0.51 (0.12). the Philippines -0. This question is complicated by the fact that the country interest rate is unlikely to be completely exogenous to the country’s domestic conditions.00).02 (0.2 To clarify ideas.1 Data like those shown in ﬁgure 5.95).1 have motivated researches to ask what fraction of observed business cycle ﬂuctuations in emerging markets is due to movements in country interest rate.58 (0.1. for example. which depicts detrended output and the country interest rate for seven developing economies between 1994 and 2001.00). 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.07 (0. Mexico -0.80 (0. Edwards (1984). let Rt denote the gross The estimated correlations (p-values) are: Argentina -0. 2 There is a large literature arguing that domestic variables aﬀect the interest rate at which emerging markets borrow externally. See.00). South Africa -0. Peru -0.Chapter 5 Interest-Rate Shocks Business cycles in emerging market economies are correlated with the interest rate that these countries face in international ﬁnancial markets.00).37 (0.67 (0.71). This observation is illustrated in ﬁgure 5. Cline 1 129 . Brazil -0.

15 Brazil Ecuador 0.1 94 95 96 97 98 99 00 01 94 95 96 97 98 Mexico 0.2 0.130 Mart´ Uribe ın Figure 5.1: Country Interest Rates and Output in Seven Emerging Countries Argentina 0.1 0.1 0.4 Philippines 0.05 0 −0.05 0 −0. Source: Uribe and Yue (2006).06 0.15 0. Data source: output.1 0. . interest rates.1 0 0 94 95 96 97 98 Peru 0. Country interest rates are real yields on dollar-denominated bonds of emerging countries issued in international ﬁnancial markets.15 0.05 94 95 96 97 98 99 00 01 0 0.05 94 95 96 97 98 99 00 01 South Africa 0. IFS.3 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.1 0.05 0 0. EMBI+.05 0.1 99 00 01 −0.05 −0.04 0.05 94 95 96 97 98 99 00 01 99 00 01 0.

and Cline and Barnes (1997). 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. consider the following example. Here. its contribution to generating business cycles is nil. furthermore.. If the country in question is a small player in international ﬁnancial markets.Lecture in Open Economy Macroeconomics. Suppose now that a domestic productivity shock induces an expansion in output. or country interest-rate premium. for instance. borrow and lend from one another.S. Thus. The researcher. the interest rate falls. is completely exogenous to the emerging country’s domestic conditions. and St denotes the gross country interest-rate spread. however. .e. may induce foreign lenders to lower spreads on believes that the country’s ability to repay its debts has improved. about the country spread St . An increase in output. We can’t say the same. in the data we ﬁnd an Argentine spread. Rus denotes the world interest rate. that Rt is countercyclical. Chapter 5 131 interest rate at which the country borrows in international markets. however. it is reasonable to assume that the world interest rate Rt . like the U. a Colombian spread. This interest rate can be expressed as Rt = Rt St . Suppose that the interest rate Rt is purely endogenous. Assume. The researcher. To see why. or the interest rate at which developed countries. foreign lenders reduce the country spread in response to expansions in aggregate activity.. who believes Rt is exogenous. In response to this output increase. etc. i. as many emerging us economies are. erroneously attributes (1995). Because the interest-rate premium is country speciﬁc. or the us country interest rate. wrongly assumes that the interest rate is purely exogenous.

and Rt denotes the gross real (emerging) country interest rate. it denotes real gross domestic us investment. of course. The right conclusion. The particular combination adopted in this chapter draws heavily from Uribe and Yue (2006). A hat on yt and it denotes log deviations from a log-linear us us trend. A hat on Rt and Rt denotes simply the log. tbyt denotes the trade balance to output ratio. We measure Rt as the 3-month gross Treasury bill rate divided by the average gross US inﬂation .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 + (5. 5.1) where yt denotes real gross domestic output.132 Mart´ Uribe ın part of the increase in output to the decline in Rt . Rt denotes the gross real US interest rate. is that all of the movement in output is due to the productivity shock. the identiﬁcation process must combine statistical methods and economic theory. Necessarily. the ﬁrst step is to identify the exogenous components of country spreads and world interest rate shocks. It follows that in order to quantify the macroeconomic eﬀects of interest rate shocks.

e. 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.Lecture in Open Economy Macroeconomics.4 An additional restriction imposed on the VAR system. . Morgan’s EMBI+ stripped spread and the US real interest rate. Because Rt and Rt appear at the bottom of the system. Also. discuss an alternative identiﬁcation strategy consisting in placing ﬁnancial variables ‘above’ real variables ﬁrst in the VAR system. t and tby t ) aﬀect ﬁnancial markets contemporaneously. and the trade balance are seasonally adjusted. for. A4i = B4i = 0.. 3 r. Uribe and Yue (2006) impose the restriction that the matrix A be lower triangular with unit diagus onal elements. conceivably. P. Output.3 We measure Rt as the sum of J. Chapter 5 133 over the past four quarters. At the same time. 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. 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. decisions such as employment and spending on durable consumption goods and investment goods take time to plan and implement. The country-interest-rate shock. This identiﬁcation strategy is a natural one. 4 Uribe and Yue (2006). investment. is that the world us interest rate Rt follows a simple univariate AR(1) process (i. To identify the shocks in the empirical model. the identiﬁcation scheme implies that real domestic shocks ( y . it seems reasonable to assume that ﬁnancial markets are able to react quickly to news about the state of the business cycle. for all i = 4). t i.

2 displays with solid lines the impulse response function implied .1). St ≡ Rt −Rt . throughout the paper we indistinctly refer to as a country interest rate shock or as a country spread shock. To see this. 5.1) the ˆ ˆ ˆ ˆ us country interest rate Rt using the deﬁnition of country spread. the estimated residual of the newly deﬁned bottom equation. 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. and the trade balance? Second. 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. how do country spreads respond to innovations in US interest rates? Third.2 Impulse Response Functions Figure 5. investment. it is obvious that the impulse response s t functions of yt . ˆt . how important are US-interest-rate shocks and country-spread shocks in explaining movements in aggregate activity in emerging countries? Fifth. us Clearly. because Rt appears as a regressor in the bottom equation of the VAR system. consider substituting in equation (5. t Moreover. call it s. After estimating the VAR system (5.134 Mart´ Uribe ın country spread shock. how and by how much do country spreads move in response to innovations in emerging-country fundamentals? Fourth. t is identical to r. t are identical to those associated r t Therefore. and tbyt associated with ˆ ı with r. how do US-interest-rate shocks and country-spread shocks aﬀect real domestic variables such as output.

4 0.1 −0.25 −0.2 0 5 10 15 20 1 0.6 −0.8 0.1 0 5 10 15 20 −1 5 0 World Interest Rate −0.5 10 15 20 Country Interest Rate 1 0. the country interest rate. and the country spread are expressed in percentage points.3 0. Chapter 5 135 Figure 5.3 5 10 15 20 0 −0.2 −0.8 0.2: Impulse Response To Country-Spread Shock Output 0 −0.4 −0.4 0.8 −1 −1.2 −0.2 5 10 15 20 Investment Trade Balance−to−GDP Ratio 1 0. The responses of the Trade Balance-to-GDP ratio. The two-standard-error bands are computed using the delta method.4 0.Lecture in Open Economy Macroeconomics. the US interest rate.5 0.2 0 5 Country Spread 10 15 20 Notes: (1) Solid lines depict point estimates of impulse responses.6 0. .15 −0.05 −0. and broken lines depict two-standard-deviation error bands.2 0. (2) The responses of Output and Investment are expressed in percent deviations from their respective log-linear trends.6 0.

This is reﬂected in the fact that the trade balance improves in the two periods following the shock. as indicated by the width of the 2-standard-deviation error bands. investment. t Broken lines depict two-standard-deviation bands. and subsequently recover gradually until they reach their pre-shock level. and the trade balance.1) to a one percentage point increase in the US interest rate shock. The half life of the country spread response is about one year. r. because of our maintained assumption that external ﬁnancial shocks take one quarter to aﬀect production and absorption.1) to a unit innovation in the country spread shock.5 In response to an unanticipated country-spread shock. The adverse spread shock produces a larger contraction in aggregate domestic absorption than in aggregate output. Figure 5. However. In the two periods following the country-spread shock. aggregate activity and gross domestic investment contract. and the trade balance-to-output ratio respond as one would expect. are qualitatively similar to those associated with an innovation in the country spread. output and investment fall. investment. The point estimates of the impulse response functions of output. . while net exports improve. however. the country spread itself increases and then quickly falls toward its steady-state level. For 5 These bands are computed using the delta method.136 Mart´ Uribe ın by the VAR system (5. They are unchanged in the period of impact. Output. t The eﬀects of US interest-rate shocks on domestic variables and coun- try spreads are measured with signiﬁcant uncertainty. That is. 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.3 displays the response of the variables included in the VAR system (5. rus .

5 1 0.5 0 −0. the country interest rate. and the US interest rate are expressed in percentage points.2 0 0 5 10 15 20 5 1 0.5 5 10 15 20 2.5 0.5 2 1.5 2 1.5 1 0.5 −2 −3 −1 −4 −5 −1.5 0. The responses of the Trade Balance-to-GDP ratio. and broken lines depict two-standard-deviation error bands. (2) The responses of Output and Investment are expressed in percent deviations from their respective log-linear trends.6 1 0.5 −1 5 Country Spread 10 15 20 Notes: (1) Solid lines depict point estimates of impulse responses. Chapter 5 137 Figure 5.3: Impulse Response To A US-Interest-Rate Shock Output 1 0 0 −1 −0.8 World Interest Rate 10 15 20 Country Interest Rate 3 2.4 0.Lecture in Open Economy Macroeconomics.5 5 10 15 20 −6 5 10 15 20 Investment Trade Balance−to−GDP Ratio 2 1.5 0 −0. .

country spreads increase signiﬁcantly in response to innovations in the US interest rate but with a short delay. the terms-oftrade. y t. We note. They can reﬂect variations in total factor productivity. However. The response of output. It is remarkable that the impulse response function of the country spread to a US-interest-rate shock displays a delayed overshooting. As a result. We now ask how innovations in output. the country spread initially falls.138 Mart´ Uribe ın instance. The response of investment is about three times as large as that . that because the models estimated by these authors are static in nature. 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. by construction.4 depicts the impulse response function to a one-percent increase in the output shock. Thus. Figure 5. the trough in the output response is twice as large under a USinterest-rate shock than under a country-spread shock. however. The overshooting of country spreads is responsible for the much larger response of domestic variables to an innovation in the US interest rate than to an innovation in the country spread of equal magnitude. etc. in the period of impact the country interest rate increases but by less than the jump in the US interest rate. investment. the country spread recovers quickly and after a couple of quarters it is more than one percentage point above its pre-shock level. The negative impact eﬀect is in line with the ﬁndings of Eichengreen and Mody (1998) and Kamin and Kleist (1999).1). The model is vague about the precise nature of output shocks. they are unable to capture the rich dynamic relation linking these two variables. impinge upon the variables of our empirical model. In eﬀect.

8 5 10 15 20 0 −0.Lecture in Open Economy Macroeconomics.2 −0. Chapter 5 139 Figure 5.1 −0.5 World Interest Rate −0.4: Impulse Response To An Output Shock Output 1 3 0. and the US interest rate are expressed in percentage points.4 −0.6 0.6 −0.5 5 10 15 20 −1 5 0 0.5 2 1.5 10 15 20 Country Interest Rate 0 −0.4 −0.2 0 5 10 15 20 0.8 0. . the country interest rate. and broken lines depict two-standard-deviation error bands.2 −0. The responses of the Trade Balance-to-GDP ratio.6 −0.3 −0.2 −0.8 5 Country Spread 10 15 20 Notes: (1) Solid lines depict point estimates of impulse response functions.4 1 0.5 Investment Trade Balance−to−GDP Ratio 1 0 −0.5 0 5 10 15 20 2. (2) The responses of Output and Investment are expressed in percent deviations from their respective log-linear trends.4 −0.

4 percent and after two quarters starts to improve. 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. 5. The half life of the country spread response is about ﬁve quarters. t Because rus t and r t are orthogonal disturbances. 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- . More interestingly.5 displays the variance decomposition of the variables contained in the VAR system (5.6 percent. the increase in output produces a signiﬁcant reduction in the country spread of about 0. t Broken lines depict the fraction of the variance of the forecasting error explained by US-interestrate shocks ( rus ). At the same time. 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 ).140 Mart´ Uribe ın of output. converging gradually to its steady-state level. the trade balance deteriorates signiﬁcantly by about 0.1) at diﬀerent horizons.3 Variance Decompositions Figure 5.

05 0 0. Chapter 5 141 Figure 5.05 5 10 15 quarters 20 0 5 10 15 quarters World Interest Rate 2 0.6 0.5 0. Broken lines depict the fraction of the variance of the forecasting error explained by rus at diﬀerent t horizons.1 5 10 15 quarters 20 0.5 0.7 0.5 0 5 10 15 quarters Country Interest Rate 20 0 5 10 15 quarters Country Spread 20 0.2 0.3 0.5 0.4 0.6 0.2 0.4 1.5: Variance Decomposition at Diﬀerent Horizons Output 0.8 0.1 0.3 1 20 Investment Trade Balances−to−GDP Ratio 0.8 0.2 0.3 0.1 0.2 0.7 0.15 0.Lecture in Open Economy Macroeconomics.4 0.3 0. .25 0.25 0.1 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.15 0.2 0.

where Cj ≡ (A−1 B)j A−1 . Then. is given by the diagonal elements of the matrix Σx. r. where Λ4 is a 5×5 matrix with all elements equal to zero except j=0 element (4. is given by h Cj t+h−j . i . the broken lines in ﬁgure 5.. The diﬀerence between the solid lines and the broken lines (i. Then. xt+h − Et xt+h .h ≡ j=0 Cj Σ Cj . 7 We observe that the estimates of y .e.h .h a particular shock. to the diagonal elements x. t explain about 20 percent of move- ments in aggregate activity in emerging countries at business cycle frequency. ≡ t h (Cj Λ4 )Σ (Cj Λ4 ) . According to our estimate of the VAR system given in equation (5. At the same time.e. we have that ˆ i rus the estimates of t will in general not be orthogonal to the estimates of y . Thus. For the purpose of the present discussion. and tbyt are excluded from the Rt equation.6. i . 1999).7 Note that as the forecasting horizon approaches inﬁnity. In turn. t t t However. Our choice of horizon falls in the middle of this window. The variance/covariance j=0 h matrix of this h-step-ahead forecasting error is given by Σx. second. the decomposition of the variance of the forecasting error coincides with the decomposition of the unconditional variance of the series in question. But because yt . . say rus . tby . t account for about 12 percent of 6 ˆ us ˆ These forecasting errors are computed as follows.h the element-by-element ratio of the diagonal elements of Σx.h of the matrix Σ for diﬀerent values of h. ˆt .1).4). this lack of orthogonality should disappear as the sample size increases. where Σ is the (diagonal) variance/covariance matrix of t . the sample residuals of t t t t the ﬁrst. country-spread shocks. the variance of the h-stepahead forecasting error of xt is simply the vector containing the diagonal elements of Σx.142 Mart´ Uribe ın ferent horizons. Researchers typically deﬁne business cycles as movements in time series of frequencies ranging from 6 quarters to 32 quarters (Stock and Watson. third. which takes the value one. rus . the fraction of the variance of the forecasting error due to r ) is t computed in a similar fashion but using the matrix Λ5 . or tby . 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. innovations in the US interest rate. that is. and r (i. The error in forecasting xt+h at time t j=0 for h > 0. one can write the MA(∞) representation of xt as xt = ∞ Cj t−j . and ﬁfth equations of the VAR system) are orthogonal to each us other. we associate business-cycle ﬂuctuations with the variance of the forecasting error at a horizon of about ﬁve years. 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.. the variance of the error of the h-step-ahead forecasting error of xt due to rus .5 are given by rus .

4 establish empirically that country spreads respond signiﬁcantly and systematically to domestic macroeconomic variables. around one third of business cycles in emerging economies is explained by disturbances in external ﬁnancial variables. Thus. we modify the Rt equation of the VAR system by setting to zero the coeﬃcients on yt−i .Lecture in Open Economy Macroeconomics. who interpret this ﬁnding as suggesting that arbitrary revisions in investors sentiments play a signiﬁcant role in explaining the behavior of country spreads. Chapter 5 143 aggregate ﬂuctuations in these countries. 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. A natural question raised by these ﬁndings is whether the feedback from endogenous domestic variables to country spreads exacerbates domestic volatility. At the same time. these two sources of uncertainty account for about 85 percent of ﬂuctuations in country spreads. Variations in country spreads are largely explained by innovations in US interest rates and innovations in country-spreads themselves. 1. We then compute ˆ i . 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. and tbyt−i for i = 0. Jointly. This last result concurs with Eichengreen and Mody (1998). Most of this fraction. Here we make a ﬁrst step at answering this question. ˆ Speciﬁcally. These disturbances play an even stronger role in explaining movements in international transactions. In eﬀect. about 60 percentage points. is attributed to country-spread shocks. The impulse response functions shown in ﬁgure 5. ˆt−i .

1 shows that the presence of feedback from domestic variables to country spreads signiﬁcantly increases domestic volatility. 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.6450 3. As such.0674 ˆ ı 14. We follow this route next. .1: Aggregate Volatility With and Without Feedback of Spreads from Domestic Variables Model Variable Feedback No Feedback Std.1060 11. 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. y ˆ 3. ˆt . Table 5.9260 4.4955 4.7696 ˆ the implied volatility of yt . In particular. Of course. The eﬀect of feedback on the cyclical behavior of the country spread itself is even stronger. the volatility of the country interest rate falls by about one third. Std. the results of table 5. In eﬀect.1 serve solely as a way to motivate a more adequate approach to the question they aim to address. We compare these volatilities to those emerging from the original VAR model.3846 3. Dev. tbyt and Rt in the modiﬁed VAR system ˆ i at business-cycle frequency (20 quarters).5198 tby R 6. Dev. this counterfactual exercise is subject to Lucas’ (1976) celebrated critique. when feedback is negated. when we shut oﬀ the endogenous feedback. 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.144 Mart´ Uribe ın Table 5.

. Thus.2 and 5.3) with those implied by the proposed theoretical framework. If the estimated shocks imply similar business cycle ﬂuctuations in the empirical as in theoretical models. Chapter 5 145 5. We depart from the canonical version of the small-open-economy RBC model along four dimensions. Accordingly. we compare estimated impulse responses (i. First. Third.g. we estimate the deep structural parameters of the model.. we develop a standard model of the business cycle in small open economies. Finally.Lecture in Open Economy Macroeconomics. we will assess the plausibility of our estimated shocks in four steps: First. we feed into the model the estimated version of the fourth and ﬁfth equations of the VAR system (5. describing the stochastic laws of motion of the US interest rate and the country spread. as in the empirical model.e.1). Second. Second.1). To assess the plausibility of these restrictions. Mendoza.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 (5. we conclude that according to the proposed theory. preferences are as- . innovations in the world interest rate or the country spread are assumed to have allocative eﬀects with a one-period lag. it is necessary to use the predictions of some theory of the business cycle as a metric. we assume that in each period. The basis of the theoretical model presented here is the standard neoclassical growth model of the small open economy (e. 1991). the identiﬁed shocks are plausible. those shown in ﬁgures 5. production and absorption decisions are made prior to the realization of that period’s world-interest-rate shock and country-spread shock.

In combination. these two frictions prevent excessive investment volatility. Habit formation has been shown to help explain asset prices and business ﬂuctuations in both developed economies (e. 5. and Fisher. ct denotes the cross-sectional ˜ average level of consumption in period t.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 . 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. Christiano. c (5. Uribe. Third.2) where ct denotes consumption in period t. Fourth. and ht denotes the fraction of time devoted to work in period t. and allows the model to predict a more realistic response of domestic output to external ﬁnancial shocks. 2001) and emerging countries (e. 1997). induce persistence. and allow for the observed nonmonotonic (hump-shaped) response of investment in response to a variety of shocks (see Uribe. 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.g.4. or catching up with the Joneses as in Abel (1990).146 Mart´ Uribe ın sumed to feature external habit formation. Households take as given the process ... 2002).g. ht ). the process of capital accumulation is assumed to be subject to gestation lags and convex adjustment costs. Boldrin.

households allocate their wealth to purchases of consumption goods. 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. We assume that households face costs of adjusting their foreign asset position. including the one used . and smooth. ut denotes the rental rate of capital. for some d > 0. decreasing in its second argument. The debt-adjustment cost function Ψ(·) is assumed ¯ ¯ ¯ to be convex and to satisfy Ψ(d) = Ψ (d) = 0. purchases of investment goods. physical capital and an internationally traded bond.3) where dt denotes the household’s debt position in period t. we compared a number of standard alternative ways to induce stationarity in the small open economy framework.Lecture in Open Economy Macroeconomics. capital rents. concave. The household’s periodby-period budget constraint is given by dt = Rt−1 dt−1 + Ψ(dt ) + ct + it − wt ht − ut kt . Rt denotes the gross interest rate faced by domestic residents in ﬁnancial markets. and purchases of ﬁnancial assets. Households have access to two types of asset. and interest income bond holdings. 1) denotes a subjective discount factor. The single-period utility index U is assumed to be increasing in its ˜ ﬁrst argument. and the parameter µ measures the intensity of external habit formation. (5. The capital stock is assumed to be owned entirely by domestic residents. kt denotes the stock of physical capital. wt denotes the wage rate. Earlier in chapter ??. Each period. The parameter β ∈ (0. and it denotes gross domestic investment. Households have three sources of income: wages. Chapter 5 147 for ct .

(5. 1) denotes the rate of depreciation of physical capital.8 The process of capital accumulation displays adjustment costs in the form of gestation lags and convex costs as in Uribe (1997). Let sit denote the number of investment projects started in t − i for i = 0. 8 .5) The stock of capital obeys the following law of motion: s3t kt kt+1 = (1 − δ)kt + kt Φ .10) below). They capture funds from foreign investors at the country rate Rt d and lend to domestic agents at the rate Rt .4) In turn. 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 ) . This digression will be of use later in the paper when we analyze the ﬁrm’s problem. dt . Suppose there is a continuum of banks of measure one that behave competitively. 1. that are increasing and convex in the volume of intermediation. Producing one unit of capital good requires investing 1/4 units of goods for four consecutive periods.6) where δ ∈ (0. which are d d given by Rt [dt − Ψ(dt )] − Rt dt . Ψ(dt ). i=0 (5. The The debt adjustment cost can be decentralized as follows. which is precisely the shadow interest rate faced by domestic agents in the centralized problem (see the Euler condition (5. 2. and conclude that they all produce virtually identical implications for business ﬂuctuations.148 Mart´ Uribe ın here. The problem of domestic banks is then to choose the volume dt so as to maximize proﬁts. Then investment in period t is given by 1 4 3 it = sit . 3. In addition. banks face operational costs. taking as given Rt and Rt . Suppose that ﬁnancial transactions between domestic and foreign residents require ﬁnancial intermediation by domestic institutions (banks). the evolution of sit is given by si+1t+1 = sit . Bank proﬁts are assumed to be distributed to domestic households in a lump-sum fashion. (5.

The Lagrangian associated with the household’s optimization problem can be written as: ∞ L = E0 t=0 β t U (ct − µ˜t−1 . and the capital stock given by equations (5. 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. s0. As discussed in chapters 3 and ??. 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 ) .Lecture in Open Economy Macroeconomics.t+1 . as deﬁned by the function Φ. where λt . k0 .7) that prevents the possibility of Ponzi schemes. concave. and a borrowing constraint of the form dt+j+1 j s=0 Rt+s j→∞ lim Et ≤0 (5. 1. and to satisfy Φ(δ) = δ and Φ (δ) = 1. 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 λt qt are the Lagrange multipliers associated with con- .3). λt νit . Households choose contingent plans {ct+1 . Rt . h0 . The introduction of capital adjustment costs is commonplace in models of the small open economy.6). The household takes as given the processes {˜t−1 . ut }∞ as well as c0 . dt+1 }∞ so as to t=0 maximize the utility function (5. and sit for c t=0 i = 0. Chapter 5 149 process of capital accumulation is assumed to be subject to adjustment costs. investment projects.4)-(5. R−1 d−1 . wt . 3. which is assumed to be strictly increasing.2) subject to the budget constraint (5. 2. the laws of motion of total investment. ht+1 .

ht+1 .9) deﬁnes the household’s labor supply schedule.3).150 Mart´ Uribe ın straints (5. by equating .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 .10) (5. and s0t+1 all reside in the information set of period t. bec cause current consumption cannot react to unanticipated changes in wealth. ht )). Et λt+1 . ht+1 ) c c Et [wt+1 λt+1 ] = −Uh (ct+1 − µ˜t .9) (5.12) (5.7) all holding with equality and Et λt+1 = Uc (ct+1 − µ˜t . respectively.8) (5.3). because of our assumed information structure. Equation (5. Equation (5.13) (5.11) (5.15) It is important to recall that. 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 (5.14) s3t+1 kt+1 s3t+1 kt+1 − + λt+1 ut+1 . and (5. (5. The optimality conditions associated with the household’s problem are (5.8) 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. (5. the variables ct+1 .4)-(5. (5.

to the discounted value of renting the unit of capital for one period and then selling it.13) show how to price investment projects at diﬀerent stages of completion.16) . thereby reducing their debt positions. qt . above its steady-state level d.15) is a pricing condition for physical capital. 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. ¯ say. when the household’s debt position is. It equates the revenue from selling one unit of capital today.4. yt = F (kt .10) is an asset pricing relation equating the intertemporal marginal rate of substitution in consumption to the rate of return on ﬁnancial assets. Chapter 5 151 the marginal disutility of eﬀort in period t + 1 to the expected utility value of the wage rate in that period. providing further incentives for households to save. 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 )].Lecture in Open Economy Macroeconomics. ht ). Equation (5. ut+1 + qt+1 .14) links the cost of producing a unit of capital to the shadow price of installed capital.11)-(5. net of depreciation and adjustment costs.2 Firms Output is produced by means of a production function that takes labor services and physical capital as inputs. Finally. Equation (5. 5. because of the presence of frictions to adjust bond holdings. we have that Ψ (dt ) > 0 so that the shadow rate of return is higher than the market rate of return. Equations (5. or Tobin’s Q. Intuitively. equation (5. qt . Note that. (5.

152 Mart´ Uribe ın where the function F is assumed to be homogeneous of degree one. and Rt ≡ Rt 1−Ψ (dt ) is the interest rate faced by nonﬁnancial domestic agents. the working-capital constraint takes the form κt ≥ ηwt ht . The debt position of the ﬁrm. Deﬁne the ﬁrm’s total net liabilities at the end of period t as at = d Rt df − κt . Then. increasing in both arguments. and concave. where κt denotes the amount of working capital held by the representative ﬁrm in period t. 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. denoted by df . 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. Formally. . ht ) + wt ht + ut kt + πt + Rt κt . we can rewrite the above expression as t d Rt − 1 d Rt at = at−1 − F (kt . as shown in footnote 8. η ≥ 0. Firms hire labor and capital services from perfectly competitive markets. evolves according to the t following expression d df = Rt−1 df − F (kt . ht ) + wt ht + ut kt + πt − κt−1 + κt .

That is. ht ) − wt ht 1 + η d Rt d Rt − 1 d Rt − u t kt .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 . the domestic residents. which is increasing d in the interest rate Rt . Using constraint (5. ht ) + wt ht 1 + η d Rt d Rt − 1 d Rt + u t kt + π t . and kt so as to maximize ∞ E0 t=0 βt λt λ0 at − at−1 + F (kt . The ﬁrm’s objective is to maximize the present discounted value of the stream of proﬁts distributed to its owners. ∞ max E0 t=0 βt λt πt .Lecture in Open Economy Macroeconomics. for otherwise the ﬁrm would incur in unnecessary ﬁnancial costs.17) to eliminate πt from the ﬁrm’s objective function the ﬁrm’s problem can be stated as choosing processes for at . which would be suboptimal. λ0 We use the household’s marginal utility of wealth as the stochastic discount factor because households own domestic ﬁrms. . Chapter 5 153 We will limit attention to the case in which the interest rate is positive at all times. subject to a no-Ponzi-game borrowing constraint of the form at+j j d s=0 Rt+s j→∞ lim Et ≤ 0. ht . 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 . (5.

For instance. as well as the amount of debt intermediated by local banks. dt represents the country’s net debt position. We assume that ﬁrms start out with no liabilities. This distortion is d larger the larger the opportunity cost of holding working capital. Then. η. The precise form taken by this wedge depends on the particular timing assumed in modeling the use of working capital.17).154 Mart´ Uribe ın The ﬁrst-order conditions associated with this problem are (5. (5.9 We also observe that any process at satisfying equation (5. (5. (Rt − d 1)/Rt . 9 . and d Rt − 1 d Rt Fh (kt . ht ) − wt ht 1 + η − u t kt In this case.17) and the ﬁrm’s no-Ponzi-game constraint is optimal. with distributed proﬁts given by d Rt − 1 d Rt πt = F (kt . 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). an optimal plan consists in holding no liabilities at all times (at = 0 for all t ≥ 0). the no-Ponzi-game constraint holding with equality.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.18) Fk (kt . Here we adopt the shopping-time timing. ht ) = wt 1 + η (5. ht ) = ut . or the higher the intensity of the working capital constraint. under a cashd in-advance timing the wedge takes the form 1 + η(Rt − 1). Alternative assumptions give rise to diﬀerent speciﬁcations of the wedge.10).

It turns out that us up to ﬁrst order.d.031. but also their contribution to business cycles in emerging economies.Lecture in Open Economy Macroeconomics.1) and is given by ˆ ˆ ˆ us ˆ us Rt = 0. one only needs to know the laws of motion of Rt and Rt to construct the coeﬃcients of the theoretical MA(∞) representation. disturbance with mean zero and standard deviation 0.4. etc.29tbyt − 0. This is because we empirically identiﬁed not only the distribution of the two shocks we wish to study. investment. Chapter 5 155 5.61ˆt−1 + 0.35Rt−1 − 0. In formal terms. the variable tbyt stands for the trade balance- .11ˆt − 0.63Rt−1 + 0.19tbyt−1 + r r t.12ˆt−1 y y ı (5. 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. we can generate the corresponding theoretical MA(∞) representation and compare it to its empirical counterpart.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 (5.79ˆt + 0.i.20) ı + 0.50Rt + 0. As indicated earlier. We therefore close our model by introducing the law of motion of the country interest rate Rt . where is an i. So using the economic model.).

22) where 0.20) us involves the world interest rate Rt . (5. innovation with mean zero and standard deviation 5. ht+1 . (5. ˜ t ≥ −1. yt . is an i. However.4 Equilibrium. it follows that both measures of private consumption are identical up to ﬁrst order.23) An equilibrium is a set of processes ct+1 . 10 . 1. Rt . λt . Accordingly.21) Because the process for the country interest rate deﬁned by equation (5.d. where the debtadjustment cost Ψ(dt ) are incurred by households. Thus. it . 3. ut . and νit for i = 0. or c t = ct . we estimate Rt as follows an AR(1) process and obtain ˆ us ˆ us Rt = 0. 1. Functional Forms. tbyt . sit+1 for ˜ d i = 0. and Parameter Values In equilibrium all households consume identical quantities. 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.007.156 to-GDP ratio and is given by:10 yt − ct − it − Ψ(dt ) . kt+1 . the national income and product accounts would measure private consumption as ct + Ψ(dt ) and not simply as ct . yt Mart´ Uribe ın tbyt = (5. wt .83Rt−1 + rus t rus t . 2 satisfying In an economy like the one described by our theoretical model. dt .4.i. ct+1 . Rt . individual consumption equals average consumption across households. which is assumed to be an exogenous random variable. 2. ¯ because of our maintained assumption that Ψ (d) = 0. qt .

given c0 . h0 . we set γ = 2. Chapter 5 157 conditions (5.18)-(5.16). and α = . 2 φ > 0. Φ(x) = x − φ (x − δ)2 . y−1 . 2 ψ ¯ (d − d)2 . both of which are in line with actual data. capital adjustment costs. i−1 .Lecture in Open Economy Macroeconomics. the time unit is meant to be one quarter. η. There is no readily available estimates for these parameters for emerging economies. and Evans (2001) and consists in choosing values . h) = k α h1−α . c − µ˜ − ω −1 hω c 1−γ 1−γ U (c − µ˜. We therefore proceed to estimate them. Our estimation procedure follows Christiano.3)-(5. a standard value in business-cycle studies. and (5. ψ. and µ. There remain four parameters to assign values to. F (k. h) = c −1 . Ψ(d) = In calibrating the model. and debt adjustment costs. technology.455. all holding with equality. Eichenbaum.22) describing the evolution of the world interest rate. (5. c−1 . φ. We adopt the following standard functional forms for preferences. i0 . We set the depreciation rate at 10 percent per year. We set the steadystate real interest rate faced by the small economy in international ﬁnancial markets at 11 percent per year. the processes for the exogenous innovations rus t and r. t and equation (5. This value is consistent with an average US interest rate of about 4 percent and an average country premium of 7 percent.32.21). Following Mendoza (1991). ω = 1.23).

the trade balance.2. φ.8.µ} [IRe − IRm (ψ. to 2 shocks (the US-interest-rate shock and the country-spread shock).2 and the corresponding impulse responses implied by the model. η = 1.11 In our exercise we consider the ﬁrst 24 quarters of the impulse response functions of 4 variables (output. φ. let IRe denote the 192×1 vector of estimated impulse response functions and IRm (ψ. and µ = 0. µ) is given by argmax{ψ.2. φ = 72. φ. whereas in Christiano et al. η. investment. φ. On the other hand.e. and the country interest rate). IR where ΣIRe is a 192×192 diagonal matrix containing the variance of the impulse response function along the diagonal. µ)]. Then our estimate of (ψ. which is a function of the four parameters we seek to estimate. the set of estimated impulse responses used in the estimation procedure are originated by a single shock. η.φ.η. Speciﬁcally. η. The resulting parameter estimates are ψ = 0. µ) the corresponding vector of impulse responses implied by the theoretical model. 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. capital adjustment costs appear as more signiﬁcant.158 Mart´ Uribe ın for the four parameters so as to minimize the distance between the estimated impulse response functions shown in ﬁgure 5. is that here the estimation procedure requires ﬁtting impulse responses to multiple sources of uncertainty (i. we are setting 4 parameter values to match 192 points. The implied debt adjustment costs are small. Thus. η. For A key diﬀerence between the exercise presented here and that in Christiano et al. 11 . country-interest-rate shocks and world-interest-rate shocks. This matrix penalizes those elements of the estimated impulse response functions associated with large error intervals. µ)] Σ−1e [IRe − IRm (ψ. For ¯ example.00042..

96 percent.204 1.2 2.6 months of wage payments. investment. a 10 percent increase in investment for one year produces an increase in the capital stock of 0. 5.8 0.html.973 2 0. Constantinides. and the country interest rate. Chapter 5 159 Table 5.edu/~mu2166/uribe_yue_jie/uribe_yue_jie. the trade balance-to-GDP ratio.Lecture in Open Economy Macroeconomics. .1 gathers all parameter values. In the absence of capital adjustment costs.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.025 1.32 72. 1990). Table 7.12 The left col12 The Matlab code used to produce theoretical impulse response functions is available on line at http://www.455 0.00042 0.columbia.2: Parameter Values Symbol β γ µ ω α φ ψ δ η R Value 0. Finally. starting in a steady-state situation. The estimated value of η implies that ﬁrms maintain a level of working capital equivalent to about 3.g..6 depicts impulse response functions of output.88 percent. 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 5. the capital stock increases by 0.

2 0 0 Response of Country Interest Rate to εr 5 10 15 20 Estimated IR -x-x.6 0.2 0.6: Theoretical and Estimated Impulse Response Functions Response of Output to εrus 0 0 −0.8 0.3 1 0.4 0.1 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 0.5 0 5 10 15 20 −0.160 Mart´ Uribe ın Figure 5.3 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.Model IR 2-std error bands around Estimated IR Note: The ﬁrst column displays impulse responses to a US interest rate shock ( rus ). .5 −1 −1. and the second column displays impulse responses to a country-spread shock ( r ).5 0.2 −0.5 5 10 15 20 Response of TB/GDP to εrus 2 0.4 1.1 −0.5 0.

Solid t lines display empirical impulse response functions.1) is indeed successful in isolating country-spread shocks and US-interest-rate shocks from the data. the trade balance improves in response to either shock. tbyt . Third. the country interest rate displays a monotonic response to a country-spread shock. ˆt . We therefore conclude that the scheme used to identify the parameters of the VAR system (5. yt .2 and 5. the exogenous country-spread shock r t (the sentiment component).3. ˆt−1 . Crossed lines depict theoretical impulse response functions. the country spread St = Rt − Rt moves in response to four types of variable: its own lagged value St−1 (the autoregressive component). Fourth. and current and past values of a ˆ ˆ set of domestic endogenous variables. yt−1 . Chapter 5 umn shows impulse responses to a US-interest-rate shock ( rus ). 5.20). The model replicates three key qualitative features of the estimated impulse response functions: First. the country interest rate displays a hump-shaped response to an innovation in the US interest rate.6 The Endogeneity of Country Spreads According to the estimated process for the country interest rate given in ˆ ˆ ˆ us equation (5. This information is reproduced from ﬁgures 5. output and investment contract in response to either a US-interest-rate shock or a country-spread shock.Lecture in Open Economy Macroeconomics. current us us and past US interest rates Rt and Rt−1 ). Second. and broken lines depict the associated two-standard-error bands. A natural ˆ ˆ ı ı question is to what extent the endogeneity of country spreads contributes . tbyt−1 . t 161 and the right column shows impulse responses to a country-spread shock ( r ).

63Rt−1 − 0. This parametrization has two properties of interest.162 to exacerbating aggregate ﬂuctuations in emerging countries. 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..29tbyt − 0. which is the negative of the coeﬃcient on the lagged country interest rate. given the past value ˆ ˆ ˆ us of the country spread. Second. To this end.19tbyt−1 + r t. The process for the US interest rate is assumed to be unchanged (i.79ˆt + 0.24) ı + 0. the current country spread. St−1 = Rt−1 − Rt−1 .12ˆt−1 y y ı (5. the above speciﬁcation of the country-interest-rate process preserves the dynamics of the model in response to country-spread shocks. Mart´ Uribe ın We address this question by means of two counterfactual exercises. St . it implies that.22)). The ﬁrst exercise aims at gauging the degree to which country spreads amplify the eﬀects of world-interest-rate shocks.61ˆt−1 + 0. given by equation (5. t . we assume that the process for the country interest rate is given by ˆ ˆ ˆ us ˆ us Rt = 0. does not directly depend upon current or past values of the US interest rate.63. Speciﬁcally. The reason is that doing so would alter the estimated process of the country spread shock r.63Rt−1 + Rt − 0. First.e.11ˆt − 0.20) only in that the coeﬃcient on the contemporaneous US interest rate is unity and the coefﬁcient on the lagged US interest rate equals -0. We note that in conducting this and the next counterfactual exercises we do not reestimate the VAR system. This process diﬀers from the one shown in equation (5.

784 0. Chapter 5 163 Table 5. or tby us ˆ.819 0.623 4.3: Endogeneity of Country Spreads and Aggregate Instability Std.420 0.319 0.983 Note: The variable S denotes the country spread and is deﬁned as S = R/Rus .639 ˆ ı 2.110 0.24)) 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. Dev. the one given in equation (5.429 4.663 0.24) and in each case compute a variance decomposition of output and other endogenous variables of interest.547 1.20) or the one given in equation (5.24)? To answer this question. due to rus Std.469 0. we feed the theoretical model ﬁrst with equation (5.983 S 2. A hat on a variable denotes log-deviation from its non-stochastic steady-state value.885 0.866 1.429 3.515 0.819 0.3. Namely..347 1.e.e. 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.Lecture in Open Economy Macroeconomics.20) and then with equation (5.580 1.175 tby 1. when Rt follows .663 0.622 2. under the process for Rt given in equation (5.547 1. Dev. changes in the endogenous and the sentiment components of the country spread process.429 3.640 4.. due to r Baseline No yt ˆ Baseline No yt ˆ us ˆ.429 4.446 R 3.245 0. This would amount to introducing two changes at the same time. We ﬁnd that when the country spread is assumed not to respond directly to variations in the US interest rate (i. The result is shown in table 5.509 1. or tby Variable Model No R ı Model No R ı y ˆ 1.

20) with the process ˆ ˆ ˆ us ˆ us Rt = 0.25) Table 5. We ﬁnd that the equilibrium volatility of output.164 Mart´ Uribe ın the process given in equation (5.20)).63Rt−1 + 0. Ideally. To this end.3 displays the outcome of this exercise.50Rt + 0. 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. we replace the process for Rt given in equation (5. 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. investment. 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. such as output and the external accounts.35Rt−1 + r t. (5. t t 13 . 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.13 We conclude that the fact that country spreads respond to the state of domestic business conditions signiﬁcantly exacerbates aggregate instability in emerging countries. This indicates that the aggregate eﬀects of US-interest-rate shocks are strongly ampliﬁed by the dependence of country spreads on US interest rates. Speciﬁcally.

Explaining this striking contrast between emerging and industrialized economies is at the top of the research agenda in small-open-economy macroeconomics.Chapter 6 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. 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. and (3) that the trade-balance-to-output ratio is signiﬁcantly more countercyclical in emerging markets than it is in developed countries. but less volatile than output in developed countries. Broadly. the available theoretical explanations fall into two categories: One is that emerging market economies are subject to more volatile shocks than are developed countries. (2) that private consumption spending is more volatile than output in emerging countries. This and the following two chapters provide a progress report on the identiﬁcation and quantiﬁcation of exogenous sources of busi165 .

the real exchange rate (rer) is deﬁned as the relative price of consumption in terms of importable goods. let Ptc denote a domestic CPI index. the terms of trade have the potential to be an important source of business cycles in developing countries. Because primary commodities display large ﬂuctuations over time. and the real exchange rate in the postwar era. In the table. It follows that for many small countries. 6. 6. The present chapter concentrates on terms-of-trade shocks. the components of aggregate demand. emerging countries specialize in exports of a few primary commodities.2 Empirical Regularities Table 6. Letting Ptx and Ptm denote indices of world prices of exports and imports for a particular country. A number of empirical regularities emerge from the table: . Then the real exchange rate is given by Ptc /Ptm . emerging countries are normally small players in the world markets for the goods they export or import. agricultural products. the terms of trade can be regarded as an exogenous source of aggregate ﬂuctuations. or oil.1 displays summary statistics relating the terms of trade to output. Speciﬁcally. the terms of trade for that country are given by tott ≡ Ptx /Ptm . At the same time.1 Deﬁning the Terms of Trade The terms of trade are deﬁned as the relative price of exports in terms of imports.166 Mart´ Uribe ın ness cycles in small open economies. Typically. such as metals.

33 0.0 0.70 0. Note: tot. and Japan. output. i) ρ(tot.77 0.38 0.28 0. UK. France.0 0. Peru. Canada. The group of developed countries is formed by the US. Germany. Venezuela. investment. Chapter 6 167 Table 6. c) ρ(tot. Algeria.40 0. Indonesia.42 Source: Mendoza (1995).74 0. The sample is 1955 to 1990 at annual frequency. tott−1 ) σ(tot)/σ(y) ρ(tot. The group of oil-exporting countries is formed by Mexico.47 0. . respectively. and tb denote. The group of developing countries is formed by Argentina.67 0. consumption.1: The Terms of Trade and Business Cycles Summary Statistic σ(tot) ρ(tott .24 0. Saudi Arabia. and Thailand. and Nigeria.52 0. tables 1 and 3-6. y. i.78 0.39 0. Mexico.45 0.07 Oil Exporting Countries 18.Lecture in Open Economy Macroeconomics. Korea. Venezuela.40 0.70 Developing Countries 10.19 0. Philippines. The terms of trade are measured as the ratio of export to import unit values with 1900=100. c. Cameroon. and the trade balance. tb) ρ(tot. Chile. rer) Developed Countries 4. All other variables are measured per capita at constant import prices. Italy. Taiwan.34 0. Brazil.30 0. India. the terms of trade. All variables are expressed in percent deviations from a HP trend constructed using a smoothing parameter of 100. y) ρ(tot.50 1.

1 TOT-TB Correlation: Two Early Explanations The eﬀects of terms-of-trade shocks on the trade balance is an old subject of investigation. This correlation is high for developed countries but almost nil for less developed countries. Later in this chapter. They are twice as procyclical in developed countries as in developing countries. The terms of trade are procyclical. 2. More than half a century ago. we attempt to answer this question by combining the empirical information contained in table 6. and 150 percent as volatile as output in oil-exporting countries.2. 3.168 Mart´ Uribe ın 1. The terms of trade are half as volatile as output in developed countries.1 with the theoretical predictions of a fully speciﬁed dynamic general equilibrium model of the open economy.1 is mute on the importance of terms of trade shocks in explaining movements in aggregate activity. The information provided in table 6. Harberger (1950) and Laursen . 6. The terms of trade display positive but small serial correlation. 4. 5. and they are almost twice as volatile in oil-exporting countries as in developing countries. 6. 75 percent as volatile as output in developing countries. The correlation between the terms of trade and the trade balance is positive but small. The terms of trade are twice as volatile in emerging countries as in developed countries. The terms of trade are positively correlated with the real exchange rate.

within the context of a keynesian model. when Obstfeld (1982) and Svensson and Razin (1983). Let us look at the HLM and ORS eﬀects in some more detail. Consider the following behavioral equations deﬁning the dynamics of each component of aggregate demand. using a dynamic optimizing model of the current account.Lecture in Open Economy Macroeconomics. This view became known as the Obstfeld-Razin-Svensson (ORS) eﬀect. where yt denotes output. This view remained more or less unchallenged until the early 1980s. ct denotes private consumption.. Chapter 6 169 and Metzler (1950) formalized. gt denotes public consumption. 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.e. and mt denotes imports. xt denotes exports. This conclusion became known as the Harberger-LaursenMetzler (HLM) eﬀect. concluded that the eﬀect of terms of trade shocks on the trade balance depends crucially on the perceived persistence of the terms of trade. it denotes private investment. Public consumption and . 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. the conclusion that rising terms of trade should be associated with an improving trade balance.

Imports are assumed to be ¯ proportional to output. we will assume that these two varibles are constant over time and given by gt = g ¯ and it = ¯ i. the value of exports in terms of importables. Thus. . respectively. with µ ∈ (0. is given by xt = tott qt .170 Mart´ Uribe ın private investment are assumed to be independent of output. the parameters α and µ are referred to as the marginal propensities to consume and import. In the jargon of the 1950s. The quantity of goods exported in period t is denoted by qt . where g and ¯ are parameters. 1) and c > 0 are parameters. xt . ¯ with α ∈ (0. For simplicity. mt = µyt . Consumption is assumed to be ¯ i an increasing linear function of output ct = c + αyt . 1). respectively. Output as well as all components of aggregate demand are expressed in terms of import goods. whereas the term c + g + ¯ is referred to as the autonomous component of ¯ ¯ i domestic absorption.

Chapter 6 171 where tott denotes the terms of trade. and mt from the national income identity. is assumed to be constant and vigen by qt = q . it . ¯ where q is a positive parameter. and the quantity of goods exported. as a larger fraction of income is used to buy foreign goods. this theory implies that an improvement in the terms of trade (an increase in tott ) gives rise to an expansion in the trade surplus. 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. The terms of trade are assumed to evolve exogenously. 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. 1+µ−α 1+µ−α tbt = Clearly. µ. the sign of the eﬀect . 1+µ−α Letting tbt ≡ xt − mt denote the trade balance. xt .Lecture in Open Economy Macroeconomics. and the smaller is the marginal propensity to consume α. Using the behavioral equations to eliminate ¯ ct . q . the smaller is the marginal propensity ¯ to import. It is worth noting that in the context of this model. Similarly. This positive relation between the terms of trade and the trade balance is stronger the larger is the volume of exports. and solving for output yields yt = ¯ c + g + ¯ + tott q ¯ ¯ i . gt . qt . we can write 1−α µ(¯ + g + ¯ c ¯ i) ¯ tott q − .

open. so it takes the evolution of tott as exogenous. We can then use the results derived in chapter 2 to draw the . or the terms of trade. Consider the small. therefore exports the totality of its endowment and imports the totality of its consumption. Let tott denote the relative world price of exported goods in terms of imported goods. The household. The economy is small in world product markets. with tott taking the place of yt . This is an economy inhabited by an inﬁnitely lived representative household with preferences described by the intertemporal utility function given in (2. The borrowing constraint given in (2. Assume for simplicity that the endowment of exportable goods is constant and normalized to unity. This is the main contrast with the Obstfeld-Razin-Svensson eﬀect.3) prevents the household from engaging in Ponzi games. Suppose that the good the household consumes is diﬀerent from the good it is endowed with.172 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. 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.5). The resource constraint is then given by dt = (1 + r)dt−1 + ct − tott . yt = 1 for all t. The model is therefore identical to the stochastic-endowment economy studied in chapter 2. endowment economy studied in chapter 2.

East Asia. This conclusion can be extended to a model with endogenous labor supply and capital accumulation. and the Middle East. In this case. Africa.Lecture in Open Economy Macroeconomics. Is the ORS eﬀect borne out in the data? If so. Figure 6. In this case.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. the productivity shock At can be interpreted as a terms-of-trade shock. an improvement in the terms of trade induces a trade balance deﬁcit. When terms of trade are nonstationary. displays no pattern. Agents save in order to ensure higher future consumption. A simple way to do this is to modify the RBC model of chapter ?? by assuming again that households do not consume the good they produce. Chapter 6 173 following conclusion: if the terms of trade are stationary then an increase in the terms of trade produces an improvement in the current account. the value of income is expected to grow over time. The cloud of points. The . An increase in the terms of trade produces an improvement in the trade balance if the terms of trade shock is transitory. 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. The 30 observations were taken from Mendoza (1995). including the G7 countries and 23 selected developing countries from Latin America. but as the serial correlation of the terms of trade shock increases. shown with circles. table 1. so agents can aﬀord assuming higher current debts without sacriﬁcing future expenditures. an improvement in the terms of trade can lead to a deterioration in the current account driven by investment expenditures.

and Argentina from the sample and is given by corr(T B.6 0. The dashed-dotted line is the OLS ﬁt after eliminating the G7 countries. 3 countries from the Middle East (Israel. The sample includes the G-7 countries (United States. Cameroon. Zaire.6 Argentina 0. and Egypt).4 −0.14ρ(T OT ). Italy.4 TOT Serial Correlation 0.6 −0.3 0. Nigeria. Peru. The dashed line is the OLS ﬁt after eliminating Argentina from the sample and is given by corr(T B.174 Mart´ Uribe ın Figure 6. Mexico. 6 countries from Latin America. The TOT serial correlation and the TB-TOT correlation are computed over the period 1955-1990.12ρ(T OT ). Canada. Venezuela). T OT ) = 0. .1 0 0. 6 countries from Asia (Taiwan. India. Korea.7 Source: Mendoza (1995).28+0.1 0. Saudi Arabia. and Japan). (Argentina. and Tunisia). France. Saudi Arabia.2 −0. Note: Each point corresponds to one country. Germany. The solid line is the OLS ﬁt and is given by corr(T B.03ρ(T OT ).8 0.23 + 0. and 8 countries from Africa (Algeria. and Thailand). Morocco. Kenya. Brazil. Philippines.2 0. T OT ) = 0. Sudan. United Kingdom. Chile. T OT ) = 0.35−0. table 1.2 0 −0.1: TOT Persistence and TB-TOT Correlations 0. Indonesia.4 TB−TOT Correlation 0.5 0.

the only country in the sample with a negative serial correlation of the terms of trade. Argentina.14. The ORS 1 Indeed. the largest economies in the world. 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.1. Does this conclusion suggest that the empirical evidence presented here is against the Obstfeld-Razin-Svensson eﬀect? Not necessarily. the terms of trade are not likely to be exogenous. . and Saudi Arabia. We conclude that the observed relationship between the TB-TOT correlation and the persistence of TOT is close to nil. For these countries.1 are likely to be large players in the world markets for the goods and services they import and/or export.03 and is shown with a dash-dotted line in ﬁgure 6. however. that the negative slope in the OLS regression is driven by a single observation. displays a small negative slope of -0.1. Eliminating these 8 countries (as well as the outlier Argentina) from the sample. The sign of the slope is indeed in line with the ORS eﬀect: As the terms of trade shocks become more persistent.1 The corresponding ﬁtted relationship is shown with a broken line on ﬁgure 6.12. Eliminating Argentina from the sample one obtains a positive OLS slope of 0. The ﬁtted line using this reduced sample has a negligible slope equal to 0. Countries in this group would include all of the G7 nations. Chapter 6 175 OLS ﬁt of the 30 points. a major oil exporter. they should be expected to induce a smaller response in the trade balance. shown with a solid line. It is apparent in the graph. Argentina is the only country whose elimination from the sample results in a positive slope.Lecture in Open Economy Macroeconomics. A number of countries in ﬁgure 6.

High world interest rates may be associated with depressed economic activity in developed and emerging economies alike. suppose that domestic technology shocks are correlated with technology shocks in another country or set of countries. Not controlling for these shocks may result in erroneously attributing part of their eﬀect on the trade balance to the terms of trade. some of these shocks may directly aﬀect both the trade balance and the terms of trade. Under this scenario.176 Mart´ Uribe ın eﬀect requires isolating the eﬀect of TOT shocks on the trade balance. In this case. as a result. the terms of trade and the trade balance are moving at the same time. but attributing all of the movement in the trade balance to changes in the terms of trade would be clearly misleading. Data analysis based purely on statistical methods will in general not result in a successful iden- . The raw data is in principle driven by a multitude of shocks. 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. At the same time. An important step in the process of isolating terms-of-trade shocks—or any kind of shock. As another example. Moreover. high world interest rates are associated with contractions in aggregate demand and improvements in the trade balance in emerging countries. with deteriorated terms of trade for the emerging countries producing those commodities. and. judging the empirical validity of the ORS eﬀect only on the grounds of raw correlations would be misplaced. of which the terms of trade is just one. for that matter—is identiﬁcation. In turn. A case in point is given by world-interest-rate shocks. low levels of economic activity in the developed world are likely to be associated with a weak demand for primary commodities.

3 Terms-of-Trade Shocks in an RBC Model Consider expanding the real-business-cycle model of chapter ?? to allow for terms-of-trade shocks. and a sector producing nontradable goods. Economic theory must be used be at center stage in the identiﬁcation process. 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. 6. a sector producing exportable goods. which follows Mendoza (1995).3.1 Households This block of the model is identical to that of the RBC model studied in chapter ??. The following exercise. Chapter 6 177 tiﬁcation of technology shocks. 6. represents an early step in the task of identifying the effects of terms-of-trade shocks on economic activity in emerging economies. 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. The main diﬀerence with the model of chapter ?? is that the model studied here features three sectors: a sector producing importable goods. The household block of the model is identical to that of the standard RBC model studied in chapter ??. we follow the work of Mendoza (1995). Similarly. A nontradable good is a good that is neither exportable nor importable. The economy is populated by a large number of identical house- . In doing this.Lecture in Open Economy Macroeconomics.

which they rent at the rate ut . and U is a period utility function taking the form [c (1 − h)ω ]1−γ . ht ). (6.3) . 1−γ U (c.178 holds with preferences described by the utility function ∞ Mart´ Uribe ın E0 t=0 θt U (ct . ht denotes labor eﬀort. The stock of capital evolves according to the following law of motion: kt+1 = (1 − δ)kt + it − Φ(kt+1 − kt ).1) where ct denotes consumption. kt . We established in chapter ?? 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. (6. h) = [1 + c(1 − h)ω ]−β . 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 . (6. ht ). Households oﬀer labor services for a wage wt and own the stock of capital.2) where the function β is assumed to take the form β(c.

The function Φ introduces capital adjustment costs.2) and (6. wt . Under these assumptions.Lecture in Open Economy Macroeconomics. which is assumed to be an importable good.2)(6.1) subject to (6. 1] denotes the capital depreciation rate. Households are subject to a no-Ponzigame constraint of the form Et dt+j ≤ 0.4). 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 ∗ . and ut are expressed in terms of importable goods. 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. Letting θt ηt and θt λt denote the Lagrange multipliers on (6.5) The household seeks to maximize the utility function (6. t which serve the role of numeraire. the period budget constraint of the household can be written as dt = (1 + r ∗ )dt−1 + pc ct + it − wt ht − ut kt . the ﬁrst-order conditions of the household’s maximization problem .5). Chapter 6 179 where it denotes gross investment. j→∞ (1 + r ∗ )j lim (6.4) The relative prices pc . the steady-state level of capital is not aﬀected by the presence of adjustment costs. t (6. As discussed in chapter ??. and is assumed to satisfy Φ(0) = Φ (0) = 0 and Φ > 0. The parameter δ ∈ [0.

ht+1 ) (6. and nontradable consumption goods.2).11) with µ > −1. (6. Formally. These ﬁrms operate a CES production function that takes tradable consumption goods.4).5) holding with equality. as inputs.2 Production of Consumption Goods The consumption good. respectively. cN .3. The ﬁrstorder conditions associated with this proﬁt-maximization problem are (6. Firms operate in perfectly competitive product and input markets. cT .6) − Uh (ct .180 are (6. ct . (6.7) (6. the relative prices of tradable and t t nontradable consumption goods in terms of importable goods. ht ) − ηt βc (ct . ht )(1 + rt )Et λt+1 (6.11) .10) 6. ht ) = λt wt λt = β(ct . They choose output and inputs to maximize proﬁts. which are given by pc ct − pT cT − pN cN . ht ) + ηt βh (ct . is produced by domestic ﬁrms. ht )Et λt+1 ut+1 + 1 − δ + Φ (kt+2 − kt+1 ) (6. and Mart´ Uribe ın Uc (ct . ht+1 ) + Et ηt+1 β(ct+1 . ht ) = λt pc t (6. t t (6. t t ct = [χ(cT )−µ + (1 − χ)(cN )−µ ]−1/µ .8) λt [1 + Φ (kt+1 − kt )] = β(ct . t t t t t where pT and pN denote.9) ηt = −Et U (ct+1 .

then the share of tradables in total consumption. t t t t t . t t t (6. cM . Firms are competitive and aim at maximizing proﬁts. (6. via a Cobbt t Douglas production function. 1) is a parameter. t t increases. given by pT cT /(pc ct ). From the second optimality condition. cT = (cX )α (cM )1−α .3. cX .12) 1 1+µ . 6.13) 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). one observes that if the elasticity of substitution between tradables and nontradables is less than unity (or µ > 0). Chapter 6 and cN t = cT t ct = cT t 1−χ χ 1 χ 181 1 1+µ pT t N pt pT t pc t 1 1+µ . 1 1+µ (6. pT /pc . are produced using importable t consumption goods.Lecture in Open Economy Macroeconomics. t t t increases as the relative price of tradables in terms of consumption. Formally. and exportable consumption goods.14) where α ∈ (0.3 Production of Tradable Consumption Goods Tradable consumption goods. denoted cT . which are given by pT cT − pX cX − cM .

3. we have that the relative price of importables in terms of the numeraire is always unity (pM = 1). T cT pt t (6. t (6. 6.15) (6.16) 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.18) . and Nontradable Goods Exportable and importable goods are produced with capital as the only input.4 Production of Importable. 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. The optimality t conditions associated with this problem are (6. the three production technologies are given by X X yt = AX (kt )αX . or the terms of trade. respectively. Exportable. Note that because the importable good plays the role of numeraire.182 Mart´ Uribe ın where pX denotes the relative price of exportable goods in terms of imt portable goods. Formally. t (6.14) and: pX cX t t =α pT cT t t cM t = 1 − α.17) M M yt = AM (kt )αM .

N . N pN yt t (6. and the vari- able hN denotes labor services employed in the nontradable sector. The factors Ai t denote exogenous and stochastic technology shocks in sectors i = X. Firms t demand input quantities to maximize proﬁts.Lecture in Open Economy Macroeconomics. which are given by X M N X M pX yt + yt + pN yt − wt hN − ut (kt + kt ). X pX yt t M ut kt = αM . yt denotes output of imN portable goods.19) X M where yt denotes output of exportable goods. M yt (6. t t 183 (6. . i The variable kt denotes the capital stock in sector i = X. Chapter 6 and N yt = AN (hN )αN .21) and wt hN t = αN .22) According to these expressions. and yt denotes output of nontradable goods. input shares are constant.20) (6. M. and as a consequence of the assumption of Cobb-Douglas technologies. t t t The optimality conditions associated with this problem are X ut kt = αX . M .

(2) The sectorial productivity shocks are assumed to be perfectly correlated.5 Market Clearing In equilibrium. 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 . . (6.3. t (6. Mendoza (1995) imposes four restrictions on the joint distribution of the exogenous shocks: (1) all four shocks share the same persistence. t t t (6.24) ht = hN . and the terms of trade. labor. t and N cN = yt . X M kt = kt + kt . the markets for capital. (3) The technology shocks aﬀecting the production of importables and exportables are assumed to be identical. and nontradables must clear. and nontradable). (4) Innovations to productivity shocks and terms-of-trade shocks are allowed to be correlated.184 Mart´ Uribe ın 6.6 Driving Forces There are four sources of uncertainty in this economy: One productivity shock in each of the three sectors (importable. We assume that all shocks follow autoregressive processes of order one.25) Also.3. That is.23) (6. exportable.26) 6.

In calibrating the driving forces of the developed-country version of the model. it . T 2 T νt ∼ N (0. pc . We are now ready to deﬁne a competitive equilibrium. Chapter 6 These assumptions give rise to the following laws of motion: p t.7 Competitive Equilibrium A stationary competitive equilibrium is a set of stationary processes {ct . AM . the parameters ρ and σ p are set to match the average serial cor- . kt . AN . and the other matching key macroeconomic relations in developing countries.3. νt ) = 0. yt . σν T ). 6. t N t = ψN T t .Lecture in Open Economy Macroeconomics. given the initial conditions t=0 k0 and d−1 and the exogenous processes {AX . wt . yt . N t . cN . yt . t t t t t=0 6.6)-(6. cM . kt . t X M N X M cX . = ψT p t T + νt .3.8 Calibration Mendoza (1995) presents two calibrations of the model.3) and (6.26). ηt . ht . pT . kt . T t . ut . t t t t t t t λt }∞ satisfying equations (6. p t 185 ln pX = ρ ln pX + t t−1 ∼ N (0. hN . pX }∞ . one matching key macroeconomic relations in developed countries. 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 . cT . dt . σ 2p ). E( p . pN .

This implies that ψT σ p = 0. This implies that 2 2 ψT σ 2p + σν T = 0. Based on correlations between Solow residuals and terms of trade in ﬁve developed countries. Using the information presented in table 1 of Mendoza (1995) yields ρ = 0.019 1 − ρ2 .047 1 − ρ2 . 2 2 ψT σ 2p + σν T Table 6.473.2 displays the parameter values implied by the above restrictions. and σ p = 0.019 and the volatility of the productivity shock in the nontraded sector at 0.014. . and ψN 2 2 ψT σ 2p + σν T = 0.165.165. Using estimates of productivity shocks in ﬁve industrialized countries by Stockman and Tesar (1995). Mendoza (1995) sets the correlation between the productivity shock in the exportable sector and the terms of trade at 0. This completes the calibration of the parameters deﬁning exogenous driving forces in the developed-country model.014 1 − ρ2 .186 Mart´ Uribe ın relation and standard deviation of the terms of trade for the group of G7 countries. Mendoza (1995) sets the volatility of productivity shocks in the importable and exportable sectors at 0.

5 0.41 -0.34 0.49 0.56 0.22 0.067 0.028 2.61 -0. Chapter 6 187 Table 6.04 0.011 0.017 0.04 0.47 0.3 2.009 .74 0.028 1.79 0.1 0.156 0.2: Calibration Parameter σp σν T ρ ψT ψN r∗ αX αM αN δ φ γ µ α ω β Developed Country 0.70 0.27 0.032 0.041 0.15 0.35 0.1 0.Lecture in Open Economy Macroeconomics.74 0.57 0.009 Developing Country 0.08 0.

12 1 − ρ2 .2.414.46. and σ p = 0.188 Mart´ Uribe ın In calibrating the driving forces of the developing-country model. This means that the parameter ψN takes the value 0.04 and their correlation with the terms of trade at -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). This completes the calibration of the exogenous driving forces for the developing-country version . This yields the restrictions: 2 2 ψT σ 2p + σν T = 0. This implies: ρ = 0.46 to match the observed average standard deviation of GDP and the correlation of GDP with TOT in developing countries. and ψT σ p = −0. 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. Mendoza sets the standard deviation of productivity shocks in the traded sectors at 0.04 1 − ρ2 .74 as in the developed-country model. 2 2 ψT σ 2p + σν T The implied parameter values are shown in table 6.

69 Model .58 .57 1 1 .14 .60 .18 -.42 .51 .39 .72 .3: Data and Model Predictions Variable σx σT OT G7 TOT Data 1 Model 1 TB Data 1.3.11 1 1 .57 Source: Mendoza DCs 1 1 1.94 ρxt .89 .28 .52 .0 RER Data 1.5 GDP Data 1.84 .39 . 2 See Mendoza.41 .32 1.66 .90 Model 2.26 .03 .70 .11 .18 .47 .82 .78 .62 .49 .xt−1 G7 DCs .62 Model 3.78 .59 Model 1.60 .79 (1995).01 I Data 1.81 .71 ρx.99 .86 C Data 1.69 .19 .49 .45 1 1 . of the model.GDP G7 DCs .43 .12 .68 .85 .38 . The following list highlights a number of empirical regularities and comments on the model’s ability to capture them.T OT G7 DCs 1 1 .41 .78 .80 .49 .32 -.96 .2 also displays the values assigned to the remaining parameters of the model.95 ρx.86 1.70 .25 .25 .44 .9 Model Performance Table 6.67 .78 .47 .32 .3 presents a number of data summary statistics from developed (G7) and developing countries (DCs) and their theoretical counterparts.33 .17 -.74 .Lecture in Open Economy Macroeconomics.38 .34 .37 .32 . Chapter 6 189 Table 6.23 .75 .2 6.08 .30 .27 1. . Table 6. 1995 for more details.25 1.47 1.44 Model .

The model fails to capture this fact. The failure of the model to capture the procyclicality of the trade balance in developed countries should be taken with caution. In the model. 6. 3. In the data. 4. although much less so in developing countries than in G7 countries. The terms of trade are less volatile than GDP. but underestimates the TB-TOT correlation. the terms of trade are procyclical. In the model. This fact is matched by construction. 2. trade-weighted average of foreign CPIs. for other authors estimate negative TB-GDP correlations for developed countries. In the data. particularly for developing countries. the trade balance is countercyclical in both. The terms of trade are positively correlated with the trade balance. 5. The model captures this empirical regularity. In the t . the real exchange rate (RER) is measured as the ratio of the domestic CPI to an exchange-rate-adjusted. recall that the parameter ρ is set to pin down the serial correlation of the terms of trade in developing and G7 countries. developed and developing countries. The trade balance is countercyclical in DCs but procyclical in G7 countries. The observed terms of trade are somewhat persistent. The model captures this fact relatively well.190 Mart´ Uribe ın 1. The model appears to overestimate the countercyclicality of the trade balance. the RER is deﬁned as the relative price of consumption in terms of importables and denoted by pc .

In the context of the model of this section.75 for both types of country. The reason for this . The RER is somewhat persistent (with a serial correlation of less than 0.45 for both developing and G7 countries). When output is measured in terms of domestic prices. In the model. Mendoza (1995) ﬁnds that when the volatility of all productivity shocks is set equal to zero in the developed-country version of the model.6 percent. the RER is highly persistent. one can run the counterfactual experiment of computing equilibrium dynamics after shutting oﬀ all sources of uncertainty other than the terms of trade themselves. shutting oﬀ the variance of the productivity shocks results in an increase in the volatility of output.1 percent to 3.10 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. Therefore. This is accomplished by setting σν T = ψT = 0. 6. the volatility of output deviations from trend measured at import prices falls from 4. the terms of trade explain about 66 percent of output movements.3. When the same experiment is performed in the context of the developingcountry version of the model. in the model the terms of trade explain about 88 percent of the volatility of output.Lecture in Open Economy Macroeconomics. although it overestimates somewhat the RER-GDP correlation. 7. with an autocorrelation above 0. The model captures this fact. the RER is procyclical. one must set all productivity shocks at their deterministic steady-state values. Chapter 6 191 data.

the contribution of the terms of trade will always be a nonnegative number no larger than 100 percent. Note that the parameter ψT must not be set to zero. One justiﬁcation for this classiﬁcation is that p t aﬀects both the terms of trade and sectoral total factor productivities. because the variance of output can be decomposed into a nonnegative fraction explained by p t T and a nonnegative part explained by νt . 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.1 Using the model presented in this section.192 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 6.and developing-country versions of the model? . Under this deﬁnition of shocks. 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 . compute a variance decomposition of output. the model with only terms of trade shocks results when σν T is set equal to zero. T while νt aﬀects sectoral total factor productivities but not the terms of trade. What fraction of output is explained by terms-oftrade shocks in the developed.2 that ψT < 0 for developing countries. An advantage of this approach is that. Exercise 6.

During downturns. One possible explanation for this phenomenon is simply that emerging countries are subject to larger shocks..e. One stresses the role of policy credibility. which exacerbates the expansionary phase of the cycle. A second explanation holds that emerging countries suﬀer from more severe economic and political distortions that amplify the cycles caused by aggregate shocks.Chapter 7 Overborrowing Business cycles in emerging countries are characterized by booms and contractions that are larger in amplitude than those observed in developed countries. which aggravates the contraction. 5. Chapters 4. It suggests that during booms emerging economies borrow and spend excessively (i. the countries ﬁnd themselves with too much debt and are forced to engage in drastic cuts in spending. The hypothesis of overborrowing belongs to this line of thought . can have deleterious eﬀects 193 . and 6 are devoted to evaluating this hypothesis. in eﬃciently). We will analyze in some detail two very diﬀerent theories of overborrowing. It shows that policies that in principle are beneﬁcial to the economy.

however. but do depend on their collective behavior. Suppose that the government. Its basic premise is quite simple. is the need for regulatory government intervention. 1 . announces the removal of a consumption tax. These variables (and therefore the cost of external borrowing) are not controlled by individual agents. or on the value of nontradable output. as banks tend to undertake immoderately risky projects and depositors have less incentives to monitor the quality of banks’ loan portfolios.1 Imperfect Policy Credibility The imperfect-credibility hypothesis is due to Calvo (1986. for example. the willingness of foreign lenders to provide funds to a given emerging economy might depend upon the countries’s aggregate level of external debt. The public. however. 1987. deposit guarantees induce moral hazard. A common theme in much of the overborrowing literature. 1988). In McKinnon’s model. For instance.194 Mart´ Uribe ın if the policymaker that is in charge of implementing it lacks credibility. possibly with good intentions. This externality can give rise to ineﬃcient borrowing. As a result deposit guarantees open the door to excessive lending and increase the likelihood of generalized bank failures. The second theory we will discuss relies in existence of pecuniary externalities in the market for external funds.1 7. which is not present in the aforementioned large-shock explanation of the emergingmarket business cycle. McKinnon’s (1973) model of deposit guarantees. that some theories argue that regulatory policy may be the cause rather than the remedy to overborrowing. We will keep this issue in mind as we present the diﬀerent theories. believes that the policy will be abandoned after a certain We note. In this brunch of the overborrowing literature. has been intensively used to understand overborrowing in the aftermath of ﬁnancial liberalization in the Southern Cone of Latin America in the 1970s. debt limits faced by individual agents depend upon variables that are exogenous to them but endogenous to the economy.

Here. they interpret the policy reform as being temporary. the consumption tax takes the form of inﬂation.Lecture in Open Economy Macroeconomics. In the aggregate. for instance. they take advantage of what they perceive to be a temporarily lower consumption tax and increase spending while the policy lasts. whereby purchases of consumption goods require holding money. Consider a perfect-foresight economy populated by a large number of inﬁnitely lived households with preferences described . Chapter 7 195 period. As a result. Calvo (1986). The particular application we analyze in this chapter is a trade reform and is due to Calvo (1988). the consumption tax also takes the form of inﬂation. it acts as an indirect tax on consumption. That is. Calvo (1987) applies the imperfect-credibility hypothesis to understand balance-of-payment crises. the spending boom is ﬁnanced by current account deﬁcits. In this case. At some point the euphoria ends (either because the tax cut is abandoned or because agents convince themselves that they are indeed permanent). and lack of credibility stems from the fact that agents disbelieve that the government will be able to cut deﬁcits to a level consistent with low inﬂation in the long run. imagine an economy in which households face a cash-inadvance constraint. and the current account experiences a possibly sharp reversal. To visualize inﬂation as a consumption tax. spending collapses. The imperfect-credibility theory of overborrowing can be applied to a variety of policy environments. Because inﬂation erodes the real value of money. studies the consequences of a temporary inﬂation stabilization program.

but can be exported. and U denotes a period utility function assumed to be increasing and strictly concave. t=0 where ct denotes consumption in period t. receive a constant endowment of goods. household receive a lump-sum transfer xt from the government each period. the terms of trade. β ∈ (0. In addition.196 by the utility function ∞ Mart´ Uribe ın β t U (ct ). The household’s sequential budget constraint is then given by dh = (1 + r)dh − y − xt + ct (1 + τt ). Debt is denominated in units of t importable goods and carries a constant interest rate r > 0. ct . 1) denotes a subjective discount factor. and must be imported from abroad. households are subject to the following borrowing constraint: dh t+j ≤ 0. Each period. For simplicity. y > 0. Consumption goods are not produced domestically. t→∞ (1 + r)j lim The fact that the period utility function is increasing implies that in the optimal plan the no-Ponzi-game constraint must hold with equality. is constant and normalized to unity. and to increase their asset position. we assume that the relative price of exportables in terms of importables. Imports are subject to a proportional tariﬀ τt . Households start each period with a stock of debt. t t−1 To prevent Ponzi games. carried over from the previous period. dh . This endowment is not consumed domestically. Households use their inherited wealth and income to purchase consumption goods. Com- .

we set β(1 + r) = 1. we assume that the subjective and pecuniary discount rates are identical. t t−1 rdg . Public outlays stem from interest payments. the optimality conditions associated with this problem are the intertemporal budget constraint and U (ct ) = λ0 (1 + τt ).Lecture in Open Economy Macroeconomics. dh }∞ to maxt+1 t=0 imize their lifetime utility function subject to this intertemporal budget constraint. That is.τt ct .1) Note that λ0 is determined in period zero but is constant over time. and the issuance of new public debt. and lump-sum transfers to households. the government has access to ﬁnancial markets. xt . 7.1 The Government Like households.1. dg − dg . The resulting sequential t−1 . Letting λ0 denote the Lagrange multiplier associated with the intertemporal budget constraint. Chapter 7 197 bining the sequential budget constraint and the no-Ponzi-game constraint holding with equality yields the following intertemporal budget constraint: ∞ (1 + r)dh −1 = t=0 1 1+r t [y + xt − ct (1 + τt )] To avoid inessential long-run dynamics. Its revenues originate in the collection of import tariﬀs. (7. The household’s problem consists in choosing sequences {ct .

198 budget constraint of the government is given by dg = (1 + r)dg − τt ct + xt . the government implements a permanent tariﬀ reform. t t−1 Mart´ Uribe ın The variable τt ct − xt represents the primary ﬁscal surplus in period t and the variable −rdg + τt ct − xt represents the secondary ﬁscal surplus. we assume that the ﬁscal policy ensures that dg t = 0. the government implements a temporary tariﬀ reform. Let dt ≡ dh + dg denote the country’s net foreign debt position. We compare two polar tariﬀ regimes. That is. In the second. This constraint states that the present discounted value of present and future primary ﬁscal surpluses must equal the government’s initial debt position. t→∞ (1 + r)t lim This condition together with the government’s sequential budget constraint implies the following intertemporal government budget constraint: ∞ (1 + r)dg = −1 t=0 1 1+r t (τt ct − xt ). We t−1 assume that ﬁscal policy is such that in the long run the government’s debt position does not grow in absolute value at a rate larger than the interest rate. Then. In one. t t combining the intertemporal budget constraints of the household and the government yields the folloiwng intertemporal economy-wide resource con- .

It follows directly from the eﬃciency condition (7. the present discounted value of the stream of current and future trade surpluses must equal the country’s initial net debt position.1. to a new level τ L < τ H . One reduces the tariﬀ permanently.2 Equilibrium Under A Permanent Trade Reform Suppose that in period 0 the government unexpectedly implements a permanent tariﬀ reform consisting in lowering τt from its initial level. (7. ¯ . the time path of τt is given by τt = τ L . Chapter 7 straint: ∞ 199 (1 + r)d−1 = t=0 1 1+r t (y − ct ).2) then implies ¯ that c is given by ¯ c = y − rd−1 . which we denote by τ H .Lecture in Open Economy Macroeconomics. A competitive equilibrium is a scalar λ0 and a sequence {ct }∞ satisfyt=0 ing (7. The focal interest of our analysis is to compare the economic eﬀects of two alternative commercial policy reforms.2). Let the equilibrium level of consumption be denoted by c.1) and (7. We begin with the analysis of a permanent trade tariﬀ reform.1) that in this case consumption is constant over time.2) According to this expression. t ≥ 0. 7. given the sequence of tariﬀs speciﬁed by the government’s tariﬀ policy and the initial debt position d−1 . Formally. The intertemporal resource constraint (7.

It is clear from equation (7.3 Equilibrium Under A Temporary Tariﬀ Reform Suppose now that in period 0 the government unexpectedly announces and implements a temporary trade liberalization. gives rise to the same level of consumption. the trade balance. then the permanent reduction in the import tariﬀ induces no changes in consumption (i. The announcement speciﬁes that the tariﬀ will be lowered from τ H to τ L between periods 0 and T − 1. the net foreign asset position. Here. T > 0 denotes the length of the commercial liberalization policy.e. Similarly. Any constant tariﬀ path.. Moreover.3) with τ L < τ H .200 Mart´ Uribe ın Note that the equilibrium level of consumption is independent of the level at which the government sets the import tariﬀ.1. the announced path of τt is given by τL for 0 ≤ t < T for t ≥ T τt = H τ . (7. implemented unexpectedly at time 0. if the economy was in a steady state before the permanent reform. ct equals c before and after the implementation of the ¯ permanent trade liberalization). and that in period T it will be raised permanently to its original level τ H . and the current account are all unaﬀected by the permanent trade liberalization. Formally. 7.1) that consumption is constant over the .

We can then write c1 = (1 + κ)c2 . Because the period utility function is concave. for someκ > 0.6). then we have that 1 + κ = 1+τ H 1+τ L 1/σ .6) The equilibrium values c1 and c2 are the solution to the two linear equations (7. (7. ¯ . Chapter 7 periods 0 ≤ t < T and t ≥ T .6) is c. We can therefore write the above expression as c = (1 − β T )c1 + β T c2 .2). U (c) = c1−σ /(1 − σ). the consumption stream must respect the intertemporal resource constraint (7. Intuitively.Lecture in Open Economy Macroeconomics. The left-hand side of equation (7. the level of ¯ consumption that results under a permanent trade reform. This restriction implies the following relationship between c1 and c2 : y − rd−1 = (1 − β T )c1 + β T c2 . Of course. For example. households substitute consumption in the low-tariﬀ period for consumption in the high-tariﬀ period. (7.5) The parameter κ is an increasing function of the intertemporal tariﬀ distortion (1 + τ H )/(1 + τ L ).5) and (7. if the period utility function is of the CRRA form.4) where c1 and c2 are two scalars determined endogenously.1) that c1 is greater than c2 . (7. We can therefore write 1 c 2 c 201 ct = for 0 ≤ t < T for t ≥ T . it follows from the eﬃciency condition (7.

Because households have concave preferences and because the present discounted value of wealth is the same in the constant-tariﬀ and the time-varying-tariﬀ economies. The planner’s objective function is to choose scalars c1 and c2 to maximize the household’s lifetime utility function evaluated at a consumption path given in (7.4) subject to the lifetime resource constraint (7. This result implies that the temporary tariﬀ reform induces households to engage in a consumption path that has the same present discounted value but is less smooth than the one associated with a constant tariﬀ. ¯ then the announcement of the temporary tariﬀ reform causes consumption to rise to c1 at time 0. ¯ This means that if the economy was at a steady state c before period t. it follows from this expression that c is a weighted ¯ average of c1 and c2 . consider the problem of a benevolent social planner trying to design a tariﬀ policy that maximizes welfare subject to the constraint that the policy must belong to the family deﬁned in (7. To state this more formally.3). we have that the time-varying tariﬀ economy must result in lower welfare for the representative household than the constant-tariﬀ economy.6). the planner’s problem is: 1 − βT βT U (c1 ) + U (c2 ) 1−β 1−β max c1 .202 Mart´ Uribe ın Because β T ∈ (0. That is. and then fall in period T to a new long-run level c2 .c2 . stay at that elevated level until period T − 1. We therefore have that c 1 > c > c2 . 1).

Clearly. These conditions are: U (c1 ) = λ. it describes a convex set of feasible pairs (c1 . and U (c2 ) = λ. Consider the behavior of the trade balance. Consequently. is nil. we have that ¯ ¯ . c2 ). Assume for simplicity that the initial net debt potion is nil (d−1 = 0). the solution to this problem is c1 = c2 . a temporary trade liberalization policy of the type studied here is welfare dominated by a constant-tariﬀ regime. c = y. It follows that the ﬁrst-order conditions of this problem are necessary and suﬃcient for a maximum. In the pre-reform equilibrium. Also.Lecture in Open Economy Macroeconomics. It then follows immediately from (7. we have that before the trade reform consumption equals output. and so is the ¯ ¯ current account. Recalling that c1 > c = y > c2 . β 1−β 1−β 1−β 203 Because the period utility function U (·) is assumed to be strictly concave. the objective function is strictly concave in c1 and c2 .1) that the set of tariﬀ schemes that implements the planner’s problem solution satisﬁes τ L = τ H . the trade balance y − cc = 0. and the net foreign asset position induced by the temporary commercial liberalization policy. y − c − rd−1 . the current account. the economy expects tariﬀs to be constant forever. Thus. because the constraint is linear in c1 and c2 . Chapter 7 subject to 1 y βT 2 1 − βT 1 − d−1 + = c + c . where λ denotes the Lagrange multiplier on the constraint.

The evolution of the net foreign debt position is given by t dt = (c1 − y) j=0 (1 + r)j > 0. in period T . This expression states that the foreign asset position is negative and deteriorates over time at an increasing rate until period T . −(dt − dt−1 ). 0 ≤ t < T. . Of course. t ≥ T. According to this expression. the path of the trade balance. the household cuts consumption to a point at which external debt stops growing and the current account experiences a sudden improvement to a balanced position. the current account is negative and deteriorates exponentially. is given by y − c1 < 0 y − c2 > 0 for 0 ≤ t < T for t ≥ T tbt = . which we denote by tbt . 0 ≤ t < T. denoted by cat and given by the change in the asset position. Therefore. The current account. evolves according to the following expression: cat = (y − c1 )(1 + r)t . Formally. these paths for the current account and the asset position are unsustainable in the long run.204 Mart´ Uribe ın for t ≥ 0. we have dt = dT −1 .

(7. the dynamics between periods 0 and T − 1 are identical as the ones associated with the temporary trade liberalization. It is worth noting that in this particular model the terms ‘temporariness’ and ‘imperfect credibility’ are equivalent in a speciﬁc sense. This is because consumption decisions during the period 0 ≤ t < T are based on expectations of a future tariﬀ increase regardless of whether the government explicitly announces it .Lecture in Open Economy Macroeconomics.3).7) The model described in this section is one of overborrowing in the sense that it delivers an initial phase (0 ≤ t < T ) in which households embark in a socially ineﬃcient spending spree with external debt growing at an increasing rate and the external accounts displaying widening imbalances. In this period. by stating that τt would be reduced to τ L for good. 205 The level of consumption that results after the demise of the trade reform. We have motivated overborrowing in the context of this model as stemming from the announcement of a temporary trade liberalization. therefore. Clearly. c2 . t ≥ T. Note that in this model overborrowing ends in a sudden stop in period T . satisﬁes dT −1 = (1 + r)dT −1 + c2 − y. Chapter 7 and cat = 0. but that the public disbelieves this announcement and instead believes that τt will follow the path given in (7. the country experiences a reversal of the current account and a sharp contraction in domestic absorption. Suppose instead that the government announces a permanent trade reform.

206 Mart´ Uribe ın or simply the public believes it will take place. moreover.2 Financial Externalities We now study a class of models of overborrowing that have in common the assumption of a ﬁnancial friction that depends upon a varible that is endogenously determined in the model but that individual agents take as exogenous in their optimization problems. it follows that c must equal c2 . That is. It follows from our previous analysis that consumption must be constant from t = T on.7). or a ﬂow. From the point of view of period T . 7. consumption must satisfy dT −1 = (1 + r)dT −1 − y + c. the government maintains the trade reform (τt = τ L ). A theme of this section is that 2 To see this. It could be a stock. the situation is one in which the tariﬀ is constant forever. This shows that the equilibrium path of consumption is identical whether the policy is temporary or permanent but imperfectly credible. such as output. such as external debt per capita. and. . Suppose now that in period T . that the policy becomes credible to the public. implies a path of dt that converges to plus or minus inﬁnity at a rate larger than r (implying a violation of the no-Ponzi-game constraint in the ﬁrst case and a suboptimal accumulation of wealth in the second) unless c is such that (1 + r)dT −1 − y + c equals d−T . for t ≥ T . The ﬁnancial friction typically takes the form of a collateral constraint imposed by foreign lenders. note that the diﬀerence equation dt = (1+r)dt−1 −y +c. a constant path of consumption is sustainable over time only if the asset position is also constant over time. A constant level of consumption can only be sustained by a constant level of assets. where c is the level of consumption prevailing for t ≥ T . And the varible in question takes various forms.2 Thus. Comparing this expression with (7. contrary to the public’s expectations.

Lecture in Open Economy Macroeconomics. The question is whether the two formulations give rise to diﬀerent equilibrium distributions of debt. and one in which foreign lenders regard the emerging country as a single investment opportunity and look only at aggregate variables in deciding whether to lend or not. where dt denotes the individual household’s net external debt in period t. Chapter 7 207 the nature of this variable is a key determinant of whether the model will or will not generate overborrowing. The key diﬀerence between these two formulations is that the consumer internalizes the ﬁrst borrowing constraint. Consider the following two alternative ¯ ¯ speciﬁcations of such constraint: dt ≤ d and Dt ≤ d.2. These two speciﬁcations are meant to capture two distinct lending environments: One in which foreign lenders base their loan decisions on each borrower’s capacity to repay.1 The No Overborrowing Result Imagine a theoretical environment in which foreign lenders impose a borrowing constraint on emerging countries. The present analysis is based on Uribe (2006). Consider ﬁrst the case of an aggregate borrowing constraint. Dt denotes the average ¯ net external debt across all households in the economy in period t. and d is a constant.e. The economy is assumed to be populated by a large number of households with . but does not internalize the second one because it features the cross-sectional average of debt. which is out of his control.. and in particular whether the one based on aggregate variables delivers more frequent crises (i. 7. episodes in which the debt limit binds). because it involves its own level of debt.

The sequential budget constraint is then given by dt = (1 + rt−1 )dt−1 + ct − ezt F (k ∗ . . such as land. and the function β is assumed to be decreasing in its ﬁrst argument and decreasing in its second argument.8) where ct denotes consumption.4. respectively. Households take the evolution of Ct and Ht as exogenous. ht ). concave. and θ denotes a subjective discount factor. and to satisfy the Inada conditions.1) and assume that θt+1 = β(Ct . the cross sectional averages of consumption and hours. we have that Ct = ct and Ht = ht . ht ). In equilibrium. Households have access to a single risk-free bond denominated in units of consumption that pays the interest rate r when held between periods t and t + 1. All of the results in this chapter go through for any speciﬁcation of θt that is exogenous to the household. Ht )θt for t ≥ 0 and θ0 = 1. The variables Ct and Ht denote.208 preferences described by the utility function ∞ Mart´ Uribe ın E0 t=0 θt U (ct . where zt is a productivity shock assumed to be exogenous and stochastic. Output is produced using hours as the sole input with the technology ezt F (k ∗ . The function F is assumed to be increasing. ht denotes hours worked. (7. For concreteness. ht ). because all households are assumed to be identical. we adopt the speciﬁcation studied in section (4. and k ∗ is a ﬁxed factor of production.

we have that in equilibrium Dt must equal ¯ dt . ht ) = β(Ct . dt } to maximize its utility function subject to the sequential budget constraint and the no-Ponzi-game constraint. when the borrowing ¯ constraint binds. ht ) = ezt Fh (k ∗ . and Uh (ct . the no-Ponzi-game constraint holding with equality. represents a pure ﬁnancial .10) Uc (ct .9) for all t The household chooses processes {ct . As a result. By contrast. in this case the domestic interest rate equals the world interest rate (rt = r). ht ) − (7. domestic agents borrow less than the total amount of funds foreign lenders are willing to invest in the domestic economy. Ht )(1 + rt )Uc (ct+1 . and consequently the domestic interest rate rt rises to a point at which everybody is happy holding ¯ exactly d units of loans. households compete for a ﬁxed amount of loans d. The quantity (rt − r)dt . At the ¯ world interest rate r. Because all agents are identical. ht ) Uc (ct . When the borrowing constraint does not bind (dt < d). ht+1 ) Debt accumulation is subject to the following aggregate constraint: ¯ Dt ≤ d.Lecture in Open Economy Macroeconomics. Chapter 7 In addition. households face a no-Ponzi-game constraint of the form 209 j→∞ lim dt+j /(1 + r)j ≤ 0 (7. ht . The ﬁrst-order conditions of this problem are the sequential budget constraint. the demand for loans exceeds d.

ht ). and that if the domestic interest rate is strictly above the world interest rate the borrowing constraint must be binding. In one. rt ≥ r.16) The last of these expressions is a slackness condition stating that if the borrowing constraint is not binding then the domestic interest rate must equal the world interest rate. There are two polar cases. ht+1 ). ht ) = β(ct . Uc (ct . In this case. ct .210 Mart´ Uribe ın rent. and ¯ (dt − d)(rt − r) = 0. It is important to specify who appropriates this rent. .15) Uc (ct . ht ). (7.11) (7. Intermediate cases are also possible. the ﬁnancial rent is appropriated by domestic banks. ¯ dt ≤ d. A stationary competitive equilibrium in the economy with an aggregate borrowing constraint is a set of processes {dt . the ﬁnancial rent generates a resource cost for the domestic economy.13) (7. the emergence of rents does not represent a loss of resources for the country.14) (7. rt . We ﬁrst study the case in which the ﬁnancial rent is appropriated domestically. In this case. ht ) = ezt Fh (k ∗ .12) (7. dt = (1 + r)dt−1 + ct − ezt F (k ∗ . In the second polar case. ht )(1 + rt )Et Uc (ct+1 . ht ) (7. the rent is appropriated by foreign banks. ht } satisfying the following conditions: − Uh (ct . which then distribute it to households in a lump-sum fashion.

we have that the optimality conditions associated with the household’s problem are the above two constraints. Consider now an environment in which the borrowing constraint is imposed at the level of the individual household. and Uc (ct .Lecture in Open Economy Macroeconomics. the labor eﬃciency condition (7. ht ) and ¯ dt ≤ d.13) features the world interest rate and not the domestic interest rate. Under the alternative polar assumption that these rents accrue to foreign investors. In this environment. the household’s optimization problem consists in maximizing the utility function (7.8) subject to the constraints dt = (1 + r)dt−1 + ct − ezt F (k ∗ . Letting θt λt and θt λt µt denote the Lagrange multipliers on the sequential budget constraint and the borrowing constraint. ht ) = λt λt − λt µt = β(Ct . Chapter 7 211 Note that the resource constraint (7. In this case. Ht )(1 + rt )λt+1 . household take explicitly into account the borrowing limit. the resource constraint would have featured the domestic interest rate rt instead. This is because we are assuming that the ﬁnancial rent that emerges when the borrowing constraint is binding stays within the country.10). respectively.

212 µt ≥ 0 ¯ (dt − d)µt = 0.

Mart´ Uribe ın

The ﬁrst and second of these expressions states that in periods in which the borrowing constraint binds (i.e., when µt > 0), the marginal utility of an extra unit of debt is not given by the marginal utility of consumption Uc (ct , ht ) but by the smaller value Uc (ct , ht )(1 − µt ). The reason behind this is that in this case an extra unit of debt tightens the borrowing constraint even more and therefore carries a shadow punishment, given by Uc (ct , ht )µt utils. A competitive equilibrium in the economy with an individual borrowing constraint is a set of processes {dt , µt , ct , ht } satisfying the following conditions: − Uh (ct , ht ) = ezt Fh (k ∗ , ht ) Uc (ct , ht ) (7.17) (7.18) (7.19) (7.20) (7.21) (7.22)

Uc (ct , ht )(1 − µt ) = β(ct , ht )(1 + r)Et Uc (ct+1 , ht+1 ) dt = (1 + r)dt−1 + ct − ezt F (k ∗ , ht ). ¯ dt ≤ d µt ≥ 0 ¯ (dt − d)µt = 0.

We wish to establish that the equilibrium behavior of external debt, consumption, and hours is identical in the economy with an aggregate borrowing constraint and in the economy with an individual borrowing constraint.

Lecture in Open Economy Macroeconomics, Chapter 7

213

That is, we want to show that the processes {dt , ct , ht } implied by the system (7.11)-(7.16) is identical to the one implied by the system (7.17)-(7.22). To this end, we will show that by performing a simple variable transformation, the system (7.17)-(7.22) can be written like the system (7.11)-(7.16). Deﬁne the shadow interest rate rt as ˜ 1+r . 1 − µt

1 + rt = ˜

(7.23)

Note that µt must be nonnegative and less than unity. This last property follows from the fact that, from equation (7.18), a value of µt greater than or equal to unity would imply that the marginal utility of consumption Uc (ct , ht ) is inﬁnite or negative. It follows that rt is greater than or equal to ˜ r. Furthermore, it is straightforward to see that µt > 0 if and only if rt > r ˜ and that µt = 0 if and only if rt = r. We can therefore write the system ˜ (7.17)-(7.22) as − Uh (ct , ht ) = ezt Fh (k ∗ , ht ) Uc (ct , ht ) (7.24) (7.25) (7.26) (7.27) (7.28) (7.29)

Uc (ct , ht ) = β(ct , ht )(1 + rt )Et Uc (ct+1 , ht+1 ) ˜ dt = (1 + r)dt−1 + ct − ezt F (k ∗ , ht ) ¯ dt ≤ d rt ≥ r ˜ ¯ r (dt − d)(˜t − r) = 0.

The systems (7.11)-(7.16) and (7.24)-(7.29) are identical, and must therefore

214

Mart´ Uribe ın

deliver identical equilibrium processes for dt , ct , and ht . This result demonstrates that whether the borrowing constraint is imposed at the aggregate or the individual level, the real allocation is the same. In other words, the imposition of the borrowing constraint at the aggregate level generates no overborrowing. What happens is that the market interest rate in the economy with the aggregate borrowing limit conveys exactly the same signal as the Lagrange multiplier µt in the economy with the individual borrowing constraint.

Resource Costs When rents from ﬁnancial rationing are appropriated by foreign lenders, the equilibrium conditions of the economy with the aggregate borrowing constraint are as before, except that the resource constraint becomes

dt = (1 + rt+1 )dt−1 + ct − ezt F (k ∗ , ht ).

The fact that the domestic interest rate, rt ≥ r, appears in the resource constraint implies that when the borrowing constraint is binding and rt > r, the country as a whole loses resources in the amount (rt − r)dt . These resources represent pure rents paid to foreign lenders. When rents are appropriated by foreign lenders, it is no longer possible to compare analytically the dynamics of external debt in the economies with the aggregate debt limit and in the economy with the individual debt limit. We therefore resort to numerical methods to characterize competitive equilibria. Preferences and technologies are parameterized as follows: U (c, h) =

**Lecture in Open Economy Macroeconomics, Chapter 7 c − ω −1 hω
**

1−σ

215

/(1−σ), β(c, h) = 1 + c − ω −1 hω

−ψ

, and F (k ∗ , h) = k ∗ α h1−α ,

where σ, ω, ψ, k ∗ , and α are ﬁxed parameters. Table 7.1 displays the values Table 7.1: Parameter Values σ 2 ω 1.455 ψ 0.0222 α 0.32 r 0.04 κ 7.83 k∗ 78.3 πHH = πLL 0.71 z H = −z L 0.0258

assigned to these parameters. The time unit is meant to be one year. The values for α, ω, σ, and R∗ are taken from Schmitt-Groh´ and Uribe (2003). e The parameter ψ is set to induce a debt-to-GDP ratio, d/y, of 50 percent in the deterministic steady state. The calibrated value of κ is such that in the economy without the debt limit, the probability that dt is larger than κ is about 15 percent. I set the parameter k ∗ so that its market price in the deterministic steady state is unity. The productivity shock is assumed to follow a two-state symmetric Markov process with mean zero. Formally, zt takes on values from the set {z 1 , z 2 } with transition probability matrix π. I assume that and z 1 , z 2 and π satisfy z 1 = −z 2 and π11 = π22 . I set π11 equal to 0.71 and z 1 equal to 0.0258. This process displays the same serial correlation (0.58) and twice as large a standard deviation (2.58 percent) as the one estimated for Canada by Mendoza (1991). The choice of a process for the productivity shock that is twice as volatile as the one observed in a developed small open economy like Canada reﬂects the view that the most salient distinction between business cycles in developed and developing countries is, as argued in chapter 1, that the latter are about twice as volatile as the former. The model is solved using the Chebyshev parameterized expectations

216

Mart´ Uribe ın

method. The state space is discretized using 1000 points for the stock of debt, dt . The parameterization of expectations uses 50 coeﬃcients. We compute the equilibrium for three model economies: An economy with no debt limit, an economy with a debt limit and ﬁnancial rents accruing to domestic residents, and an economy with a debt limit and ﬁnancial rents ﬂowing abroad.3 The Matlab code that implements the numerical results reported in this section are available at http://www.columbia.edu/~mu2166/ overborrowing/overborrowing.html. Figure 7.1 displays with a solid line the equilibrium probability distribution of external debt in the economy with an aggregate debt limit and ﬁnancial rents from rationing accruing to domestic agents. According to the no-overborrowing result obtained earlier, this economy is identical to the one with a household-speciﬁc debt limit. The ﬁgure shows with a dashcrossed line the distribution of debt in the economy with an aggregate debt limit and ﬁnancial rents accruing to foreign lenders. As a reference, the ﬁgure also displays, with a dashed line, the debt distribution in an economy without a debt limit. The main result conveyed by the ﬁgure is that the distribution of debt in the economy with a debt limit is virtually unaﬀected by whether ﬁnancial rents are assumed to ﬂow abroad or stay within the country’s limits.4 The reason behind this result is that the resource cost

The procedure approximates the equilibrium with reasonable accuracy. The DenHaanMarcet test for 5-percent left and right tails yields (0.047,0.046) for the economy without a debt limit, (0.043,0.056) for the economy with a debt limit and rents owned domestically, and (0.048,0.056) for the economy with a debt limit and rents ﬂowing abroad. This test was conducted using 1000 simulations of 5000 years each, dropping the ﬁrst 1000 periods. 4 This no-overborrowing result is robust to a more stringent debt limit. We experimented lowering the value of κ by 25 percent, from 7.8 to 5.9. This smaller value of the debt limit is such that in the unconstrained economy the probability that at is larger than κ is about 30 percent. Under this parameterization, We continue to ﬁnd no overbor3

Lecture in Open Economy Macroeconomics, Chapter 7

217

**Figure 7.1: Equilibrium Distribution of External Debt
**

0.07 No Resource Costs Resource Costs No Debt Limit 0.06

0.05

Probability

0.04

0.03

0.02

0.01

0 −10

−5

0

5 external debt

10

15

20

as they tend to exacerbate the contractionary eﬀects of negative aggregate shocks. is the result of two properties of the equilibrium dynamics. we have limited attention to a constant debt limit. This implication. Speciﬁcally. in turn. about 0. .218 Mart´ Uribe ın incurred by the economy when ﬁnancial rents belong to foreigners is fairly small.02 × 100−1 percent of GDP per year. as demonstrated earlier. such as land or structures. This is because states in which the collateral constraint is binding are associated with sharp declines in stock prices and ﬁre sales of collateral (see. The debt constraint binds on average less than once every one hundred years. Second. Theoretically. it produces a country interest-rate premium of less than 2 percent on average. debt limits take the form of collateral constraints limiting the size of debt to a fraction of the market value of an asset. The Role of Asset Prices Thus far. This observation implies that the average cost of remitting ﬁnancial rents abroad is less than 0.008 = 40 × 0. when the debt limit does bind. for examrowing. which. In practice. the economy seldom hits the debt limit. and the external debt is about 40 percent of GDP when the economy hits the debt limit.008 percent of annual GDP. this type of borrowing limit have been shown to help explain observed macroeconomic dynamics during sudden stops. is identical to the debt distribution in the economy with an individual borrowing limit. the debt distribution in the economy with an aggregate borrowing limit and rents accruing to foreign lenders is virtually identical to the distribution of debt in the economy with an aggregate debt limit and rents accruing to domestic households. First. This is because agents engage in precautionary saving to mitigate the likelihood of ﬁnding themselves holding too much debt in periods in which the interest rate is above the world interest rate.

yt = ezt F (kt . The sequential budget constraint of the household is given by dt = (1 + rt−1 )dt−1 + ct + qt (kt+1 − kt ) + ezt F (kt . Chapter 7 219 ple. In that period. Individual households do not internalize this constraint. The household’s optimization problem consists in choosing processes {ct . here we will assume that there is a market for land. To model a time-varying collateral constraint. kt+1 } to maxi- . The central question for the issues being analyzed in this section is whether these ﬁre sales are more or less severe when the collateral constraint is imposed at an aggregate level as opposed to at the level of the individual borrower. assume that output is produced via an homogeneous-of-degree-one function F that takes labor and land as inputs. Let qt denote the market price of land in terms of consumption goods. kt . the household chooses the amount of land that it will use in production in period t + 1.Lecture in Open Economy Macroeconomics. Suppose further that the aggregate per capita supply of land is ﬁxed and given by k ∗ > 0. Unlike in the previous section. Mendoza. the household’s land holdings.30) Note that in period t. are predetermined. ht ) (7. Formally. 2010). it chooses kt+1 . that is. dt . ht . Consider ﬁrst the case in which foreign investors impose a collateral constraint at the country level of the form Dt ≤ κqt k ∗ . ht ).

t+1 [qt+1 + ezt+1 Fk (k ∗ .9). ht+j ). is given by qt = Et {Λt.31) where Λt. and dt ≤ κqt k ∗ . A stationary competitive equilibrium with an aggregate collateral constraint and ﬁnancial rents accruing domestically is given by a set of stationary stochastic processes {dt . evaluated at equilibrium quantities. The ﬁrst-order conditions associated with this optimization problem with respect to consumption. (7. Λt.34) .31).t+1 }∞ satisfying (7. t=0 (7. however. (7.30) and the no-Ponzi-game constraint (7.15). hours. rt .t+j ezt+j Fk (k ∗ .11)-(7. (7. this expression states that the price of land equals the present discounted value of its future expected marginal products. This condition. ht .8) subject to the sequential budget constraint (7. there is an additional ﬁrst-order condition.t+j ≡ Uc (ct+j . In the present environment. qt .220 Mart´ Uribe ın mize the utility function (7. It is an Euler equation associated with the use of land. (7. ht+s ) s=0 (7. ht+1 )]} . and debt are identical to those of its no-land-market counterpart. ct .32) is a stochastic discount factor given by the equilibrium marginal rate of consumption substitution between periods t and t + j.13). ht+j ) Uc (ct . (7.33) Intuitively. ht ) j−1 β(ct+s . Iterating this expression forward yields ∞ qt = Et j=1 Λt.32).

The pricing equation for land takes the form 1 1 − 1 + r 1 + rt ˜ qt 1 − κ = Et {Λt.t+1 [qt+1 + ezt+1 Fk (k ∗ . The household internalizes this borrowing limit. ht+j ) j−1 s=0 1−κ 1 1+r − 1 1+˜t+s r . dt ≤ κqt kt+1 .36) where rt ≥ r denotes the shadow interest rate as deﬁned by equation (7. all external loans are extended at the world interest rate r. we observe that the fact that the shadow value of collateral. Comparing this expression with its counterpart in the economy with an aggregate borrowing constraint (equation (7. as ﬁrst noted by Auernheimer and Garc´ ıa-Saltos (2000).Lecture in Open Economy Macroeconomics. Chapter 7 and (rt − r)(dt − κqt k ∗ ) = 0. This is because the individual household.33)). ˜ Iterate this expression forward to obtain ∞ q t = Et j=1 Λt. In this case. 221 (7. being a price takier. qt . is nonnegative ˜ .t+j ezt+j Fk (k ∗ . ht+1 )]} . given by 1/(1 + r) − 1/(1 + rt ).35) Consider now the case in which the collateral constraint is imposed at the level of each borrower. That is. However. the presence of the price of land on the right-hand side of this borrowing constraint introduces an externality. takes as exogenous the evolution of the price of land.23). (7.

Note that the above expression for qt does not state that the price of land should be higher in periods in which the shadow interest rate rt exceeds r. Λt. r (7. (7. ˜ we will see shortly that the real value of land falls dramatically during such periods. In turn. collateral. ˜ (7.222 Mart´ Uribe ın implies.32). namely.34). intuitively one should not expect the no-overborrowing result of the previous section to be overturned by the introduction of a time-varying collateral constraint of the type considered in this section. qt . and (˜t − r)(dt − κqt k ∗ ) = 0.24)-(7.28). (7. ˜ all other things equal. dt .36). (7.26). A stationary competitive equilibrium with an individual collateral constraint and ﬁnancial rents accruing domestically is given by a set of stationary stochastic processes {dt . rt . This is because when the collateral constraint is internalized the individual agent values the ﬁnancial service provided by land. the value of land is likely to be higher in the economy in which the collateral constraint is internalized than in an economy in which it is not. the fact that in the economy with an internalized collateral constraint the value of collateral is higher than it is in an economy with an aggregate collateral constraint suggests that borrowing is less limited when the collateral constraint is internalized. Thus.t+1 } satisfying (7. that the individual agent discounts future marginal products of land less heavily in the economy with the internalized borrowing constraint than in the economy with an aggregate borrowing constraint.37) . ceteris paribus. Indeed. ht . The above expression does state that when rt is larger than r. ct .

048. I calibrate the economy as in the previous subsection. This test is conducted using 5000 simulations of 5000 years each.06) for the economy with an aggregate collateral constraint. so that κqt k∗ = 7. To ascertain whether the imposition of an aggregate collateral constraint induces external overborrowing.0.043. we compute equilibrium dynamics numerically. whether the collateral constraint is imposed at the individual or the aggregate levels appears to make no diﬀerence for the equilibrium dynamics of stock prices or consumption.edu/ ~mu2166/overborrowing/overborrowing. as shown in the top-right and bottom-left panels of the ﬁgure. qt . But the collapse of land prices is In the deterministic steady state we have that qt = 1.2 displays the unconditional distribution of external debt.83.061) for the economy with an individual collateral constraint. which now takes the value 0.1. Chapter 7 223 The only diﬀerence between the equilibrium conditions of the economy with an aggregate collateral constraint and those of the economy with an individual collateral constraint is the Euler condition for land (equation (7.31) versus equation (7. A solid line corresponds to the economy with an internal collateral constraint.6 The top-left panel of ﬁgure 7.html.0. and a dashed line corresponds to the economy with an aggregate collateral constraint.Lecture in Open Economy Macroeconomics. Similarly.5 The model is solved using the Chebyshev parameterized expectations method. Note that when the stock of debt is high agents engage in ﬁre sales of land resulting in sharp declines in its market price.columbia.36)). and (0. 5 . which is the value assigned to κ in the economy with the constant debt limit. except for the parameter κ. The Matlab code that implements the numerical results reported in this section are available at http://www. 6 The DenHaan-Marcet test for 5-percent left and right tails yields (0. dropping the ﬁrst 1000 periods. The distribution of debt is virtually identical in the economy with an individual collateral constraint and in the economy with an aggregate collateral constraint.

8 Ec t 9.04 1.97 0.02 0.035 1.055 1.04 probability 0.05 1.2: Equilibrium Under a Time-Varying Collateral Constraint Unconditional Distribution of Debt 0.05 0.95 −5 0 External Debt 5 10 0. .6 9.96 0.045 ER t Average Stock Prices 10 Average Consumption 10.03 −5 Indiv CC Agg CC 0 5 External Debt 10 Note: ‘Indiv CC’ stands for Individual Collateral Constraint.01 0 −10 Indiv CC Agg CC 1.4 9. and ‘Agg CC’ stands for Aggregate Collateral Constraint.2 10 9.224 Mart´ Uribe ın Figure 7.94 −5 Indiv CC Agg CC 0 5 External Debt Average Interest Rate 1.01 1 0.98 0.03 Eqt 0.2 −5 Indiv CC Agg CC 0 5 External Debt 10 1.99 0.

The fact that land prices are slightly higher in the economy with the individual borrowing limit means that in this economy the value of collateral is on average higher than in the economy with an aggregate borrowing limit. rt replaces rt . households internalize the fact that holding land relaxes the borrowing limit.Lecture in Open Economy Macroeconomics.t+1 . depressing the value of the asset. when the borrowing constraint is imposed at the aggregate level.t+1 . in line ˜ with the intuition developed earlier in this section. The contraction in real estate prices is caused by the increase in interest rates as the economy approaches the debt limit. As a result. the ﬁre sale of land is driven by a drop in the stochastic discount factor Λt. To see that in a crisis the stochastic discount factor falls. This allows households in the economy with the individual borrowing limit to actually hold on average more debt than households in the economy with an aggregate borrowing limit (see the top left panel of the ﬁgure 7. When it is imposed at the individual level. During a crisis. The intuition behind this result is that in the economy with the individual borrowing limit. Chapter 7 225 quantitatively similar in the two economies. note that the Euler equation for bond holdings is given by 1 = (1 + rt )Et Λt. generating expectations of a decline in Λt.2). Finally. land prices are indeed higher in the economy with an individual debt limit. rt and ˜ rt increase. When Λt. future expected marginal products of land are discounted more heavily.t+1 .t+1 falls. but this diﬀerence is quantitatively small. the demand for land is higher in the economy . It follows that the current model predicts underborrowing. Formally.

given by the pseudo interest rate rt .3. is constant and equal to the world interest rate r except when the debt ˜ ceiling is binding. that the no-overborrowing result is robust to allowing for a debt limit that is increasing in the market value of a ﬁxed factor of production.226 Mart´ Uribe ın with an individual debt limit than in the economy in which the collateral service provided by land is not internalized. in the economy with the individual debt constraint. The absence of either of the abovementioned two features may cause the market price of foreign funds to be below the social price. This property is a consequence of the assumption of homogeneity across economic agents. thereby inducing overborrowing. the shadow price of funds equals the world interest rate even as households operate arbitrarily close to the debt ceiling. It follows from the analysis of this section. The reason why in this class of models households do not have a larger propensity to borrow under an aggregate debt limit is that the market and social prices of international liquidity are identical (in the case of a constant debt limit) or almost identical (in the case of a debt limit that depends on the price of land). Two features of the economies studied in this section are crucial in generating the equality of market and social prices of debt. First. This causes land prices to be higher in the economy with the internalized borrowing constraint allowing households to borrow more. Second. We will return to the issue of underborrowing in section 7. .2. when the debt ceiling binds. Importantly. The next section explores this issue in more detail. when the borrowing limit is internalized the shadow price of funds. it does so for all agents simultaneously.

Heterogeneous Agents The following example. and in the third the collateral is a ﬂow that depends on the relative price of nontradables. is a continuous function of aggregate spending. There is a continuum of agents of measure one. endowment economy without uncertainty. The economy faces a constant debt ceiling κ per capita. Suppose that in the absence of a debt ceiling households with high expected endowment consume ca > y +κ units in period 1 and the rest of the households consume .2 The Case of Overborrowing The present section provides three variations of the environment studied thus far that give rise to overborrowing. In the ﬁrst example. Here. and agents are heterogeneous. Speciﬁcally. The central result obtains under a variety of sources of heterogeneity. which. in the second agents face a debt-elastic interest rate. I assume that agents are identical in all respects except for their period-2 endowments. Agents receiving the larger future endowment have a stronger incentive to borrow in period 1 to smooth consumption over time. such as diﬀerences in endowments. describes a situation in which overborrowing occurs because debt limits do not bind for all agents at the same time. agents are heterogeneous in endowments. preferences. The example is in the context of a two-period.Lecture in Open Economy Macroeconomics.2. whereas in period 2 half of the households receive an endowment y a > y and the other half receive a smaller endowment y b < y a . or initial asset positions. taken from Uribe (2007). Chapter 7 227 7. in period 1 all households receive the same endowment y. in turn.

and . When the borrowing ceiling κ is imposed at the level of each individual household. the equilibrium interest rate equals the world interest rate r. Now suppose that the debt ceiling is imposed at the aggregate level. when du ≤ κ. Aggregate external debt per capita equals di = (κ + cb − y)/2 < du . In this case. That is. Figure 7. whereas the other households are unconstrained and consume cb . In the unconstrained equilibrium.3 depicts the equilibrium in the absence of a debt constraint. One is that the aggregate debt limit is not binding. This case takes place when in the absence of a debt constraint debt per capita does not exceed the ceiling κ. Two alternative situations are possible.228 Mart´ Uribe ın Figure 7. we also have that di < κ. half of the households—those with high period-2 endowment— are constrained and consume y + κ units. Clearly.3: Overborrowing in an Economy with Heterogeneous Agents cb < y + κ units. aggregate external debt per capita equals du = (ca + cb )/2 − y.

is not priced in the competitive equilibrium. there is heterogeneity in the provision of liquidity across assets. Overborrowing occurs because of a ﬁnancial externality. if the aggregate level of external debt in the unconstrained environment exceeds the ceiling (i. external borrowing is higher when the debt ceiling is imposed at the aggregate level. however. then the economy is ﬁnancially rationed. regardless of whether the aggregate debt limit is binding or not.Lecture in Open Economy Macroeconomics. while other assets serve as collateral only to domestic lenders. is also the root cause of overborrowing in the dual-liquidity model of emerging-market crisis developed by Caballero and Krishnamurthy (2001). Chapter 7 229 consumption of each agent equals the level attained in the absence of any borrowing constraint. intertemporal marginal rates of substitution are not equalized across households.e. if du > κ). the combination of heterogeneous consumers and a debt limit imposed at the aggregate level induces overborrowing in equilibrium. Alternatively. External debt is given by da = du > di . Speciﬁcally. It follows that. economic heterogeneity. This service. although of a diﬀerent nature. When the debt limit κ is imposed at the individual level. Some assets are recognized as liquid collateral by both domestic and foreign lenders. 7 . In their model. and aggregate borrowing per capita is given by da = κ > di . the group of more frugal consumers provides a ﬁnancial service to the group of more lavish consumers by placing comparatively less pressure on the aggregate borrowing constraint. Caballero and Krishnamurthy show that in ﬁnancially underdeveloped economies this type of heterogeneity produces an externality whereby the market price of international liquidity is below its social marginal cost.. That is. the domestic interest rate exceeds the world interest rate.7 This overborrowing result relies on the absence of a domestic ﬁnancial market. Suppose that a domestic ﬁnancial market existed in which frugal households (households with relatively low future endowments) could borrow exInterestingly.

foreign lenders may realistically impose that total debt held by lavish households (i. This reemergence of the nooverborrowing result relies. referring to a suboptimal amount of external ﬁnancing. When agents are heterogeneous it is not necessarily the case that the debt distribution associated with the economy featuring an aggregate debt limit is less desirable than the one implied by the economy with an individual debt limit.. for instance. on the assumption that foreign lenders will keep the external debt limit on lavish agents equal to κ.230 Mart´ Uribe ın ternally and lend internally to lavish households. a domestic ﬁnancial market would be ineﬀective in eliminating overborrowing. In this case. It follows that with a domestic ﬁnancial market. In this case. In the models with homogeneous agents and a constant debt limit studied earlier in this paper. in equilibrium. however. This is because the competitive equilibrium associated with the economy featuring an individual debt limit coincides with the optimal allocation chosen by a social planner that internalizes the debt limit. Alternatively. suppose. or overborrowing. To see this. the sum of external and domestic debt) be limited by κ. intertemporal marginal rates of substitution would be equal across households and the consumption allocation would equal the one emerging under an aggregate collateral constraint. that in the economy analyzed in this section the social planner cared only about the well being . The term overborrowing has a negative connotation. A comment on the concept of overborrowing when agents are heterogeneous is in order. one can safely interpret any excess external debt in the economy with an aggregate debt limit over the economy with an individual debt limit as suboptimal.e. the overborrowing result disappears.

as exogenous. We study this class of models in chapter 8. let the country interest rate be given by an increasing function of aggregate external debt of the form rt = r + ρ(Dt ). the social planner would favor the equilibrium associated with an aggregate borrowing limit over the one associated with an individual debt limit. (7.38) . Formally. we have that dt = Dt .Lecture in Open Economy Macroeconomics.4. Chapter 7 231 of agents with high period-2 endowments. In this case. Dt . Then d∗ must satisfy the condition 1 = (1 + r + ρ(d∗ ))β. where dt is the individual level of debt held by the representative household. As a result. Debt-Elastic Country Premium In section 4. they do not internalize the dependence of the interest rate on their individual debt positions. an externality emerges whereby the economy assumes more external debt than it is socially optimal. we studied a way to induce stationarity in the small-openeconomy business-cycle model consisting in making the country interest rate a function of the cross sectional average of external debt per capita. Because individual households take the evolution of the aggregate debt position. Let d∗ > 0 denote the steady-state value of debt in this economy. agents do not internalize that their individual borrowing contributes to increasing the cost of external funds. The reason why the cost of funds is debt elastic is unspeciﬁed in this simple setting. but it could be due to the presence of default risk as in models of sovereign debt.2. In this case. with ρ > 0. In equilibrium.

g. as usual. in the Auernheimer-Garc´ ıa-Saltos model an externality emerges even in this latter case. Auernheimer and Garc´ ıa-Saltos (2000) derive a similar result in a model in which the interest rate depends on the leverage ratio. (7. As noted earlier. That is. because agents do not internalize the eﬀect that their borrowing behavior has on the price of land. Schmitt-Groh´ and Uribe. it is clear that because ρ > 0. . is strictly lower than the social cost of foreign funds. That is. qt .39). β is a constant subjective discount factor. Let d∗∗ > 0 denote the steady-state level of external debt in this economy.232 Mart´ Uribe ın where. the economy with the ﬁnancial externality generates overborrowing. we have that d∗ > d∗∗ .38) and (7. 2003). given by r + ρ(dt ) + dt ρ (dt ). which is key in generating overborrowing. so that rt = r + ρ(dt ). This steadystate condition arises in virtually all formulations of the small open economy with utility-maximizing households (e. It can be shown that d∗∗ is determined by the condition 1 − βd∗∗ ρ (d∗∗ ) = (1 + r + ρ(d∗∗ ))β. and rt = ρ( qt kt+1 ) in the case of an individual debt limit. e Assume now that the debt-elastic interest-rate schedule is imposed at the level of each individual household. This discrepancy. rt = ρ Dt qt k ∗ dt in the case of an aggregate debt limit.39) Comparing equations (7.. is absent in the economy of the previous sections. We note that in the economy with the aggregate debt limit the market price of foreign funds. r + ρ(dt ).

The consumption good is assumed to be a composite of tradable and nontradable consumption as follows: 1−1/η 1−1/η 1/(1−1/η) ct = A(cT . internationally-traded bond that pays the constant interest rate r. extends the Korinek model to an inﬁnite-horizon quantitative setting. .Lecture in Open Economy Macroeconomics. The period utility function takes the form U (c) = (c1−σ − 1)/(1 − σ). Bianchi (2010). the main mechanism for inducing overborrowing is the real exchange rate. Suppose that due to a negative shock. Our exposition follows Bianchi’s formulation.40) with the usual notation. deﬁned as the relative price of nontradable goods in terms of tradables. This mechanism was ﬁrst studied by Korinek (2010). Households are assumed to have access to a single. aggregate demand falls causing the relative price of nontradables to collapse. In this case.41) where cT denotes consumption of tradables and cN denotes consumption of t t nontradables. in the context of a three-period model. (7. risk-free. Chapter 7 Nontraded Output As Collateral 233 Consider now a variation of the model analyzed thus far in which the object that serves as collateral is output. cN ) ≡ ωcT t t t + (1 − ω)cN t . (7. the borrowing constraint tightens exacerbating the economic contraction. one-period. Consider a small open endowment economy in which households have preferences of the form ∞ E0 t=0 β t U (ct ). Here. Since this variable serves as collateral. the value of nontradable output in terms of tradable goods falls.

Households choose a set of processes {cT . cN ) = λt . t t pN = t 1−ω ω cT t cN t 1/η . cN .41)-(7. just as in the case in which the value of land is used as collateral. dt } to maximize (7.42) where dt denotes the amount of debt assumed in period t and maturing in t + 1.43) and G(cT .40) t t T N subject to (7. because each individual household takes the real exchange rate pN as exogenously determined. yt . yt } and the initial debt t position d−1 . Households internalize this borrowing limit. However. t T N and yt and yt denote the endowments of tradables and nontradables. κN > 0 are parameters. The ﬁrst-order conditions of this problem are (7. .41)-(7. Both endowments are assumed to be exogenous and stochastic. even though their collective t absorptions of nontradable goods is a key determinant of this relative price. this borrowing constraint introduces an externality ` la Auernheimer a and Garc´ ıa-Saltos (2000). The borrowing constraint takes the form T N dt ≤ κT yt + κN pN yt .43). ct . t t t (7.43) where κT . given the processes {pN . t (7. re- spectively. pN denotes the relative price of nontradables in terms of tradables.234 The household’s sequential budget constraint is given by Mart´ Uribe ın T N dt = (1 + r)dt−1 + cT − yt + pN (cN − yt ).

Chapter 7 λt = β(1 + r)Et λt+1 + µt . t t+1 (7. In equilibrium. As usual the Euler equation (7. a competitive equilibrium is a set of processes {cT . when the collateral constraint does not bind. and T N µt (dt − κT yt − κN pN yt ) = 0.43). t Then. reﬂecting a shadow penalty for trying to increase debt when the collateral constraint is binding.cN ) 1/η t t cT t denotes the marginal util- ity of tradable consumption. During tranquil times.44) where G(cT . N cN = yt . µt ≥ 0. equates the marginal beneﬁt of assuming more debt with its marginal cost. t t which in turn equals λt . the marginal utility of increasing debt is unchanged. but the marginal cost increases to β(1 + r)Et λt+1 + µt . the beneﬁt of increasing dt by one unit is the marginal utility of tradables G(cT . β(1+r)Et λt+1 . and λt and µt denote the Lagrange multipliers on the sequential budget constraint (7. respectively. t 235 (7.42) and the collateral constraint (7. the market for nontradables must clear.44). The marginal cost of an extra unit of debt is the present discounted value of the payment that it generates in the next period. During ﬁnancial crises. cN ) ≡ ωU (A(cT .45) . That is. when the collateral constraint binds. cN )) t t t t A(cT . yt+1 ) + µt .Lecture in Open Economy Macroeconomics. µt } satisfying t N N G(cT . yt ) = β(1 + r)Et G(cT . cN ). dt .

cT t cT t 1/η N 1−1/η yt T µt dt − κT yt − κN 1/η N 1−1/η yt = 0. yt )) t . A benevolent government would be interested in designing policies that induce households to internalize the ﬁnancial externality.49) T N given processes {yt .47) means that during contractions in which the absorption of tradables falls the collateral constraint endogenously tightens. yt } and the initial condition d−1 . Individual agents do not take this eﬀect into account in choosing their consumption plans. The Socially Optimal Equilibrium The socially optimal policy aims at attaining the allocation implied by an optimization problem that takes into account the dependence of collateral on the aggregate consumption of tradables.46) (7.48) (7.dt } t max E0 t=0 N β t U (A(cT . This is the nature of the ﬁnancial externality in this model. (7. The fact that cT appears on the right-hand-side of the equilibrium version t of the collateral constraint (7.47) 1−ω ω 1−ω ω µt ≥ 0.236 T dt = (1 + r)dt−1 + cT − yt t T dt ≤ κT yt + κN Mart´ Uribe ın (7. Such allocation is the solution to the following social planner’s problem: ∞ {cT . .

The ﬁrst-order conditions associated with this problem are the above two constraints and N N N N G(cT . Chapter 7 237 subject to the equilibrium conditions (7.47). yt )+µt Γ(cT . N and Γ(cT .51) and µt ≥ 0.47).46) and (7. which we reproduce here for convenience: T dt = (1 + r)dt−1 + cT − yt t T dt ≤ κT yt + κN 1−ω ω cT t 1/η N 1−1/η yt . In . (7. yt+1 ) + µt+1 Γ(cT . cT . whose evolution is taken as endogenous by the t social planner but as exogenous by the individual household. The right-hand side of this expression contains the equilibrium level of consumption of tradables.50) T µt dt − κT yt − κN 1−ω ω cT t 1/η N 1−1/η yt =0 (7. yt+1 )] +µt t t t+1 t+1 (7. the social planner internalizes the fact that in the competitive equilibrium the market for nontraded goods clears at all times.Lecture in Open Economy Macroeconomics. This internalization implies that the endogenous variable cN . is replaced everywhere by the exogenous variable t N yt . yt ) = β(1+r)Et [G(cT . Also. yt ) ≡ t κN η 1−ω ω N yt cT t 1−1/η denotes the amount by which an extra unit of tradable consumption relaxes the borrowing constraint (7.47).52) where µt denotes the Lagrange multiplier on the borrowing constraint (7.

the marginal utility of tradable consumption from a social point of view is given by the sum of the N N direct marginal utility G(cT . It turns out that such optimal ﬁscal policy takes the form of a proportional tax on external debt and a lump-sum transfer that rebates the entire proceeds from the debt tax equally among households. however. µt } t satisfying (7. dt . the budget constraint of he household . yt ) reﬂecting the t t fact that an extra unit of consumption of tradables rises the relative price of nontradables. The social-planner equilibrium is then given by a set of processes {cT . yt ) and the factor µt Γ(cT .52). Shortly.238 Mart´ Uribe ın periods in which the borrowing constraint is binding. thereby making the collateral constraint marginally less tight. (7.45) and (7. we will compare the macroeconomic dynamics induced by the competitive equilibrium and the social-planner equilibrium.47). Optimal Fiscal Policy The competitive equilibrium conditions and the social-planner equilibrium conditions diﬀer only in the Euler equation (compare equations (7. we wish to design ﬁscal instruments capable of supporting the socialplanner allocation as the outcome of a competitive equilibrium.46). Before doing so. The question we entertain here is whether there exists a ﬁscalpolicy scheme that induces households to internalize their collective eﬀect on the credit limit and thereby makes the competitive-equilibrium conditions identical to the social-planner equilibrium conditions.50)). and (7. Then. let τt denote the tax on debt and st denote the lump-sum transfer. Speciﬁcally. That is. we wish to address the issue of how to implement the latter.50)-(7.

τt . ct . st } and the initial debt position d−1 . the government rebates the tax to households in a lump-sum fashion by adopting a balanced-budget rule of the form st = τt−1 (1 + r)dt−1 . equals t t the marginal cost of debt.Lecture in Open Economy Macroeconomics.41) and (7. cN . t t t Households choose processes {cT . cN ). β(1 + r)(1 + τt )Et G(cT . cN ). The Euler equation t associated with this problem is of the form G(cT .53) .40) subject to t t this budget constraint and conditions (7.49). which are common to both the competitive equilibrium without taxes and the social planner’s equilibrium. the more t+1 t+1 costly it is for households to accumulate debt. dt } to maximize (7. given by the current marginal utility of tradable consumption G(cT . It follows that the higher the tax rate on debt. Chapter 7 is given by 239 T N dt = (1 + r)(1 + τt−1 )dt−1 + cT − yt + pN (cN − yt ) − st . yt+1 ) + µt . dt .46)-(7. and N N G(cT . given the processes T N {pN . µt } t satisfying (7. This expression says that households choose a level of t+1 t+1 consumption at which the marginal beneﬁt of an extra unit of debt. t t+1 (7. Of course. cN ). cN ) = β(1 + r)(1 + t t τt )Et G(cT . yt . given by the expected present discounted value of the after-tax interest payment measured in terms of utils. yt . yt ) = β(1 + r)(1 + τt )Et G(cT .43). this tax cost of debt is only an individual perception aimed at distorting household’s spending behavior. In the aggregate. It is straightforward to establish that a competitive equilibrium in this economy is given by processes {cT .

50). Thus.53) and (7. the government does not tax external debt (τt = 0). when the collateral constraint does not bind in the current period but has some probability of binding in the next period (µt = 0 and µt+1 > 0 in some states of t + 1). in which the borrowing constraint does not bind (µt = 0) and is not expected to bind in the next period (µt+1 = 0 in all states of period t + 1).240 Mart´ Uribe ın T N given a ﬁscal policy τt . When the borrowing constraint does bind in the current period or when it does not bind either currently or in any state of the next period. yt }. the government taxes debt holdings to discourage excessive spending and external borrowing.54) According to this expression. In periods of uncertainty. It is easy to establish that the optimal tax process is given by τt = N N β(1 + r)Et µt+1 Γ(cT . and the initial condi- tion d−1 . the tax rate is set at zero. The numerical exercise conducted by Bianchi (2010) considers an approximately optimal tax policy in which the government levies a tax on debt equal to the one prescribed by the optimal tax policy only in periods in which the borrowing constraint is not binding in the current period but does bind in at least one of the possible states of period t + 1. exogenous processes {yt . The model is calibrated at an annual frequency. given by (7. The optimal ﬁscal policy then is the tax process {τt } that makes these two Euler equations equal to each other. yt ) t t+1 . yt+1 ) t+1 (7. yt+1 ) − µt Γ(cT . the competitive equilibrium conditions in the economy with taxes and the social planner’s equilibrium conditions diﬀer only in their respective Euler equations. in tranquil periods. N β(1 + r)Et G(cT . The output .

yt ) in the competiT N tive equilibrium (CE) and dt = D s (dt−1 . . D1 > 0) when the borrowing constraint is not binding but are des creasing in the level of debt (D1 < 0. This means that the CE and SP economies experience capital outﬂows (dt − dt−1 < 0) when the borrowing constraint binds. There are also signiﬁcant diﬀerences between the debt-accumulation behavior in the CE and SP economies: First. and nontraded output as the diﬀerence between GDP and GDP in the traded sector. ∼ N (0.31. especially when the borrowing limit is not binding in either the CE or SP economies or when it is binding in the CE but not in the SP economy. ω = 0. D is larger than D s . 2.46 0. Using data from the World Development Indicators from 1965 to 2007. η = 0. Let the T N optimal debt accumulation rule be dt = D(dt−1 .90 0. and r = 0. D1 < 0) when the borrowing constraint is binding. yt . σ = 2.8 The remaining parameters are calibrated as follows: κT = κN = 0.22 0. yt . Chapter 7 241 process is assumed to follow a bivariate AR(1) process. given the 8 Unfortunately.83.45 0.04. The functions D and D s are increasing in the level of debt s (D1 > 0. and mining sectors.022 yt−1 T yt−1 1 t 2 t . the 2010 version of Bianchi’s paper does not specify how the output series were detrended. and in which units they are expressed. Bianchi ﬁnds that the optimal debt accumulation decision is quite different when the borrowing constraint binds and when it does not.Lecture in Open Economy Macroeconomics. Second.91.035 0. β = 0.047 = + N −0. An empirical measure of traded output is deﬁned as GDP in the manufacturing. yt ) in the social-planner (SP) equilibrium. agriculture. the output process is estimated to be with i t T yt N yt 0 0.32. 1) for i = 1.

the borrowing constraint binds at a higher level of debt in the SP economy than in the CE economy. The required tax on external debt that corrects this ineﬃciency. under the present calibration. is 5. Moreover. Bianchi ﬁnds that on average the representative household of the CE economy requires an increase of only one tenth of one percent of its consumption stream to be as happy as the household of the SP economy. This modest welfare cost of overborrowing is due to the fact that in this economy output is assumed to be exogenous.5 percent in the SP economy. and is increasing in the level of debt. the presence of overborrowing in this economy has quantitatively important eﬀects on the frequency and size of economic crises. The probability of a crisis.5 percent in the CE and 0. The next section shows that relaxing this strong assumption can lead to a surprising result. there is a 15 percent probability that debt in the CE is higher than the upper bound of the support of the stationary distribution of debt in the SP economy. deﬁned as a situation in which the borrowing constraint is binding and capital outﬂows (dt − dt−1 < 0) are larger than one standard deviation. But these crises do not appear to inﬂict too much pain to the representative household.242 Mart´ Uribe ın levels of sectoral output. These predictions of the model point at a signiﬁcant degree of overborrowing. Third. the initial contraction in consumption in a crisis is higher in the CE economy than in the SP economy. Finally. . is 5 percent on average.

This simple and realistic extension can fundamentally alter the nature of the ﬁnancial externality.) can have crucial consequences for whether or not the presence of the borrowing constraint causes overborrowing.3 The Case of Underborrowing One lesson we can derive from the models of overborrowing based on collateral constraints is that the assumption of what precise object is used as collateral (a constant. deﬁned as a situation in which the competitive equilibrium produces more debt accumulation than the social planner equilibrium. which. a way to relax the borrowing constraint is to supply more hours of work. thereby increasing income at any given wage rate.2. tightening the borrowing limit. if the elasticity of substitution of tradables for nontradables is low (less than unity). as in the Bianchi (2010) model. and Young (2009). Chen. For an increase in the labor supply causes the supply of nontradables to rise. may in turn cause a fall in the value of nontraded income in terms of tradables. suppose that. the market value of some asset. From a private point of view. this can have a deleterious eﬀect on the country’s ability to borrow. Rebucci. however. etc. The key modiﬁcations introduced by BCORY (2009) is to assume that the labor supply is endogenous and that nontradables are produced using labor—recall that Bianchi’s model assumes that output is exogenous. . borrowing is limited by income. Chapter 7 243 7. In this section we stress this idea by considering a modiﬁcation of Bianchi’s (2010) model due to Benigno. nontraded output. To see this. traded output. Otrok. that under plausible calibrations can produce underborrowing.Lecture in Open Economy Macroeconomics. In equilibrium. hereafter BCORY.

ht . (7.41). t t t (7. this borrowing limit contains on its right-hand side a variable that is endogenous to households. As before. The period utility function. Borrowing in international markets is limited by income: T dt ≤ κ(yt + wt ht ). The endowT ment of tradables. the consumption good is a composite of tradables and nontradables and the aggregation technology is given by equation (7.244 Mart´ Uribe ın Formally. Note that.56) where wt denotes the real wage expressed in terms of tradables. ht ). namely. 1−σ is assumed to be increasing in consumption and decreasing in hours worked. (7. is assumed to be exogenous and stochastic. given by c− U (c. the representative household is assumed to have preferences described by the utility function ∞ E0 t=0 β t U (ct . The sequential budget constraint is of the form T dt = (1 + r)dt−1 + cT − yt + pN cN − wt ht . In particular. yt . unlike the Bianchi (2010) model.55) where ht denotes hours worked in period t.in the present environment households internalize the beneﬁt that working longer hours has on their ability . h) = hδ t δ 1−σ −1 .57) where κ > 0 is a parameter.

as explained earlier. except for the emergence of a labor supply expression of the form: −Uh (ct . and µt denotes the Lagrange multiplier associated with the borrowing constraint. given by wt +κµt /λt . This expression shows that periods in which the borrowing constraint is binding (µt > 0). cT . the shadow real wage. The problem of the ﬁrm consists in choosing ht to maximize N proﬁts. Chapter 7 to borrow. as before. λt denotes the Lagrange multiplier associated with the sequential budget constraint. The ﬁrst-order conditions associated with this problem are identical to those derived for the Bianchi (2010) model. subject to the production technology. and the borrowing constraint (7. an increase in hours raises labor income thereby enlarging the value of collateral and relaxing the borrowing limit. output of tradables is given N by yt = ht . in the fact that at the going wage rate. dt } to maximize the utility function (7.Lecture in Open Economy Macroeconomics.56). This . were. 245 The household’s optimization problem consists in choosing processes {ct . giving households an extra incentive to supply hours to the labor market.57). This incentive to work originates. given by pN yt − wt ht . the sequential budget constraint (7. exceeds the market wage rate wt .55) subject to the agt t gregation technology (7. Speciﬁcally. ht ) = λt (wt + κµt /λt ). Firms produce nontradables using a linear technology that takes labor as the sole input of production. cN . We t assume that free entry guarantees zero proﬁts in the nontraded sector.41). ht .

Noting N that the zero-proﬁt condition implies that wt ht = pN yt at all times. households take the variables that appear on the right-hand side of this expression as exogenous. in Bianchi’s model it holds both in equilibrium and at the level of the individual household. the market-clearing condition yt = cN t t N N to get rid of cN .43) in yet another way by using the zero-proﬁt condition wt = pN to get rid of wt . and the technological relation yt = ht get rid of yt . when κT = κN = κ. we can t write equation (7. we can rewrite the borrowing constraint (7.56) as: T N dt ≤ κ(yt + pN yt ). This t . Mart´ Uribe ın It is of interest to examine the shape of the borrowing limit. The key diﬀerence between the two models is that whereas in the present model the above expression holds only in equilibrium.43). the eﬃciency condition t pN = t 1−ω ω cT t cN t 1/η N to get rid of pN . To understand why the incentives of households to work and in that way relax the borrowing constraint can be counterproductive in equilibrium. but from an equilibrium perspective. now not from the household’s perspective. given by equation (7.246 means that p N = wt t at all times. t which is identical to the borrowing constraint assumed by Bianchi (2010). Note that both in the present model and in Bianchi’s.

the model generates underborrowing. This will be the case when tradables and nontradables are poor substitutes. that an increase in hours expands the borrowing limit.57) is internalized. Ostry and Reinhart. or. . an increase in hours produces a oneto-one increase in the supply of nontradables which drives the relative price of this type of goods down so much that the value of nontraded output ends up falling. They then compute both the competitive equilibrium and the social planner’s equilibrium. a BCORY (2009) set η at 0. on the right-hand side of (7. for instance. is less than unity. That is. more precisely.Lecture in Open Economy Macroeconomics. The latter is. 2003). Existing estimates of the elasticity of substitution η for emerging countries yield values ranging from slightly below 0. η.83 (see. unlike the perception of the individual household. as before.5 to about 0. when the elasticity of substitution between these two types of good. BCORY (2009) ﬁnd that the competitive equilibrium yields a lower average debt-to-output ratio and a smaller probability that the borrowing constraint will bind than the social planner’s equilibrium. Chapter 7 yields the following equilibrium borrowing limit 1 − ω T 1/η 1−1/η ht c ω t 247 T dt ≤ κ yt + This expression says that.e. i. In this case. 1992. BCORY (2009) argue that this is indeed a realistic possibility. and Neumeyer and Gonz´lez-Rozada. the solution to a maximization problem in which the endogeneity of the wage rate. in equilibrium increasing hours might actually shrink the borrowing possibility set.76 and calibrate the rest of the parameters of the model using values that are plausible for the typical emerging economy. wt ..

248

Mart´ Uribe ın

7.2.4

Discussion

We close this section by reiterating that in models in which borrowing is limited by a collateral constraint, the emergence of overborrowing depends crucially on what objects are assumed to serve as collateral. This adds an element of discretionality to the analysis because in the class of models reviewed here the collateral constraint is assumed in an ad hoc fashion. This conclusion suggests two priorities for future research in this area. One is empirical and should aim at answering a simple question: what do foreign lenders accept as collateral from borrowers residing in emerging countries? The second line of research suggested as important by the current state of the overborrowing literature is theoretical. It concerns the production of microfoundations for the type of borrowing constrains analyzed in this chapter. The hope here is that through such foundations we will be able to narrow the type of items that can sensibly be included on the right-hand side of the borrowing constraint.

Chapter 8

Sovereign Debt

Why do countries pay their international debts? This is a fundamental question in open-economy macroeconomics. 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. Thus, 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. At the international level the situation is quite diﬀerent. 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. If agents have no incentives to pay their international debts, then lenders should have no reason to lend internationally to begin with. Yet, we do observe a signiﬁcant amount of borrowing and lending across nations. It follows that international borrowers must have reasons to repay their debts other than 249

250 pure legal enforcement.

Mart´ Uribe ın

Two main reasons are typically oﬀered for why countries honer their international debts: economic sanctions and reputation. Economic sanctions may take many forms, such as seizures of debtor country’s assets located abroad, trade embargoes, import tariﬀs and quotas, etc.1 Intuitively, the stronger is the ability of creditor countries to impose economic sanctions, the weaker the incentives for debtor countries to default. 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. Being isolated from international ﬁnancial markets is costly, as it precludes use of the current account to smooth out consumption in response to aggregate domestic income shocks. As a result, countries may choose to repay their debts simply to preserve their reputation and thereby maintain access to international ﬁnancing. This chapter investigates whether the existing theories of sovereign debt are capable of explaining the observed levels of sovereign debt. Before plunging into theoretical models of country debt, however, we will present some stylized facts about international lending and default that will guide us in evaluating the existing theories.

The use of force by one country or a group of countries to collect debt from another country was not uncommon until the beginning of the twentieth century. In 1902, an attempt by Great Britain, Germany, and Italy to collect the public debt of Venezuela by force prompted the Argentine jurist Luis-Mar´ Drago, who at the time was serving ıa as minister of foreign aﬀairs of Argentina, to articulate a doctrine stating that no public debt should be collected from a sovereign American state by armed force or through the occupation of American territory by a foreign power. The Drago doctrine was approved by the Hague Conference of 1907.

1

Lecture in Open Economy Macroeconomics, Chapter 8

251

8.1

Empirical Regularities

Table 8.1 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. The table also displays average debt-to-GNP ratios at the beginning of default or restructuring episodes. The data suggest that at the time of default debt-to-GNP ratios are signiﬁcantly above average. In eﬀect, for the countries considered in the sample, the debt-to-GNP ratio at the onset of a default or restructuring episode was on average 14 percentage points above normal times. The information provided in the table is silent, however, about whether the higher debt-to-GNP 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 or both. Table 8.1 also shows the country premium paid by the 9 emerging countries listed over a period starting on average in 1996 and ending in 2002. During this period, 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. There is evidence that country spreads are higher the higher the debt-to-GNP ratio. Akitoby and Stratmann (2006), estimate a semielasticity of the spread with respect to the debt-to-GNP ratio of 1.3 (see their table 3, column 4). That is, they estimate ∂ log country spread = 1.3. ∂debt-to-GNP ratio

252

Mart´ Uribe ın

Table 8.1: Debt-to-GNP Ratios and Country Premiums Among Defaulters Average Debt-to-GNP Ratio 37.1 30.7 58.4 33.6 70.6 38.2 55.2 31.5 41.3 44.1 Debt-to-GNP Ratio at Year of Default 54.4 50.1 63.7 112.0 46.7 70.6 21.0 46.3 58.1 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. Debt-to-GNP ratios are averages over the period 1970-2000. Country spreads are measured by EMBI country spreads, produced by J.P. Morgan, and expressed in basis points, and are averages through 2002, with varying starting dates as follows: Argentina 1993; Brazil, Mexico, and Venezuela, 1992; Chile, Colombia, and Turkey, 1999; Egypt 2002; Philippines, 1997. 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; Brazil 1983; Chile 1972 and 1983; Egypt 1984; Mexico 1982; Philippines 1983; Turkey 1978; Venezuela 1982 and 1995. Colombia did not register an external default or restructuring episode between 1970 and 2002. Source: Own calculations based on Reinhart, Rogoﬀ, and Savastano (2003), tables 3 and 6.

Lecture in Open Economy Macroeconomics, Chapter 8

253

Thus, if, as documented in table 8.1, the debt-to-GNP ratio is 14 percentage points higher at the beginning of a debt default or restructuring episode than during normal times, and the average country spread during normal times is about 640 basis points, it follows that at the beginning of a default episode the country premium increases on average by 1.3 × 640 × 0.14 = 116 basis points, or 1.16 percent. This increase in spreads might seem small for a country that is at the brink of default. We note, however, 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. Country spreads are known to respond systematically to other variables that may take diﬀerent values during normal and default times. For instance, 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. Because around default periods inﬂation tends to be high and foreign reserves tend to be low, these two factors contribute to higher country spreads in periods immediately preceding a default episode. Table 8.2 displays empirical probabilities of default for 9 emerging countries over the period 1824-1999. On average, the probability of default is about 3 percent per year. That is, countries defaulted on average once every 33 years. Table 8.2 also reports the average number of years countries are in state of default or restructuring after a default or restructuring episode. After a debt crisis, countries are in state of default for about 11 years on average. If one assumes that while in state of default countries have limited access to fresh funds from international markets, one would conclude that

011 0.051 0. the average probability is conditional on at least one default in the sample period.046 0. Source: Own calculations based on Reinhart. .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 1999.017 0. and Savastano (2003).034 0.040 0.006 0. table 1.254 Mart´ Uribe ın Table 8.040 0. Therefore. Rogoﬀ.2: Probability of Default and Length of Default State 1824-1999 Probability of Default per year 0.023 0.

Default episodes are also associated with disruptions in international trade. Sturzenegger (2003) ﬁnds that after controlling for a number of factors that explain economic growth. does not necessarily preclude the possibility of obtaining new loans from other lenders with which the borrower has no unpaid debts. For being in state of default with a set of lenders. Rose estimates an equation of the form M ln Tijt = β0 + βXijt + m=0 φm Rijt−m + ijt . 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 countries.Lecture in Open Economy Macroeconomics. But the connection between being in state of default and being in ﬁnancial autarky should not be taken too far. the cumulative output loss associated with default or debt restructuring episodes in the 1980s was of about 4 percent over four years. Chapter 8 255 default causes countries to be in ﬁnancial autarky for about a decade. then maintaining access to international trade could represent a reason why countries tend to honor their international ﬁnancial obligations. Rose (2005) investigates this issue empirically. 8. where Tijt is a measure of average bilateral trade between countries i and .2 The Cost of Default Default and debt restructuring episodes are typically accompanied by signiﬁcant declines in aggregate activity.

one. The Paris Club is an informal association of creditor-country ﬁnance ministers and central bankers that meets to negotiate bilateral debt rescheduling agreements with debtorcountry governments. Rose identiﬁes default with dates in which a country enters a debt restructuring deal with the Paris Club. The sample contains 283 Paris-Club debt-restructuring deals. The variable Xijt is a vector of regressors including (current and possibly lagged) characteristics of the country pair ij at time t such as combined output. or 2. or membership in a regional free trade agreement trade more. etc. Landlocked countries and islands trade . combined population. combined area. Rose ﬁnds sensible estimates of the parameters pertaining to the gravity model. The regressor Rijt is a proxy for default. sharing of a common language. countries that are more distant geographically trade less. Specifically. if neither. The vector Xijt also includes current and lagged values of a variable IM Fijt . It 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. a common language. Countries that share a common currency. distance. or both countries i and j engaged in an IMF program at time t. whereas high-income country pairs trade more. membership to the same free trade agreement. respectively. that takes the values 0. 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. 1. country-pair-speciﬁc dummies.256 Mart´ Uribe ın j in period t. Rose’s empirical model belongs to the family of gravity models. a common border. The main focus of Rose’s work is the estimation of the coeﬃcients φm .

07. and the lag length. Do the estimated values of φm really capture the eﬀect of trade sanctions imposed by creditor countries to defaulting countries? Countries undergoing default or restructuring of their external ﬁnancial obligations typically are subject to severe economic distress.e. M . To disentangle these two eﬀects. Chapter 8 257 less. Rose estimates the parameter φm to be on average about -0. 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.07. .. which may be associated with a general decline in international trade that is unrelated to trade sanctions by the creditor country. If this is indeed the case. Default. the cumulative eﬀect of default on trade is about one year worth of trade in the long run. i. M m=0 φm 1+M = −0. Thus. then the coeﬃcients φm would be picking up the combined eﬀects of trade sanctions and of general economic distress during default episodes. The inception of IMF programs is associated with an cumulative contraction in trade of about 10 percent over three years. as measured by the debt restructuring variable Rijt has a signiﬁcant and negative eﬀect on bilateral trade. Rose concludes that one reason why countries pay back their international ﬁnancial obligations is fear of trade disruptions in the case of default. and most of the colonial eﬀects are large and positive. to be about 15 years. Based on this ﬁnding.Lecture in Open Economy Macroeconomics. Mart´ ınez and Sandleris (2008) estimate the following variant of Rose’s gravity model: M M ln Tijt = β0 + βXijt + m=0 φm Rijt−m + m=0 γm Gijt−m + ijt .

and evidence of a general eﬀect of default on trade would require a negative and signiﬁcant estimate of Sandleris estimate 15 m=0 M m=0 γm .01. The sign of the point estimates are robust to setting the number of lags. when a country enters in default its international trade falls by about 40 percent over 15 years with all countries. and zero otherwise. at 0.41. Mart´ ınez and γm to be -0. unlike variable Rijt . or 10. variable Gijt is unity as long as one of the countries is a renegotiating debtor. In this version of the gravity model. One is of the form M M ln Tijt = β0 + βXijt + m=0 φm CREDijt−m + m=0 γm Gijt−m + ijt .258 Mart´ Uribe ın where Gijt−m is a binary variable taking the value one if either country i or country j is a debtor country renegotiating in the context of the Paris Club in period t. Notice that. This regressor is meant to capture the general eﬀect of default on trade with all trading countries. evidence of trade sanctions would require a point estimate for M m=0 φm that is negative and signiﬁcant. However. That is. More importantly. they obtain a point estimate of 15 m=0 φm that is positive and equal to 0. M . . regardless of whether or not the other country in the pair is the renegotiating creditor. not just with those with which the debtor country is renegotiating debt arrears. This result would point at the absence of trade sanctions if creditor countries acted in isolation against defaulters. if creditors behave collectively by applying sanctions to defaulters whether or not they are directly aﬀected then the γm coeﬃcients might in part be capturing sanction eﬀects. 5. Mart´ ınez and Sandleris control for collective-sanction eﬀects by estimating two additional variants of the gravity model.

and N OT CREDijt are all binary variables taking the values 1 or 0.Lecture in Open Economy Macroeconomics.19. independently of whether or not it is renegotiating with the debtor country in the pair. respectively. The variable ACREDijt takes the value 1 if one of the countries in the pair ij is a defaulter negotiating its debt in the context of the Paris Club in period t and the other country is the negotiating Paris Club member (the aﬀected creditor). It takes the form: M M ln Tijt = β0 + βXijt + m=0 M φm ACREDijt−m + m=0 ijt . ξm N ACREDijt−m + m=0 γm N OT CREDijt−m + Here. The point estimate of M m=0 M m=0 φm to φm turns out to be sensitive to the lag length considered.09. 5. 0. Chapter 8 259 where CREDijt is a binary variable that takes the value 1 if one of the countries in the pair ij is a debtor renegotiating its debt and the other is a creditor. N ACREDijt . But when the lag length is set at 15 years.01. and 0. At lag lengths of 0. and 10 years the point estimate is positive and equal to 0.19. The variable N ACREDijt takes the value 1 if one of the countries in the pair ij is a defaulter negotiating its debt in the context of the Paris Club in period t and the other country is a nonnegotiating Paris Club member (a nonaﬀected creditor). The second variant of the gravity model considered aims at disentangling the individual and collective punishment eﬀects. ACREDijt . the point estimate turns negative and equal to -0. Evidence of trade sanctions would require be negative and signiﬁcant. The variable N OT CREDijt takes the value 1 if one of the countries in the pair ij is a .

In this variant of the model.0631. 0. Speciﬁcally. and 15. 5. We interpret the work of Mart´ ınez and Sandleris as suggesting that the importance of trade sanctions as a cost of default depends crucially upon one’s beliefs regarding the magnitude of the delay with which creditors are able or willing to punish defaulter debtors.0854.5629. it takes the values -0. respectively. if one believes that creditors have good memory and are capable of castigating defaulting debtors many years (more than a decade) after a default episode. given by M m=0 φm .260 Mart´ Uribe ın defaulter negotiating its debt in the context of the Paris Club in period t and the other country is not a member of the Paris Club. However. the point estimate is 0. is consistently negative and robust across lag lengths. 10. then the gravity model identiﬁes a signiﬁcant punishment component in the observed decline in . -0.0246. then the gravity model oﬀers little evidence of trade sanctions to defaulters. given by M m=0 ξm . and 15 years. and -0. respectively. at lag lengths of 0. -0.4675. Virtually all of the observed decline in the bilateral trade of debtors after a default episode can be attributed to economic distress and not to punishment inﬂicted by creditors.0119.2314. is again sensitive to the speciﬁed lag length. and 0. 5. If one believes that a reasonable period over which creditors apply trade sanctions to defaulting debtors is less than a decade.3916 at lag lengths of 0. 10. Speciﬁcally. However. the cumulative eﬀect of default on trade between defaulters and directly aﬀected creditors. evidence of individual and collective trade sanctions would require both M m=0 φm and M m=0 ξm to be negative and signiﬁcant. taking positive values at short and medium lag lengths and turning negative at long lag lengths. The cumulative eﬀect of default on trade between defaulters and nonaﬀected creditors. 0.

8. where s = 1. without loss of generality. . Speciﬁcally. . and in states in which ds is positive. Let us assume. . In states in which ds is negative. The variable ds can take positive or negative values. that 1 < 2 < ··· < S. these debt contracts stipulate that the country must pay ds units of goods to foreign lenders in state s. the borrowing country makes a payment to the rest 2 A similar exposition appears in Obstfeld and Rogoﬀ (1996). Chapter 8 261 bilateral trade following default episodes of about 50 percent of the trade volume cumulated over 15 years. S denotes the state of nature. y > 0 is a constant. . Before the realization of the state of nature. the household can buy insurance from foreign lenders in the form of state-contingent debt contracts. The material in this section draws from the inﬂuential work of Grossman and Van Huyck (1988). and s is a random endowment shock with mean zero.Lecture in Open Economy Macroeconomics.2 Consider a one-period economy facing a stochastic endowment given by ys = y + s.3 Default Incentives With State-Contingent Contracts The focus of this section is to analyze the structure of international debt contracts when agents have accesss to state-contingent ﬁnancial instruments but may lack commitment to honor debt obligations. the borrowing country receives a payment from the rest of the world. . There are S > 1 possible states of nature.

3. We begin by considering the case in which households can commit to honor their promises. · · · .1 Optimal Debt Contract With Commitment The household’s problem consists in choosing a state-continget debt contract ds .2) subject to the partic- .262 Mart´ Uribe ın of the world. 8. In each state of nature s = 1. the household’s budget constraint is given by cs = y + s − ds (8. s=1 (8. and to face an opportunity cost of funds equal to zero. S. the zero-expected-proﬁt condition can be written as S πs ds = 0. s=1 (8. letting πs denote the probability of occurrence of state s. to operate in a perfectly competitive market. Formally.2) where cs denotes consumption in state of nature s and u(·) denotes a strictly increasing and strictly concave utility index. · · · . Foreign lenders are assumed to be risk neutral. s = 1. These assumptions imply that debt contracts carrying an expected payment of zero are suﬃcient to ensure the participation of foreign investors.1) The representative household in the domestic economy seeks to maximize the utility function S πs u(cs ). to maximize the utility function (8. S.3) We are now ready to characterize the form of the optimal external debt contract.

That is. Under the optimal contract. and u (cs ) = λ.3). the optimal external debt contract achieves perfect consumption smoothing. under the optimal contract consumption equals the average endowment in all states. (8. or cs = y.3). In particular. Chapter 8 263 ipation constraint (8.1) and to the budget constraint (8. ds = s. The ﬁrst-order conditions associated with the representative household’s problem are (8. in exchange. The Lagrangian associated with this problem can be written as S L= s=1 πs [u(y + s − ds ) + λds ] . domestic risk-averse households transfer all of their income uncertainty to risk-neutral foreign lenders. In this way.1). where λ denotes the Lagrange multiplier associated with the participation constraint (8. Note that λ is not state contingent. the domestic households receive payments from the rest of the world when the endowment realization is low and. for all s. The associated debt payments are exactly equal to the endowment shocks. this expression states that consumption is constant across states of nature. Noting that the multiplier λ is independent of the state of nature.Lecture in Open Economy Macroeconomics. transfer resources to the rest .1).

in our one-period world.4) . d s which means that transfer payments to the rest of the world move one to one with income innovations. so debtors cannot be punished for having defaulted. The perfect-risk-sharing equilibrium we analyzed in the previous subsection was built on the basis that the sovereign can resist the temptation to default. Indeed.2 Optimal Debt Contract Without Commitment In the economy under analysis. the country would default in any state in which the contract stipulates a payment to the rest of the world. debtors have incentives not to pay. payments to (from) the rest of the world are larger the farther is output above (below) its mean value.. the debtor country would be better oﬀ consuming the resources it owes. Furthermore. in states in which the endowment is above average). the world simply ends. In addition.e. What if this commitment to honoring debts was absent? Clearly. In eﬀect. in any state of the world in which the contract stipulates a payment to foreign lenders (i.3. (8. there are no negative consequences for not paying debt obligations.264 Mart´ Uribe ın of the world when the domestic endowment is high. It then follows that any debt contract must include the additional incentive-compatibility constraint ds ≤ 0. After consuming this resources. we have dds = 1. 8.

. No meaningful debt contract can be supported in equilibrium. Chapter 8 265 for all s. The representative household’s problem then consists in maximizing the utility function (8. for all s. namely.3).4) together state that debt payments must be zero on average and never positive.1). It follows that under lack of commitment international risk sharing breaks down. The only debt contract that can satisfy these two requirements simultaneously is clearly ds = 0. Because in the economy without commitment international transfers are . to the budget constraint (8. This is a trivial contract stipulating no transfers of any sort in any state.2) subject to the participation constraint (8. and to the incentive-compatibility constraint (8. As a result. Restrictions (8.e.1) and (8. the consumption plan under commitment is constant across states.Lecture in Open Economy Macroeconomics. Put diﬀerently. for all s.4). the country is in complete ﬁnancial autarky and must consume its endowment in every state cs = y + s. This consumption proﬁle has the same mean as the one that can be supported with commitment. y ≡ Eys . However. whereas the one associated with autarky inherits the volatility of the endowment process. commitment is welfare increasing. households with concave preferences) are worse oﬀ in the ﬁnancially autarkic economy. It follows immediately that risk-averse households (i.

etc. Formally. It follows that the borrower will honor all debts not exceeding k in value. In that case. What is the shape of the optimal debt contract that emerges in this type of environment? We model direct sanctions by assuming that in the case of default lenders can seize k > 0 units of goods from the delinquent debtor.).2) subject to the participation constraint (8.266 constant (and equal to zero) across states. and to the incentive-compatibility constraint (8. . exports. (8.1). this means that the incentive-compatibility constraint now takes the form ds ≤ k. we have that dds = 0. to the budget constraint (8. Direct Sanctions Suppose that foreign lenders (or their representative governments) could punish defaulting sovereigns by seizing national property (such as ﬁnancial assets.5).3). the derivative of debt payments with respect to the endowment is unity at all endowment levels. d s Mart´ Uribe ın This result is in sharp contrast with what we obtained under full commitment.5) The representative household’s problem then consists in maximizing the utility function (8. One would expect that this type of actions would deter borrowers from defaulting at least as long as debt obligations do not exceed the value of the seizure.

6) that the marginal utility of consumption equals λ for all states of nature in which the incentive compatibility constraint is not binding. i.e. This means that consumption is constant across all states in which the incentive-compatibility constraint does not bind.7) (k − ds )γs = 0.5)..6) γs ≥ 0. (8. and the slackness condition (8. and u (cs ) = λ − γs .8) states that the Lagrange multiplier γs must vanish. The ﬁrst-order conditions associated with the representative household’s problem are (8. (8. where λ denotes the Lagrange multiplier associated with the participation constraint and γs denotes the Lagrange multiplier associated with the incentive-compatibility constraint in state s (there are S such multipliers in total. when ds < k. The budget constraint (8. one for each state of nature). Chapter 8 The Lagrangian associated with this problem can be written as S 267 L= s=1 πs [u(y + s − ds ) + λds + γs (k − ds )] .8) In states in which the incentive-compatibility constraint does not bind.Lecture in Open Economy Macroeconomics. the slackness condition (8.3).3) implies that across states in which the .1). (8. It then follows from optimality condition (8. (8.

Formally. where d is a constant.9) We will show shortly that the debt contract described by this expression is indeed continuous in the endowment. states in which the incentive-compatibility constraint does not . we have ¯ d+ k ds = s for for s s <¯ ≥¯ . Then. In particular. Let ¯ denote the smallest endowment level at which the incentive compatibility constraint binds. we will show that as S → ∞. s by only a constant.10) ¯ We will also show that if this condition holds. (8. Based on our analysis of the case with commitment. payments to foreign lenders must diﬀer from the endowment innovation we have that ¯ ds = d + s. it is natural to conjecture that the optimal contract will feature the incentive compatibility constraint binding at relatively high levels of income and not binding at relatively low levels of income. (8.268 Mart´ Uribe ın incentive-compatibility constraint is not binding. we have that ¯ d + ¯ = k. This means that under the optimal debt contract without commitment but with direct sanctions the borrower enjoys less insurance than in the case of full commitment. in which large payments to the rest of the world take place in states of nature featuring large endowments. then the constant d is indeed positive. This is because in relatively low-endowment states (i. ¯ for all s in which ds < k..e.

1).Lecture in Open Economy Macroeconomics. we made use of the conjecture that the debt contract is continuous in the endowment. i. as long as the incentive compatibility constraint binds.1) and to the restriction that the debt contract .2) subject to the participation constraint (8.. which indicates that debt payments must be nil on average. given simply by s. with density π( ). Clearly. The optimal ¯ contract sets d and ¯ to maximize the utility function (8. To show that this conjecture indeed holds. as ¯ πs (d + s <¯ 0 = = s <¯ s) + s ≥¯ πs k ¯ πs (¯ + d) s ≥¯ ¯ πs (d + πs s <¯ s) + ¯ = d+ ¯ = d− s ≥¯ s + s ≥¯ πs ¯ πs ( s − ¯). Chapter 8 269 ¯ bind) the borrower must pay d+ s . ¯ In showing that d is positive.e. it is convenient to assume that there is a continuum of states. H ]. write the participation constraint (8. with L < H.10) holds then d is positive. and therefore d > 0. which implies that ¯ d= s ≥¯ πs ( s − ¯). we have that the right-hand side of the above expression is ¯ positive. as long as ¯ ≤ S. ¯ To see that if condition (8. in the interval [ L . which is a larger sum than the one that is stipulated for the same state in the optimal contract with full commitment.

this expression is satisﬁed if d + ¯ = k.9). ¯ Diﬀerentiating this expression with respect to d and ¯ and setting the result equal to zero. Our analysis shows that the case with no commitment and direct sanctions falls in between the case with full commitment and the case with no . we wish to show that this condition emerges as ¯ part of the optimal contract. given by ¯ L ¯ (d + )π( )d + [1 − F (¯)]k = 0. then the implied transfer payment is continuous in . Note that we are not imposing the continuity restriction (8. This result proves that that if the optimal contract problem has a local maximum. ¯ Clearly.11) and (8.12) we obtain ¯ ¯ ¯ −u (y − d)[(y − d) − (y + ¯ − k)] + [u(y − d) − u(y + ¯ − k)] = 0. the utility function is given by ¯ L H ¯ u(y − d)π( )d + ¯ u(y + − k)π( )d . ¯ L (8. At the same time. Given d and ¯.12) Combining equations (8.270 Mart´ Uribe ın take the form given in (8. we obtain ¯ ¯ [(y + ¯ − k) − (y − d)]π(¯)d + F (¯)dd = 0 (8.10). diﬀerentiating the participation constraint.11) where F (¯) ≡ π( )d denotes the probability that is less than ¯. Rather. yields ¯ ¯ ¯ − u (y − d)F (¯)dd + [u(y − d) − u(y + ¯ − k)]π(¯)d¯ = 0.

the higher the welfare of the debtor countries themselves. Figure 8. Note also that if the sanction is suﬃciently large—speciﬁcally. Chapter 8 271 commitment and no direct sanctions. consumption is perfectly smooth across states and equal to the average endowment.Lecture in Open Economy Macroeconomics. the derivative of the optimal payment with respect to the endowment. In particular. if k > then ¯ > S S— and the optimal contract is identical to the one that results in the case of full commitment. As mentioned earlier. if k = 0—then ¯ = 1 and the optimal contract stipulates ﬁnancial autarky as in the case with neither commitment nor direct sanctions. it is of interest to compare the consumption proﬁles across states in the model with commitment and in the model with direct sanctions and no commitment. In the absence of commitment. consumption smoothing is a direct function of the ability of the lender to punish debtors in the case of default. if the ability of creditors to impose sanctions is suﬃciently limited—speciﬁcally. that the larger the ability of creditors to punish debtor countries in case of default. By contrast. dds /d s s equals unity for s < ¯ (as in the case with full commitment). perhaps paradoxically. Finally. It follows. in this case the risk-averse debtor country transfers all of the risk to risk neutral lenders. Consumption is ﬂat in low-endowment states (from 1 to ¯) and . and equals zero for larger than ¯ (as in the case without commitment and no direct sanctions): 1 dds = d s 0 s s <¯ ≥¯ . In the model with commitment.1 provides a graphical representation of this comparison.

and s denotes the endowment shock. ys denotes output.272 Mart´ Uribe ın Figure 8. .1: Consumption Proﬁles Under Full Commitment and No Commitment With Direct Sanctions ys k cs s cc = y s ¯ d 0 ¯ ε s Note: cc and cs denotes the levels of consumption under full coms s mitment and direct sanctions. respectively.

. . Clearly. Consider designing a debt contract with the following three characteristics: First. the contract stipulates that in . That is.Lecture in Open Economy Macroeconomics. First. 2. ﬁnancial exclusion has the potential to support international lending. assume that creditors have the ability to exclude delinquent debtors from ﬁnancial markets. Time is at the center of any reputational model of debt. the model we have in mind can no longer be a one-period model like the one studied thus far. Instead. . payments are state contingent. Second. Chapter 8 increasing in the endowment in high-endowment states (from ¯ to S ). Second. Reputation Suppose now that creditors do not have access to direct sanctions to punish debtors who choose to default. the ﬂat segment of the consumption proﬁle is lower than the level of consumption achieved under full commitment. 273 The reduced ability of the risk averse agent to transfer risk to risk neutral lenders is reﬂected in two features of the consumption proﬁle. Accordingly. but time independent. . Because ﬁnancial autarky entails the cost of an elevated consumption volatility. Debtor countries pay their obligations to maintain their performing status. and it stipulates precise payments in every date t ≥ 0 and state s ∈ {1. it is signed before period 0. S}. the proﬁle is no longer ﬂat across all states of nature. we will assume that the debtor country lives forever and has preferences described by the utility function ∞ S β t=0 t s=1 πs u(cs ).

13) The left-hand side of this expression is the welfare level associated with maintaining a good standing. the policy of honoring the debt contract in date/state (s. it suﬃces to maximize the period utility index..274 Mart´ Uribe ın any state s. t). S s=1 πs u(y+ s −ds ). the country must pay ds to foreign lenders independently of t. Because the problem is stationary. s=1 (8. These restrictions imply that the following incentive compatibility constraint must hold in each state of nature: β 1−β S u(y + s − ds ) + πs u(y + s=1 s − ds ) ≥ u(y + s ) + β 1−β S πs u(y + s ).1). t) and in every possible subsequent date/date— welfare dominates the policy of defaulting in (s. The Lagrangian associated with this problem is S L = πs u(y + s=1 S s − ds ) + λ s=1 S πs ds β 1−β S + s=1 γs u(y + s − ds ) + πs u(y + s=1 s − ds ) − u(y + s) − β 1−β S πs u(y + s=1 s) The ﬁrst-order conditions associated with the problem of choosing the trans- . the policy of never defaulting— i.e. The right-hand side represents the welfare level associated with ﬁnancial autarky. subject to the above incentive-compatibility constraint and to the participation constraint (8. for every state and date pair (s. t). in the sense that the optimal contract is time independent. Third.

(8.15) stipulates that the Lagrange multiplier γs must vanish.Lecture in Open Economy Macroeconomics.15) is not binding—low-endowment states in which the incentives to default are small. Note that over these states. Chapter 8 fer schedule ds are λ 1+ γs πs 275 u (y + s − ds ) = + β 1−β S s=1 γs (8. the optimality condition (8. implies that consumption is constant ¯ across these states. we have that the marginal utility of consumption must be constant across these states. as payments to creditors are either negative or small—the slackness condition (8. This. In these states.15) In states in which the incentive compatibility constraint (8. and that transfers are of the form ds = d + s. where ¯ d is a constant.14) becomes λ 1+ β 1−β S s=1 γs u (y + s − ds ) = Because the right-hand side of this expression is independent of the state of nature s. What is the pattern of transfers in states in which the incentive compat- . in turn. consumption and payments to or from the rest of the world behave exactly as in the case with direct sanctions: domestic risk-averse agents transfer their endowment shock plus a constant to risk-neutral foreign lenders.14) and the slackness condition β 1−β S γs u(y + s − ds ) + πs u(y + s=1 s − ds ) − u(y + s) − β 1−β S πs u(y + s=1 s) = 0.

showing the incentive β 1−β S s=1 compatibility constraint holding with equality. restriction (8. Because the incentive compatibility constraint binds only when the riskaverse agent must make payments (ds > 0)—there are no incentives to default when the risk-averse agent receives income from the risk-neutral lender—and because the utility index is strictly concave. the terms s) πs u(y + s and s β 1−β S s=1 πs u(y + s − ds ) are both unchanged as we move from to .and by how much? To answer these questions. then we can approximate the eﬀect on the optimal transfer of moving from one state to the other by diﬀerentiating (8. Only the ﬁrst terms on the right.16). s=1 (8.16) with respect to the current transfer and the current endowment. s and s in which the incentive compatibility constraint binds and such that s > s. consider two states.16) How does the optimal transfer ds vary across states in which the collateral constraint is binding? Does it increase as one moves from low. which yields dds u (y + s − ds ) − u (y + = d s u (y + s − ds ) s) . it follows that u (y + s − ds ) > u (y + s) in all states in which the incentive compati- .16) change s as we evalluate that expression at diﬀerent states of nature.and left-hand sides of (8. Notice that in equation (8.to highendowment states. If and s are very close to each other.13) holds with equality: β u(y + s − ds ) + 1−β S s=1 β πs u(y + s − ds ) = u(y + s ) + 1−β S πs u(y + s ).276 Mart´ Uribe ın ibility constraint is binding? In these states.

the higher is the level of current consumption that can be achieved upon default. (This is because the period utility function is assumed to be strictly concave.Lecture in Open Economy Macroeconomics. This implies that when the incentive compatibility constraint binds we have 0< dds < 1. The incentive compatibility constraint tends to bind in relatively high-endowment states. but less than one for one. s.) It follows that in states in which ds > 0. ds > 0. The intuition behind the direct relation between income and payments is that given a positive level of current payments. After all. by more than it raises the utility associated with the alternative of defaulting. We close by noting that the positive slope of the payment schedule with respect to the endowment (when the incentive compatibility constraint is binding) presents a contrast with the pattern that emerges in the case of . payments to the foreign lender increase with the level of income. the higher is the level of payments to foreign lenders that can be supported without inducing default. It might seem counterintuitive that as the current endowment increases the payment to creditors that can be supported without default also increases. This does not mean that default incentives are weaker the higher is the level of the endowment. u(y + − ds ). the higher is the current level of endowment. Chapter 8 277 bility constraint binds. Recall that the analysis in this paragraph is restricted to states in which the incentive compatibility constraint is binding. d s That is. u(y + s ). a small increase in current endowment. raises the current-period utility associ- ated with not defaulting. the higher is the current endowment.

which implies that the slope of the payment schedule equals zero. we focus on the polar case of a single non-state-contingent asset. the optimal ﬁnancial contract facilitates a smooth level of consumption across states of nature. An implication of this result is that default incentives are stronger in high-income states and weaker in low-income states. optimal risk-sharing arrangements stipulate positive payoﬀs in low-income states and negative payoﬀs in high-income states.4 Default Incentives With Non-State-Contingent Contracts In a world with complete ﬁnancial markets. when the incentive compatibility constraint binds payments equal the maximum punishment k. In that case. In this environment. To this end. as documented earlier in this chapter.278 Mart´ Uribe ın direct sanctions. 8. debts assumed in the current period impose ﬁnancial obligations in the next period that are independent of whether income in that period is high or low. debtors facing high debt obligations and low endowments will have strong incentives to default. As a result. represents the ﬁrst formalization of this idea. In the real world. countries tend to default during economic contractions. which we study in this section. we remove the assumption that ﬁnancial markets are complete. The pioneer model of Eaton and Gersowitz (1981). One of our goals in this section is to explain this empirical regularity. The debtor is no longer able to design debt contracts that pay a high interest rate in good states and a low interest rate in bad states. In this way. Our version of the Eaton-Gersowitz model . Indeed. however.

and i. As a result. 279 Consider a small open economy populated by a large number of identical individuals. If the household is in bad ﬁnancial standing. 1) is a parameter denoting the subjective discount factor. the household is forced to consume its endowment. If it chooses to honor its debt. with a distribution featuring a bounded support Y ≡ [y.i. If it chooses to default. it can choose to default on its debt obligations or to honor its debt. the representative household is endowed with yt units of consumption goods. If the household is in good ﬁnancial standing. Chapter 8 follows Arellano (2005).d. consumption of a household in bad ﬁnancial standing is given by c = y.Lecture in Open Economy Macroeconomics. stochastic. . then it is prevented from borrowing or lending in ﬁnancial markets. ¯ At the beginning of each period. the household can be either in good ﬁnancial standing or in bad ﬁnancial standing.. then it immediately acquires a bad ﬁnancial status. Formally. where ct denotes consumption in period t. We drop the time subscript in expressions where all variables are dated in the current period. y ]. u is a period utility function assumed to be strictly increasing and strictly concave. This endowment is assumed to be exogenous. Preferences are described by the utility function ∞ E0 t=0 β t u(ct ). Each period t ≥ 0. and β ∈ (0.

If instead we had assumed that y was serially correlated.280 Mart´ Uribe ın then it maintains its good ﬁnancial standing until the beginning of the next period. . The household enters in bad standing when it defaults on its ﬁnancial obligations.d. then bond prices would depend on the level of the current endowment. it remains in that status forever. and q(d ) denotes the market price of the household’s debt. We assume that ‘bad ﬁnancial standing’ is an absorbent state. This is because the default decision in the next period depends on the amount of debt due then. 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 ). which implies that its current value conveys no information about future expected endowment levels. but not on the level of debt acquired in the previous period and due in the current period. This means that once the household falls into bad standing. Notice also that q(·) is independent of the current level of output. d denotes the debt acquired in the current period and due in the next period. This is because of the assumed i. where d denotes the household’s debt due in the current period. Households that are in good standing and choose not to default face the following budget constraint: c + d = y + q(d )d . d . d. Note that the price of debt depends on the amount of debt acquired in the current period and due next period. nature of the endowment.i.

In what follows. y) = max{v b (y). We do so in steps. not defaulting) is denoted by v c (d. the value function associated with continuing to participate in capital markets (i. d subject to ¯ d ≤ d.Lecture in Open Economy Macroeconomics. we demonstrate that this intuition is in fact correct. v c (d. y) = max u(y + q(d )d − d) + βEv 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. and is given by v g (d. y denotes next period’s endowment. y) denotes the value function associated with being in good ﬁnancial standing. In this economy. ¯ The parameter d > 0 is a debt limit that prevents agents from engaging in Ponzi games. y)}. y) and is given by v c (d. 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. where v g (d. Chapter 8 Here.. We denote the default set by . The Default Set The default set contains all endowment levels at which a household chooses to default given a particular level of debt. y ) .e. 281 For an agent in good standing.

d satisﬁes d ≤ d. then if it is to continue to participate in the . v c (d. The ﬁrst inequality follows from the fact that d is a feasible point of the constrained maximization problem that appears on the right¯ hand side of the ﬁrst line of the above expression—i. This proposition states that if the household has a level of debt that puts it at risk of default. y ) ¯ d <d ≥ u(y + q d d − d) + βEv g (d. an economy that chooses not to default will run a trade surplus. y ) ≥ u(y) + βEv b (y ) ≡ v b (y).. The second inequality holds because. Because it is never in the agent’s interest to default when its asset position is nonnegative (or d ≤ 0).1 If D(d) = ∅. for all y ∈ Y . v g (d. y ) ≥ v b (y ). y) ≡ max u(y + q(d )d − d) + βEv g (d . it follows that D(d) is empty for all d ≤ 0. Then. ¯ Proof: Suppose that q d d − d ≥ 0 for some d ≤ d. The following proposition shows that at debt levels for which the default set is not empty. It follows that if q(d)d − d ≥ 0 for some ¯ d ≤ d.282 D(d). then q(d )d − d < 0 for all d ≤ d. then D(d) = ∅. the default set is deﬁned by Mart´ Uribe ın D(d) = {y ∈ Y : v b (y) > v c (d. by deﬁnition.e. q(d)d − d ≥ 0 and because. by assumption. y)}. ¯ Proposition 8. Formally.

if the default set is not empty then it is indeed an interval with lower bound given by the lowest endowment level y. Put diﬀerently. if y2 ∈ D(d). then v b (y1 ) > v c (d. This means that if v b (y2 ) > v c (d.e. b c vy (y) = u (y) and vy (d.2 If y2 ∈ D(d) and y1 < y2 . Proof: Suppose D(d) = ∅.. the proposition states that in this model the economy runs trade deﬁcits only when its debt positon is such that there is no endowment realization that would induce it to default (i. then y1 ∈ D(d) for any y1 < y2 . Proposition 8. We have shown that the default set is an interval with a lower bound given by the lowest endowment y. Chapter 8 283 ﬁnancial market. That is. By proposition 8. it will devote part of its current endowment to servicing the debt. We now show that the default set D(d) is a larger interval the larger the stock of debt. when d is such that D(d) = ∅). v b (y) − v c (d.1. y)/∂y. we ¯ have that q(d )d − d < 0 for all d ≤ d. the higher . that u (y + q(d )d − d) > u (y). It follows that vy (y) − vy (d. y2 ) for y2 ∈ D(d). by strict concavity b c of u. Equivalently. y) = u (y + q(d )d − d). y1 ) for y1 < y2 . In other worlds. y) < 0.Lecture in Open Economy Macroeconomics. This implies. y) is a decreasing function of y for all y ∈ D(d). then y1 ∈ D(d). y) ≡ ∂v c (d. Put diﬀerently. Consider any y ∈ Y such that y ∈ D(d). Speciﬁcally. 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 a lower level of income. By the envelope theorem. We now establish that in this economy households tend to default in bad times. for all y ∈ D(d). Let b c vy (y) ≡ ∂v b (y)/∂y and vy (d.

y) > 0. y ∗ (d)) < c 0. At the same time. y) < 0 for all y ∈ D(d). b c we showed that vy (y) − vy (d. default is more likely the lower the level of output.284 the debt.3 If D(d) = ∅. Second. Using the deﬁnition of vd (d. we have obtained two important results: First. c dd vy (y (d)) − vy (d. the higher the probability of default. [y. Diﬀerentiating ¯ this expression yields v c (d. We have shown that vy (y ∗ (d))−vy (d. y ∗ (d)) = −u (y ∗ (d) + q(d )d − d) < 0. where y ∗ (d) is increasing in d if y ∗ (d) < y . We then conclude that dy ∗ (d) > 0. dd as stated in the proposition. By deﬁnition. y ∗ (d)) dy ∗ (d) = b ∗ d . indicating that at the time of default countries tend to display above-average debt-to-GNP ratios (see table 8. It follows that y ∗ (d) is given either by y or (implicitly) by v b (y ∗ (d)) = v c (d. y) ≡ ∂v c (d. the larger the stock of debt. y ∗ (d)]. These two results are in line with the stylized facts presented earlier in this chapter. y)/∂d. the larger the probability of default. y ∗ (d)). given the stock of debt. Mart´ Uribe ın Proposition 8. every y ∈ D(d) satisﬁes v b (y) − v c (d.1). y) and applying the envelope theorem. Summarizing. y ∗ (d)) c b c where vd (d. . then D(d) is an interval. ¯ Proof: We already proved that the default set D(d) is an interval. it folc lows that vd (d.

It follows that the expected rate of return on the country’s debt must equal r ∗ . foreign lenders receive 1/q(d ) units of goods per unit lent. given by the diﬀerence between 1/q(d ) and 1 + r ∗ . . We assume that foreign lenders are risk neutral and perfectly competitive. one obtains Prob {y ≥ y ∗ (d )} . foreign lenders receive nothing. equating the expected rate of return on the domestic debt to the risk-free world interest rate. If the country does not default. y ∗ (d ) ≥ 0. is nondecreasing in the stock of debt.Lecture in Open Economy Macroeconomics. If the country does default. Therefore. we can write 1 − F (y ∗ (d )) . by proposition 8. 1 + r∗ q(d ) = Taking derivative with respect to next period’s debt yields dq(d ) −F (y ∗ (d ))y ∗ (d ) = ≤ 0. Letting F (y) denote the cumulative density function of the endowment shock.3. It follows that the country spread. Chapter 8 Default Risk and the Country Premium 285 We now characterize the behavior of the country interest-rate premium in this economy. dd 1 + r∗ The inequality follows because by deﬁnition F ≥ 0 and because. Let the world interest rate be constant and equal to r ∗ > 0. q(d ) 1 + r∗ = The numerator on the right side of this expression is the probability that the country will not default next period.

8. it might seem that what is important is that defaulters be precluded from borrowing in international ﬁnancial markets. If delinquent countries were not allowed to borrow but could run current account surpluses. but that saving in these markets is allowed after default.286 We summarize this result in the following proposition: Mart´ Uribe ın Proposition 8.5 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 it from the possibility to borrow or lend in international ﬁnancial markets. given by 1/q(d ) − 1 − r ∗ is nondecreasing in the stock of debt. At a ﬁrst glance. Suppose that a reputational equilibrium supports a path for external debt given by {dt }∞ . a . no lending at all could be supported on reputational grounds alone. Why should defaulting countries not be allowed to save? Bullow and Rogoﬀ (1989) have shown that prohibiting defaulters to lend to foreign agents (or save in foreign assets) is crucial for the reputational model to work.3 Assume that default is punished with perpetual exclusion from borrowing in international ﬁnancial markets. see Eaton and Fern´ndez (1995). consider a deterministic economy. This assumption 3 For an example of a deterministic model with sovereign debt supported by reputation. To illustrate this insight in a simple setting.4 The country spread. where dt denotes the level of external debt ast=0 sumed in period t and due in period t + 1.

notice that if the country defaults at some date T > 0. That is. Chapter 8 287 and the fact that the economy operates under perfect foresight imply that any reputational equilibrium featuring positive debt in at least one date must be characterized by no default. then no foreign investor would want to lend to this country in period T − 1. dt ≤ 0 for all t ≥ 0. The evolution of the equilibrium level of debt is then given by dt = (1 + r ∗ )dt−1 − tbt . Thus. for t ≥ 0. In turn. dT −1 ≤ 0. then it will have no incentives to honor any debts outstanding in that period. To see . To see this. because the probability of default is nil. Does it pay for the country to honor this debt? The answer is no. if the country is excluded from borrowing starting in period T − 1. since default would occur for sure one period later. no foreign investor will be willing to lend to the country in period T − 2.Lecture in Open Economy Macroeconomics. 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 maintian a level of consumption no lower than the ones that would have obtained in the absence of default. That is. dT −2 ≤ 0. we arrive at the conclusion that default in period T implies no debt at any time. As a result. It follows from this result that in an equilibrium with positive external debt the interest rate must equal the world interest rate r ∗ > 0. Continuing with this logic. Let dT be the maximum level of external debt in this equilibrium sequence. dT ≥ dt for all t ≥ −1. where tbt ≡ yt − ct denotes the trade balance in period t. The country premium is therefore also nil.

That is. We have that −tbT +1 = dT +1 −(1+r ∗ )dT . where. which implies that dT +1 = dT +1 − (1 + r ∗ )dT . where tbT +1 is the trade balance prevailing in period T +1 under the original debt sequence {dt }. we obtain dT +2 = dT +2 − (1 + r ∗ )2 dT < 0. We have shown that the defaulting strategy can achieve the no-default level of . Using (8.17) Because by assumption dT ≥ dT +1 and r ∗ > 0.17) and the fact that tbT +2 = (1 + r ∗ )dT +1 − dT +2 . we have that dT +1 < 0. The inequality follows because by assumption dT +2 ≤ dT and r ∗ > 0. tbT +2 is the trade balance prevailing in period T + 2 under the original (no default) debt sequence {dt }. let dt for t > T denote the post-default path of external debt. (8. Let the external debt position in period T + 2 in the default strategy be dT +2 = (1 + r ∗ )dT +1 − tbT +2 . without having to borrow internationally. Let the debt position acquired in the period of default be dT +1 = −tbT +1 . in period T + 1 the country can achieve the save level of trade balance (and hence consumption) as under the no-default strategy. again.288 Mart´ Uribe ın this.

It follows that it pays for the country to default immediately after reaching the largest debt level dT . which is strictly negative for all t ≥ T + 1. . can be supported by the debt path dt satisfying dt = dt − (1 + r ∗ )t−T dT . tbt for t ≥ T + 1. But the result also holds in a stochastic environment (see Bullow and Rogoﬀ.Lecture in Open Economy Macroeconomics. we derived this breakdown result using a model without uncertainty. The fact that the entire postdefault debt path is negative implies that the country could also implement a post default path of trade balances tbt satisfying tbt ≤ tbt for t ≥ T + 1 and tbt < tbt for at least one t ≥ T + 1 and still generate no positive debt at any date t ≥ T + 1. This new path for the trade balance would be strictly preferred to the no-default path because it would allow consumption to be strictly higher than under the no-default strategy in at least one period and to be at least as high as under the no-default strategy in all other periods (recall that tbt = yt − ct ). Chapter 8 289 trade balance in period t + 2 without requiring any international borrowing. 1989). For simplicity. Continuing in this way. In other words. Therefore. But we showed that default in this perfect foresight economy implies zero debt at all times. no external debt can be supported in equilibrium. alloweing the country to save in international markets after default implies that no equilibrium featuring strictly positive levels of debt can be supported on reputational gourds alone. one obtains that the no-default sequence of trade balances.

290 Mart´ Uribe ın .

Andrew. Stratmann. Christopher Otrok. December 2005.” Journal of Political Economy (115). C.” Review of Economic Studies (58). Gianluca. and T. University of Minnesota. 69-102. 291 . H. C. January 2006. Arellano. Bond. Huigang Chen. L.” working paper. Aguiar. October 2000.” IMF working paper wp/06/16. “Estimation of Dynamic Models With Error Components. “Some Tests of Speciﬁcation for Panel Data: Monte Carlo Evidence and An Application to Employment Equations. Akitoby. 1990.” American Economic Review (80). 1991. Auernheimer. “International Debt and the Price of Domestic Assets. 2007. 277-297. Anderson... “Fiscal Policy and Financial Markets. T.” IMF Working Paper wp/00/177. Benigno. 38-42. “Asset Prices Under Habit Formation and Catching Up With The Joneses. Alessandro Rebucci.” Journal of the American Statistical Association (76). and Hsiao. and R. M. Mark. pp. Arellano. and S. B. Garc´ ıa-Saltos. “Emerging Market Business Cycles: The Cycle is the Trend. “Default Risk and Income Fluctuations in Emerging Markets.References Abel. 1981. 598-606. and Gita Gopinath.

Bullow.” American Economic Review (79). 513-548. University of Maryland. Guillermo. Javier. “Does Saving Anticipate Declining Labor Income? An Alternative Test of the Permanent Income Hypothesis. June. Lawrence J. 245-261. “Debt. The World Bank. Bianchi. Ricardo and Arvind Krishnamurthy.” Journal of Monetary Economics (48). Boldrin. J. Caballero.A. “Asset Pricing Lessons for Modeling Business Cycles. 1987..” American Economic Review (91). 149-166.” manuscript. 1986. “Balance of Payments Crises in a Cash-in-Advance Economy. 1319-1329.” Journal of Political Economy (93).. “International and Domestic Collateral Constraints in a Model of Emerging Market Crises. pp. 1987. Young. University of Virginia.A.” Econometrica (55).” Journal of Money. Olivier. March 1989. November 2009. “Characterizing the Business Cycles o e of Emerging Economies..” manuscript. John Y. G. Calder´n. Blanchard. Michele. Christiano. 43-50.” Journal of Development Economics (27). and K. “Revisiting Overborrowing and its Policy Implications. 2001. . 2010. and Finite Horizons. 1249-1273. G. 2010. Calvo. Campbell. “Sovereign Debt: Is to Forgive to Forget?. Credit and Banking (19). “On the costs of temporary policy.” Journal of Political Economy (94).292 Mart´ Uribe ın and Eric R.” Unpublished. December 2001. 1985. “Temporary Stabilization: Predetermined Exchange Rates. Calvo. 223-247. and Jonas Fisher. C´sar and Rodrigo Fuentes. November 1987. pp. Rogoﬀ. 19-32. Calvo. “Overborrowing and Systemic Externalities in the Business Cycle. Deﬁcits..

98. 10891113. Martin Eichenbaum. Cline. and Kevin S. and F. 411-434.. “Debt with Potential Repudiation: Theoretical and Empirical Analysis. “The Dynamics of a Small Open Economy in Response to Monetary. Chapter 8 293 Cardia. Institute for International Finance/ 1995. Fern´ndez. 1997. Fiscal. P. William R. DC. 1995. 519-543.” Review of Economic Studies (48). Emanuela.” Journal of Monetary Economics (28). Washington. . Institute for International Finance. Correia. Eaton.Neves. “Default Episodes in the 1980s and 1990s: What have We Learned?. J.” European Economic Review (39). “Habit Persistence: A Resolution of the Equity Premium Puzzle. Washington. International Debt Reexamined. Isabel. pages 2031-2077. Eaton.” NBER working paper No.. and Charles Evans. William R. 1991. “Nominal Rigidities and the Dynamic Eﬀects of a Shock to Monetary Policy. Lawrence J. DC. Amsterdam. Chuhan. The Netherlands: Elsevier Science.” manuscript. Handbook of International Economics III. Cline. Barnes. July 2001.” Journal of Political Economy (1990). Constantinides. and Sergio Rebelo. November 24. 8403. and R.Lecture in Open Economy Macroeconomics. and Productivity Shocks. Grossman and K. Jonathan and Mark Gersovitz.. 1995. “Business Cycles in a a Small Open Economy. Sturzenegger. a Rogoﬀ editors.” in G. Universidad Torcuato Di Tella.” Research paper 97-1. “Sovereign Debt. 2003. Jo˜o C. “Spreads and Risk in Emerging Market Lending. April 1981. Christiano. chapter 39. 289-309. George M.

” NBER Working Paper No. Van Huyck. and Mart´ Uribe.” Economic Letters (65).” Journal of Political Economy (86). Arnold C. Sebastian. Garc´ ıa-Cicco. February 1998. Kamin. 1999. Steven and Karsten von Kleist. “The Evolution and Determinants of Emerging Market Credit Spreads in the 1990s. “LDC Foreign Borrowing and Default Risk: An Empirical Investigation. and Mart´ Uribe. 68. 2009. 47-60. and Ann L. “Real Business ın Cycles in Emerging Countries?. Herschel I. Roberto Pancazi. Hall. “Stochastic Implications of the Life Cycle-Permanent Income Hypothesis: Theory and Evidence. Javier.” American Economic Review (78). and John B. 971-987. pp. 6408. Columbia University. “Real Business ın Cycles in Emerging Countries?: Expanded Version. May 1999. Javier. 1978.” Journal of Political Economy (58). Ruth A.” American Economic Review (100). “What Explains Changing Spreads on Emerging-Market Debt: Fundamentals or Market Sentiment?.” American Economic Review (74). 25102531. Judson. Income. 1950. December 2010. Barry and Ashoka Mody. Robert.” unpublished. . Harberger. and the Balance of Trade. Eichengreen.. December 1988. Repudiation.294 Mart´ Uribe ın Edwards. Roberto Pancazi. Grossman. Owen. 1984. 726-734. and Reputation. Garc´ ıa-Cicco. “Estimating Dynamic Panel Data Models A Guide For Macroeconomists. “Sovereign Debt as a Contingent Claim: Excusable Dafault. 1088-1097. “Currency Depreciation. 9-15.” BIS Working Paper No.

Sunghyun Henry and Ayhan Kose. DC/ 1973. Krugman. “Is the Business Cycle of Argentina Diﬀerent?. 2000. November 1950. Universidad Torcuato Di Tella. Anton. Zarazaga. “Dynamics of Open Economy Business Cycle Models: Understanding the Role of the Discount Factor.” Journal of Money.” CarnegieRochester Conference Series on Public Policy (1). Paul. Mendoza. “Excessive Dollar Borrowing in Emerging Markets: Balance Sheet Eﬀects and Macroeconomic Externalities. Laursen. 1979. “Using the Generalized Schur Form to Solve a Multivariate Linear Rational Expectations Model. McKinnon.” Review of Economics and Statistics (32). “Is It Punishment? Sovereign Dee fault and the Decline in Trade. Lucas. 1976.” Macroeconomic Dynamics (7). “Flexible Exchange Rates and the Theory of Employment. Svend and Lloyd A. 19-46. Finn E. Credit and Banking (11). Paul R.. R. Klein.” Amer- . J.” Journal of Economic Dynamics and Control (24). and Guido Sandleris. 281-299. Brookings Institution Press. 311-25. Mart´ ınez.” Economic Review (Federal Reserve Bank of Dallas).. Jos´ V. and Carlos E. Kydland. Money and Capital in Economic Development. 1405-1423.” working paper 01-2008. “Real Business cycles in a small-open economy. 263-90.. 2010. 2008. Washington. Jr. Korinek. 21-36. University of Maryland. Ronald I. Metzler. pp. “Econometric Policy Evaluation: A Critique.. “A Model of Balance of Payments Crises. 2001. Enrique. Chapter 8 295 Kim.” manuscript. 1997.E. M.Lecture in Open Economy Macroeconomics..

Maurice. Mendoza. and John H.” Journal of Monetary Economics (26). November 2003. and dynamics.” Department of Economics Working Papers 027. Enrique G. James M. Obstfeld. impatience.” Carnegie-Rochester Conference Series on Public Policy (53). 101-137. and Fabrizio Perri. Rogers. 159-187.” American Economic Review. Maurice.296 ican Economic Review (81). “Sudden Stops. “Rational and Self-Fulﬁlling Balance of Payments Crises. 239-96. Obstfeld. Maurice. the Real Exchange Rate. May 1982. “Intertemporal dependence. Enrique. “Business Cycles in Emerging Markets:The Role of Interest Rates. 797-818.” Quarterly Journal of Economics (97). 251-270. “Aggregate Spending and the Terms of Trade: Is There a Laursen-Metzler Eﬀect?.. Case Study: Argentina. Universidad Torcuato Di Tella. Pablo Andres and Martin Gonz´lez Rozada. February 1995. Nason. “The Present-Value Model of the Current Account Has Been Rejected: Round Up the Usual Suspects. Pablo A. December 2000. Mendoza. Neumeyer. “The Terms of Trade. 2010.” Journal of International Economics (68). 1990. 1991. Neumeyer.” .” International Economic Review (36). and Economic Fluctuations. Financial Crises and Leverage. November 2001. “The elasticity of a Substitution in demand for Non tradable Goods in Latin America. January 2006. Mart´ Uribe ın Mendoza.” manuscript. New York University. 45-75. forthcoming. “Devaluation Risk and the Businessın Cycle Implications of Exchange-Rate Management. Enrique and Mart´ Uribe. Obstfeld.

297 Obstfeld. and Miguel Savastano. Carmen M. 357-378. The MIT Press. Bank of France (43). business cycles on the Canadian economy. “Nominal interest rates. 189-213. Maurice. “The Logic of Currency Crises. Foundations of International Macroeconomics. Schmitt-Groh´.” Journal of International Economics (44).” Cahiers Economiques et Monetaires. Maurice. 1994. Schmitt-Groh´. Reinhart.” Journal of Development Economics (46). 1996. April 1995. 189-206.” Journal of Money. “Solving Dynamic General e ın Equilibrium Models Using a Second-Order Approximation to the Policy Function. May 2001.” Journal of Economic Dynamics and Control (vol.” Journal of International Economics (61). Obstfeld. Stephanie and Mart´ Uribe. Rose. and Banking (33). and Kenneth Rogoﬀ. “One Reason Countries Pay Their Debts: Renegotiation and International Trade. “The international transmission of economic ﬂuce tuations: Eﬀects of U. V´gh. Andrew K. 1986. 1-74. Kenneth Rogoﬀ. cone sumption booms. Stephanie and Mart´ Uribe.Lecture in Open Economy Macroeconomics. Credit. April 1998. pp. Chapter 8 American Economic Review (76). “Closing Small Open Economy e ın Models. .. Carmen M.” Journal of Development Economics (77(1)).S. 163-185 . 2003.. “Stabilization Policy and the e ın Costs of Dollarization. 2005.” Brookings Papers On Economic Activity (2003(1)). Massachusetts. Schmitt-Groh´. 482-509. October 2003. Stephanie. Reinhart. Stephanie and Mart´ Uribe. 28).. 257-287. 72-81. Schmitt-Groh´. Cambridge. and Carlos A. and lack of credibility: A quantitative examination. “Debt Intolerance.

755-775. Uribe. Stockman. 168-85. Lars E. Handbook of Macroeconomics. Mart´ “The Price-Consumption Puzzle of Currency Pegs. Tesar. 417-421.” Journal of Political Economy (91). Taylor and Michael Woodford. of Monetary Economics (49). Yale University/ 1996. Uribe. 533-569. February. “Country Spreads and Emerging Counın tries: Who Drives Whom?. p. 197-221. and Assaf Razin.” Journal of Monetary Economics (39).” Journal of International Economics (69). “Business Cycle Fluctuations in US Macroeconomic Time Series. University of Pennsylvania. .. “Adjustment of a Small Open Economy to External Shocks..” American Economic Review Papers ın. eﬀects of credible plans. Svensson. James H. Mart´ Uribe ın Senhadji. Watson. Vivian Yue. Sims. 1995. “The Terms of Trade and the Current Account: The Harberger-Laursen-Metzler Eﬀect. Amsterdam.” Journal ın. June 2006. 1983. Uribe. May 2006. Christopher.” Dissertation.” in John B. Alan C. 1999. Mart´ and Z. Abdelhak S. 6-36. April 2002. 3-64. pp..O. 1994. 1997. Stock. “Tastes and Technology in a TwoCountry Model of the Business Cycle: Explaining International Comovements. 97-125. eds. Mart´ “Exchange rate based inﬂation stabilization: the initial real ın. The Netherlands: Elsevier Science.298 January 2004. and Proceedings (96).” mimeo. Uribe.” American Economic Review (85). Mart´ “On Overborrowing. “Solving Linear Rational Expectations Models. and Mark W. and Linda L.

485-504. Capital and Growth: Papers in Honor of Sir John Hicks.Lecture in Open Economy Macroeconomics. Value. Mart´ “Individual Versus Aggregate Collateral Constraints and the ın.” NBER working paper 12260. Uzawa.N. 1968.). H. the Consumption Function and Optimum Asset Holdings. J. Overborrowing Syndrome.” in Wolfe. . (Ed. pp. Chapter 8 299 Uribe. “Time Preference. The University of Edinburgh Press Edinburgh. May 2007..

Sign up to vote on this title

UsefulNot useful- LTexpDyPol15september2012.pdf
- WEF_GCR_CountryProfilHighlights_2011-12
- Transition to Market Economy - Alfred Levinson
- Cities and Economic Prosperity
- GtE Course Compendium
- HOUSE HEARING, 110TH CONGRESS - FULL COMMITTEE HEARING ON LEGISLATION TO PROVIDE FUNDING LEVELS
- Lecture 37
- Memo to Congress from Larry Summers on the American Economy
- The moral economy of the peasant:
- ECON2123 the Goods Market
- The Elder Scrolls IV Oblivion Economy
- Nightly Business Report - April 26 2013
- c4.pdf
- 02 Chapter 2
- 18_GABRIELA_DOBROTA
- 2013 Zambia Economic Report
- REAL 215-724 Class 02 01-21-2015 (5)-2
- Northern Ireland Knowledge Economy Index and Metrics 2014 Northern Ireland Science Park
- GEC Newsletter February 2010
- Atul Auto_Auto and Ancillary_Mahim Dwivedi
- Industry Overview
- Informal Economy Final
- CA modelling
- Water and Green Economy Report
- Fao Sse Def1
- 978f8947-10df-4ee7-a308-518ba4075acd
- 07EconomicDevelopment-goalsstatus_201305090928201253
- Commodity Mcx Trading Tips and News
- [6] Annual Worth Analysis
- Equalisation Levy
- Lecture Notes

Are you sure?

This action might not be possible to undo. Are you sure you want to continue?

We've moved you to where you read on your other device.

Get the full title to continue

Get the full title to continue reading from where you left off, or restart the preview.

scribd