Financial Integration, Growth, and Volatility

Transcription

1 W/05/67 Financial Integration, Growth, an Volatility Anne paular an Aue ommeret

2 005 International Monetary Fun W/05/67 IMF Working aper IMF Institute Financial Integration, Growth, an Volatility repare by Anne paular an Aue ommeret 1 Authorize for istribution by Anrew Feltenstein April 005 Abstract This Working aper shoul not be reporte as representing the views of the IMF. The views expresse in this Working aper are those of the author(s) an o not necessarily represent those of the IMF or IMF policy. Working apers escribe research in progress by the author(s) an are publishe to elicit comments an to further ebate. The aim of this paper is to evaluate the welfare gains from financial integration for eveloping an emerging market economies. To o so, we buil a stochastic enogenous growth moel for a small open economy that can (i) borrow from the rest of the worl, (ii) invest in foreign assets, an (iii) receive foreign irect investment (FDI). The moel is calibrate on 3 emerging market an eveloping economies for which we evaluate the upper boun for the welfare gain from financial integration. For plausible values of preference parameters an actual levels of financial integration, the mean welfare gain from financial integration is about 10 percent of initial wealth. Compare with financial autarky, actual levels of financial integration translate into slightly higher annual growth rates (aroun 0.4 percentage point per year.) JL Classification Numbers: 13, F0, F36, F3, F43, O41 Keywors: Financial Integration, Risk-sharing, nogenous Growth, Stochastic Growth Author(s) -Mail Aress: Anne.paular@auphine.fr; Aue.ommeret@hec.unil.ch 1 Anne paular is with the University of aris Dauphine, an Aue ommeret is with the University of Lausanne. The authors starte working on this paper while Anne paular was a senior economist in the Asian Division of the IMF Institute. They benefite from comments from hilippe Bacchetta, Jean-ierre Danthine, Anrew Feltenstein, Olivier Jeanne, Françoise Le Gall, ascal Saint-Amour, an participants at the IMF Institute seminar, an the 003 SD conference in aris.

3 - - Contents age I. Introuction... 4 II. Financial Integration: Facts an mpirical vience... 5 A. Measuring Financial Integration... 5 B. mpirical vience... 7 III. The Small Open conomy Stochastic nogenous Growth Moel... 8 A. Technology an the Sharing of GD between the Representative Agent an Owners of FDI... 9 B. Other Features of the Moel IV. Solving the Moel... 1 A. ortfolio Allocation...1 B. Growth C. Volatility...14 D. Welfare Gain From Financial Integration V. Measuring the Welfare Gain From Financial Integration A. The Data B. Calibration Issues C. valuating the Welfare Gain From Financial Integration D. Checking for Robustness...0 VI. Conclusion... Appenices 1. Solving the Moel Data Constructing Capital an Debt Stocks More Integrate an Less Integrate conomies References Figures 1. Splitting Up the Welfare Gain From Financial Integration...8. Dynamics of Financially Integrate an Financial Autarky conomies...9

4 - 3 - Tables 1. Measuring Financial Integration for 3 Developing an merging conomies in Common arameters Country-Specific arameters Calibrate arameters uner the Borrowing Constraint Hypothesis Welfare Gain From Financial Integration (average country) Sensitivity to reference arameters an Interest Rate Calibrate Risk remium on Foreign Debt Welfare Gain From Financial Integration Risk remium Hypothesis Welfare Gain From Financial Integration when FDI Boosts Overall rouctivity... 7 Appenix Tables 10. Data Sources an Definitions Constructing Inices of Financial Integration Distinguishing Between LIs (Less Integrate conomies) an MIs (More Integrate conomies)...35

5 - 4 - I. INTRODUCTION What are the benefits of financial integration for eveloping an emerging market economies? conomic theory tells us that financial integration brings four types of benefits: First, financial integration potentially allows risk sharing. As in the basic portfolio allocation moel, global iversification reuces the portfolio risk for a given expecte rate of return. In turn, this reuction in risk affects omestic saving. Secon, financial integration potentially eases the capital scarcity constraint a eveloping/emerging economy might face. Thir, financial integration may bring foreign irect investments (FDI) into the country. art of the income generate with this FDI remains in the country because of taxes or other trickle-own effects. Finally, financial integration may boost prouctivity through a number of channels. For instance, when financial integration increases FDI, foreign investors bring into the country not only capital, but their technology as well. However, there is mixe empirical evience on whether capital account liberalization an financial integration have resulte in increase long-run economic growth in eveloping economies (see the recent survey by ison, Klein, Ricci, an Sløk, 00a). How can we reconcile theoretical preiction with empirical evience? The absence of clear empirical evience on the benefits of financial integration may result from the fact that the benefits liste above exist but are small, either because financial integration is not complete yet or because there is not much to gain even uner complete financial integration. The gap between theoretical preiction an empirical evience can also come from the fact that, in aition to the benefits pointe out by economic theory, financial integration brings potential losses (associate, for instance, with suen stops or reversals of capital flows) that cancel out the benefits. The aim of this paper is to give a measure of the potential benefits that countries coul have seen given the nature of their financial integration an its level. To o so, we buil an calibrate a stochastic growth moel that encompasses most of the features that are suppose to convert financial integration into economic growth for 3 eveloping an emerging economies. The aim of this calibration is to erive an upper boun (inee, we consier only the potential benefits of financial integration an completely neglect potential losses that can arise following suen stops, reverse capital flows, etc.) for the benefits of financial integration of emerging an eveloping economies, given their actual egree of financial integration. This upper boun provies a benchmark for econometric stuies on the effect of financial integration. To be more precise, we buil a stochastic enogenous growth moel for a small open economy that can (i) borrow from the rest of the worl so as to relax its capital constraint, (ii) invest in foreign assets so as to benefit from risk sharing, an (iii) receive FDI. In aition, there are two sources of uncertainty in the moel prouctivity shocks an foreign prices shocks. We erive

6 - 5 - a close-form solution for the risk premium that the country is facing, the optimal portfolio choice of the omestic investor, the growth rate of the economy, an the welfare gain from financial integration. The moel is then calibrate on 3 emerging an eveloping economies for which we can evaluate the upper boun for the welfare gain from financial integration. For plausible values of the preference parameters, we conclue that the gains from financial integration are not huge. These gains are nevertheless significant since they represent about 10 percent of the existing wealth an, in terms of growth ifferential, actual financial integration brings about an aitional 0.4 percentage points compare with growth uner financial autarky. Note that our assessment leas to gains that are a lot higher than those compute by Gourinchas an Jeanne (004). Comparing actual economies with their fully integrate counterpart, they obtain a gain from financial integration that represents only aroun 1 percent of consumption each year along the growth path of the economy. The paper is organize as follows. Section II provies measures of financial integration of emerging an eveloping economies an briefly iscusses the empirical evience on the effects of financial integration on economic growth an volatility. In Section III, we evelop an solve a simple stochastic enogenous growth moel that encompasses the main theoretical features linking financial integration, growth, an volatility. In Section IV the moel is calibrate on 3 emerging an eveloping economies; we then compute an upper boun for the effect of observe financial integration on economic growth. Finally, in Section V, we check the robustness of our evaluation of the welfare gain from financial integration by calibrating the moel uner ifferent assumptions. Section VI conclues. II. FINANCIAL INTGRATION: FACTS AND MIRICAL VIDNC A. Measuring Financial Integration There is not yet a measure of the intensity of financial integration that is wiely use by economists. xisting measures of financial integration, incluing that of Quinn (1997) that looks at capital account regulations, focus on potential financial integration rather than on actual financial integration. Using Quinn s measure of financial integration, a country that impose no restrictions on capital flows, but that receive or sent little capital flows, woul qualify as financially integrate without actually benefiting from financial integration. Table 1 presents two measures of actual financial integration for 3 emerging an eveloping economies. The first one (inex 1) is the ratio of the sum of all claims an liabilities of a country to its GD. Its components are also given for G-3 economies (the Unite-States, Japan, an Germany) so as to provie a benchmark against which actual financial integration of eveloping an emerging economies can be compare. The secon measure (inex ) exclues government an monetary authority claims an liabilities; it is the ratio of claims on foreign assets (excluing foreign exchange reserves) an liabilities (excluing public an publicly-guarantee ebt) of a country to its GD. Inex aims

7 - 6 - to measure the part of financial integration that is not channele by the government an the monetary authorities of the country. Within our set of 3 emerging an eveloping economies, we istinguish two subsets. The first one inclues economies that are more financially integrate than the average country of our sample; the secon one inclues economies that are less financially integrate than the average country of our sample. In what follows, we will refer to these two groups as LIs (less integrate economies) an MIs (more integrate economies). Figures in Table 1 show that the intensity an nature of financial integration iffer a lot from one group to another, an across regions. Inex 1 varies from 0.4 for LIs to 0.88 for MIs. For both LIs an MIs, the main component of inex 1 is government ebt an governmentguarantee ebt (31 percent of GD for LIs an 35 percent of GD for MIs). For MIs, the secon-largest component is claims on foreign assets (excluing foreign exchange reserves), which amount to 33 percent of GD. For LIs, it is foreign exchange reserves. As measure by inex 1, financial integration is as high in MIs as it is in G-3 economies. However, the nature of financial integration iffers consierably between these emerging economies an G-3 economies. The typical MI has a foreign ebt of more than 50 percent of GD an receives FDI, whereas G-3 economies exhibit a high inex of financial integration thanks to their holings of foreign assets Inex, which exclues government an monetary authorities, ranges from 0.6 for LIs to 0.75 for MIs. The main component of inex is the stock of FDI for both LIs (1 percent of GD) an MIs (19 percent of GD). For MIs, the stock of private ebt (18 percent of GD) is close to that of FDI, an equity liabilities amount to a mere 3 percent of GD. For LIs, private foreign ebt is only 6 percent of GD, an equity liabilities amount to only 1 percent of GD. ut together, these figures show that for eveloping an emerging economies, actual financial integration takes place through FDI an ebt. Therefore, gains of financial integration shoul mainly be relate to these two components of financial integration. Moreover, with equity liabilities that are only 3 percent of GD (at most), stock market liberalization, although a potential source of gains, shoul not be responsible for a large share of the benefit of actual financial integration. Finally, as shown in the bottom line of Table 1, eveloping an emerging economies in our sample are net borrowers; at the country level, because foreign assets are more than offset by ebt, there is no risk sharing. See Appenix 4.

8 - 7 - B. mpirical vience Research aiming to measure benefits of financial integration usually procees by comparing two economies that iffer in their level of financial integration. But there are at least three types of comparisons. The comparison can be between an economy with financial autarky an its fully financially integrate counterpart. This is the approach followe by most papers base on the calibration of small theoretical moels; it focuses on the potential gain of full financial integration (see, for example, Obtsfel, 1994, or Athanasoulis an Van Wincoop, 000). A secon approach consists in comparing an actual economy with its fully financially integrate counterpart. This is the approach followe by Gourinchas an Jeanne (004). Finally, a thir approach consists in comparing the economic performance of an actual economy (with its current level of financial integration) with its counterpart uner a lower level of financial integration or uner financial autarky. This is the approach followe in econometric stuies. This paper follows this approach as well. valuating the potential gain of financial integration by calibrating small theoretical moels, Obtsfel (1994) an Athanasoulis an Van Wincoop (000) foun that risk-sharing gains that woul come with financial integration are huge. However, in a eterministic growth moel in which, by construction, risk sharing oes not bring any benefit, Gourinchas an Jeanne (004) show that the welfare gain from capital account liberalization is rather small (aroun 1 percent of consumption each year along the growth path of the economy). As far as the effect of capital account liberalization on volatility is concerne, Kose, rasa, an Terrones (003) conclue by reviewing the literature on international business cycles, that theoretical moels (RBC moels, NOM moels) o not provie a clear guie to the effects of financial integration on macroeconomic volatility. A recent paper by Tille (004) confirms this conclusion. In a twocountry NOM moel, he shows that the impact of integration is not universally beneficial (it epens on the egree to which exchange rate fluctuations are passe through to consumer prices). The comparison of economies at ifferent stages of their financial integration process provies mixe empirical evience that capital account liberalization promotes long-run economic growth in eveloping economies (see the survey by ison, Klein, Ricci, an Sløk, 00). erforming econometric estimations using a new ataset, ison, Klein, Ricci, an Sløk (00b) o not fin any significant positive effect of financial integration on economic growth. However, Henry (003) reports empirical evience that stock market liberalizations (a component of capital account liberalization an financial integration) are followe by lower cost of capital, higher investment, an higher growth. Finally, Bekaert, Harvey, an Lunbla (001, 004a) report that, on average, equity market liberalizations lea to a 1 percent increase in economic growth over a five-year perio. mpirical literature is also mixe regaring the effect of financial openness on macroeconomic volatility. For example, Razin an Rose (1994) fin no evience of a link between trae an financial openness an macroeconomic volatility. A recent paper by asterly, Islam, an Stiglitz (000) conclues that neither financial openness nor volatility of capital flows has a significant

9 - 8 - impact on macroeconomic volatility. They show, however, that a higher level of evelopment of the omestic financial sector (as measure by private creit to GD) reuces growth volatility. Buch, Dopke, an ierzioch (00) o not fin any consistent empirical relationship between financial openness an the volatility of output. Nevertheless, Bekaert, Harvey an Lunbla (004b) fin a significant ecrease in both GD an consumption growth variability after equity market liberalizations. In this paper, we follow the approach that consists in comparing an actual economy to an economy that is less financially integrate. However, we o not rely on econometric estimations, but rather on the calibration of a theoretical moel. Therefore, we provie a theoretical framework to unerstan the mixe empirical results reviewe above. Moreover, our moel consiers simultaneously the potential effects of FDI, of the openness of financial markets, an of risk sharing on both growth an volatility. III. TH SMALL ON CONOM STOCHASTIC NDOGNOUS GROWTH MODL To measure the gain from financial integration, we exten a stanar macroeconomic moel to account for financial integration: First, in the financially integrate economy, the representative agent has access to global financial markets. She can hol riskless foreign bons an contract ebt (either to consume or invest in omestic capital). Both foreign bons an ebt are enominate in foreign currency. The agent chooses the amount of ebt an bons that she hols. Depening on her net position, the agent pays or receives interest. Note that, in our moel, the representative agent oes not have access to foreign risky capital. Secon, the financially integrate economy receives FDI, the amount of which is not ecie by the representative agent. This FDI is converte into prouctive capital an use for prouction. The income generate by this FDI goes partly to the foreign owners; the rest stays in the country. An equilibrium conition ensures that for the foreign investors, the certainty equivalent of the return on FDI matches that of international riskless bons. The rest of the moel is a stanar AK moel with technological shocks. In aition, the representative agent consumes omestic an importe goos, foreign prices are subject to shocks, an part of the omestic prouction is exporte. The trae balance, an current an capital accounts simultaneously ajust to ensure balance of payment equilibrium. Note that later in the paper the term autarky economy refers to an economy that is in financial autarky but traes with the rest of the worl. The moel is presente below.

10 - 9 - Technology A. Technology an the Sharing of GD Between the Representative Agent an Owners of FDI The small open economy prouces one goo using capital. The technology is AK, an temporary technological shocks perturb the prouction process. Over the perio (t, t+t), the flow of output is: = K( µ t + σ z) (1) where z is the increment of a stanar Wiener process ( z = η ( t) t; η( t) N(0,1)). quation (1) asserts that the flow of output over the perio (t+t) consists of two components: a eterministic component ( Kµ t ) an a stochastic one ( Kσ z ) reflecting the ranom influences that impact on the prouction. Thus, as far as prouctivity is concerne, shocks are neither correlate nor persistent. In the absence of financial integration, the whole stock of prouctive capital installe in the country is owne by the omestic representative agent. With financial integration, aitional prouctive capital (K*) comes from FDI. This FDI, owne by foreign investors, is use for prouction. The rest of the stock of capital (K) is owne by the representative agent so K = K + K*. We efine φ* to be the ratio of foreign-owne capital to omestically owne capital (φ*= K*/K). In the autarky economy, φ* = 0. In the financially integrate economy, we suppose that φ* is constant over time, an assumption that permits us to erive a close-form solution for the representative agent s intertemporal allocation problem. Finally, we will later 3 allow the eterministic prouctivity to epen on the ratio of foreignowne capital to omestically owne capital, so µ y = µ y (φ*). Sharing Income from rouction Income generate with capital owne by the omestic agent goes to the omestic agent (who still has to pay interest on his foreign ebt). In aition, a fraction τ of the income generate with foreign-owne capital (FDI) goes to the omestic agent to account, for instance, for taxes or any trickle-own effect; 4 the rest of the income generate by FDI goes to the foreign owners. Thus, before interest payments/revenues on foreign ebt/claims, the omestic agent income is given by: 3 See Section IV.D of the paper. 4 It can be wages pai to resient as well.

11 [ µ σ z ][ 1+ τφ *] K +. y Foreign owners of FDI receive the fraction (1-τ ) of the income generate by their capital that is not kept by the omestic agent. Therefore, the revenue from their investment is: ( µ + σ z ) K* = (1 τ )( µ σ z ) * K ( 1 τ ) + φ. y This income is stochastic. We impose the conition that for FDI owners, the certaintyequivalent return on risky omestic capital is equal to the riskless rate on foreign bons (i*): ( ) ( ) ( ) 1 τ µ γ 1 τ σ / = i * + µ, () y where γ is the foreign investor s risk aversion, an µ is the expecte rate of epreciation of the omestic nominal exchange rate against the foreign investor currency. Note that conition () is not sufficient to etermine enogenously the amount foreign investors are willing to invest in the country. It just says that foreign investors are inifferent between holing riskless foreign bons an holing risky omestic prouctive capital. This conition satisfie, investors can invest any amount in the small open economy. Their participation is measure by φ *, the ratio of foreign-owne capital to omestically owne capital. We take φ * as an exogenous parameter. Moreover, uner the assumption that this ratio is constant over time, the flow of foreign investment into the economy is proportional to the investment of omestic agents in omestic capital 5. B. Other Features of the Moel The representative househol consumes a omestically prouce goo ( C D ) an an importe one ( C M ). Total consumption in terms of omestic currency then epens on both the price of the importe goo enominate in foreign currency an the exchange rate. y 5 We o not moel the behavior of the foreign investor but make sure he is inifferent between investing in riskless bons in his country an risky FDI in the eveloping country. The assumption that his participation to the economy is constant over time implies that the stock of FDI he owns grows at the eveloping economy growth rate. In orer to moel the behavior of the foreign investor one woul have to consier a multicountry moel. This is not one in this paper.

12 The Utility Function We consier a recursive utility function that isentangles risk aversion an intertemporal substitution: (1 γ ) ε ε 1 ε 1 θ 1 θ ε 1 ε 1 Dt Mt ( n δ ) t ( γ ) 1 γ ε nt t t+ t Le 0 C C (1 γ) Ut ( ) = t+ e (1 ) U with ε > 0, ε 1, anγ 1, Where γ is the relative risk aversion coefficient of the omestic agent (an we suppose it is equal to that of the foreign investor),ε is the inter-temporal elasticity of substitution, L 0 is the population size at the initial ate which is assume to be unitary ( L 0 = 1), an n is the population growth rate. Domestically prouce an importe consumptions are aggregate θ θ accoring to a Cobb-Douglas function, so C 1 gives the aggregate consumption. Trae, Real an Nominal xchange Rates DtC Mt Let be the price of the importe goo enominate in foreign currency. We assume that follows a geometric Brownian motion: = µ t + σ z, where z is the increment of a stanar Wiener process. Therefore, the price of the importe goos grows at a constant rate µ continuously perturbe by shocks. The instantaneous stanar eviation of the growth rate is enote by σ. We assume that prouctivity shocks an shocks on are uncorrelate ( z = 0 ). z Total consumption in terms of omestic currency or omestic goos (the price of the omestically prouce goo is taken as numeraire) is CD + CM, where is the nominal exchange rate efine as the number of omestic currency units per unit of foreign currency. is therefore the price of the importe goo in terms of omestic currency. The nominal exchange rate evolves at a constant an eterministic rate µ consistent with conition (): with: = µ t, ( µ γ ( 1 τ ) / ) i * µ = (1 τ ) σ, The evolution of, the real exchange rate, is then given by: ( ) = ( µ + µ ) t + σz, ( ) (3)

13 - 1 - Wealth Accumulation, Capital Mobility, an the Flow of FDI In the autarky economy, the representative agent has no access to foreign financial market, an the only way of saving is investment in risky omestic capital. Wealth evolution is then given by: ( ) W = K = µ K C + C t + Kσ z D M When the economy is financially integrate, the representative agent can hol equity claims on omestic capital an riskless foreign bons (B>0). She can also contract ebt enominate in foreign currency (B<0). Therefore, omestic wealth is split into two components: capital owne by omestic agents ( K ) an foreign ebt/assets (B): W = K + B, Wealth evolution is as follows: W = + K + B i + C + C t + + K z * (1 τφ *) µ ( µ ) ( D M) (1 τφ*) σ Balance of ayments The balance of payments equilibrium is always satisfie: ( µ t + σ z) K K AWt (1 τ ) φ * K ( µ t + σ z) + B( i * + µ ) + φ * K ( B) Trae Balance Current Account Balance of ayments = [ K ( B) ] = 0 = ( µ t + σ z) K AWt (1 τ ) φ * K ( µ t + σ z) + B( i * + µ ) W W IV. SOLVING TH MODL The resolution of the moel is given in Appenix 1. In this section, we present the main finings on portfolio allocation, growth, volatility, an, finally, the welfare gain from financial integration. A. ortfolio Allocation In the autarky economy, there is no portfolio choice since the only asset available to the representative agent is omestic capital.

14 Uner financial integration, maximizing the utility function subject to the wealth accumulation equation provies expressions for the share of the two assets (as well as for consumption of the importe goo an omestically prouce goo) in the composition of the wealth of the representative agent. These shares are (see Appenix 1): n K K 1 = = + W γ(1 + τφ*) σ [ i ] (1 + τφ *) µ γ(1 + τφ*) σ / ) * + µ (3) n B = = 1, W B n k As in any other portfolio choice, the optimal portfolio composition epens on expecte returns an risk. The share of omestic capital in the omestic agent s portfolio can be larger than 1, meaning that the omestic investor borrows abroa to invest in omestic capital. The fact that the economy benefits from FDI (τ >0) increases the eterminist part of the return on omestic capital for the omestic agent an thus his incentive to borrow abroa to invest at home. It increases as well the volatility of this return, thus limiting the incentive to borrow. In the case where the economy oes not benefit from foreign investment ( τ = 0 ), conition () implies that the expecte return on omestic capital ajuste for risk is the same as the one on claims on foreign assets. Therefore, the omestic agent hols half of her wealth as omestic capital, the other half being investe abroa for iversification purposes. B. Growth In the autarky economy, the eterministic part of the GD growth rate is: A A g = µ A, where A A A A is the propensity to consume wealth, equal to: A = εδ n+ (1 ε ) Cq, where Ceq A is the certainty equivalent of the return on wealth. This expression for the marginal propensity to consume is stanar. Note that, thanks to the recursive utility specification, it is clear that the effect on A A of the certainty equivalent Ceq A epens on the intertemporal elasticity of substitution. When the intertemporal elasticity of substitution is less than 1, a higher Ceq A increases the propensity to consume, an hence reuces the growth rate of the economy. In turn, risk aversion oes affect the relationship between risk an Ceq A. : Cq σ σ ( ) (( ) ). A = µ y γ (1 θ ) µ + µ 1 θ (1 γ) + 1 When the economy is financially integrate, the eterministic part of the economy growth rate is: L L g = (1 + τφ *) n µ + (1 n )( i* + µ ) A ; (4.a) K K

15 A L, the constant propensity to consume wealth, is equal to: L L A = εδ n+ (1 ε ) Cq, (4.b) where Ceq L is the certainty equivalent of the return on wealth, which epens on the portfolio allocation. It is equal to: C L q = n k (1 + τφ*) µ y γn k σ (1 + τφ*) + (1 n k )( i * + µ ) (1 ) θ ( µ + µ ) σ [(1 θ )(1 γ ) + 1] (4.c) In the financially integrate economy, part of the certainty equivalent of the return on wealth comes from foreign ebt/assets, an part of it comes from omestic capital income (net of the income that goes to foreign investors). In a net borrowing country ( n K > 1), the stock of physical capital is higher both because omestic agents borrow abroa to invest at home an because there are FDI flows. The component of the certainty equivalent of the return on omestic activity ue to eterministic prouctivity increases, but the component linke to the stochastic part (which reuces Ceq L ) rises as well. In the case of net leners ( n K < 1), it is even more ambiguous since more wealth comes from foreign investors, but omestic agents now evote part of their wealth to invest abroa. The conition for international financial integration to spur growth is that g conition is not met for any set of parameters. L A > g. This In the simplest case where the intertemporal elasticity of substitution is equal to 1, an the country is neither a lener nor a borrower, financial integration spurs growth (one can show that g L g A = τφ µ y > 0) through FDI. Moreover, when the intertemporal elasticity of substitution is equal to 1, one can show that the erivatives of g L g A to n k are always positive, which means that the more ebt the country hols, the higher its growth rate. Let us recall, though, that the optimal amount of ebt is not a parameter but is enogenously etermine within the moel. When the intertemporal elasticity of substitution is ifferent from 1, the effect of financial integration on growth is ambiguous an cannot be analyze analytically. Therefore, we rely on calibration experiments in the next section. Volatility in the Growth Rate of Wealth C. Volatility L Wealth volatility in a financially integrate economy (enote vol ) is:

16 vol L ( 1+τφ*) nkσ =. Comparing this volatility with that of wealth in the autarky economy (enote L vol = ( 1+τφ*) n A K. (5) vol A vol ) gives: In the case of a net borrower country ( n K > 1), financial integration increases total wealth volatility. Moreover, the larger the stock of FDI, the higher the volatility. For countries that are net leners ( n K < 1), current account liberalization may either increase or ecrease wealth volatility. Volatility in the Growth Rate of Consumption Solving the maximization problem provies expressions for the consumption of importe an omestic goos (see Appenix 1): C i D = i AθW i i C = A (1 θ ) W with i = A, L M The volatility of the growth rate of total nominal consumption ( CD + CM ) an that of omestic consumption are the same as that of wealth since they both are a constant part of wealth. This is not the case for the volatility of importe consumption, which is (see Appenix 1): vol C = vol + i = A L. i i ( M) σ with, Volatility of the growth rate of aggregate consumption is then (see Appenix 1): vol θ 1 θ i i ( C C ) = vol + 1 θ ) σ D M ( with i=a,l. Consumption volatility is higher than that of output owing to foreign price volatility. Moreover, consumption volatility is higher in the financially integrate economy than in the autarky economy as soon as wealth volatility is higher in the financially integrate economy. D. Welfare Gain from Financial Integration Total Welfare Gain from Financial Integration The welfare gain from financial integration is compute as the percentage of current wealth that the omestic representative agent shoul receive to be as well off in the autarky economy as she

17 is in the financially integrate one 6. We enote this percentage by k. Note that at the time t=s of the switch, the wealth of the agent is W ( s) = K ( s) + B( s) = K( s), which means that the wealth is at the same level before an after the switch, but that its composition iffers. After the switch, it is only mae up of omestic capital. The value functions for the optimal program or the inirect utilities of the two economies are as follows (see the appenix 1). One can show that: 1 γ (1 γ)(1 θ) 1 γ A θ ( θk) 1 ε (1 γ ) nt ( ) A V = A e 1 θ (1 γ) ( ) ( K + B) 1 γ 1 γ (1 γ)(1 θ) L θ θ 1 ε (1 γ ) nt L V = A e 1 θ (1 γ) k 1 1 ε L A = 1 A A. (6) Splitting-Up the Welfare Gain From Financial Integration To appraise whether the welfare gain from financial integration comes from FDI or from the openness to foreign ebt/assets, we break own the total welfare gain compute previously. To split the total welfare gain from financial integration, we consier three ifferent economies. Two of them have alreay been consiere above the autarky economy an the actual economy. We consier a thir economy that receives FDI but is close to other capital flows. We then procee in two stages as illustrate in Figure 1. First, we consier the switch from the actual economy to an economy with FDI but no access to global financial markets; this allows us to compute the welfare gain from access to global financial markets. Next, we consier the switch from the latter economy to autarky, an this gives us the welfare gain of FDI. L Let us note V, the optimal value of the program of an economy with FDI but no other capital flows. One can show that it is: L A = εδ n + ( 1 ε) L Cq V Cq L L = ( A L ) 1 γ 1 σ θ 1 θ (1 γ )(1 θ ) 1 γ [ θk ] (1 γ ) nt 1 γ [ µ γ (1 + τφ*) σ / ] (1 θ )[ µ + µ (( 1 θ )( 1 γ ) 1) σ / ] = (1 + τφ *) + 6 The efinition we use is the compensating variation. e.

18 The welfare gain from access to financial markets is then compute as the percentage of wealth that the omestic representative agent shoul receive to be as well off in the economy where there is FDI but no access to global financial markets as she is in the actual economy. It is: 1 1 L ε AFM A k = 1. (7) L A Finally, the welfare gain from FDI is the percentage of wealth that the omestic representative agent shoul receive to be as well off in the autarky economy as she is in the economy with FDI but no other capital flows. It is then: L ε FDI A k = 1 (8) A A The three welfare gains are such that: 1 1 FDI AFM (1 + k) = 1 + k 1+ k V. MASURING TH WLFAR GAIN FROM FINANCIAL INTGRATION A. The Data To measure the upper boun for the welfare effect of financial integration, we calibrate our moel on 3 eveloping an emerging economies over the perio Our aim is to measure the welfare gain of a country representative agent that woul result from the observe level of financial integration of that country in the specific (an optimistic) case where no costly sie effects of financial integration hurt the country. Tables an 3 contain common an country-specific parameters use in the calibration. Common parameters (Table ) are set to stanar values with a iscount rate of percent; an international interest rate of 4 percent, an intertemporal elasticity of substitution set at 0.5, an a risk aversion of 5. Robustness of our results to preference parameters an the international interest rate is checke for aitional values inicate in Table. Country-specific parameters (see Table 3) have been compute using the enn Worl table (HS 6.1) an the atabase on stock of international wealth put together by Lane an Milesi-Ferreti (001). For our evaluation, the two most important parameters are φ *, the ratio of foreignowne capital to omestically owne capital, an n K, the share of omestic capital in the representative agent s wealth. The mean value of n K is 1.09, just above unity an ranges from 0.73 (Botswana) to 1.4 (akistan). Among the 3 countries we consier, only 7 of them are net leners (Botswana, China, gypt, Malaysia, anama, South Africa, an Venezuela). The ratio of foreign-owne capital to omestically owne capital varies from 0 (Syria), which means that

19 there is no FDI in this country, to 0. (Malaysia), which means that 18 percent of the physical capital installe in Malaysia is owne an manage by foreigners. B. Calibration Issues Our theoretical moel has been solve uner the hypothesis that the representative agent maximizes her intertemporal utility. In oing so, she allocates her wealth between omestic capital an ebt/assets abroa accoring to her tastes, returns, an risk. She also etermines her savings an consumption accoring to her tastes. We ha no ifficulties calibrating the intertemporal choice of the omestic agents of the countries we stuie. The portfolio choices were trickier to reprouce. Actual ata show that the volatility of emerging an eveloping economies, although higher than that of inustrial economies, is low compare with the excess return of omestic investment for the representative agent. This shoul lea emerging an eveloping economies to borrow a lot abroa to invest at home, or, in our setting, a high n k. This is not what we observe in the ata. Numbers reporte in Table 3 for actual n k show that it is above but close to unity, much lower than what one woul expect. Our calibration strategy is then to stick to the n k observe in the ata an consier that this low level of inebteness results from borrowing constraints that prevent emerging countries from holing their optimal portfolio. Consequences of this strategy will be iscusse later in the paper. Note, however, that since portfolio choices an consumption-saving ecisions are inepenent, the value function of the programs still hols uner portfolio constraint. Thus, the value function of the program can be use to evaluate the welfare gain from financial integration. Calibration runs as follows. We use equation () an the system compose of equations (4.a)- (4.c) to fin the parameter τ, which measures the share of prouction using FDI that goes to omestic agent, the parameter µ, the expecte change in the nominal exchange rate, an the prouctivity parameter µ. Note that conition () still hols when the omestic agent is constraine in her borrowings. Inee, this conition is erive from the perspective of foreign agents owning FDI in the country who are not affecte by the borrowing constraint. It is the expression of the optimal share n K (equation (3)) that is no longer vali uner the borrowing constraint. The growth rate of the economy (g L ), the volatility (σ ), an the portfolio allocation n k are taken from the ata (see Table 3). We calibrate the moel for each country as well as for the average country, which we efine as a country with country-specific parameters set to the mean of the sample. Table 4 contains the outcome of this calibration process. Recall that the parameter τ measures the extent to which a country benefits from FDI. Our calibration inicates that for the average country of our sample, 37 percent of the prouction on foreign-owne plants goes to the

20 omestic agent; in our sample, it ranges from a mere percent to 51 percent, with a mean of 5 percent. The expecte annual epreciation of emerging market currency against the ollar ranges from 0 percent to 8 percent, lower than actual between 1990 an C. valuating the Welfare Gain From Financial Integration We now use the calibrate moel to compute the welfare gain from financial integration. Recall that we compare actual economies with economies uner financial autarky. An the question we answer is the following: by how much shoul the total wealth of the representative agent be increase for her to accept to switch back to financial autarky? This is what we call the welfare gain from financial integration. It is compute using equation (6).This gain is then split into two components: the gain from FDI (equation (8)) an the gain from access to global financial markets (the ability to borrow an/or to invest abroa, equation (7)). Finally, we compute the ifference in the growth rates of the two economies (the actual one an the one in financial autarky) an their relative volatility (equation (5)). Results are shown in Table 5. For the average country of our sample, the gain from financial integration is aroun 11 percent of the representative agent s wealth, an the gain comes equally from FDI (4.9 percent of wealth) an from access to global financial markets (5.6 percent). Compare with autarky economies, we calculate that, on average, actual economies enjoy an aitional 0.31 percentage points of annual growth. This comes at the cost of more volatility, which is 5 percent higher in actual economies compare with the autarky situation. Figure illustrates the ynamics of prouction, consumption, an capital accumulation in the two economies. The soli line refers to actual economies as escribe in our moel, an the otte line refers to the autarky economy. Let us consier what happens to the economy at the time the representative agent agrees to switch back to autarky an is compensate so as to keep intertemporal utility the same. At the beginning, the representative agent is richer (her wealth increases by the compensating amount she receives), but there is less prouctive capital available in the autarky economy since it is close to FDI. As a result, GD is lower in the autarky economy, while gross national income (GD less interest an ivien payments), is almost the same in the two economies (note that in the autarky economy, GNI is equal to GD). ropensity to consume wealth is higher in the actual economy than in the autarky one, but because wealth is initially higher in the autarky economy, it turns out that initial consumption is higher too in the autarky economy. From this initial situation, all macroeconomic aggregates in each economy grow at the same rate, which is lower in the autarky economy, an eventually the two economies iverge. The ivergence process is slow. In our calibration, because the ifference in the annual growth rates is small (aroun 0.3 percent per year), it takes aroun twenty years for consumption in the integrate economy to catch up with that in the autarky economy. The ashe line in Figure illustrates the ynamics of the autarky economy in which no compensation is provie to the representative agent at the time she leaves the financially integrate economy. This economy exhibits the same growth rate as the autarky economy with compensation pai, but starts at a lower level. The comparison of these two economies gives us a measure of the compensation in terms of permanent consumption. It is straightforwar to show that it is equal to the percentage compensation in terms of initial wealth. Thus, the welfare

21 - 0 - gain of financial integration can also be evaluate at aroun 11 percent of permanent consumption. It is of interest to istinguish the welfare gain from financial integration by level of financial integration an by region. It is no surprise that the gain from financial integration is higher for countries that are more financially integrate, since we measure the gain from complete financial autarky to the actual level of financial integration. However, one can notice that the gain from financial integration in terms of annual growth rates is not that high. Our calibration says that more financially integrate economies woul grow by an aitional 0.4 percentage point compare with autarky, while less financially integrate economies grow by an aitional 0.16 percentages point. The ifference between these two figures is only 0.6 percentage point per year. This rather small number may explain why econometric stuies so far have not provie clear evience on the effect of financial integration on economic growth. Turning to the gain from financial integration by region reporte in Table 5, we see that the average Asian country benefits more from financial integration than countries in any other region. Financial integration inexes reporte in Table 1 inicate that Latin America was the most financially integrate region, yet our results show that it benefits less from financial integration than Asia; the welfare gain from financial integration is equal to 11.6 percent for Asia compare with 9.7 percent for Latin America. This unerscores that the egree of financial integration is not the only component of the welfare gain from financial integration; the nature of financial integration (FDI, ebt) an the capacity of the country to benefit from FDI (parameter τ) also matter. D. Checking for Robustness Sensitivity to reference arameters an Interest Rate To check for the robustness of the above evaluation of the welfare gain from financial integration, we start by moifying preference parameters (risk aversion an intertemporal elasticity of substitution) an international interest rates. Table 6 contains the results. The welfare gain from financial integration is a ecreasing function of all these parameters. An international interest rate 1 percent higher reuces the overall evaluation of the welfare gain from financial integration by 1 percent (from 10.9 percent to 9.8 percent). A higher risk aversion translates into a smaller welfare gain. This comes from the fact that financial integration, an the foreign borrowing that accompanies it in most emerging economies, increases the volatility of gross national income. As a consequence, the more risk averse the representative agent, the smaller the gain from financial integration. In our calibration, changing the risk aversion parameter from to 5 reuces the welfare gain from 10.9 percent to 8.7 percent. Finally, a higher intertemporal elasticity of substitution reuces the sensitivity of the marginal propensity to consume with respect to financial integration. Therefore, the larger the intertemporal elasticity of substitution, the smaller the gains from financial integration.

22 - 1 - Alternative Calibration Strategy So far our calibration strategy has been to consier that, because of borrowing constraints, countries were unable to hol their optimal portfolio. As sai before, given the moel parameters an the relatively low volatility in most emerging countries, our moel preicte that emerging countries woul want to borrow a lot abroa to invest at home. This is not what is observe in the ata, which show that emerging countries are inebte in reasonable proportions (with nk slightly above 1). This iscrepancy between observe an preicte portfolios coul also come from the existence of a risk premium on foreign borrowing. Inee, we have assume so far that investors require a risk premium only on risky FDI, but not on lening that was consiere as risk-free for the foreign investor. We now relax this hypothesis an calibrate a risk premium (R) on emerging country borrowing such that the observe n K is optimal. Uner this risk premium assumption, equation (5) becomes: W = [( 1+ τφ *) µ K + B( i * + µ + R) ( CD + CM )] t + (1 + τφ*) σ K z, an the optimal portfolio share is now given by: n K [(1 + τφ*) µ γ (1 + τφ*) σ / ] [ i * + µ + R] 1 = +. (9) γ (1 + τφ*) σ The growth rate of the financially integrate economy is: with g L = ( 1+ µ ) L τφ *) nk µ + (1 nk )( i * + + R A, an Cq L = n K L L = εδ n + ( 1 ε Cq, A ) σ (1 + τφ*) µ (1 + *) + (1 γ τφ n )( i * + µ + R) σ (1 θ ) ( µ + ) (( 1 )(1 ) + 1). µ θ γ Note that equation (), which expresses the require return on FDI, remains the same. To calibrate this new version of our moel, we keep all the other parameters equal to their value in the previous calibration an calculate the R that exactly solves equation (9). The risk premium that we obtain on foreign borrowing is given in Table 7. For the average country, it is aroun 3 percent an varies from 8 percent to 1 percent. It is higher for more financially integrate economies than for less integrate ones. Finally, it is relatively stable across regions. K

23 - - Table 8 shows that the welfare gain from financial integration is much lower (aroun 5 percent of initial wealth) when we introuce a risk premium, although small, on foreign borrowing. The ecomposition of the welfare gain shows that the gain from FDI is unchange compare to with previous calibration. On the contrary, the gain from access to global financial markets becomes a lot smaller almost negligible. This is not surprising. The risk premium emerging countries have to pay on their ebt reuces the benefit they get from borrowing. This was not the case in the previous calibration in which they only bear the riskless international interest rate augmente for currency epreciation. Allowing for a Higher rouctivity of FDI In the introuction of this paper, we mentione that financial integration may provie aitional benefits to the country when FDI brings in new technology that raises the overall prouctivity of the economy. So far, our moel an our calibration strategy o not allow for such an effect. The calibration can be easily moifie to take it into account an compute its impact on the welfare gain from financial integration. To o so, let us assume that FDI makes capital use in the country (FDI itself + omestic capital) more prouctive. The larger the FDI, the more important this prouctivity gain. Therefore, we assume that the gain in prouctivity is proportional to the size of the FDI: D µ = ( 1+ X φ*) µ, where µ is the prouctivity in the autarky economy. It is less than µ as soon as X is positive. In terms of computing the welfare cost of financial integration, we use the same calibration for the financially integrate economy, but reuce to µ the prouctivity parameter in the autarky economy. In the calibration below we set X=0.5, which correspons, for the average country in our sample, to a 9 percent increase in the overall prouctivity in the financially-integrate economy. Results of that experiment are shown in Table 9. Of course, the fact that FDI affects the overall prouctivity oes not change the welfare gain from access to global financial markets. The welfare gain from FDI is significantly higher. VI. CONCLUSION In this paper, we have calibrate a theoretical moel in orer to appraise the welfare gains from actual financial integration in eveloping an emerging economies. We provie an upper boun to the gains from financial integration since we focus on the potential benefits an neglect potential losses. Gains we have compute are not huge, but significant, since they represent about 10 percent of existing wealth. In particular, the gains in terms of growth ifferential cannot be ignore; actual financial integration brings about 0.3 percentage point of growth per year. Moreover, we show that the welfare gains from financial integration are nearly equally split between the gains from access to global financial markets an those ue to FDI. Finally, when we allow for FDI to raise omestic prouctivity we obtain a large welfare gain, with financial integration translating into 0.5 percentage point of growth per year.