Undervalued Currency in Spite of Growth: The Role of Current Account Imbalances

Transcription

1 Undervalued Currency in Spite of Growth: The Role of Current Account Imbalances Byoung Hoon Seok y Preliminary February 15, 2011 Abstract For the past 30 years, China has grown more than ve percent per year and surprisingly has run a persistent current account surplus with little appreciation in its real exchange rate. This fact contradicts the predictions of the standard growth theories. This paper attempts to resolve this contradiction by using learning-by-exporting. Many empirical studies nd that exporters of developing countries improve their productivity through technology spillover from foreign countries. Based on these empirical studies, this paper builds up a two country general equilibrium growth model with learningby-exporting in China. The model is calibrated to match relevant data moments of the U.S. and China in 1981 and simulated for a transition to a new steady state. In this setup, China has a strong incentive to raise its amount of export by undervaluing its currency in order to take advantage of learning-by-exporting. One way China undervalues its currency is accumulating a current account surplus, as we observe in the data. This paper shows that China s currency undervaluation policy improves the welfare of the U.S. as well as China and that China needs to make an adjustment to its currency undervaluation policy because the present level of its current account surplus is too excessive. Keywords: Current Account, Learning-by-Exporting, Real Exchange Rate JEL Classi cations: F31, F32, F41, F43, O24 I am deeply indebted to Mark Aguiar for his support and encouragement. I also thank Mark Bils, Yongsung Chang, William Hawkins, Jay Hong, Ronald Jones, Nobuhiro Kiyotaki, Ryan Michaels, Ronni Pavan, Aleh Tsyvinski, Hye Mi You, and seminar participants at the University of Rochester for their valuable comments. All errors are mine. y Dept. of Economics, University of Rochester, Rochester, NY byoung.seok@rochester.edu 1

2 1 Introduction For the past 30 years, China has grown more than ve percent per year 1 and surprisingly has run a persistent current account surplus during that time 2. This feature contradicts the prediction of the neoclassical growth model that countries with faster productivity growth should receive more net capital in ows because of higher return to capital. Over the same period, China s real exchange rate did not appreciate 3. This fact contrasts with the Balassa (1964)-Samuelson (1964) hypothesis that growth driven by productivity gains in traded goods sector implies the real exchange rate appreciation. This feature is named China s real exchange rate puzzle by Tyers and Golley (2008). This paper attempts to resolve the contradictions between the two facts and the predictions of standard theories. I build a two country general equilibrium growth model with learning-by-exporting in China. The empirical studies nd that exporters of developing countries improve their productivity through imitation and technology spillover from foreign countries 4. In this setup, China has a strong incentive to raise its amount of export by undervaluing its currency in order to take advantage of learning-by-exporting. One way China undervalues its currency is accumulating a current account surplus, as we observe in the data. This shows that China s currency undervaluation policy improves the welfare of the U.S. as well as China; China is better o due to rapid economic growth through learning-by-exporting, while the U.S. receives a welfare gain from improved terms of trade. In addition, the paper indicates that China needs to make an adjustment to its currency undervaluation policy because the present level of its current account surplus is too excessive. 1 See Figure 2. 2 See Figure 1. 3 See Figure 2. 4 These studies include Kraay (1999), Park, Yang, Shi, and Jiang (2010) for China, Aw, Chung, and Roberts (2000), Hallward-Driemeier, Iarossi, and Sokolo (2002), Blalock and Gertler (2004), Aw, Roberts, and Winston (2007), Aw, Roberts, and Xu (2010) for East Asian countries, Van Biesebroeck (2005) for Aftrican countries, De Loecker (2007) for Slovenia, and Fernandes and Isgut (2009) for Colombia. 2

3 In this paper, I solve the world planner s problem to obtain the rst-best allocation and then compare it with a competitive equilibrium allocation in which no one recognizes the positive externality of China s export. The model is calibrated to match relevant data moments of the U.S. and China in 1981 and simulated for a transition to a new steady state. This shows that the world planner reduces China s consumptions for both non-traded goods and exporting goods, and increases its consumption for importing goods in order to raise China s export. This policy makes China grow faster while preventing its real exchange rate from appreciating. Moving from the rst-best outcomes to the competitive equilibrium outcomes reduces per-period consumptions of the U.S. and China by 2.7 percent and 32.5 percent, respectively. However, in the rst-best outcome, China runs a current account de cit. This implies that the present level of its current account surplus is too excessive even if we consider China s learning-by-exporting. Therefore, China needs to make an adjustment to its currency undervaluation policy. To the best of my knowledge, this paper is one of the rst to use learning-by-exporting in order to explain both China s real exchange rate puzzle and its current account surplus simultaneously. Thus, it is at the junction of two distinct lines of literature. This paper is related to the papers on the worldwide current account imbalances. Caballero, Farhi, and Gourinchas (2008) and Mendoza, Quadrini, and Rios-Rull (2009) claim that the lack of - nancial assets in developing countries has generated current account imbalances between the U.S. and developing countries. Fogli and Perri (2006) argue that the great moderation has raised the U.S. current account de cit by reducing their incentive to accumulate precautionary savings. Gourinchas and Jeanne (2009) nd that fast growing developing countries have run a current account surplus. Since this fact contradicts the neoclassical growth theory, they named it the "allocation puzzle". Aguiar and Amador (2010) insist that since capital will not be invested in a country with high debt, because of the risk of expropriation, only politically stable developing countries can grow by reducing sovereign debt which attracts foreign capital. Song, Storesletten, and Zilibotti (2009) argue that the expansion of nancially constrained rms in China has made it grow rapidly and accumulate a current account 3

4 surplus. However, these papers are silent on the undervalued currency of rapidly growing developing countries. This paper is also related to the literature on currency undervaluation and economic growth. Rodrik (2008) argues that currency undervaluation boosts economic growth in developing countries due to their severe market failures and institutional weaknesses in the traded goods sector. However, there is no indication as to why fast growing developing countries run a current account surplus. The remainder of the paper is organized as follows. In Section 2, I describe the model. Section 3 explains the calibration of the model. In Section 4, I show the results and do the welfare analysis. Section 5 concludes my ndings. 2 Model The model I present is a two country general equilibrium growth model with learning-byexporting in the home country. This section is organized as follows. Section 2.1 describes environment. In Section 2.2, in order to gure out the rst-best allocation, I explain a world planner s problem. Section 2.3 explains the way I decentralize the rst-best allocation. Section 2.4 describes a competitive equilibrium in which no one recognize that there is learning-by-exporting in the home country. 2.1 Environment There are two countries in this economy. Let a superscript denote a foreign country. The foreign country which corresponds to the U.S. owns the highest level of human capital stock H. The home country, which is equivalent to China, has the inferior human capital H t 2 (1; H ]; where 2 (0; 1], and can grow through learning-by-exporting. Thus, there is a positive externality in the home country s export. There are a single non-traded commodity in each country and two traded goods denoted by z 2 f1; 2g. There is one internationally traded bond b t with a return denoted by 1 + r t. 4

5 A representative worker supplies labor inelastically for domestic rms and gets utility ows from consumption of a non-traded commodity c t and two traded goods c t (z) where i 1 z 2 f1; 2g : The worker discounts the future with a discount factor 2 0; 1+r t and has preferences: where 1X t u(c t ); t=0 u(c t ) C1 t 1 1 ; C t c 1 t c t (1) c t (2) (1 ) ; ; 2 (0; 1) : Trade goods sector rms use labor n t (z) to produce y t (z) according to a production function y t (z) = A t (z) n t (z); where A t (z) (H t ) exp(z) ; z 2 f1; 2g : Note that the home country has a comparative advantage in the rst traded commodity production: A t (2) A t (1) = (H t) exp(2) (H t ) A (2) exp(1) A (1) = (H ) exp(2) (H ) : exp(1) A non-traded goods sector rm uses labor n t to produce y t according to a production function y t = n t : The home country s human capital stock evolves according to the law of motion for human capital H t+1 = H t + (H H t ) (EX t ) ; where (EX t ) 1 exp EXt ; EX t max f(y t (1) c t (1)) ; 0g + A (1) A (2) max f(y t(2) c t (2)) ; 0g : 5

6 The second term in the right side of the law of motion, (H H t ) (EX t ) ; represents the learning-by-exporting. The home country s human capital can grow up to H through learning-by-exporting and captures the degree of learning-by-exporting. 2.2 World Planner s Problem A world planner maximizes the world utility by taking advantage of the home country s learning-by-exporting. Thus, the world planner solves 1X max t fu(c t ) + (1 ) u(ct )g ; t=0 subject to c t = n t ; c t = n t ; c t (z) + c t (z) = A t (z) n t (z) + A (z) n t (z); n t (z) 0; n t (z) 0; z 2 f1; 2g ; n t + n t (1) + n t (2) = N t ; n t + n t (1) + n t (2) = Nt ; H t+1 = H t + (H H t ) (EX t ) ; where is the home country s Pareto weight. If the home country produces the rst traded commodity, the world utility maximizing behavior of the planner implies a rst-order condition: u 0 (C t t (1) + t (H H t ) 0 (EX t t (1) u 0 (C t ) = (1 ) u 0 (C t t (1) ; t (1) A t (1) = u 0 (C t t ; (2) where t is a multiplier for the law of motion for human capital. The second term in the left side of the condition (1), t (H H t ) 0 (EX t(1) ; shows up due to the learning-byexporting and is a negative number. Thus, this condition implies that when the home country exports the rst traded commodity, the planner reduces the home country s consumption 6

7 of exporting goods in order to take advantage of the learning-by-exporting. The condition (2) shows that there is no wedge between the home country s consumption of exporting goods and that of non-traded goods. Therefore, the conditions (1) and (2) imply that the planner decreases not only the home country s consumption of exporting goods but also its non-traded goods consumption. This means that the planner subsidizes the home country s consumption of importing goods in order to take advantage of home country s learning-byexporting. 2.3 Decentralization In this section I explain the way I nd prices and wedges which imply the rst-best allocation. Assume that there are no wedges in the foreign country s prices and let the foreign country s wage w be a numeraire. Then, the price of traded commodity z 2 f1; 2g is de ned by p t (z) u0 (C t t (z) u 0 (C t t Since the home country has a comparative advantage in the rst traded commodity production, it exports the rst traded commodity. Therefore, by the rst traded commodity sector rm s pro t maximization, the home country s wage is Each country s non-traded goods prices are w t p t (1)A t (1) : p t w t ; p w = 1 by the non-traded goods sector rms pro t maximization. By the foreign country s Euler equations, the world interest rate is de ned by : where P t 1 p r t+1 P t+1 u 0 (Ct ) 1; Pt u 0 Ct+1 1 pt (1) 7 pt (2) (1 ) (1 ) :

8 Then, the home country s real exchange rate is where e R t P t P t P t 1 pt = pt p 1 = 1 pt (1) wt w 1 = (wt ) 1 ; pt (2) (1 ) (1 ) : Since the law of one price holds in the traded goods sector, the home country s real exchange rate is de ned by the ratio of each country s non-traded commodity price. A wedge r t+1 in the home country s domestic interest rate and a wedge p t in the home country s domestic relative price of importing goods are de ned by r t+1 P t+1 u 0 (C t) P tu 0 (C t+1 ) 1 r t+1 1 () P t+1 u 0 (C t ) P t u 0 (C t+1 ) p t p t(2) u 0 (C t(1) p t (1) u 0 (C t(2) 2.4 Competitive Equilibrium 1 () u0 (C t(2) u 0 (C t(1) = 1 = 1 + r t+1 rt+1 ; p t (2) (1 + p t ) p t (1) : Assume that no one recognizes that there is learning-by-exporting in the home country. A representative worker in the home country solves subject to Traded goods sector rms solve max fc t;c t(1);c t(2);b t+1 g 1X t u(c t ); t=0 p t c t + p t (1)c t (1) + p t (2)c t (2) + b t+1 = w t + f1 + r t g b t : max fp t (z)a t (z) n t (z) fn t(z)g w t n t (z)g ; z 2 f1; 2g and a non-traded goods sector rm in the home country solves max fn tg fp tn t w t n t g : Then, a competitive equilibrium consists of a set of allocations fc t ; c t (z); c t ; c t (z); b t ; b t ; H t g and a set of prices fp t ; p ; p t (z); w t ; w ; r t g such that: 8

9 1. given prices, workers maximize utilities; 2. rms maximize pro ts; 3. the home country s human capital evolves according to the law of motion for human capital; 4. goods markets clear c t = y t ; c t = y t ; c t (z) + c t (z) = y t (z) + y t (z) ; z 2 f1; 2g ; 5. labor markets clear n t + n t (1) + n t (2) = N t ; n t + n t (1) + n t (2) = N t ; 6. a bond market clears b t = b t : 3 Calibration This section explains how I set parameter values described in Table 1. I set a set of parameters using the U.S. data and the standard values used in a related literature. The model period is one year and discount factor is set to 0:95. Both the home country s labor N and the foreign country s labor N are 1. Among parameters related to preferences, the coe cient of risk aversion, ; is set to 2 and the expenditure share of traded goods, ; is 0:32 which is the average of the U.S. real personal consumption expenditure share of goods from 1981 to Another set of parameters is selected so that the model can replicate relevant data moments of the U.S. and China. A parameter which governs the degree of learning-byexporting is chosen to match the average real GDP per capita growth rate of China from 9

10 1981 to The foreign country s human capital stock H is set to 3:56 which implies the real GDP per capita of the U.S. to that of China ratio in The expenditure share of traded commodity 1 is calibrated by 0:34 which maximizes the length of periods when both countries are completely specialized. The home country s Pareto weight is picked so that the home country s current account is balanced at the new steady state. 4 Result In this section, I explain the quantitative results and do the welfare analysis. This section is organized as follows. Section 4.1 shows the result of a world planner s problem. In Section 4.2, I explain a competitive equilibrium allocation. Section 4.3 compares the welfare of the rst-best allocation with that of the competitive equilibrium allocation. 4.1 World Planner s Problem From period 1 to 12, the home country specializes on traded commodity 1 production and the foreign country produces both trade goods because the home country s human capital stock is much less than that of the foreign country. Then, as the home country s human capital stock grows due to learning-by-exporting, both countries are completely specialized from period 13 to 23. From period 24, the home country produces both traded goods and the foreign country specializes on trade commodity 2 production. At period 40, the new steady state begins. Figure 3 shows that the world planner reduces the home country s consumption for both nontraded commodity and exporting goods, and raises its consumption for importing goods in order to take advantage of learning-by-exporting. Both countries consumption grows as the home country s human capital grows due to learning-by-exporting. According to Figure 4, the world planner subsidizes the home country s import and saving. By subsidizing the home country s import and saving, the world planner can make its labor move from the non-traded goods sector to the exporting goods sector and make the home country consume importing goods instead of exporting goods. Therefore, the home country can grow faster through 10

11 learning-by-exporting. Figure 5 shows that when both countries are completely specialized, the home country s human capital and real GDP grow rapidly. However, over the same period, its real exchange rate appreciates slowly. If both countries produce same traded goods, then the home country s relative wage is determined by the its labor productivity to the foreign country s labor productivity ratio. However, when both countries are completely specialized, the world planner can reduce the home country s relative wage by moving its labor from non-traded goods sector to exporting goods sector. Therefore, the planner make the home country grow fast with little appreciation in its real exchange rate. According to Figure 6, the home country runs a current account de cit. This implies that the present level of China s current account surplus is too excessive even if we consider its learning-byexporting. Therefore, China needs to make an adjustment to its currency undervaluation policy. 4.2 Competitive Equilibrium Since no one recognizes the positive externality of the home country s export, the transition to the new steady state in the competitive equilibrium takes two times longer than that in the world planner s problem. Figure 7 indicates that the home country does not reduce its consumption for non-traded goods. Thus, it does not take advantage of learning-byexporting. According to Figure 8 and 9, as the home country s human capital and real GDP grow, its real exchange rate appreciates rapidly and it runs a current account de cit. Note that in the competitive equilibrium, the home country s saving is two times less than that in the world planner s problem. Therefore, the home country cannot undervalue its currency. 4.3 Welfare Analysis According to Table 2, when the planner takes advantage of the home country s learningby-exporting, both countries are better o. Moving from the rst-best outcomes to the competitive equilibrium outcomes reduces per-period consumptions of the foreign country and the home country by 2.7 percent and 32.5 percent, respectively. This shows that China s 11

12 currency undervaluation policy improves the welfare of the U.S. as well as China; China is better o due to rapid economic growth through learning-by-exporting, while the U.S. receives a welfare gain from improved terms of trade. 5 Conclusion This paper presents a two country general equilibrium growth model with learning-byexporting in China in order to explain China s rapid economic growth, current account surplus, and undervalued currency. In this setup, China has a strong incentive to raise its amount of export by undervaluing its currency in order to take advantage of learningby-exporting. One way China undervalues its currency is accumulating a current account surplus, as we observe in the data. This paper shows that China s currency undervaluation policy improves the welfare of the U.S. as well as China and that China needs to make an adjustment to its currency undervaluation policy because the present level of its current account surplus is too excessive. References Aguiar, M., and M. Amador (2010): Growth in the Shadow of Expropriation, Quarterly Journal of Economics, Forthcoming. Aw, B. Y., S. Chung, and M. J. Roberts (2000): Productivity and Turnover in the Export Market: Micro-level Evidence from the Republic of Korea and Taiwan (China), The World Bank Economic Review, 14(1), Aw, B. Y., M. J. Roberts, and T. Winston (2007): Export Market Participation, Investments in R&D and Worker Training, and the Evolution of Firm Productivity, The World Economy, 30(1), Aw, B. Y., M. J. Roberts, and D. Y. Xu (2010): R&D Investment, Exporting, and Productivity Dynamics, Mimeo. 12

13 Balassa, B. (1964): The Purchasing Power Parity Doctrine: a Reappraisal, Journal of Political Economy, 72(6), Blalock, G., and P. J. Gertler (2004): Learning from Exporting Revisited in a Less Developed Setting, Journal of Development Economics, 75(2), Caballero, R., E. Farhi, and P.-O. Gourinchas (2008): An Equilibrium Model of aµrglobal ¾ Imbalancesaś ¾ and Low Interest Rates, American Economic Review, 98, De Loecker, J. (2007): Do exports generate higher productivity? Evidence from Slovenia, Journal of International Economics, 73, Fernandes, A. M., and A. E. Isgut (2009): Learning-by-Exporting E ects: Are They for Real?, Mimeo. Fogli, A., and F. Perri (2006): The Great Moderation and the US External Imbalance, Monetary and Economic Studies, 24, Gourinchas, P.-O., and O. Jeanne (2009): Capital Flows to Developing Countries: The Allocation Puzzle, Mimeo. Hallward-Driemeier, M., G. Iarossi, and K. L. Sokoloff (2002): Exports and Manufacturing Productivity in East Asia: A Comparative Analysis with Firm-Level Data, NBER Working Paper No Kraay, A. (1999): Exportations et Performances Economiques: Etude d un Panel d Entreprises Chinoises, Revue d Economie Du Developpement, 1-2, Mendoza, E. G., V. Quadrini, and J.-V. Rios-Rull (2009): Financial Integration, Financial Development and Global Imbalances, Journal of Political Economy, 117(3),

14 Park, A., D. Yang, X. Shi, and Y. Jiang (2010): Exporting and Firm Performance: Chinese Exporters and the Asian Financial Crisis, Review of Economics and Statistics, Forthcoming. Rodrik, D. (2008): The Real Exchange Rate and Economic Growth, Brookings Papers on Economic Activity, pp Samuelson, P. A. (1964): Theoretical Notes on Trade Problems, Review of Economics and Statistics, 46(2), Song, Z., K. Storesletten, and F. Zilibotti (2009): Growing Like China, Mimeo. Tyers, R., and J. Golley (2008): Chinaaŕs ¾ Real Exchange Rate Puzzle, Journal of Economic Integration, 23(3), Van Biesebroeck, J. (2005): Exporting Raises Productivity in Sub-Saharan African Manufacturing Firms, Journal of International Economics, 67(2),

15 Table 1: Parameter Values Parameter Description Source/ Target = 0:95 Discount factor = 2 Coe cient of risk aversion = 0:32 Expenditure share of traded goods U:S:(81 07) = 0:34 Expenditure share of traded commodity 1 Home country s = 0:28 CA T = 0 Pareto weight = 20:94 Learning-by-exporting Y N = 0:07 Foreign country s H = 3:56 Y =N = 25:47 Y=N human capital 81 N = 1 Home country s labor N = 1 Foreign country s labor Table 2: Welfare Analysis Laissez-faire First-best Welfare Gain (%) Home P 80 t=0 t u(c t ) Foreign P 80 t=0 t u(ct )

16 Figure 1: Current Account Imbalances of the U.S. and China Figure 2: China s Real Exchange Rate and Economic Growth 16

17 Figure 3: Consumption (World Planner s Problem) Figure 4: Wedges (World Planner s Problem) 17

18 Figure 5: Growth and Real Exchange Rate (World Planner s Problem) Figure 6: Current Account Imbalances (World Planner s Problem) 18

19 Figure 7: Consumption (Competitive Equilibrium) Figure 8: Growth and Real Exchange Rate (Competitive Equilibrium) 19

20 Figure 9: Current Account Imbalances (Competitive Equilibrium) 20