1. East Asian Currency Union October 2006 Jong-Wha Lee Korea University and Robert J. Barro Harvard University

2. Motivation Are Current Exchange Rate Arrangements in East Asia Appropriate? Before the crisis, most East Asian economies adopted de facto U.S. dollar peg systems. Before the crisis, most East Asian economies adopted de facto U.S. dollar peg systems. After the financial crisis, most of the crisis-affected East Asian economies switched to floating exchange-rate regimes. After the financial crisis, most of the crisis-affected East Asian economies switched to floating exchange-rate regimes. But, they tend to intervene heavily in the foreign exchange market. But, they tend to intervene heavily in the foreign exchange market. ― ‘fear of floating’ (Calvo and Reinhart, 2002), ― ‘fear of floating’ (Calvo and Reinhart, 2002), ― ‘export-led growth strategy’ (Dooley et al., 2003) ― ‘export-led growth strategy’ (Dooley et al., 2003) The instability of US dollar and the need for intra- region exchange rate stability ask for a new viable regime. The instability of US dollar and the need for intra- region exchange rate stability ask for a new viable regime.

3. Motivation (Continued) Can a Currency Union be a Choice for East Asia? McKinnon and Mundell have proposed a currency union with the US dollar as the common currency. McKinnon and Mundell have proposed a currency union with the US dollar as the common currency. Others suggest East Asia can emulate the European experience of monetary integration, leading eventually to the formation of an East Asia-wide Monetary Union with a new currency. Others suggest East Asia can emulate the European experience of monetary integration, leading eventually to the formation of an East Asia-wide Monetary Union with a new currency. There has been a trend toward evolution of currency unions in the world due to: There has been a trend toward evolution of currency unions in the world due to: (i) trade and financial globalization, and (i) trade and financial globalization, and (ii) emphasis by monetary authorities of price stability over macroeconomic stabilization. (Alesina, Barro and Tenreyro (2002)) (ii) emphasis by monetary authorities of price stability over macroeconomic stabilization. (Alesina, Barro and Tenreyro (2002))

4. Purpose of the Paper (I) Assess the feasibility of a common currency arrangement in East Asia, particularly compared to the euro area Assess the feasibility of a common currency arrangement in East Asia, particularly compared to the euro area – We investigate whether East Asia meets the traditional OCA criteria: trade integration, the symmetry of output and price shocks across countries, and commitment to price stability. – We will also discuss financial integration and political proximity among the East Asian countries, particularly compared to those of European countries.

5. Purpose of the Paper (II) Estimate the welfare effects of joining a currency union for individual East Asian economies. Estimate the welfare effects of joining a currency union for individual East Asian economies. – We calculate welfare effects of various types of currency unions, such as a dollar bloc, Euro bloc, Yen bloc, and a new regional currency bloc, for individual East Asian economies. – We compare potential benefits from increasing trade with potential costs from the loss of monetary policy independence. – We present estimation results from calibrations of a representative consumer model.

6. Benefits and Costs of Currency Unions Benefits (of currency unions or credible pegs) : Benefits (of currency unions or credible pegs) : – Provides a nominal anchor for monetary policy and a more credible commitment to price stability. – Promotes trade and investment by reducing uncertainty and transaction costs. – Prevents competitive devaluation. – Avoids speculative bubbles in exchange rates. – Helps a developing country to increase its access to long-term financing. Costs: Costs: – Causes loss of independent monetary policy – Hinders automatic adjustment to external shocks. – Loss of seigniorage revenue and a lender-of-last-resort capability.

7. Optimum Currency Area (OCA) Criteria Trade openness Trade openness Symmetry of shocks Symmetry of shocks Labor mobility Labor mobility Speed of adjustment Speed of adjustment History of inflation and variability of relative prices History of inflation and variability of relative prices Financial integration Financial integration Political proximity Political proximity

8. Which Currency Union for East Asia? We want to evaluate the costs and benefits of joining a currency union. A potential client country in East Asia may either adopt an existing major currency, such as the US dollar, Euro, or Japanese yen, or create a new regional currency. We want to evaluate the costs and benefits of joining a currency union. A potential client country in East Asia may either adopt an existing major currency, such as the US dollar, Euro, or Japanese yen, or create a new regional currency. We adopt the approach suggested by Alesina, Barro and Tenreyro (2002) and uses updated data to assess the feasibility of East Asian currency unions. We adopt the approach suggested by Alesina, Barro and Tenreyro (2002) and uses updated data to assess the feasibility of East Asian currency unions. Financial integration is measured by the size of bilateral international financial asset holdings, based on the IMF, the Coordinated Portfolio Investment Survey (CPIS). Financial integration is measured by the size of bilateral international financial asset holdings, based on the IMF, the Coordinated Portfolio Investment Survey (CPIS). Political proximity is measured by bilateral vote correlations at the United Nations General Assembly. Political proximity is measured by bilateral vote correlations at the United Nations General Assembly.

9. Table 1. Average Trade-to-GDP Ratio for East Asia and Europe, 1990-2003 (percent) Trade Partners Country USAEuropeJapanEast AsiaWorld China3.93.43.713.320.6 Hong Kong12.814.010.663.6119.8 Indonesia3.34.16.114.724.7 Japan2.11.4.3.28.3 Korea5.53.64.611.527.7 Malaysia15.311.713.950.882.6 Philippines9.65.37.320.335.7 Singapore22.318.816.672.0141.0 Taiwan11.26.39.225.451.1 Thailand6.76.68.621.341.8 Average9.37.59.029.655.3 Europe Average2.321.10.92.531.2 United States.1.91.13.09.0

10. Table 2. Mean and Standard Deviation of Annual Inflation Rates for East Asia and Europe Country1975-19891990-2003 means.d.means.d. China3.83.95.36.7 Hong Kong, China8.74.12.66.1 Indonesia13.08.913.817.8 Japan3.62.4-0.21.6 Korea, Rep.13.38.55.53.4 Malaysia3.95.53.32.5 Philippines13.311.78.23.3 Singapore3.23.31.22.9 Taiwan4.94.81.52.2 Thailand5.63.23.53.1 Average7.35.64.55.0 Europe Average10.54.33.32.2 United States5.62.62.20.8

11. Table 3. Co-Movements of Prices for East Asia and Europe, 1975-1989 and 1990-2000 1975-19891990-2000 CountryUSEuropeJapan East Asia USEuropeJapan East Asia China0.1090.1380.1560.1050.1480.1990.1720.195 Hong Kong0.0640.0970.1300.0820.0320.1080.1050.138 Indonesia0.1150.1790.1890.1320.2980.3080.2570.238 Japan0.1330.105.0.1390.0990.149.0.133 Korea, Rep.0.0900.1200.1230.0880.1440.1590.1110.119 Malaysia0.0550.1100.1360.0840.1490.1940.1560.141 Philippines0.0730.1270.1530.0830.1200.1260.1230.126 Singapore0.0410.1040.1250.0780.0700.0990.0860.109 Taiwan0.0870.1090.1240.0870.0690.1030.0960.110 Thailand0.0510.0980.1190.0720.0860.1100.0880.110 Average0.0820.1190.1390.0950.1220.1550.1330.142 Europe Average0.1050.0640.1050.1180.1130.0550.1490.155 United States.0.1050.1330.082.0.1130.0990.122

12. Table 4. Co-Movements of Outputs for East Asia and Europe, 1975-1989 and 1990-2003 1975-19891990-2003 CountryUSEuropeJapan East Asia USEuropeJapan East Asia China0.0620.0610.0590.0650.0400.0430.0400.050 Hong Kong0.0400.0490.0500.0490.0420.0480.0350.039 Indonesia0.0320.0400.0310.0410.0520.0560.0430.039 Japan0.0300.025.0.0380.0250.027.0.038 Korea, Rep.0.0340.0440.0380.0470.057 0.0490.046 Malaysia0.0340.0360.0380.0420.0290.0350.0280.034 Philippines0.0500.0420.0350.0460.0330.0370.0310.039 Singapore0.041 0.0340.0420.0430.0520.0430.052 Taiwan0.0250.0370.0360.0400.0240.0320.0290.041 Thailand0.0280.0330.0190.0370.0530.0570.0450.040 Average0.0380.0410.0380.0450.0400.0440.0380.042 Europe Average0.032 0.0250.0410.0230.0260.0270.044 United States.0.0320.0300.038.0.0230.0250.040

13. Table 5. International Portfolio Asset Holdings by East Asia and Europe in 2003. (Percent in GDP) Destination Country/Region Source Country USAEuropeJapanEast AsiaWorld China--- Hong Kong29.857.66.734.9213.8 Indonesia0.20.10.00.10.9 Japan14.414.10.00.540.0 Korea1.30.50.00.22.9 Malaysia0.30.40.00.71.6 Philippines3.10.90.00.34.6 Singapore24.761.33.831.9157.5 Taiwan--- Thailand1.20.40.00.11.9 Average9.416.91.38.652.9 Europe Average21.262.12.53.6108.0 United States.15.12.74.128.6

14. Table 6. International Portfolio Assets Invested in East Asia and Europe in 2003 ( Percent in GDP) Source Country/Region Destination Country USAEuropeJapanEast AsiaWorld China1.00.70.21.83.8 Hong Kong24.024.24.67.765.0 Indonesia2.41.50.11.48.5 Japan6.86.20.00.314.9 Korea8.85.90.93.520.5 Malaysia7.77.81.613.031.4 Philippines6.36.01.64.218.4 Singapore27.416.53.010.161 Taiwan--- Thailand5.1 0.75.117.8 Average9.88.11.45.026.4 Europe Average14.159.05.76.990.0 United States.14.65.76.425.8

15. Table 7. Political Proximity for East Asia and Europe, Average over 1985-1990 and 2000-2005 1985-19902000-2005 Country USEuropeJapan East Asia USEuropeJapan East Asia China0.1500.4870.4890.7910.1060.5320.5910.787 Hong Kong........ Indonesia0.1250.4730.4550.8230.1200.5670.6010.821 Japan0.3170.757.0.4910.2920.836.0.653 Korea, Rep.....0.2780.8050.8530.633 Malaysia0.1430.4950.4960.8490.1210.5710.6100.827 Philippines0.1420.5130.4880.8350.1420.5870.6260.828 Singapore0.1540.5240.5290.8280.1370.5930.6320.818 Taiwan........ Thailand0.1320.4970.4900.8360.1320.6150.6560.832 Average0.1660.5350.4910.7790.1660.6380.6530.775 Europe Average0.3390.7620.7570.5350.3650.9220.8360.639 United States.0.3390.3170.166.0.3670.2920.166

16. Figure 1. Political Proximity for Selected Country-Pairs, 1980-2005

17. Figure 1. Continued

18. Summary: Is East Asia an OCA? East Asia is quite favorable for a currency union in terms of its substantial degree of intra-region trade. East Asia is quite favorable for a currency union in terms of its substantial degree of intra-region trade. East Asia has less favorable conditions for a currency union, particularly compared to Europe, in terms of business cycle synchronization and financial integration. East Asia has less favorable conditions for a currency union, particularly compared to Europe, in terms of business cycle synchronization and financial integration. Political proximity in East Asia (except Japan) seems fairly high (comparable to that in Europe in the 1980s). Political proximity in East Asia (except Japan) seems fairly high (comparable to that in Europe in the 1980s). It is not clear which currency, if they opt for a currency union, East Asian economies should choose as an anchor. It is not clear which currency, if they opt for a currency union, East Asian economies should choose as an anchor. The endogeineity of OCA criteria: country characteristics affecting OCA conditions can be endogenously determined. The endogeineity of OCA criteria: country characteristics affecting OCA conditions can be endogenously determined.

19. Quantitative Analysis of the Welfare Effects of East Asian Currency Unions We calculate welfare effects of a dollar bloc, Euro bloc, Yen bloc, and a new regional currency bloc for East Asia. We calculate welfare effects of a dollar bloc, Euro bloc, Yen bloc, and a new regional currency bloc for East Asia. We compare their potential benefits from joining a currency union with their potential costs (relative to those of floating exchange rate regime). We compare their potential benefits from joining a currency union with their potential costs (relative to those of floating exchange rate regime). A currency union, by creating additional trade between member countries, increases growth of output and consumption. The loss of monetary policy independence leads to higher degree of consumption fluctuations. A currency union, by creating additional trade between member countries, increases growth of output and consumption. The loss of monetary policy independence leads to higher degree of consumption fluctuations. Based on calibrations of a representative consumer model, we estimate the welfare gains from increasing consumption growth and the welfare costs of increasing consumption volatility from both normal disturbances and rare disasters. Based on calibrations of a representative consumer model, we estimate the welfare gains from increasing consumption growth and the welfare costs of increasing consumption volatility from both normal disturbances and rare disasters.

20. A Model of a Representative Consumer A representative consumer with a CRRA utility function, A representative consumer with a CRRA utility function, A stochastic consumption stream follows a random walk process: A stochastic consumption stream follows a random walk process: - t is ‘normal’ disturbance, iid log-normal. - ε t is ‘normal’ disturbance, iid log-normal. - v t is low-probability disasters, such as World War, and financial crises, with a small probability (p), but big consumption contraction (b) (Rietz, 1988 and Barro, 2006). - v t is low-probability disasters, such as World War, and financial crises, with a small probability (p), but big consumption contraction (b) (Rietz, 1988 and Barro, 2006). probability e -p : v t = 0, probability 1- e -p : v t = log(1-b). probability e -p : v t = 0, probability 1- e -p : v t = log(1-b). The expected utility is rewritten by The expected utility is rewritten by U= V∙(C) 1-θ /(1-θ), U t = V∙(C t ) 1-θ /(1-θ), 1/V = ρ + (θ-1)∙γ – (1/2)∙(θ-1)∙σ 2 – p∙((1-b) 1-θ – 1) 1/V = ρ + (θ-1)∙γ – (1/2)∙(θ-1) 2 ∙σ 2 – p∙((1-b) 1-θ – 1)

21. A Model of a Representative Consumer: Continued Welfare loss from an increase in volatility (σ*> σ ) is measured by compensation parameter λ (Lucas, 1987, 2003): Welfare loss from an increase in volatility (σ*> σ ) is measured by compensation parameter λ (Lucas, 1987, 2003): V (σ) ∙((1+ )C t ) 1-θ = V* (σ*) ∙(C t ) 1-θ At the margin, the compensation parameter is, At the margin, the compensation parameter is, The compensation for eliminating all normal disturbances: The compensation for eliminating all normal disturbances: (σ*=0) = [1+ V (σ) ∙(1/2)∙(θ-1) 2 ∙σ 2 ] 1/(1-θ) (σ*=0) = [1+ V (σ) ∙(1/2)∙(θ-1) 2 ∙σ 2 ] 1/(1-θ)

22. Effects of a Currency Union on Growth and Volatility An increase in growth rate by trade expansion (Rose, Frankel and Romer) An increase in growth rate by trade expansion (Rose, Frankel and Romer) - For benchmark calibration,  =0.5 and ψ=0.01 - For benchmark calibration,  =0.5 and ψ=0.01 A (potential) increase in volatility due to loss of monetary stabilization A (potential) increase in volatility due to loss of monetary stabilization -The monetary authority can stabilize a constant fraction, χ, of consumption fluctuations (e.g. nominal shocks). -The monetary authority can stabilize a constant fraction, χ, of consumption fluctuations (e.g. nominal shocks). -When country i adopts country j’s currency, the volatility of consumption of country i increases by -When country i adopts country j’s currency, the volatility of consumption of country i increases by - For benchmark calibration,  =0.3 (Lucas, 2003) - For benchmark calibration,  =0.3 (Lucas, 2003)

23. Table 8. Mean and Standard Deviation of Annual Per Capita Real Consumption, 1960-1997 Country Mean growth rate S. D. of growth rate shocks S.D. of growth rate shocks after a currency union US dollar EuroYen East Asia10 China 0.03670.05320.06140.06170.06340.0603 Hong Kong 0.05690.04620.05250.05230.05400.0527 Indonesia 0.03590.05930.06870.06980.07370.0680 Japan 0.04580.03220.03780.03650.03220.0372 Korea 0.04950.03670.04330.04240.04640.0410 Malaysia 0.03260.04410.05320.05030.05190.0497 Philippines 0.01390.03380.04050.03850.04100.0379 Singapore 0.04800.03810.04320.04240.04460.0428 Taiwan 0.06280.03190.03710.03620.04040.0365 Thailand0.04110.0360 0.04200.04120.04500.0406 Average0.04230.0410 0.04780.04700.05000.0465

24. Table 9. Declines of 10% or More in Real Per Capita GDP in East Asia, 1915-2000 CountryYears %fall CountryYears %fall Great DepressionWorld War II China1932-3410China1936-5018 Indonesia1929-3514Indonesia1941-4936 Malaysia1929-3217Japan1943-4552 Taiwan1928-3112Korea1937-4558 Financial CrisesMalaysia1942-4736 Indonesia1997-9912Philippines1939-4660 Philippines1982-8617Taiwan1942-4551 Thailand1997-9812 Mean of 14 major contractions 29

25. Table 10. Compensating parameter Proportionate change in initial consumption (1)(2)(3) Marginal change in:Welfare effectNumerical value normal s.d. (σ)V∙(θ-1)∙σ1.1 disaster probability (p)V∙[(1-b) 1-θ –1]/(θ-1)5.1 disaster size (b) V  p∙(1-b) -θ 0.6 growth rate (γ)-V-8.6 (1)(2)(3) Change in:Welfare effectNumerical value normal s.d. (σ=0)[1+ V∙(1/2)∙(θ-1) 2 ∙σ 2 ] 1/(1-θ) 2.1 disaster probability (p=0)[1+ V∙ p∙((1-b) 1-θ – 1)] 1/(1-θ) 8.5 normal s.d. and disaster probability (σ=b=0) [1+ V∙(1/2)∙(θ-1) 2 ∙σ 2 + V∙ p∙((1-b) 1-θ – 1)] 1/(1-θ) 10.3 A. Marginal Effect B. Total Effect

26. Table 11. Estimates of Welfare Effects, Baseline Scenario: (ρ=0.03,  =4, α=0.5, ψ=0.01,  =0.5, p=0.018, b=0.29) A. Welfare Effect from Trade Creation CountryUS dollarEuroYen East Asia C.U. 9 East Asia C.U. 10 China 0.200.180.190.500.69 Hong Kong 0.400.440.331.621.93 Indonesia 0.180.230.340.480.81 Japan 0.080.050.00 0.12 Korea 0.200.130.160.240.41 Malaysia 0.870.670.792.042.77 Philippines 1.360.761.041.832.80 Singapore 0.810.690.611.972.53 Taiwan 0.310.170.250.440.69 Thailand 0.290.280.370.550.91

27. Table 11. Continued B. Welfare Effect of Increasing Volatility CountryUS dollarEuroYen East Asia C.U. 9 East Asia C.U. 10 China -1.54-1.60-1.94-1.30 Hong Kong -0.60-0.58-0.75-0.64-0.61 Indonesia -2.11-2.37-3.44-1.84-1.92 Japan -0.46-0.340.00 -0.40 Korea -0.58-0.49-0.88-0.35-0.36 Malaysia -1.58-1.04-1.32-0.95-0.92 Philippines -2.29-1.52-2.48-1.41-1.34 Singapore -0.47-0.39-0.61-0.45-0.42 Taiwan -0.30-0.24-0.51-0.27-0.26 Thailand -0.61-0.53-0.96-0.46

28. Table 11. Continued C. Total Welfare Effect of Currency Union(A+B) CountryUS dollarEuroYen East Asia C.U. 9 East Asia C.U. 10 China -1.32-1.41-1.73-0.78-0.58 Hong Kong -0.19-0.13-0.411.011.36 Indonesia -1.91-2.12-3.05-1.33-1.05 Japan -0.38-0.290.00 -0.28 Korea -0.38-0.36-0.71-0.100.05 Malaysia -0.66-0.35-0.491.161.94 Philippines -0.80-0.71-1.330.521.61 Singapore 0.360.310.011.562.15 Taiwan 0.01-0.07-0.250.180.44 Thailand -0.32-0.24-0.580.090.46

29. Table 12. Sensitivity of Welfare Estimates (1) θ = 3 (2) χ = 0.5 (3)  =0.009 (4) p=0.008 (5) b=0.19 A. US Dollar Currency Union China -0.95-2.39-1.154.594.14 Hong Kong 0.01-0.590.133.633.36 Indonesia -1.38-3.41-1.754.383.90 Japan -0.29-0.69-0.313.913.58 Korea -0.24-0.76-0.223.733.42 Malaysia -0.16-1.730.066.125.67 Philippines -0.05-2.350.3313.3112.45 Singapore 0.660.051.005.014.70 Taiwan 0.15-0.190.263.343.10 Thailand -0.14-0.73-0.084.634.27

30. Table 12. Continued (1) θ = 3 (2) χ = 0.5 (3)  =0.009 (4) p=0.008 (5) b=0.19 B. Currency Union of 10 East Asian Economies China-0.18-1.46-0.015.595.16 Hong Kong 2.020.972.836.235.98 Indonesia -0.48-2.37-0.385.585.12 Japan -0.18-0.54-0.184.043.71 Korea 0.22-0.190.384.293.99 Malaysia 2.631.384.029.579.18 Philippines 1.840.783.7315.5114.74 Singapore 2.921.894.037.867.59 Taiwan 0.700.270.984.063.82 Thailand 0.770.161.185.795.44

31. Main Results from the Estimation The welfare gain from increasing growth rate due to trade creation is estimated to be substantial. The welfare gain from increasing growth rate due to trade creation is estimated to be substantial. Because of their substantial degree of intra-region trade, an East Asian economy would benefit most from forming an East Asia-wide currency union, compared to joining a US dollar bloc, Euro bloc, or Yen bloc independently. Because of their substantial degree of intra-region trade, an East Asian economy would benefit most from forming an East Asia-wide currency union, compared to joining a US dollar bloc, Euro bloc, or Yen bloc independently. The potential welfare cost of increasing volatility due to loss of independent monetary policy is also substantial as shocks have a permanent effect on consumption. The potential welfare cost of increasing volatility due to loss of independent monetary policy is also substantial as shocks have a permanent effect on consumption. China, Indonesia, Japan, and Korea may lose even with joining an East Asia-wide currency union. China, Indonesia, Japan, and Korea may lose even with joining an East Asia-wide currency union. If joining a East Asia currency union eliminates the probability or size of disasters completely, all East Asian economies would get a net positive welfare gain. If joining a East Asia currency union eliminates the probability or size of disasters completely, all East Asian economies would get a net positive welfare gain.

32. Concluding Remarks Judging from OCA criteria, East Asia has less favorable conditions for a currency union, particularly compared to Europe, in terms of business cycle synchronization and financial integration. Judging from OCA criteria, East Asia has less favorable conditions for a currency union, particularly compared to Europe, in terms of business cycle synchronization and financial integration. The calibration results shows that most countries in East Asia would obtain a net welfare gain from forming a currency union involving a broad group of East Asian economies. The calibration results shows that most countries in East Asia would obtain a net welfare gain from forming a currency union involving a broad group of East Asian economies. The possible negative effect of increasing volatility on growth may cause net welfare losses to larger East Asian economies, especially China, Indonesia, Korea and Japan, even when they join an East Asia-wide currency union. The possible negative effect of increasing volatility on growth may cause net welfare losses to larger East Asian economies, especially China, Indonesia, Korea and Japan, even when they join an East Asia-wide currency union. The important welfare consequences of a currency union involve its influence on the probability and size of disasters such as wars and financial crises in East Asia. The important welfare consequences of a currency union involve its influence on the probability and size of disasters such as wars and financial crises in East Asia. The prospect for an East Asian currency union will hinge on future developments of both economic and political conditions, rather than current environments. The prospect for an East Asian currency union will hinge on future developments of both economic and political conditions, rather than current environments.