RDP 9203: Real Exchange Rates and the Globalisation of Financial Markets Tables

Adrian Blundell-Wignall and Frank Browne

March 1992

Table 1: Unit Root ADF Statistics
Real Exchange Rates, Real Interest Differentials and Cumulative Current Accounts
(Sample Period: 1974Q1 to 1990Q4)
  Real Exchange Rates Real Long-Term Interest Differential Cumulated Current Balances
Yen/$ −1.6 −1.7 −0.5
DM/$ −2.2 −1.9 −0.05
£/DM −1.9 −1.6 −1.3
FF/DM −4.1* −1.2 1.1

Note: The augmented Dickey-Fuller (ADF) unit root statistic is employed to test the null hypothesis that the variable concerned possesses a unit root. Critical values are reported in Fuller (1976), Table 8.5.2. An asterisk denotes rejection of the null hypothesis at the 5 per cent level.
Table 2: Cointegration Tests For Bilateral Real Exchange Rates
(Sample Period: 1974Q1 to 1990Q4)
  Eigenvalues
λ		 Conditiobr nal Hypothesis
Max λ Tests		 Unconditional Hypothesis
Trace Tests 		Long-run Coefficients
Real Interest Differential R*−R Cum. Current Account Surplus
yen/$ 0.000 Ho:r=2 0.0 Ho:r≤2 0.0  
  0.224 Ho:r=1 17.2** Ho:r≤1 17.2** 0.125 −0.585
  0.319 Ho:r=0 26.1** Ho:r≤0 43.4** 0.027 −0.302
DM/$ 0.005 Ho:r=2 0.4 Ho:r≤2 0.4  
  0.073 Ho:r=1 5.1 Ho:r≤1 5.5  
  0.212 Ho:r=0 16.2 Ho:r≤0 21.6 −0.008 −0.102
FF/DM 0.002 Ho:r=2 0.1 Ho:r≤2 0.1  
  0.072 Ho:r=1 5.0 Ho:r≤1 5.1  
  0.393 Ho:r=0 33.5** Ho:r≤0 38.6** 0.007 −0.043
£/DM 0.001 Ho:r=2 0.1 Ho:r≤2 0.1  
  0.092 Ho:r=1 6.6 Ho:r≤1 6.7  
  0.279 Ho:r=0 22.3** Ho:r≤0 28.9* 0.064 −0.426

Note: The conditional maximum eigenvalue test is based on the largest squared eigenvalue. The unconditional trace test is based on the null hypothesis that there are k or less cointegrating vectors. Critical values are given in Johansen and Juselius (1990). One asterisk denotes rejection of the null hypothesis at the 10 per cent level and two asterisks denotes significance at the 5 per cent level.
Table 3: Alternative Suggested Measures of Financial Market Integration
Measure Context Test Equation Typically Used Definitions Perfect Capital Mobility Null Hypothesis
1
Closed interest parity		 a) Same currency
b) Different country (i.e. “onshore”-“offshore”) 		i = α1+β11i*+β12(prp) i = Domestic
interest rate
i* = Foreign interest rate
prp = Political risk premium 		α1 = 0
β11 = 1
βl2 = 0
2
Covered interest parity (CIP)		 a) Different currency
b) Same or different country
c) Investors cover
themselves in the forward market 		i = α2+β21i*+β22(fpd)+β23(prp) fpd = Forward premium or discount α2 = 0
β21 = β22 = 1
β23 = 0
(Required if different country)
3
Uncovered interest parity (UIP)		 a) Different currency
b) Same or different country
c) Investors take open positions in foreign currency 		i = α3+β31i*+β32E(ΔS)+β33(prp)+β34(erp) erp = Exchange risk premium
E(ΔS) = Expected change in the spot exchange rate 		α3 = 0
β31 = β32 = 1
β33 = 0
(Required if different country)
β34 = 0 (Risk neutrality)
4
Real interest parity		 a) Different currency
b) Same or different country
c) Different commodity bundles
d) Investors take open positions in foreign currency 		R = α4+β41R*+β42E(ΔRS)+β43(prp)+β44(rerp) R = Domestic real interest rate
R* = Foreign real interest rate
E(ΔRS) = Expected change in the real exchange rate
rerp = Real exchange rate risk premium 		α4 =0
β41 = 1
β42 = 0
Because
E(ΔRS) = 0)
β43 = 0
(Required if different country)
β44 = 0 (Risk neutrality)
5
Independence of domestic saving and investment ratios		 See Section 3 of paper for further discussion (I/Y) = α5+β51(NS/Y)
(I/Y) = a−bR+u		 I = Investment
NS = National saving
Y = GNP
u = Other factors affecting domestic investment other than R 		β51 = 0, all conditions required for 4 i.e. R=R* and, in addition, R*exogenous, ρ(u,NS/Y) = 0, and no non-traded goods

Note: This table summarises algebraically the various definitions of international financial market integration which have been proposed and sets out the conditions required to be fulfilled by each definition if capital markets are indeed fully integrated (refer to column entitled “Perfect Capital Mobility Null Hypothesis”). Uncovered interest parity (UIP) is rarely tested in the full form in which it is presented in the table. The table serves to highlight the number of hypotheses that are required to be maintained for the Feldstein-Horioka national savings-investment measure to be a valid measure of the degree of international financial market integration.
Table 4: Real Interest Rate Linkages With the United States: Three-Month Interest Rates and Consumer Prices
(Absolute t Values in Parentheses)
(The Data are Monthly)
  First Period Second Period Third Period
α γ α γ α γ
Japan 0.595
(10.17)		 0.316
(4.48)		 0.659
(8.14)		 0.291
(2.93)		 0.357
(2.04)		 0.634
(6.36)
Germany 0.219
(4.12)		 0.199
(1.15)		 0.900
(25.77)		 0.022
(0.60)		 0.548
(7.24)		 0.595
(5.64)
France −0.16
(1.34)		 0.566
(3.50)		 0.54
(4.46)		 0.424
(3.31)		 1.27
(17.72)		 0.007
(0.09)
Italy −0.40
(2.78)		 0.944
(7.20)		 1.77
(28.95)		 −0.500
(5.26)		 1.65
(25.18)		 −0.261
(2.48)
United Kingdom −0.54
(1.55)		 1.920
(3.01)		 0.425
(4.11)		 0.730
(8.00)		 0.824
(10.68)		 0.757
(7.41)
Canada 0.480
(5.54)		 1.374
(14.75)		 0.860
(8.99)		 0.397
(3.40)		 0.874
(6.49)		 0.524
(2.63)
Netherlands −0.046
(0.27)		 0.937
(2.69)		 1.004
(9.52)		 −0.064
(0.58)		 1.155
(6.14)		 0.189
(0.79)
Switzerland 0.406
(2.91)		 1.024
(3.51)		 0.408
(7.15)		 0.059
(0.89)		 0.175
(1.48)		 0.739
(5.3)

Note: See Annex for an explanation of the model used. The hypothesis that foreign and domestic ex ante real rates move together and thus that the domestic and foreign markets are completely integratedimplies γ=1. γ =0 implies complete disintegration. The first, second and third periods are August 1974 to October 1979, November 1979 to February 1990 and January 1986 to February 1990, respectively.
Table 5: Real Interest Rate Linkages With Germany: Three-Month Interest Rates and Consumer Prices
(Absolute t Values in Parentheses)
  First Period Second Period Third Period
α γ α γ α γ
France 0.357
(5.13)		 0.476
(2.52)		 0.645
(2.25)		 0.331
(1.02)		 0.827
(3.69)		 0.363
(1.68)
Italy 0.758
(3.66)		 −0.398
(1.81)		 1.361
(8.84)		 0.058
(0.35)		 1.305
(7.96)		 0.083
(0.50)
United Kingdom −1.745
(7.26)		 3.27
(4.27)		 1.351
(3.76)		 −0.383
(0.91)		 0.460
(3.26)		 0.865
(6.00)
Netherlands −0.732
(7.77)		 1.301
(12.32)		 0.356
(1.70)		 0.699
(2.97)		 0.162
(1.77)		 1.014
(10.91)
Switzerland 0.031
(0.79)		 0.966
(6.73)		 −0.399
(1.65)		 0.923
(3.50)		 −0.487
(8.98)		 1.101
(17.28)

Note: See Annex for an explanation of the model used. The hypothesis that foreign and domestic ex ante real rates move together and thus that the domestic and foreign markets are completely integrated implies γ = 1. γ = 0 implies complete disintegration. γ =1 and α =O imply equality of domestic and foreign rates.