Capital Controls and Interest Rate Parity

Evidences from China, 1999-2004

LIU Li-Gang&

Ichiro Otani

Recent Discussions on Capital Controls

• During and after the Asian Financial Crises• Example. Malaysia• Impossible Trinity

--Fixed exchange rate--Free capital mobility--Independent monetary policy

Previous Empirical Studies on China’ Capital Controls

• Mostly legalistic interpretation of rules and regulations w.r.t. effectiveness and market integration.

• Quantitative studies—just a few.(1) Cheung, Chinn and Fujii (2003)(2) Ma, Ho, and McCauley (2004)

Interest Rate Parity Theory

• Jeff Frankel (1992): Theory provides a useful framework for analyzing capital mobility quantitatively.

• Strictly speaking, the covered interest rate parity condition does not hold in the real world.

Deviations from Covered Interest Rate Parity

[ kti -

*kti - ( tktt sf −+, )] = [( k

ti - *kti ) - e

kts +∆ ] + ( ekts + - ft,t+k)

Components of Deviations from the Parity Condition

• Deviations represent the cost of transactions

• Total, TC,• Of which:

--TCr,Arbitrate risks and exchange risks--TCp, Pure transaction costs

Comparisons of Market Places

• Hong Kong: known to be one of the most open markets.--TC,hk = TCp, hk

• Onshore market in China.--TC,on = TCp,on +TCr,on.

• Cross-border market in China.--TC,cb = TCp,cb + TCr,cb.

• Approximation, TCp,hk = TCp,on = TCp,cb

Chart 1: Deviations from Parity,Q4 1999-Q4 2004

Chart 1: Deviations from Interest Rate Parity, Q4 1999 - Q4 2004

-0.5

0

0.5

1

1.5

2

1999

Q420

00Q1

2000

Q220

00Q3

2000

Q420

01Q1

2001

Q220

01Q3

2001

Q420

02Q1

2002

Q220

02Q3

2002

Q420

03Q1

2003

Q220

03Q3

2003

Q420

04Q1

2004

Q220

04Q3

2004

Q4in

per

cent

Hong Hong Onshore Cross-Border

Deviations from Parity

• Negligible deviations for Hong Kong.Free capital mobility.

• Negative deviations for China’s cross-border in early 1999. Controls to discourage outflows from China.

• Positive deviations for China since 2000: Controls to discourage inflows into China.

Chart 1a: Cost of Transactions,Absolute Values of Deviations

Chart 1a: Cost of Transactions, Q4 1999 - Q4 2004

0

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

1.8

2

1999

Q420

00Q1

2000

Q220

00Q3

2000

Q420

01Q1

2001

Q220

01Q3

2001

Q420

02Q1

2002

Q220

02Q3

2002

Q420

03Q1

2003

Q220

03Q3

2003

Q420

04Q1

2004

Q220

04Q3

2004

Q4

in te

rcen

t

Hong Kong Onshore Cross-Border

Cost of Transactions• Hong Kong:

Very small: 0.013% at quarterly rate, mainly fees and charges.

• China:Dominated by market pressures.Cross-border TC > Onshore TCAdditional cost for cross-border transactions, driving a wedge between RMB interest rate for onshore and for cross-border transactions.

Chart 1b: Cost Differentials btw Cross-Border and Onshore,

Q4 1999 – Q4 2004

-0.05

0

0.05

0.1

0.15

0.2

0.25

1999

Q4

2000

Q1

2000

Q2

2000

Q3

2000

Q4

2001

Q1

2001

Q2

2001

Q3

2001

Q4

2002

Q1

2002

Q2

2002

Q3

2002

Q4

2003

Q1

2003

Q2

2003

Q3

2003

Q4

2004

Q1

2004

Q2

2004

Q3

2004

Q4

in p

erce

nt

Series1

Table 3: Cost of Capital Controls and E&O, 2000-04

Table 3: Cost of Capital Controls and E&O, 1H 2000 - 2H 2004

TC of Capital E&O as % ofControls, in % Trade

1H 2000 0.25 1.802H 2000 0.20 3.201H 2001 0.25 3.602H 2001 0.11 1.871H 2002 0.18 2.002H 2002 0.19 0.731H 2003 0.11 1.282H 2003 0.12 3.961H 2004 0.12 1.432H 2004 1/ 0.07 4.43

Sources; SAFE, PRC and The authors' calculation

1/ Eatimates

Some Observations

• Transaction costs affected by capital controls on a declining trend.

• Apparent negative correlation between the intensity of capital controls and the degree of illegal capital movements

Chart 2: E&O vs Cost of Capital Controls, 1H 2000 – 2H 2004

0.00

0.50

1.00

1.50

2.00

2.50

3.00

3.50

4.00

4.50

5.00

0 0.05 0.1 0.15 0.2 0.25 0.3

Cost or Capital Controls

E&O

as

% tr

ade

Series1Linear (Series1)

Assessing Money Market Integration: Uncovered Interest

Rate Differentials

)()]([)( ,,** e

ktktttkttkt

kt

ekt

kt

kt sfsfiisii ++++ −+−−−=∆−−

Properties

• A) Are uncovered interest differentials mean reverting?

• B) Are the differentials predictable?

• C) Are absolute differentials shrinking?

A: Are uncovered Interest Differentials mean reverting?

• Unit root test:• Dicky-Fuller test: Phillips-Perron Test

t-statistics: t-statistics:DUIP (12): -0.6 DUIP(12): -1.85DUIP (24): -0.75 DUIP(24): -1.28DUIP (48): -0.22 DUIP(48): -1.1

B: Are the differentials predictable?

• Predictability regression results:• AR(1): -0.12 (-4.16)***• AR(3): -0.1 (-3.56)***• AR(5): 0.1 (3.56)***• AR(7): 0.006(0.07)• AR(9): -0.03(-1.3)• AR(11): -0.07 (-2.62)**

C: Are the differentials shrinking?

• Dependent variable: Absolute value of Differentials:

Reg. (1) Reg. (2)Time trend: 0.00005(14.3)*** Year99 -.003(-6.43)***Year00 -.005 (-16.4)***Year01 -.01 (-33.5)***Year02 -.009(-32.3)***Year03 -.005(-16.1)*Adjusted R SQ: 0.13 0.54

Effectiveness of Capital Controlsin Recent Years: Summary

• Qualitatively speaking, changes in capital controls had desired impact on illegal capital movements (as represented by Es&Os).

• Quantitatively, the impact was not significant and seems to be waning over time.

• Capital controls drove a wedge btw onshore and offshore interest rates.

• Deviations from uncovered interest rate parity are not random and predictable, but arbitrages could not make onshore and offshore interest rates to converge.

• Some signs of shrinking differentials.

Road Map for Further Integration of the Capital Market

• Increase RMB interest rate and exchange rate flexibility and deepening of markets for RMB assets and Forex.

• Reduce impediments of financial movements associated with:– current transactions,– long-term equity transactions, &– Short-term portfolio transactions.