measure the intensity c.q. effectiveness of capital controls. The closed and adjusted covered nominal interest
differentials -- we will refer to as country risk premia -- primarily reflect the joint influence of existing
and expected capital controls (i.e. political risks) (Aliber 1973, p. 1453). They indicate agents’ ability to
move financial assets across national borders.
Capital controls play two major roles. First, where capital controls are effective, they produce a bias against
either capital inflows or capital outflows.8 In general, a negative domestic onshore-offshore interest differential
or a negative adjusted domestic covered nominal interest rate differential is indicative of capital export
restrictions in the domestic country (c.q. capital import restrictions in the foreign country). Second, they
tend to slow down the process of convergence of closed c.q. adjusted covered nominal interest differentials.
Therefore, we formulate a partial adjustment model for the closed c.q. adjusted covered nominal interest
rate differential. The partial adjustment model for the closed nominal interest rate differential may be derived
as follows (suppressing the time and country subscripts) (see Harvey, 1990, p. 230-231):
(3)
i - i Euro= (i - iEuro ) _ ɪ + ψ ( ( ι - i Euro ) - ( i - iEuro ) _ ɪ ) + e
Equation (3) shows that the actual interest rate differential in period t equals the actual interest rate differential
in the previous period t-1 and the proportion (ψ) in which the difference between the desired interest rate
differential in period t (bar) and the actual interest rate differential in the previous period t-1 (no bar) is
erased. Finally, ε denotes an error term. Rewriting equation (3) leads to equation (4):
( _ ( Euro ( ( ɪ _ ψ ) ( i _ ( E4r0 ) ɪ + ψ ( j _ j Euro ) + ɛ (4)
Assuming the desired domestic interest rate differential is determined by a constant (δ1) and other determinants
(x)9 ___________
( - ( (5) (5 (5 ɪ + Ô ( ( (5)
and combining (4) and (5) leads to
i - ( E6ro (6 ψ Ô 1 + ( 1 - ψ ) ( ( - ( E6ro ) _ 1 + ψ δ 2 x + e (6)
Similarly, the partial adjustment model for the adjusted domestic covered nominal interest rate differential
may be formulated as follows:
( i - ( ( (Euro + f - ∣S ) ) = ψ δ 1 + ( 1 - ψ ) ( ( - ( ( (Euro + f - S ) ) .1 + ψ δ ( X + € (7)
Annual series for closed nominal interest rate parity may be derived as follows (see Appendix B):
for Germany, France, Netherlands and the United Kingdom.10 Unfortunately, Euro-interest rates are not
available for Austria, Belgium, Denmark, Finland, Italy, Spain and Sweden. Therefore, for these countries
f Year _
Ψ Domestic ~
7V12 [½u÷3 -½7÷
J = I 12
(8)
8 This reasoning applies to both direct or quantitative measures (consisting of outright restrictions or prohibitions of certain capital
transactions) and indirect or cost measures (effecting the operations of the banking and nonbanking sector). Numerous qualitative
studies exist that describe the introduction and workings of capital control measures.
9 In section 3, we identify relevant determinants.
10 Data for Germany refer to the Federal Republic of Germany before the unification of Germany.