the Global Development Finance. All these sources have presented many lacks in
the data in their series, forcing us quite often to choose between the number of
countries in the sample and the number of variables to include in the equations.
In fact, according to the IFS de...nition, the total number of “developing coun-
tries” would be 157: of those 64 had to be excluded due missing data, both in the
control and in the dependent variables. We assume that the resulting countries
selection is not endogenous. The variables we have decided to use are described
in Table 3 below. We decided to choose among the most signi.cant regressors we
found in the literature on both IMF arrangements and debt rescheduling, that is:
Table 3: Variables de.nition (a)
Variable |
De...nition____________________________ |
Units___________ |
Expctd. sign |
IMF |
Approval of an IMF arrangements |
Binary | |
DRES |
Total amount of Debt Rescheduled |
Binary | |
BB |
Baker and Brady countries |
Binary |
+ |
BEF |
Previous Fund arrangement |
Binary |
+ |
GDPPC |
GDP at market prices per capita |
% Pop. |
— |
GDI |
Gross domestic investment |
% GDP |
— |
EDT |
Total external debt stock |
% GDP |
+ |
TDS |
Total debt service |
% exports |
+ |
RES |
Total reserves minus gold |
% imports |
— |
INFL |
Consumer price index |
% rate of change |
+ |
BOP |
Balance of Payment |
% GDP |
— |
CA |
Current Account |
% GDP |
— |
EXP |
Exports of goods and services |
% imports |
— |
IAR |
Interest arrears on long-term debt |
% exports |
-/+ |
PAR |
Principal arrears on long-term debt |
% reserves |
-/+ |
GGC |
General government consumption |
% rate of change |
— |
(a) the Appendix contai ns all the calculation details.
As for the control variables they are basically traditional “ratio variable”,
quite common regressors among those we have found in other studies on the
determinants of both IMF arrangements and debt rescheduling. These variables
capture both domestic and external factors. They will be more carefully analysed
in the next two subsections.
21