DESCALZI RICARDO LUIS
Congresos y reuniones científicas
Título:
Probability of Current Account Reversals in Argentina and other Latin
Lugar:
Montevideo
Reunión:
Jornada; XXIII Jornadas Anuales de Economía. Banco Central del Uruguay.; 2008
Institución organizadora:
Banco Central de Uruguay
Resumen:
The cumulative effect of selected shocks on the probability of a Current Account Reversal is
measured as a quantification of the external vulnerability of Latin American economies. In the
framework of the intertemporal approach the explanation of reversals in the LACs is focused on
external solvency as the restrictive determinant of the threshold of overborrowing –in contrast with
perspectives stressing fiscal solvency-. Following the empirical strategy suggested by Milesi-
Ferreti and Razin and others, a random effects probit panel model is run to find the probability of
occurrence of CARs in LA countries. Relevant variables are GDP growth, changes in terms of
trade and contagion. A statistical model is used to build a case of a representative LAC:
simulations with estimated coefficients suggests that the presence of low growth and contagion
may be absorbed by the economy when these factors operate individually, but their simultaneous
occurrence drive the probability of reversals rapidly up above 40%. A policy implication is that
solvency may vanish rapidly, out of control of domestic authorities, and that the limit of
overborrowing is uncertain. Orders of magnitudes for safe scenarios may be quantitatively
established providing a guide for limits in external indebtedness.
measured as a quantification of the external vulnerability of Latin American economies. In the
framework of the intertemporal approach the explanation of reversals in the LACs is focused on
external solvency as the restrictive determinant of the threshold of overborrowing –in contrast with
perspectives stressing fiscal solvency-. Following the empirical strategy suggested by Milesi-
Ferreti and Razin and others, a random effects probit panel model is run to find the probability of
occurrence of CARs in LA countries. Relevant variables are GDP growth, changes in terms of
trade and contagion. A statistical model is used to build a case of a representative LAC:
simulations with estimated coefficients suggests that the presence of low growth and contagion
may be absorbed by the economy when these factors operate individually, but their simultaneous
occurrence drive the probability of reversals rapidly up above 40%. A policy implication is that
solvency may vanish rapidly, out of control of domestic authorities, and that the limit of
overborrowing is uncertain. Orders of magnitudes for safe scenarios may be quantitatively
established providing a guide for limits in external indebtedness.
