How OECD policies affected Latin America in the 1980s

The authors assess the effects of OECD monetary and fiscal policies on Latin America by means of simulation studies using the LBS/NIESR Global Econometric Model and a new empirical model of Latin America. The Latin American model pays special attention to the supply-side determination of natural rate of output and to the effects of asset accumulation. The Latin American model and its properties are presented by both empirical simulations and by means of a simple analytical representation. This model of Latin America is used in conjunction with the Global Econometric Model to study the macroeconomic interactions between Latin America and the rest of the world. The assumption in policy simulations is that G-3 exchange rates are forward-looking while Latin America pegs its currency to the U.S. dollar. It is postulated that Latin American fiscal policy adjustments target a baseline current account balance, in the face of external shocks. The simulation results reflect a number of important international links, which can be quantified as multiplier properties of the linked system of models. A permanent 5 percent contraction in the U.S. money supply induces a contraction of about the same order in Latin American GDP and capital stock. This is caused by higher U.S. interest rates and diminished Latin American competitiveness in third markets, reinforcing the fall in U.S. demand. Similarly, a combined monetary contraction in G-3 countries on a permanent footing - a contraction like the one in 1978-80 (U.S., 5.2 percent, German, 11.9 percent, and Japanese, 1.7 percent) hurts Latin America. Latin American GDP remains depressed by 4 percent and capital stock by 5 percent even after 10 years. The effects of negative income and interest rates emanating from G-3 countries are mutually reinforcing. U.S. fiscal expansion equal to 1 percent of baseline GDP, sustained over five years, transmits negatively to Latin America, where GDP falls 0.6 percent in the short run and remains depressed by 0.3 percent even after 10 years. The negative effects of higher interest rates and diminished competitiveness dominate the positive effects (which are short-lived) of expanded U.S. demand for Latin American exports. Similarly, G-3 fiscal spending shocks, which aregradually built up over five years, then reversed the next two years, have a mild negative effect on Latin American GDP. The G-3 fiscal shocks administered were set to their actual magnitudes relative to baseline GDP, as observed in 1980-85 (U.S. expansion of 3.5 percent but contraction in German and Japanese spending of 4.4 percent and 3.5 percent, respectively). Latin American GDP is lower than baseline GDP by 0.5 percent when the shocks peak at the end of five years, but continues to remain depressed 0.3 percent by the end of 10 years. The simulated effects of G-3 monetary and fiscal policies, with the shocks constructed to reflect their actual sizes in the early 1980s, suggest that Latin America's adjustment problems in that period cannot be attributed to G-3 fiscal imbalances that arose because of failures of G-3 fiscal policy coordination. But concerted G-3 monetary contraction in response to the second oil shock imposed heavy costs on Latin America; without it, Latin American GDP would have been 5 percent higher in the 1980s.