Latin America in the XXth Century: Stagnation, then Collapse.∗

Most Latin American countries experienced their last peak in output per capita relative to the United States ’ between 1971 and 1982. Prior to this peak per capita output was rapidly catching up to the developed world. Twenty years after the peak the average country’s relative per capita output was 68 % of its peak level. A growth accounting exercise shows that between 1960 and 1985 the contribution of physical capital to growth, at 74%, was more than twice the world’s average. There is an investment/productivity puzzle since capital accumulation was among the highest in the world and productivity growth one of the lowest. Import Substitution Industrialization and targeted investment subsidies may be the key to understanding Latin America’s lack of development. The theory of economic growth predicts that poor countries should grow faster than richer ones so that their living standards converge. In the last half of the twentieth century, many market economies in different regions of the world (such as Western and Southern Europe and East Asia) have indeed enjoyed remarkable growth experiences significantly reducing their development gap. Latin American countries on the other hand have stagnated relative to the productive frontier represented by the US until the 1970’s. Thereafter, their economies collapsed. This paper is about Latin America’s growth performance in the last fifty years. Its goal is modest as it only seeks to describe Latin America’s development patterns in this period. Explaining the facts described here is a challenge faced by Latin American social scientists. This study of Latin America’s development experience is based on a com-parative analysis of relative per capita income across the world over the period following 1950 and a subsequent growth accounting exercise that relates the evolution of relative per capita income to that of the accumulation of human and physical capital, as well as to the rate of growth of total factor productivity. The last sections of the paper contain some conjectures about a plausible theory to explain the facts. We thank Lee O’Hanian and Harold Cole for useful conversations