اثرات شوک های داخلی و خارجی بر بهره وری کل عوامل

This paper examines structural changes that occur in the total factor productivity (TFP) within countries. It is possible that some episodes of high economic growth or economic decline are associated with permanent productivity shocks; therefore, this research has two objectives. The first one is to estimate the structural changes present in TFP for a sample of 77 countries between 1950 (1960) and 2000. The second one is to identify possible explanations for breaks. Two sources were analyzed: (i) episodes in political and economic history; (ii) changes in international trade – a measure of absorption of technology. The results suggest that about one-third of the TFP time-series present at least one structural break. Downwards breaks are more common, indicating that after a break the TFP has much difficulty to recover. When we investigated factors related with structural change, developed countries presented a break near the first oil shock while the developing countries’ breaks are more spread along the decades. Thus, external strikes seem to be more relevant for developed countries. However, for each country and break date, it was possible to find an event close to the break date endogenously detected. Last, the relevance of international trade, measured by trade share percentage of GDP, seems to be limited to explain abrupt changes in TFP.

One of the main characteristics of modern economies is the large differences in per capita income among countries. Explaining these differences and their evolution over time is an extremely important issue. Economists have recognized that total factor productivity (TFP) acts as a determinant factor in the growth process. Hall and Jones, 1999 and Parente and Prescott, 1999, Prescott (1998), Klenow and Rodriguez-Clare (1997), among others, show that there is strong evidence that TFP is considerably responsible for the differences in per capita income across countries.

The purpose of this work is to present estimates for structural breaks in TFP within countries, and to identify, whenever possible, episodes in the history of these countries that may explain the structural breaks in question. The results suggest that about one-third of the TFP time-series present at least one structural break. Downwards breaks are more common than upwards breaks. Also, the breaks are spread among developed and developing countries. When we analyzed factors related with structural change, developed countries presented breaks near the first oil shock while the developing countries’ breaks are more spread over the time. Thus, external strikes seem to be more relevant for developed countries. On the other hand, we investigated internal factors potentially related with structural changes. We considered political and economic events, besides any type of conflict. For each country and break date, it was possible to find one such event close to the estimated break date.