Working Paper, University of Massachusetts, Department of Economics 2014-11
Using dynamic panel models with data for 62 developing countries, this paper examines whether growth in agriculture elicits growth in manufacturing. For identification, I use population-weighted, average temperature as an instrument for growth in agriculture. I identify large short-run effects: An increase in growth in agriculture by one percentage point is estimated to raise contemporaneous growth in manufacturing by between 0.47 and 0.56 percentage points. The baseline models also imply sizable long-run effects of permanent increases in growth in agriculture. Extensions of the empirical model suggest that growth in agriculture benefits the manufacturing sector by improving its domestic terms of trade, by increasing the share of investment and saving in GDP, and by increasing the capacity to import industrial inputs. The paper makes two main contributions. First, it joins a growing literature using climate data to identify supply shocks in agriculture, establishing a robust empirical relation between these shocks and growth in manufacturing. Second, it includes a stylized two-sector model to illuminate the macroeconomic channels behind this complementarity. Together, these contributions lend support to the notion that agriculture plays key macroeconomic roles in the industrialization of developing countries by relieving saving, aggregate demand, _scale, and foreign exchange constraints on the industrial sector.