140 likes | 272 Views
Trade Inequality in Developing Countries: A General Equilibrium Analysis. Susan Zhu (Michigan State University) Dan Trefler (University of Toronto and CIAR) June 5, 2014. The Model. Follow Dornbusch, Fischer and Samuelson (1980) as closely as possible.
E N D
Trade Inequality in Developing Countries: A General Equilibrium Analysis Susan Zhu (Michigan State University) Dan Trefler (University of Toronto and CIAR) June 5, 2014
The Model • Follow Dornbusch, Fischer and Samuelson (1980) as closely as possible. • 2 regions, North (N) and South (S). • 2 factors, unskilled labor (L) and skilled labor (H). • Factor prices: wNL, wNH, wSL, andwSH . • Measure of inequality: wS = wHS/wLSandwN = wHN/wLN . • A continuum of goods indexed byzwith 0 < z < 1. No factor intensity reversals so that a largerzis indexes a more skill-intensive good. • CRS, perfect competition, no international barriers to trade in goods, Cobb-Douglas preferences, balanced trade.
CS(wS) CN(wN) z– North exports South exports
The Southern Wage Inequality Equation Southern Demand for Skilled Labour =g(z– , wS ) Southern Demand for Unskilled Labour Southern Supply of Skilled Labour = HS / LS Southern Supply of Unskilled Labour • Dependent Variable • Southern Inequality: dln(wS) • Independent Variables • Trade Cut-Off: dln(z–) • Endowments: dln( HS / LS )
The Export Share Shift Equation In general equilibrium (z– , wS, wN) are endogenous. Totally differentiate the 3 g.e. equations to solve for dz– , dwS , and dwN . Dependent Variable: dz– Independent Variables: • Southern Catch-up • Endowments • dln( HS / LS ) • dln( LS / LN ) • dln( HN / LN )
Measuring wS = wHS / wLS • Freeman and Oostendorp (2001) NBER database on wages by occupation and industry for 1983-97. • Use changes over the four periods 1983-86, 1986-89, 1990-93, and 1993-97. • Criterion for inclusion in dataset: • Per capita GDP below $14,000 in 1980. • Country had observations on a fixed set of manufacturing occupations for at least two periods. • Consistent data for both non-production occupations (managers, professionals, technicians, and clerks) and production occupations (craft workers, operators, and laborers). • 20 countries and 58 observations.
Conclusions • Exports by themselves have no impact on inequality. I devised a novel measure of the changing skill intensity of exports. • Southern catch-up shifts the South's export shares towards more skill-intensive goods. • The resulting shift in export shares increases the level of wage inequality. • Taken together, these conclusions mean that Southern catch-up has contributed to rising wage inequality in the South. • The model's exclusion restrictions are accepted by the data i.e., Southern catch-up raises wage inequality only indirectly by shifting export shares.