Openness, imported commodities and the Phillips curve

This paper derives a Phillips curve with imported commodities as an additional input in the production process. Given greater reliance on exogenously priced imported commodities in production then changes in output lead to a reduced impact on marginal costs and prices. The Phillips curve becomes ‡atter relative to the benchmark New Keynesian case. Empirical evidence supports the hypothesis that greater imported commodity intensity in production increases the sacri…ce ratio. Econometrically controlling for imported commodity intensity also doubles the explanatory power of openness in determining the sacri…ce ratio, as conjectured by Romer (1993). JEL classi…cations: E31, E32, F41 Keywords: openness, imported commodties, sacri…ce ratio 1 Introduction This paper investigates the e¤ects of imported materials costs and trade openness on the slope of the Phillips curve. The Phillips curve represents the feasible set of combinations of output growth and in‡ation in the short-run; a ‡atter slope means less in‡ation for a given real expansion and equivalently, disin‡ation becomes more costly in terms of reduced output. Romer (1993) conjectures that the negative relation between in‡ation and openness is driven by steeper Phillips curves in more open economies, but the subsequent empirical literature has been far from supportive of this hypothesis. To resolve theory and data we put forward a new explanation of the link between the slope of the Phillips curve and openness. In particular, we consider the impact of imported commodities in the production process. In the standard New Keynesian model increased output increases marginal costs because of upward sloping labour supply. Given the …xed markup of monopolisitically-competitive …rms, increased prices and in‡ation eventually ensue. However, marginal production costs also depend on inputs other than labour. Imported commodities such as oil and gas, natural resources in general, and intermediate production goods such as iron, steel, chemicals and textiles can all play an important role in determining marginal production costs, and the prices of these commodities are plausibly exogenous for most countries unlike wages. In small open economies, changes in production will not change the prices of these commodities. It is also the case that real commodity costs have risen in recent years, and that producers have pointed to these as important drivers of their day-today pricing and output decisions. The crucial point is that when imported commodities are important in the production process then the link between increased output and marginal