On the distributive effects of terms of trade shocks: the role of non-tradable goods

In a recent paper, jointly with Sebastian Galiani and Daniel Heymann, we analyze the effects of the 2008 commodities’ price spikes on income distribution. We mainly focus on resource-rich economies (especially those related to foodstuff) but our analysis is general enough to encompass a broader set of economies—as shown. We put special emphasis on the role played by the non-tradable sector. Since in one way or another the reaction of governments has been tariffs, export taxes, or quantitative restriction (just to cite a few), we incorporate this into the analysis. Importantly, the model is able to explain the implications of the distributive tensions observed. Among the most salient ones is the redistributive conflicts between urban labor (skilled vs. unskilled), as opposed to the traditional study of tensions between landlords and urban labor. A summary follows.

An improvement in international terms of trade clearly benefits the representative agent of a small open economy. Higher returns for the exportable sector and an increased aggregate spending capacity in terms of traded goods are likely to encourage economic growth. However, this does not imply that every group in a society is necessarily made better off.

The potential for distributive impacts of changes in international prices has been traditionally recognized in the literature, at least as far back as the Stolper-Samuelson theorem (1941). In turn, those distributive impacts may trigger social conflict, which could be deleterious for economic growth (see, among others, Rodrik, 1999). A recent instance of the tensions that can be generated by large shifts in world prices occurred in early 2008, following a sharp rise in the prices of commodities–and food in particular. This lead to social and political unrest in a large number of developing countries (with riots in some 30 cases), and to policy responses in the form of subsidies, price fixing, and export restrictions. Such reactions were observed both in food importers and in economies that export commodity foodstuffs (see The Economist, 2008).

To explain the emergence of those types of events we introduce non-tradable goods to an otherwise standard neoclassical trade model, in the Heckscher-Ohlin-Samuelson (HOS) tradition. This allows us to study the distributive effects of terms of trade shocks for a wide arrange of configurations.

We show that in a two-sector economy (comprised of exportable and non-tradable goods) there are no redistributive effects of external terms of trade shifts since relative prices remain the same. This neutrality implies the non-existence of Stolper-Samuelson effects in this set up.

By extending the model to the domestic production of manufactures, distributional tensions arise. Conditional on factor intensities, some income earners benefit at expense of the others in response to the terms of trade shock. Interestingly, important distributional conflicts arise within urban labor (skilled vs. unskilled) as opposed to the “traditional” rural vs. urban factors disputes.

Then we analyze the implications of export taxes as a mechanism to re-distribute the effects of external shocks. We find that the ability of the government to cushion the impact of the terms of terms shift on the economy’s income distribution depends not only on how it re-distributes the revenues, but also in how much it “saves.” Conditional on factor intensities and market shares, interesting alliances among different income groups might arise. Again, the center of the distributional tensions and gains and losses resulting from the income redistribution policy concentrate within urban labor. As an example, we show conditions in our benchmark model (that we may dub as a standard resource-rich economy) under which skilled (urban) labor favors landowners against unskilled (urban) labor regarding export tax policy.

The analytical framework can be used to describe the motivations and incentives of different groups in political economy games. In this connection, our arguments contribute to a broader literature that emphasizes the role of trade on domestic political cleavages and domestic policies and institutions (see, for example, Rogowski (1989)). Here, we specifically analyze the consequences of taxes on foreign trade on factor incomes—in the setups described above—to discuss the effects of international prices’ shifts and income redistribution. The introduction of the non-traded goods in the analysis implies that foreign trade taxes need not imply protection of import-competing activities (since these may be economically insignificant, as in the two-sector case) and that their implications for factor prices may depend strongly on the use of the tax revenues, apart from the effect of policies on the relative prices of traded goods, which is the change on which the typical HOS analysis concentrates.