Exchange Rates and Monetary Policy with Heterogeneous Agents: Sizing up the Real Income Channel

Introducing heterogeneous households with realistic, empirically-documented marginal propensities to consume into an otherwise-canonical small open economy New Keynesian model (the representative-agent, complete-markets “RA-CM” model of Galí and Monacelli 2005) changes how depreciations and monetary policy affect output. Beyond the standard expenditure-switching channel, in which a cheaper currency shifts domestic and foreign spending toward home goods, the heterogeneous-agent (HA) model adds a “real income channel,” through which a depreciation’s rise in import prices lowers households’ real income and induces them to cut consumption, and a Keynesian multiplier that feeds any output change back into income. The balance between these forces is governed by the trade elasticity χ (the sum of the import and export price elasticities): at χ = 1, the real income channel and multiplier exactly cancel and household heterogeneity is irrelevant to the exchange rate shock’s effects; below χ = 1, the real income channel can dominate, and for a sufficiently low trade elasticity, output falls on impact – a “contractionary depreciation” – something the paper shows is quantitatively powerful only when high marginal propensities to consume are combined with incomplete markets (the HA-IM case), not in representative-agent or two-agent models with the same average MPC. An analogous neutrality result holds for domestic monetary policy shocks at χ = 2−α (nesting the Cole-Obstfeld unitary-elasticity case); away from it, monetary easing can “steal demand from the future” by financing a current spending boom with a current account deficit that must later be repaid. A calibrated quantitative extension – adding delayed substitution (a Calvo-style adjustment friction that generates a rising, J-curve-shaped trade elasticity), sticky import/export prices, non-homothetic consumption baskets, and unequal incidence of aggregate income shocks across households, calibrated broadly to Mexico – finds that depreciations are contractionary for about a year and expansionary thereafter, and that the resulting policy dilemma (whether a central bank facing capital outflows should hike rates to defend the currency or cut rates to support demand) can go either way depending on the trade elasticity.

Martin Souchier | Macro Paper Warehouse

Exchange Rates and Monetary Policy with Heterogeneous Agents: Sizing up the Real Income Channel