Exchange rate effects on agricultural exports

Abstract

Purpose The purpose of this paper is to extend empirical investigations of the relationship between real exchange rates and agricultural exports to the firm-product-country level with the use of disaggregated panel data of China’s food industry. In particular, the study aims to explore heterogeneities in the export response to real exchange rates across firms, destinations and products, as well as to differentiate responses on the intensive and extensive margins. Design/methodology/approach This paper utilizes a merged panel data set of firm-product-country level transaction records of China’s agricultural exports with firm-level survey data of the food industry. Panel regression models are constructed to identify empirical relationships. Findings Real appreciations are found to reduce export quantities and the probability to enter destination markets. These impacts are enhanced in 2005 when China unexpectedly depegged yuan from the USD. In addition, real appreciations in 2005 also reduced the yuan-denominated export price and increased firms’ probability to exit destination markets. Taking the exchange rate reform as a natural experiment, evidence suggests that the negative exchange rate effects on exports are robust to the endogeneity issue. Finally, heterogeneous export responses are identified with respect to firm productivities and ownerships, income levels and locations of destination markets, as well as product groups. Originality/value This paper provides first-hand evidence on how real exchange rates influence agricultural exports at the firm-product-country level. A featured contribution is that China’s exchange rate reform in 2005 is utilized to alleviate the typical concern of endogeneity. Findings may benefit policy makers, for example, by identifying firms most vulnerable to real appreciations.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 15 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.