Portal de Periódicos da CAPES

Do transport costs have a differential effect on trade at the sectoral level?

2008; Taylor & Francis; Volume: 40; Issue: 24 Linguagem: Inglês

10.1080/00036840600994179

1466-4283

Inmaculada Martínez‐Zarzoso, Eva María Pérez-García, Celestino Suárez Burguet,

Global Trade and Competitiveness

Abstract This article aims to analyse the determinants of transport costs and to investigate their influence in international trade with a sample of disaggregate trade data. First, we estimate a transport-cost function using cross-section data on maritime and overland transport for four sectors: agro-industry, ceramic tiles, motor vehicle parts and accessories, and electrical and mechanical household appliances, obtained from interviews held with Spanish exporters and logistics operators in 2001. Second, we study the relationship between transport costs and trade and estimate the elasticity of trade with respect to transport costs for each sector. Important differences for high value- and low value-added sectors are observed. The trade-equation estimation shows that higher transport costs significantly deter trade, especially in high value-added sectors. Acknowledgements Financial support from Fundación Caja Castellón-Bancaja, Generalitat Valenciana and the Spanish Ministry of Education is gratefully acknowledged (P1-1B2005-33, Grupos 03-151, INTECO; Research Projects GV04B-030, SEJ 2005-01163 and ACOMP06/047). We would also like to thank two anonymous referees and the participants in the 6th INFER Workshop on Economic Policy in Reus (Spain) and in the I International Transport Congress hold in Castellón (Spain) for their comments and suggestions. We thank Elena Sanjuan-Lucas for excellent research assistance. Notes 1Martínez-Zarzoso et al . (Citation2003) analysed the impact of transport cost on trade in the ceramic sector. 2Cif stands for 'cost, insurance and freight'; fob stands for 'free on board.' 3Hummels (Citation1999a). 4See Appendix 3 for additional details of how the data were collected and from whom. 5Given the complexity of studying all Spanish export trade flows, the aim of selecting four sectors was to achieve a significant overview of transport cost and trade determinants by researching a representative sample of the Spanish production framework. 6The variable INF ij has been initially constructed as an index (by taking information on roads, paved roads, railroads and number of telephone lines) differentiating between importer and transit countries' infrastructure as explanatory variables of transport costs. This index is comparable to that of Limao and Venables (Citation2001), but opposite signed. 7Results are available upon request. 8Bayoumi and Eichengreen (Citation1997, p. 142). 9We used a Wald test to test for the equality of slopes in the sectoral-distance elasticities. The test is included at the end of Table 2 and the result indicates rejection of the null hypothesis (equality of slopes). 10Sectoral dummies are excluded since they were nonsignificant in Model 5.