Estimating Trade Elasticities: Demand Composition and the Trade Collapse of 2008–2009

Abstract

This paper introduces a new methodology for the estimation of income trade elasticities based on an import intensity-adjusted measure of aggregate demand. It provides an empirical illustration of this new approach for a panel of 18 OECD countries, paying particular attention to the 2008-09 Great Trade Collapse, which standard empirical trade models fail to account for. In this paper, we argue that the composition of demand plays a key role in the collapse of trade during crises because of a relatively bigger fall in the most import-intensive categories of expenditure (especially investment, but also private consumption), which has a large downward impact on the quantity of imports from the rest of the world. In addition, the fragmentation of production implies high import content of exports and, in turn, strongly synchronized trade ‡uctuations across countries. We provide evidence in favor of these factors based on the analysis of the new OECD input-output tables and building on a stylized theoretical model. Importantly, we show that our new intensityweighted measure of demand outperforms alternative measures, during crises but also in normal times, providing import elasticities of demand that are much less volatile across the cycle. JEL Codes: F10, F15, F17.

Exploring Price and Non-Price Determinants of Trade Flows in the Largest Euro-Area Countries

“…Government expenditure has a low import content, given that it generally includes goods or services which are not traded internationally. Differently from Bussière et al (2013)'s results which refer to a panel of OECD countries, the fit of the import model does not improve for any of the four countries under study (Table 7). We find that the role of exports in explaining imports in Italy increases substantially relative to the baseline model.…”

“…The empirical literature has often focused solely on the export equation (for instance, Ca ' Zorzi and Schnatz, 2007;European Commission, 2010;Bayoumi et al, 2011) or on the import equation (for example, Barrell and Dées, 2005;Stirböck, 2006;Bussière et al, 2013) with only a few studies estimating both models (Hooper, Johnson and Marquez, 1995;Allard et al, 2005;MartinezMongay and Maza Lasierr, 2009). We estimated both export and import equations over a longer time horizon than many existing studies, in particular including the period subsequent to the outbreak of the recent financial crisis, and referring to a four-country sample (Italy, Germany, France and Spain), as opposed to the euro area as a whole 12 .…”

“…Amongst the other economies, non-price competitiveness proves a crucial determinant of export growth in Germany and, even more so, in Spain; accordingly, also in these countries enhancing structural reforms would be a key policy action to nurture their export performance. -2011Q2 -4.5 -4.9 -3.8 -5.2 -3.8 -6.6 -4.5 -3.4 -15.0 -4.7 2011Q2 -2012Q2 -3 Source: Authors' calculations on data provided in Bussière et al (2013). Notes: The import contents have been normalised to sum to unity.…”

“…Demand component data are taken from Istat and Eurostat; the import contents are computed on the basis of the OECD Input-Output Database following Bussière et al (2013) 29 . Since I-O tables are available only every five years, we linearly interpolated the weights to obtain quarterly series.…”

“…Moreover, the significant coefficient of pricecompetitiveness indicators for Italy proves stable over the long-run in a rolling regression framework. Finally, we also correct the demand components included in the import equations by their import intensity, as in Bussière et al (2013), in order to better take into account the increasing internationalization of firms' production processes.…”