What are the main results achieved in BATModel?
First, we developed a literature review and a conceptual framework to examine the distributional effects of trade and trade policies, with a focus on the so-called hidden costs of globalisation, such as labour market issues, income and wage effects, and environmental and health effects. Each of these dimensions of trade liberalisation has been examined through specific case studies, using both
econometric models and computable general (and partial) equilibrium models. The evidence suggests that the interaction between trade liberalisation and the environment is quite small but heterogeneous in the EU agricultural sector. Trade shocks (imports and exports) have a significant impact on labour markets, especially in the EU food industry. In addition, trade shocks seem to have a negative impact on health status, in particular by increasing obesity.
Second, we have provided solutions to the problem that trade models cannot make counterfactual predictions of positive flows when those flows are initially null. Various modelling approaches have been developed to represent trade in a more non-linear way, thus allowing for possible discontinuities, such as trade flows that become zero and vice versa. An additional advantage of these approaches is that they allow the modelling of agri-food products at a detailed level, which allows for much more realistic and useful trade policy simulations.
Third, the impact on firms' export is examined, including firm characteristics on the supply side, such as productivity, intermediate inputs and product quality, as well as market characteristics on the demand side, like consumer tastes. Theoretical developments and empirical case studies were carried out to analyse the relationship between supply and demand side characteristics and their impact on firms' exports. The results are used to provide a numerical prototype model to guide the consideration of product quality differentiation in existing applied simulation models.
Fourth, we have improved data and models to analyse the impact of GVC on the agri-food sector. New evidence shows the link between GVC participation and trade policyNew data have been collected at both micro (firm) and macro level, providing micro-based indicators of GVC positioning and integration for agri-food exports, which are then used to improve models for analysing GVC dynamics. The governance of GVCs has also been taken into account to analyse how it may affect firms' participation and performance in the chain. Finally, significant progress was made in integrating GVC analysis into CGE models through the creation of new GVC-related modules.
Fifth, using the latest data sources at a highly disaggregated level, we have examined how non-tariff measures (NTMs) affect trade, including trade in value-added, and welfare. The results suggest a high degree of heterogeneity in both the effects of NTMs and the types of measures themselves, including at-the-border and behind-the-border measures, and regulations such as those on new plant engineering techniques. They also confirm that harmonisation and regulatory cooperation within preferential trade agreements help to reduce the trade-distorting effects of NTMs. More recently, ad valorem equivalents of NTMs have been calculated and further used in simulations.
Sixth, we empirically and theoretically analysed the welfare and trade effects of Geographical Indications (GIs) at different levels of aggregation and for different stakeholders. Based on new datasets, empirical analyses show the positive impact of GI protection both at the regional level for the home countries and at the firm level, as GI exporting firms sell more products and have better access to destination markets. The results of the simulations suggest that GI protection in the CETA agreement can increase export volumes by facilitating the entry of GI-producing firms into the Canadian market.
Seventh, we made the methodological advances of BATModel available to the wider modelling community as well-tested, concisely coded and clearly documented modules. We also performed the analysis of a free-trade agreement (FTA) in all the improved models in a policy-relevant test case. In most cases, the newly developed modules did not trigger visible changes in terms of macroeconomic impacts of the CETA on the involved countries. The development of the new indicators on jobs, wages and wage-income distribution helps to broaden the model assessment of upcoming FTAs, and the addition of product details (products & GVCs) increases the expected response of the FTA.
All the public deliverables are available on the BATModel website (
https://www.batmodel.eu/deliverables/(s’ouvre dans une nouvelle fenêtre)). The materials presented during the summer school to the end users of our results are also available (
https://www.batmodel.eu/dissemination-materials/(s’ouvre dans une nouvelle fenêtre)). The BATModules with accompanying documentation are available at GitHub (
https://github.com/BATModules(s’ouvre dans une nouvelle fenêtre)).
