.png){kind=icon}
.png){kind=icon}
Return
Preferential trading in agriculture: New insights from a structural gravity analysis and machine learning
.png){kind=icon}
PRE
AI
D
S
DOI：10.1002/ajae.70077.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Preferential trade agreements (PTAs) contain various non-tariff provisions, yet identifying their trade effects remains challenging because these commitments are high-dimensional and strongly correlated within agreements. We estimated a theory-consistent structural gravity model with domestic flows for 26 agricultural subsectors over 1988–2017 and employed a plug-in Elastic Net-penalized PPML approach to isolate the provisions that systematically shape agricultural trade. The procedure identifies 13 provisions that are particularly important for agricultural trade, spanning trade facilitation, SPS/TBT disciplines, and rules on competition and subsidies. After accounting for the impact of preferential tariff rates, a generic PTA indicator is still estimated to increase trade by 20%. This effect is absorbed once the identified provision bundle is included, demonstrating that trade patterns under PTAs are primarily shaped by specific non-tariff commitments. The results point to clear channels: facilitation and transparency provisions are trade-promoting, whereas stronger reliance on international standards is associated with lower trade, consistent with compliance-cost mechanisms.
Keywords:
agricultural trade
Elastic Net-penalized PPML
non-tariff provisions
preferential trade agreements
structural gravity model
Journal
IF:
3.3
Papers:
6.6K
Citations:
8.9K
Organization
Cited Papers