Price transmission in Indonesia’s CPO chain: an ARDL-UECM assessment

Abstract

This study analyzes price transmission along Indonesia’s cooking-oil value chain using Autoregressive Distributed Lag (ARDL) models expressed in Unrestricted Error-Correction (UECM) form, which separates short-run dynamics from long-run equilibrium relationships. Monthly data from January 2016 to May 2023 are employed, covering stable conditions as well as the 2022 cooking-oil crisis. The dataset comprises international crude palm oil (CPO) benchmark prices (deflated and JISDOR FX rate-adjusted), domestic wholesale bulk prices, and retail bulk prices, alongside policy dummies for the export-ban/DMO window and the post-2022 MGKR regime. Results reveal a sharp asymmetry between upstream and downstream stages. In the upstream link, CPO price shocks are transmitted quickly but only partially: the long-run pass-through is β = 0.420, with an immediate impact of 0.131 and a very high adjustment speed (λ = –0.639), implying a half-life of only 0.68 months. By contrast, the downstream link is characterized by amplification and sluggish adjustment. Retail prices exhibit an exaggerated long-run pass-through (β = 2.34), weak contemporaneous transmission (0.104), and slow convergence (λ = –0.143), with a half-life of 4.49 months and overshooting in lagged responses. These findings underscore the resilience of wholesale markets and the vulnerability of retail segments.