Determinants of inflation expectations in Colombia: a VAR-X analysis

Purpose: This study aims to investigate the determinants of inflation expectations in Colombia through a vector autoregression model with exogenous variables (VAR-X) and uses quarterly data for survey-based inflation expectations and different supply shocks. Design/methodology/approach: We propose a VAR-X model. Despite data unavailability, we gathered quarterly data for the period 2005–2022 for the following variables: oil price, real exchange rate, headline inflation, output gap, policy interest rate and inflation expectations. Findings: We identified significant responses to inflation expectations in the first quarter. Although we found a positive response of inflation expectations to the interest rate, the robustness tests show that the interest rate negatively affects inflation expectations in the long run. Additionally, we detected a pass-through effect regarding the positive response of inflation expectations to a real exchange rate shock and the inertia of inflation expectations to their own innovations. Research limitations/implications: We must emphasize that reliable data from households would be preferred to follow the trend in international research and thus make feasible comparisons. Practical implications: Inflation expectations play an important role in an inflation targeting scheme. Specifically, this scheme allows monitoring of how those approach the proposed target and how they change in the face of changes in total inflation, demand and supply shocks. Originality/value: The inclusion of exogenous variables contributed to the stability of the model specification by capturing supply shocks not previously considered in the literature.