This study explores the impact of crude oil price fluctuations on Pakistan's domestic markets for refined petroleum products, focusing on petrol, diesel, and kerosene. The retail prices of these products are determined by government-administered periodic revisions rather than continuous market-driven adjustments. The study uses weekly data from January 2013 to November 2018, totaling 307 observations for each series. A combined Vector Autoregressive (VAR) and Dynamic Conditional Correlation GARCH (DCC-GARCH) framework models short-term price dynamics and time-varying volatility dependence. Augmented Dickey–Fuller tests confirm that all series are integrated of order one, while Johansen cointegration tests indicate no stable long-term equilibrium relationship among the four price series, supporting the estimation in first differences. Granger causality analysis reveals strong bidirectional predictability among petrol, diesel, and kerosene prices, aligning with coordinated administrative price revisions, and a consistent unidirectional influence from crude oil prices to petrol prices across all lag specifications. Crude oil shows no predictive power for diesel at any lag, whereas its influence on kerosene emerges only at extended lags. Impulse responses diminish to statistical insignificance within four weeks of the announcement. The DCC-GARCH estimates show significant volatility persistence across series, with GARCH parameter sums exceeding 0.88, and demonstrate that conditional correlations between crude oil and refined products surged in mid-2018 before returning to long-term levels. The low unconditional correlations between crude oil and products in the first differences, compared to high inter-product correlations, reflect the constraints imposed by the administered pricing regime. These findings have implications for fiscal planning, price revision policies, and energy market risk management in Pakistan.
Keywords: Crude Oil Price Shocks; Energy Sector; Pakistan; VAR; Dynamic conditional correlation GARCH; Volatility Spillovers; Granger Causality