The Asymmetric Effects of Oil-Price Shocks on Inflation in Cambodia: A Nonlinear Autoregressive Distributed Lag Model

In this study, a nonlinear autoregressive distributed lag (NARDL) was used to analyze asymmetric effects of oil-price shocks on consumer prices. The Augmented Dickey–Fuller test results indicate that consumer price index (CPI) shocks, positive and negative oil-price shocks, and the broad money supply ar I(1), whereas the foreign exchange rate is I(0), thereby favoring the use of the NARDL framework. The bounds test asserts that there is a statistically significant long-run relation between the variables at a 5% level. Diagnostic tests provide additional evidence that the estimated model is free of serial correlation and heteroskedasticity with normally distributed residuals that are stable over the sample period. The error-correction coefficient estimation is −0.122 and significant, which means that approximately 12.2% of the short-run disequilibrium is corrected per period. Over the long run, the CPI is affected to a significant degree by positive oil-price shocks, with a 1% cumulative rise in oil prices resulting in an approximate 0.053% increase in the CPI. On the contrary, a negative oil-price shock has no statistically significant impact. The existence of cointegration between oil-price increase shocks and inflation is a necessary but only partial condition for the presence of asymmetry under such conditions, and a Wald test provides evidence that long-run asymmetries, in the form of stronger pass-through from positive rather than negative oil price movements, exist with high significance as well. The CPI is positively influenced by the broad money supply, wherein the exchange rate also has a significant, long-run relationship with inflation. In general, the results indicate asymmetry in oil-price pass-through and significant effects of monetary and exchange-rate conditions on inflation dynamics.