Purpose – While price studies such as Jawadi et al. generally focus on the relationships between oil and stock markets through the study of oil price on stock markets, this paper takes a different perspective to the linkages between oil and stock markets. This study sets out to investigate the efficiency hypothesis for oil markets while testing for whether oil price dynamics depend on stock market fluctuations or not. Design/methodology/approach – Using nonlinear econometric modeling, this paper investigates the oil market adjustment dynamics for four developed and emerging countries: France, the USA, Mexico and the Philippines. Our findings show strong evidence of significant linkages between oil and stock markets for all the countries under consideration. Findings – As in Jawadi et al. who focus on stock price dynamics regarding oil price, the findings of this present paper, which focuses more on the oil industry, also point to an asymmetrical mean-reversion between oil and stock markets that occurs in a nonlinear manner. They reject the informational efficiency hypothesis for oil markets. Indeed, while the previous literature often highlights the stock markets' dependence on the oil industry, this study contributes to the literature by concluding in favor of significant feedback from stock to oil markets, which is not compatible with the efficiency principle according to Fama. Research limitations/implications – This paper develops a new nonlinear framework that should improve the investigation of oil-stock market linkages. Future research could check the forecasting properties of this model to forecast the future dynamics of oil prices. Originality/value – This paper adds to the literature by suggesting that it is not only oil shocks that affect stock markets, but that the latter also have a strong nonlinear impact on oil markets, reducing the diversification benefits of oil-stock portfolios.