This paper attempts to investigate the possibility of structural change in tanker freight volatilities pre-and during the financial crisis. The aim is to apply a Markov-switching general autoregressive conditional heteroskedasticity (MS-GARCH) model that identifies and estimates the parameters of high and low volatility states, which are associated with different stages in the business cycle. Time varying volatility models, proposed by Engle (1982) and Bollerslev (1986) show that volatility is driven by shocks. Estimates of the persistence of shocks in time varying volatility models have been very high, particularly for financial data. This led to the introduction of integrated general autoregressive conditional heteroscedasticity (IGARCH) models by Engle and Bollerslev (1986), with unit persistence implying that market shocks do not die out over time. However, it has been suggested that the cause of high persistence of shocks within market volatilities may be due to structural shifts in the unconditional variance of the time series. Diebold (1986) argues that volatility persistence can be decomposed into two components, namely shocks persistence and persistence due to regime switching in the parameters of the variance model. Based on these findings we investigated the possibility of tanker freight volatilities being state dependent. Empirical findings show that tanker freight volatilities are clustered which may indicate that volatilities switch between distinct states. Assuming conditional volatilities of tanker freight rates switch simultaneously between a high volatility state and a low volatility state, and by measuring the magnitude and duration of these volatilities shocks, this paper attempts to explore the usefulness of such an implication to shipping freight risk and trading strategies, during booms and busts. The validity and comparative performance of the models is investigated with a set of diagnostics that discriminate between models on the basis of conservatism, accuracy and efficiency. Thus, this study contributes to the literature by: 1) investigating the possibility of state dependence of tanker freight volatilities. 2) measuring the duration and magnitude of high and low tanker freight volatilities shocks. 3) proposing a dynamic approach to measure Dynamic Risk and Volatility in Tanker Shipping Markets: A Markov-switching application Insert the paper’s ID 14 IAME 2013 Conference, July 3-5– Marseille, France 2 long-term risk exposure through a Markov-Switching Conditional Variance-Value-at-Risk (VaR) model.