Corporate finance comprises investment, financing and dividend policies aimed at maximizing shareholder value. In particular, the results of commodity producers and, consequently, the value to their shareholders are subject to high volatility, resulting from the variation of prices of these products in the global market. However, the risk of this variation can be mitigated by exploiting the broad derivatives market that is generally available for commodities. This work proposes to calculate the value increase that a commodity-producing company can provide to its shareholders through the use of an optimal derivatives management policy, by buying or selling forward contracts. To this end, it seeks to maximize shareholder returns via dividends in a risk-averse environment. The model assumes that the commodity price follows a discrete state Markov process. Since the model is applied in several stages, the problem becomes quite complex, and it is necessary to use a decomposition method to obtain the solution, so we used the method known as stochastic dynamic dual programming. The results show that by trading forward contracts, a company increases the value perceived by the shareholder, measured by the payment of dividends, to any level of risk aversion. The average value increase, considering different levels of risk aversion and an unbiased pricing assumption, is higher than 320 per cent when compared to companies that do not have access to such instruments. In addition to measuring the value increase, we also analyzed which factors determine the optimal derivatives management policy. It was possible to identify that the derivatives management policy is very determined by the prices, which in turn are associated with the state of the Markov chain in force at each stage.