Since the beginning of this century, China’s annual GDP growth is over 9%. This growth is fueled by large increases in energy consumption, led by a coal-dominated energy structure, and associated with higher sulfur dioxide emissions and industry dust. In 2008, China accounted for over 17% of the world’s total primary energy consumption and accounts for nearly three-quarters of global energy growth. At an average annual energy growth rate over 12% since 2000, China’s future share of primary energy consumption will continue to increase. A consequence of this growth is China becoming the global leader in sulfur and carbon dioxide emissions. To deal with these energy and environmental challenges, the government set energy saving and pollution reduction target objectives in the 11th Five Year Plan (2006-2010): relative to 2005 by 2010, saving national energy use per unit of GDP by 20% and reducing the country’s primary pollution emissions by 10%. These targets were then disaggregated into energy saving targets for each province. With this disaggregated scheme, similar to country’s target, 20 provinces were assigned a 20% energy saving target, seven provinces were assigned targets below 20%, varying from 12% to 17%, and four provinces were given targets above 20%. These allocation were generally not guided by technical or economic efficiency, and thus may not be optimal from the perspectives of equity and efficiency. Historically less energy efficiency provinces may have more potential to reduce their energy consumption and pollution emissions, while higher efficiency provinces may have less potential. The major objective is to determine the optimal targets for each province required to comply with the national Five Year Plan target. A comparison of the estimated optimal with the current government targets will then reveal the value of incorporating economic theory into the decision calculation of setting disaggregate targets. Determining optimal targets requires consideration of both desirable and undesirable comes from alternative feasible targets. An objective is then to delineate these comes as criterion for selection. The procedure employed is a parametric hyperbolic distance function approach with a translog specification. This procedure provides the flexibility of using energy, labor, and capital stock as inputs to produce the desirable output (GDP) and the undesirable output (sulfur dioxide emissions). The procedure will address the objectives by simultaneously estimating both the desirable and undesirable comes. Specifically, the production frontier and environmental productivity efficiency are estimated for each province. The hyperbolic distance function enables the estimation of efficiency scores by incorporating all types of inputs and outputs, and only requires information on input and outputs quantities but not on prices, making it possible to model the emissions in the production process, given nonmarket characteristics of emissions. Based on these parametric estimations, the optimal targets are determined. The trajectory of obtaining these optimal targets for each province is determined by estimating how each province can improve its productive performance through increasing its desirable output and reducing its undesirable output, while simultaneously saving energy inputs. The results provide an empirical measurement of energy efficiency with maximum potential of energy saving for each province at a given technology considering the diverse economic, industry, and energy consumption patterns in the provinces. With a panel data of 29 provinces in China from 2000-2007, the hyperbolic distance function allows us to measure environmental productivity change over time, and then decompose this environmental productivity change into efficiency change, which is the movement toward the frontier, and technical change, which is the shift of the frontier. These further analyses help us identify potential different contributions of productivity growth for each province in China, and examine how the energy saving program will affect the environmental productivity growth for each province.