If power plant efficiency is an option EPA proposes for state consideration, the question then may be asked how a fleet-wide improvement program could be achieved in United States. One possible approach might be to follow NETL’s suggestion of using the top decile of CFPP efficiency as a benchmark for U.S. fleet efficiency, used with an efficiency frontier. Using statistical methods, benchmarks could be used to improve efficiency of the CFPP fleet. NETL observes that while some improvements could be “relatively inexpensive” (for example,improved O&M, more frequent or pro-active maintenance), other improvements could be “very expensive” (for example, improvements bundled with a new SO2 scrubber, or turbine overhauls or heat exchanger replacement). But NETL notes that “if each plant achieved their maximum efficiency each year, 5% reduction in CFPP carbon dioxide emissions” could result.