EPA's existing source performance standards (ESPS) proposal from Monday claims it will achieve 30% emissions reductions from the power sector by 2030. That reduction is relative to 2005 emissions - a target that's important because of the president's stated goal in Copenhagen of 17% economy-wide emissions reductions from 2005 by 2020. If that goal is achieved, early reductions from the ESPS will play a major role (though given the rule's timeline, the 2020 target looks ambitious).
Focusing on this headline emissions reduction target, many observers have claimed that picking a 2005 base year instead of 2012 means the rule is less stringent than feared (or hoped, depending on where you sit). This is because US emissions have actually declined substantially since 2005. That means a big slice of the needed reductions are already in the bank. To pick one example, Bloomberg's Mathew Philips claims "EPA did the power industry a big favor by using 2005 levels" for this reason.
Former EPA General Counsel Roger Martella, when asked about this in an interview this week, pointed out that even if this is true for the country as a whole, it's not true for every state. Some states increased their emissions between 2005 and 2012. He points to Texas, but EIA data available through 2011 indicates that Texas emissions have gone down slightly (about 5%). However, 8 states' emissions did increase: Oklahoma, Wyoming, Kansas, North Dakota, Iown, South Dakota, Arkansas, and Nebraska - the latter three by 10-20%. Choosing a 2005 base year therefore seems to disadvantage those states.
However, all of this analysis appears to be wrong since EPA did not actually use 2005 as the base year. All of the calculations in EPAs proposal used to create state-level emissions rate improvement targest are based on 2012, not 2005 (or in the case of heat rates at existing coal plants, a 2002-2012 average). 2005 is only used to get the headline emissions reduction figure. Moreover, these goals are stated in terms of emissions rate - that is, the tons of emissions per megawatt hour generated. That means that running fossil power plants less due to reduced demand, a major component of the decline in US emissions, has no effect on the target.
You can debate whether the reason for nevertheless picking 2005 as the base year for the headline emissions cuts numbers is to inflate the emissions reductions from the proposal or, as EPA claims, to match up with international emissions reduction commitments. In any case, claims that using 2005 as the base year for calculating these overall results helped the industry, or helped or hurt any state, don't hold water. Everything is based on the most recent numbers available - 2012.
If state targets are still based on 2012 numbers in 2020 when they become enforceable, it's of course possible that changes in states' emissions in the intervening 8 years will mean some have more work to do than others. But that's much less of a problem - first because EPA could update the targets with new data, and second because states are now on notice that these requirements will be in place.