Yesterday the Environmental Protection Agency released its tabulation of greenhouse gases (GHGs) from large facilities in the US. In perusing the data I couldn't help thinking of the quote attributed to Willie Sutton concerning why he robbed banks. Even if he never actually said, "Because that's where the money is," the simple logic of that analysis transfers neatly to the question of why we might be interested in assessing and ultimately managing GHG emissions from such installations. While there are other important sources, notably including motor vehicles and aircraft, the more than 6,000 sites reported in the agency's online registry account for roughly half of all US GHG emissions. Furthermore, just a quarter of these sites--power plants--contribute nearly three-fourths of US emissions from large facilities. That's where the emissions are and where US climate policy should focus.

Although that doesn't dictate that we should entirely ignore all the other facilities, it certainly raises serious questions about the threshold of reporting for the hundreds of installations emitting less than 10,000 tons of CO2-equivalent gases per year, compared to the top-100 facilities, the smallest of which emitted nearly twice that much every day.

It should also challenge the belief systems of some members of Congress concerning the relative importance of different sectors. The highest-emitting oil refinery in the country is also one of the biggest in the world by throughput capacity, at 573,000 barrels per day. Yet it comes in at #45 on the list, with only one other refinery appearing in the top 100. The entire refining sector, comprising 145 plants, emitted around 5.7% of the total GHGs represented in the registry, and thus less than 3% of the US total. Why does that matter as more than an industry talking-point? Because reducing emissions from refineries by 10%--no easy task when they are already roughly 90% efficient in terms of their total energy output vs. inputs--would be lost in the rounding in our national emissions statistics. We won't get very far chasing expensive diminishing returns.

By comparison, reducing emissions from the 1,555 power plants on the list by an average of 10% would reduce US emissions by more than 3%. And because we are blessed with many more processes for generating electricity than for refining oil, this could be achieved in a variety of ways, nor does 10% represent any kind of ceiling for what might be possible. One option would be to retire the least-efficient coal-fired plants and take up the slack at existing gas-turbine power plants, plus some additional renewables. That may happen anyway, as a consequence of other EPA regulations. We could also replace the worst coal plants with near-zero-emission nuclear power plants of advanced design, such as the AP-1000 reactor that won NRC approval late last year, or the various modular nuclear reactors now under development. Capturing and sequestering the CO2 from coal-fired power plants would be another option, if it can be perfected at a reasonable cost.

I would never suggest that climate policy could be truly simple, but the numbers the EPA just reported, combined with what we know about the lifecycle emissions from the petroleum value chain, indicate that the scope of the US climate policy debate could usefully be narrowed to focus on just two main emissions sources: power plants and the end-use combustion of hydrocarbon fuels. On the scale of overall US emissions, almost everything else is noise. Of course that leaves plenty of room for discussion and disagreement on the most effective ways to address these emissions at the lowest cost and least disruption to an already-fragile economy. We can still argue endlessly about the relative merits of putting a price on emissions, providing incentives for emission-reducing technologies, and setting command-and-control regulations. Yet when we contrast the potential effectiveness of such a limited approach with the intricacy and distortions entailed in "comprehensive" efforts like the failed Waxman-Markey climate bill of 2009, it looks like a very helpful simplification to pursue.