If we are to solve the climate problem, our current generation of conventional, CO2 intensive coal plants must be our last.
With its high CO2 emissions and central place in the US (and global) energy infrastructure, 'fixing' coal is arguably at the center of a successful resolution to the climate challenge in the United States and the world. At present, coal is responsible not only for significant CO2 emissions, but also for environmental damages ranging from mountain top removal to mercury emissions. However, it offers a plentiful and low cost fuel that today provides more than 50 percent of U.S. electricity production, and we do not, at present, have either the financial capital or technological alternatives within reasonable costs to call for an immediate phase-out.
If we are to solve the climate problem, our current generation of conventional, CO2 intensive coal plants must be our last. A price on carbon as part of a larger strategy to move into non-emitting sources is clearly critical. However, the urgency of the climate change challenge means we cannot wait for the coal problem to be solved through such pricing mechanisms alone. For the next 20-30 years, we must pursue a set of complementary strategies that will transition the electricity grid away from coal over the long-term while reducing its destructive environmental impacts as much as possible in the short term. Carbon capture and storage  (CCS) is a critical part of such a strategy. In practice, this means that new coal plants should only go forward under a narrow set of conditions.
First, on a generator-by-generator basis, new coal plants should be the option of last resort. Even under optimistic assumptions, CCS is projected to capture well below 100% of CO2 emissions; many believe it will capture less than 80% even under optimistic scenarios . Thus, even with CCS, coal will always be a liability from the perspective of climate change. Cleaner renewable energy sources and demand reduction through energy efficiency should be the first alternatives whenever possible.
Second---and as soon as possible---all new coal plants should be designed and engineered to capture the majority of their CO2 emissions for either long-term storage or industrial use. This requirement would facilitate a rapid transition to broad-scale underground CO2 storage if and when the technology and infrastructure make it possible. Currently, "carbon capture ready" is poorly defined; often it means only that a utility has set aside acreage for capture facilities. Carbon capture needs to be built into the plant design, and implemented on day one.
Third, we should not build new coal plants in locations where the surrounding geology is not conducive to long-term underground CO2 sequestration, or where a lack of CO2 pipeline infrastructure would mean massive and costly delays for adequate CO2 storage. Nationally, the U.S. has huge sequestration potential---some have called it the "Saudi Arabia of sequestration." But just as wind is not universally feasible, neither is sequestration. New large-scale CO2 pipelines are not currently being developed, and are likely to be prohibitively expensive in many cases.
Finally and most important, we must immediately embark on a "crash program" to develop and deploy carbon storage capability on a massive, global scale. Underground storage is the only option on the table for dealing with CO2 emissions from fossil fuel power plants. We now know enough about CCS siting, regulatory and liability challenges  to quickly move towards industry-scale demonstrations. But CCS will require billions---not millions---in research funding. The G8 goal of 20 CCS demonstration projects requires funding on the order of $1-1.5 billion per project. Investments on that scale will not happen fast enough without public subsidies, which should be a priority for the next Administration.