We analyze how uncertain future US carbon regulations shape the current choice of the type of power plant to build. Our focus is on two coal-fired technologies; pulverized coal (PC) and integrated coal gasification combined cycle technology (IGCC). The PC technology is cheapest-assuming there is no need to control carbon emissions. The IGCC technology may be cheaper if carbon must be captured. Since power plants last many years and future regulations are uncertain, a US electric utility faces a standard decision under uncertainty. A company will confront the range of possible outcomes, assigning its best estimate of the probability of each scenario, averaging the results and determining the power plant technology with the lowest possible cost inclusive of expected future carbon related costs, whether those costs be in the form of emissions charges paid or capital expenditures for retrofitting to capture carbon. If the company assigns high probability to no regulation or to less stringent regulation of carbon, then it makes sense for it to build the PC plant. But if it assigns sufficient probability to scenarios with more stringent regulation, then the IGCC technology is warranted. We provide some useful benchmarks for possible future regulation and show how these relate back to the relative costs of the two technologies and the optimal technology choice. Few of the policy proposals widely referenced in the public discussion warrant the choice of the IGCC technology. Instead, the PC technology remains the least costly. However, recent carbon prices in the European Emissions Trading System are higher than these benchmarks. If it is any guide to possible future penalties for emissions in the US, then current investment in the IGCC technology is warranted. Of course, other factors need to be factored into the decision as well. (c) 2006 Elsevier Ltd. All rights reserved.