The CO2 concentration in exhaust gas is less than in the output of a water-shift reactor, true, but only by a few percent. It's more investment, of course, but a hell of a lot less than starting from scratch with a IGCC. Costs you no energy because you use the waste heat of the power plant.

My understanding is that retrofitting CO2 extraction is a very costly exercise, to the point where it probably won't happen.

New plants, again my understanding is an IGCC with CO2 extraction is a lower cost proposition than a supercritical steam coal plant with CO2 extraction.

Do you have any good data on costs? Any cites?