Fusion reactions can only be self-sustaining at incredibly high temperatures and pressures. If the containment mechanism loses power, the reaction stops. Unlike fission reactions, at least as used in current reactors, which have the potential to keep on going unless we do something active to stop them. Yeah, the modern, well-designed ones have lots of designed-in safeguards against this, void coefficients and all, but it's still safer still to be running fusion.
Hydrogen and it's isotopes in, with other isotopes or helium out is better than assorted radioactive heavy metal isotopes with relatively long half-lives. The only possibility for dangerous waste from a fusion reactor is any transmutation that takes place in reactor components from the neutron flux. Seems likely to be better, but I don't think anybody has really investigated what this part will look like in a running fusion reactor with reaction rates high enough for commercial power production.
http://en.wikipedia.org/wiki/Fusion_power#Waste_management
A lot safer too!
http://en.wikipedia.org/wiki/Cost_of_electricity_by_source
What I don't know is how much of the cost of nuclear power is the much higher level of safety precautions they need.
I haven't seen any projected costs for DEMO which might well be the first fusion power plant (DEMO is the proposed successor to ITER):