Recent work on consequentialism has revealed it to be more flexible than previously thought. Consequentialists have shown how their theory can accommodate certain features with which it has long been considered incompatible, such as agent-centered constraints. This flexibility is usually thought to work in consequentialism's favor. I want to cast doubt on this assumption. I begin by putting forward the strongest statement of consequentialism's flexibility: the claim that, whatever set of intuitions the best non-consequentialist theory accommodates, we can construct a consequentialist theory that can do the same while still retaining whatever is compelling about consequentialism. I argue that if this is true then most likely the non-conseque...