Oxford macroeconomist Simon Wren-Lewis has a great post on how economics' extreme bias towards Pareto optimality is, of course, not value-free.
I'll add this: You often hear from economists that Pareto optimality is value-free and no one can object to it because no one is hurt by the Pareto option.
But here's the huge problem with that. What you're immediately doing when you say that is saying we will consider only two options, and rule out as unacceptable every other. We decide every other option is unacceptable except the two libertarian options, Pareto and nothing. THAT's value-free? That's positive and unideological? To rule out all of the many other options, including the utilitarian one, immediately as unacceptable, and only consider permissible for analysis the two libertarian ones? That's positive value-free? You can't even positively state, without endorsement, this is what the total societal utils optimum is, so people can know it and consider it?
This is a source of extreme, pervasive, and profound bias in economics toward libertarianism, as well as plutocracy. And I discussed this in a 2012 post, which Simon cited.