Consider this a response after a fashion to Henry Farrell's latest.
1: I completely agree that the orthodox libertarian conception of freedom is nonsense (as I said in the first place).
2: But that still does leave us with the question of economic efficiency or, rather, I would say social welfare.
3: Simply pointing out that an "economics 101" model of frictionless labor markets and full employment is an abstraction does not eliminate its analytic utility in pursuing these questions. If anything, the existence of frictions makes me more skeptical of labor market regulations. In a frictionless marketplace regulations would reduce wages but with frictions it's likely to increase long-term unemployment which is worse in social welfare terms.
4: Since the libertarian conception of freedom is nonsense and there's more to the good society that economic efficiency, there's nonetheless sometimes a decent case for labor market regulation. But you have to make the case! For example, homophobia is bad. And it seems clear that one potent tool for fighting prejudice against gays and lesbians is for people to recognize that friends, family members, and colleagues who people know and like have been gay all along. Yet a collective action problem exists in a homophobic society whereby coming out is socially useful, but in some ways individually costly (in other ways individually liberating, of course). So legal protections for people to be openly gay in the labor market (or rental housing market or other markets) can be a great way of fighting prejudice perhaps at the cost of incurring some kind of excess legal overhead. Similarly, introducing rules against sexual harassment at the workplace is a vital tool in promoting women's social and economic quality relative to a brutally unequal system that had been entrenching itself for centuries. The law (in the United States) against racial discrimination in public accommodations was, by the same token, a specific response to slavery and Jim Crow. You can't fire a woman for being pregnant both as a matter of gender equity and because we conceive of American society as a multi-generational enterprise that aspires to endure.
5: The general schema of a sound proposal here seems to me to be that you're asking businessmen to put up with some annoyance in order to advance a policy goal that isn't narrowly economic in nature. Pollution regulations, by the same token, really do hurt not only the owners of pollution-intensive industries but also the workers in said industries. Nonetheless, limiting pollution is important. Dismantling Jim Crow was important. Fighting for women's social equality is important.
6: But this generally isn't a good way to fight the specific problem of economic inequality or limited economic opportunities. Poor people need more money. Unemployed people need adequate aggregate demand. People with limited skills need better schools and training programs. Workers at all skill levels have more freedom and more choices when there's full employment and it's relatively easy to switch jobs.
7: Some regulations that are worth having are plausibly bad for poor people. If bosses were given greater freedom to coerce their employees' political behavior, that would almost certainly increase incomes at the bottom end (in effect you'd be letting poor people sell votes) but corruption is bad and economic elites have too much rather than too little sway over the political system.
8: As a forward-looking example, employment discrimination against ex-convicts seems like a very rational business strategy in a wide variety of situations. But the United States has a significant violent crime problem, and while ex-convicts might prefer cash transfers to improved labor market prospects cash transfers don't actually address the issue of the social marginalization of offenders or help break the cycle of marginalization and crime.
9: Real-world labor market regulations often do precisely the wrong thing. Where prejudice against gays and lesbians is widespread, what you often see are pushes to legally exclude openly gay people from employment opportunities (banning them from teaching, for example). Today one frequent form of labor market regulation is rules preventing ex-offenders from obtaining occupational licenses. In Ohio recently they had a rare moment of bipartisanship to partially ease such restrictions, but they're still appallingly widespread in many states.
10: In general, I think raising the question of "workplace freedom" is an interesting narrow rhetorical move against highly ideological libertarians. But that only gets you so far in advancing an argument for a specific proposal. Married life is "less free" than single life and the fact that marriages are voluntarily entered into doesn't change that. And yet that hardly makes the case for ending no-fault divorce. Minor differences in phrasing sometimes tend to do too much work in discussions of freedom. Should the boss be allowed to fire you for refusing to take a drug test? Freedom says no! Should an unemployed former recovering drug addict be prohibited from agreeing to regular drug testing as a way to get a job? Freedom says yes! I don't think this is a helpful way of trying to address the merits of banning employment-linked drug tests.