When the words “labor union” are spoken, many libertarians will rightfully cringe. The student of economics understands the economic complications that unions create and the student of politics recognizes the influence on government that unions exert. In this regard, the free market libertarian approach appears to be clearly anti-union. At least until we begin to explore some of the nuances of the debate.

In the debate on unions,  the first distinction to make is between “forced association” and “free association.” Because unions make up some of the largest political contributors, the overwhelming influence of unions – both economically and politically – arises from the forced association group of unions. Teacher’s unions (depending on the state), the AFL-CIO (the American Postal Worker’s Union), etc. all operate by forcing employees to become members and forcing employers to only negotiate with them provided they get a majority of employees on their side at some point with the “vote to unionize” or have the government recognize them explicitly.

These forced association unions as we know them are like any other cartel: beneficial to members at the expense of non-members. The union members get above-market wages, so long as the union can prevent any workers from accepting lower wages, thus limiting the supply of labor. That’s their whole purpose, which is why they can’t function without the ability to forcibly exclude competition. They can’t be beneficial to “workers,” if we include non-unionized or anti-union laborers in the definition of “workers.” Non-union workers are actually put out of a job by union-imposed policies, and workers in other industries often end up with lower wages due to higher competition from those now unemployed workers that were previously in the now-unionized industry.

They also use other techniques employed by cartels. For example, they deliberately suppress productivity. In the Post Office, for example, overly productive workers during the busy holiday season are disciplined because a high rate of productivity translates to less overtime pay. This will benefit the individual unionized employee, at the expense of the consumer overall. Thus, it would also be correct to say that labor unions raise the price of goods for consumers in various ways, both indirectly (as we see in the above Post Office example) and directly (mandating pay scales, for example). They produce less, increase the employer’s costs, and get cozy with the State to ensure high prices are protected by such policies as tariffs and trade quotas. It is important to note that up to this point I am describing common unions in the modern sense, not unions such as Industrial Workers of the World (IWW) who are more “free market” and not in bed with the state.

In this context of forced association unionization, the libertarian position is clearly to oppose unions. However, many libertarians aren’t opposed to unions, per se, in the context of the free market. This is where the concept of “free association” applies. If the State is not involved, how can a libertarian oppose the voluntary association of laborers for the purpose of bargaining collectively?

However, looking past the obvious motte and bailey here (where a supporter of all unions would hedge and say they only want the market-based union if challenged), a union is currently privileged by the State to force association between employers and themselves and employees. In union states without “right-to-work laws”, employers can’t refuse to associate nor can employees choose another union or forgo one entirely. Without that power (for instance, in a free market) a union is – although not immoral – arguably pointless. It can accomplish nothing of significance.

Let’s assume that we are living in a stateless society and that unions exist but (obviously) without any ability to exert state-sponsored coercion. It is entirely plausible that workers may still decide that their wages are unfair or the working conditions are too poor (they can decide this even if they receive high wages and great working conditions relative to real-world employees, since we know that humans always act in an attempt to improve their situation). As a result, they organize a labor union and utilize tactics advocated by groups such as IWW, a union that is not politically connected to the State.

So let’s assume this purely free union pulls strategy from pamphlets like “How to Fire Your Boss” and sponsors picketing, information campaigns engaging in slowdowns, work-to-rule, and other on-the-job direct action that does not utilize the threat of government force. Let’s assume that they use pressure – instead of legislation – to gain membership and discourage scab workers, none of which violates the libertarian principle of non-aggression.

In this hypothetical, I see no reason to oppose the right to unionize, even if I don’t want to join the union myself. Furthermore, I see no significant degree of harm from such unionization because, without the State, the union is still subject to market regulation. If they win too much from their employers, it will give competing businesses a market advantage. As a result, overly-greedy unions will drive their employers out of business, effectively dissolving the more harmful unions. Because “scabs” can be pressured not to replace picketers, but not forced nor removed from the job when the strike ends, unions that are vying for power in an otherwise well-accepted work environment will have more trouble limiting replacement labor from taking their jobs. It is clear to see how unions can exist in a truly free market in a way that would balance the power between the union and the business.

Obviously we don’t live in a stateless society and in this context, I am willing to side with the libertarian left, at least to a degree. This is because present-day state institutions promote wage labor by, among other things, erecting barriers to entry into other ways of making a living. The State does make it so bigger companies can be more influential when it comes to steering and influencing bargaining power, especially when workers are easily replaceable. And it is hard for competitive entrepreneurs to enter the market to either compete directly or bid for labor. Unlike state sponsored unions, unions like IWW aren’t using legislative power to erect barriers of entry, and we aren’t using bailouts to prop-up business who cede more than they can afford to the non-privileged unions.

Unfortunately, even though I do sympathize with left-libertarians on this issue, I still must part ways although the issue isn’t a simple one. One of the reasons that businesses cave is that they don’t want to risk an official unionization effort, so it’s kind of a contract signed under duress. The effectiveness of a union like IWW is, as I understand it, not based in legal unions backed by the State, but it certainly benefits anyway.

To put it another way, If I could legally beat you up to take your money, but I choose not to – though you know that I can change my mind at any time – would this not affect the way you respond to my demands? Real-world private unions can choose to become legal unions and “beat you up.” Because they choose not to, I can’t ideologically oppose their right to organize. But it would be intellectually dishonest to assume that they exist in the same context as a union formed in a purely free market, such as our earlier hypothetical. This pressure is similar to the examples of people will give the government easements onto their land to avoid an eminent domain battle. Even if a contract is ostensibly signed voluntarily in a heavily regulated market, there’s often duress somewhere. The overarching threat of government force is omnipresent.

One could argue that the state apparatus through taxes, regulations, zoning laws, etc, makes self employment more difficult; that people are strongly incentivized to work and in many cases work for corporations. Thus, in summation, force has already been used to lower your bargaining power for wages or personal autonomy.

This is accurate and something anarcho-capitalists will acknowledge. However, the difference is that it doesn’t do this on behalf of employers, per se. Taxes, fees, licensing, etc., benefit the state. Additionally, they get lobbyists – so there are particular lobbying groups, which aren’t all businesses, trying to force American goods, etc. that are basically funding politicians. That’s a big part of it too.

Unless you can point to a specific case where you are pushed to wage labor at the behest of a specific company, then you’re kind of attacking a guy who benefits almost by accident. Like going after pawn shops because thieves sometimes sell them stolen goods that can’t be traced back to a legitimate sale.

Although it is tricky, it is my belief that the State has given the illusion of unions being necessary which is why I would argue that unions are not a solution to state-created problems, but by cutting back the State and freeing the market the illusion disappears and unions would be essentially unnecessary. What people need to realize is that unions always have an incentive to restrict supply, like any cartel. So if the problem is mass unemployment due to minimum wage laws, unions only exacerbate the problem. If the problem is not enough taxis, the solution isn’t to cap the number of licenses or to let the taxi medallion owners determine the supply of medallions. If the problem is not enough doctors, the solution isn’t to let the AMA cap the number of diplomas granted each year.

This leads to the question of should we achieve a free market, would unions be necessary? I argue that they wouldn’t, but this is where anarcho-capitalists like myself and left-libertarians differ. It’s suggested that free market philosophy and unionism have a long and rich intertwining history and both help empower the individual worker and unions would be necessary. As Roderick Long states in “Beyond the Boss: Protection from Business in a Free Nation” in arguing the benefits of a union:

“[…] bargaining power will have shifted to favor the employee. Since prosperous economies generally see an increase in the number of new ventures but a decrease in the birth rate, jobs will be chasing workers rather than vice versa. Employees will not feel coerced into accepting mistreatment because it will be so much easier to find a new job. And workers will have more clout, when initially hired, to demand a contract which rules out certain treatment, mandates reasonable notice for layoffs, stipulates parental leave, or whatever. And the kind of horizontal coordination made possible by telecommunications networking opens up the prospect that unions could become effective at collective bargaining without having to surrender authority to a union boss.”

The majority of that is the benefits of an economy growing faster than population. Insofar as his assumptions are correct, the conclusions do follow. At least until he gets to collective bargaining, where he simply assumes that being able to coordinate with your coworkers automatically means you find it in your interest to bargain collectively. It’s really easy to get in contact with coworkers now and coordinate efforts. But why bargain collectively? What would I get out of it? It makes no sense. Plus, any benefits of the union to its members necessarily come at the expense of non-members: consumers, employers, and non-unionized workers.

Why would a stateless society have no need for unions? Let us return to our previous hypothetical of the freely formed union in our stateless society. One day, the union rep announces he’s going to the boss to demand a raise for everybody, “or else.” Now, if my skill set and experience are such that I have better prospects elsewhere, I don’t need a union — I’ll negotiate a better wage for myself, with this employer or another. And if they’re not, a rather lousy gamble is being made on my behalf, and knowing this, I would probably never have joined the union in the first place. Not only this, but presumably union dues are being taken out of my check every month, for a deal where my only bargaining chip is to quit and be replaced by an unemployed guy for probably even less than what I’m making. The alternative option is for me to leave the union, save the union dues, and use the artificially created limitation to labor to negotiate better wages for myself as a scab.

Once workers catch on to this, union members drop, disempowering the union. If businesses are able to individually negotiate wages too low, the possibility of unionization is still a possibility keeping the employer in check. For the same reasons I previously mentioned that free market unions would be rather harmless, they would also be entirely unnecessary. Market regulations lead to a balance of power between employee and employer, and this comes at no expense to any third party.

So it’s fair to say that I support the right of a union to form in the absence of state coercion. But as a student of economics, I can’t envision a scenario in which I would support a union as a worker. Were one to form, I would likely be the scab who takes advantage of the new opening, and I believe the scab is more beneficial to society as a whole.

In light of what I argued above, some left-libertarians will obviously disagree. To play devil’s advocate, strikes aren’t the only “peaceful” tactics bottom-up unions use. Other things including slow downs, work-to-rule, picketing, boycott campaigns, etc., can and have been utilized in the past and are currently used by groups such as the IWW, the Coalition of Immokolee Workers, OURWalMart, and others and are useful in dealing with employee abuse, unsafe working conditions, and other unjust working practices.

This is not a difficult argument to refute. If an employee is easy to replace, walking off the job is a poor negotiating tactic because he will simply be replaced. If not, it’s a pointless tactic, as he can negotiate individually without threatening or injuring his employer. Unless the union has special legal protections, a slowdown exposes everybody to the very serious risk of being fired and replaced.

The same goes for all of these. In actual negotiations, each party focuses on the benefits to the other party. These more closely resemble blackmail than negotiation. “Capitulate, or else” except that without special legal protection it’s not much of a threat. I would not be inclined to capitulate, even if I dread the prospect of having to stop operations to replace everybody because I can know with reasonable certainty that they’ll do this again, next time with more optimism and less inclination to budge.

It’s also argued by left-libertarians, such as Kevin Carson, that the cost of replacing workers is too great. In Kevin Carson’s “Labor Struggle in a Free Market he states:

“The primary reason for the effectiveness of a strike is not the exclusion of scabs, but the transaction costs involved in hiring and training replacement workers, and the steep loss of productivity entailed in the disruption of human capital, institutional memory, and tacit knowledge.

With the strike is organized in depth, with multiple lines of defense — those sympathy and boycott strikes at every stage of production — the cost and disruption have a multiplier effect far beyond that of a strike in a single plant. Under such conditions, even a large minority of workers walking off the job at each stage of production can be quite effective.”

As far as this point is concerned, many of the costs of hiring come from compliance with government regulations which go out the window in a stateless society. Plus, Carson’s claim depends entirely on what the job is. Sure, if a bunch of software developers strike, you might have an issue training new people. For factory workers or fast food employees? Not really. This is why most strikes are operated in a manner as to discourage scabs and even forcibly prevent them from crossing the picket line and taking the now open jobs.

Carson is taking a small possibility – one that is already accounted for by market forces, as with a smaller labor pool, employees offer better wages anyway – and generalizing it to all strikes. But can he really say that a bunch of stock boys and cashiers walking out on Wal-Mart would matter? Of course not. They’d just hire new ones (even if new hires come at an increased cost, we’ve already established that this would often be preferable to avoid future strikes, thus maintaining the balance of negotiating power). If Carson’s arguments were true, unions wouldn’t rely so heavily on these tactics that prevent the hiring of replacement workers. Carson also seems to be missing the concept of a market wage (as if wages are arbitrary here – that the supply and demand for labor is not dependent on the wage rates involved). The incentive of the employer is to discover and pay that wage. Any higher and he’s throwing away money. Any lower and he’ll lose employees to better pay. Unions try to get higher than market wages – they aren’t needed to get market wages. So while it may be troublesome and expensive to replace your workforce, at market wages there’s no reason to expect to have to do that over and over, but if you capitulate to the union, your costs will continue to increase well beyond market wages until you go broke or finally decide to fire everybody.

The important thing to realize though is that, whether or not a union is moral, the economic implications change very little. Even if a union is 100% voluntary and a company voluntarily gave it a closed shop contract under no threat of state duress, that union would still constrict supply and raise prices as any cartel does if the demand curve is inelastic. Typically, cartels don’t form without barriers to entry, but, as Rothbard points out in one of his Volcker Fund memos, you could theoretically have a voluntary closed shop. Unions are still going to lead to an incentive to automate while at the same time opposing automation where they can, as well as a pricing out of non-union employees, regardless of whether or not it is formed through free association.

Even in a purely free economy, unions suffer from the same economic problems that government faces – even assuming the government has perfect knowledge and perfect morality: the cost of administration. A union must be funded by dues and be headed by salaried union bosses. To be successful in a free market, it would have to negotiate better wages and working conditions in a way that exceeded the cost of union dues, did not harm non-unionized citizens, and it would have to do all of this while paying for an administration to facilitate all of it. In short, it is economically impossible for unions to have a net benefit for society.

As economist Thomas Sowell points out, there are two occupations that are paid to produce nothing: the bureaucrat and the union boss. So while I will adamantly defend your right to unionize as free association, I would point out how it’s time consuming and completely unnecessary. Thus, it’s through the free market that it will be seen as pointless, hence it being abolished and naturally dissolved. Compared to a free market, the union is a far inferior tool on behalf of the worker.




Thanks to those who contributed to the discussion and insight: Chris Calton, Rocco Stanzione, Kevin Carson, Andrew Criscione, Jason Lee Byas