Minimum wage, measurements and incarceration rates

A few weeks ago, I published a blog post about how incarceration rates affect our measurement of the relative economic conditions of Blacks in America. My claim was that the statistics are hiding a reversal of the painfully achieved advances secured between 1870 and 1960. Basically, my claim was that those who (in greater numbers) found their ways to a prison cell tended to be at the bottom of the income distribution, were more susceptible to be unemployed and had lower wages. This creates a composition effect whereby the official surveys cream-skim the top of black wage, income and employment distributions.

But, could this problem also affect our measurement of the effects of minimum wage? Let me be clear before you continue ahead, I am just asking this question because I could find no satisfactory answer to (or even mention of) this issue.

In recent times, minimum wage surveys have tended to find some gains in earnings for some workers following increases in minimum wage rates. Regardless of how you look at the prison population, it increases  – albeit at a decelerating rate since the early 2000s – since the 1980s. Coincidentally, that starting point is also the point at which the famous Minimum Wage Study Commission was published (1981). That report basically cemented the point made by George Stigler (i.e. minimum wages are not desirable). That report surveyed the entire literature to summarize the amplitude of the effects. That literature encompassed articles written between the end of the Second World War and … well… 1981. If you look below at the graph, incarceration rates were more or less constant during that regime. Thus, if there were composition effects associated with surveys of wages, incomes and employment, they were more moderate than after 1981 when incarceration rates surged.


But, its also after 1981 that some papers began to find some positive effects of minimum wage increases. These studies took place under a growing composition problem in surveys of wages, incomes and employment. Take the famous Dube, Lester and Reich paper in the Review of Economics and Statistics  who used data from 1990 to 2006. During that period, the male incarceration rate surged from 297 per 100,000 to 501 per 100,000. I understand that DLR used a time fixed effect method, but would that be sufficient to at least deal with the issue of shifting labour supplied (it won’t for the data bias issues described notably by Bruce Western)

If we assume that those who are plausibly affected by minimum wages (i.e. lower income individuals) are also those more likely to end up in jail in the United States, then there is clearly a bias. As they are dropped from the labor market (or as they join the prison population), they leave only the workers least affected by the minimum wage inside the samples. That is one possibility.

The other possibility – which is that surveys do not suffer from a large composition, but which is not mutually exclusive to that composition problem – is that the growing prison population represents a year-over-year reduction in the labor supply which offsets the effects of hikes in the minimum wage (or maybe even eliminates them entirely if the shift is big enough).


I have tried many variations of this google scholar research and went back to my copy of the Handbook of Labor Economics and my Economics of Inequality, Poverty and Discrimination  (a book worth reading by the way) and I found very little on this point. Very few scholars have considered the possibility of this problem (which implies a shift of the labor supply curve concurrent with minimum wage hikes and a composition problem where those affected are simply not measured anymore). Yet, I feel like this is a defensible claim. In England, where some studies also show minimal effects or positive effects of the minimum wage, there has also been an increase in the prison population. In contrast, Canada – whose prison population is declining moderately (meaning that the labor supply is increasing as the minimum wage is being increased – the studies do tend to find the “conventional” result.

Am I crazy or is this a case of poor measurement? Personally, I feel that there must an answer, but please tell me I did not just stumble on this!

Reply to Matthew Strebe on Hoppe and Immigration

In response to my recent post on immigration, fellow Notewriter Matthew Strebe asserts two main objections to my defense of open borders: freedom of movement is not a right in the first place so immigration restrictions are not violating anyone’s rights, and that citizenship is relevant to the debate over the impact of policies. In order to make the case for the first, Strebe relies on Hans-Hermann Hoppe’s anarcho-capitalist counterfactual, which has been debated at length over the past twenty or so years by libertarians. I think the second argument is inconsistent with the first and winds up begging the question. Before I delve into the argument, I should mention that I neglected to bring it up in my first post because I hadn’t really seen many people bring it up in recent years and was addressing arguments which I considered most pertinent to the current discussion on this topic among libertarians. That said, it is worth addressing again since Mr. Strebe has brought it up.

The Anarcho-Capitalist Counterfactual

Mr. Strebe presents the argument that the right to freedom of movement is not really a right at all because, in the absence of a state, people would not have the right to traverse across owned property, and ultimately free movement would become a privilege to be handed out at the whims of private property owners, an argument originally made by Hans-Hermann Hoppe.

This argument gets off to a bad start when Mr. Strebe, quoting Hoppe, claims the argument for free movement to work “it is – implicitly – assumed that the territory in question is unowned, and that the immigrants enter virgin territory (open frontier).” This assumption need not be made. Typically, when an immigrant seeks to come to another country they want to buy or rent some previously owned property which is being voluntarily sold or rented out to them. This would be perfectly acceptable under anarchism as, without a state to control who can contract with whom, there is little reason why person A can’t contract with person B just because a third party forces A away. Even from Hoppe’s perspective, borders are arbitrary; and this is hinted at when Hoppe himself calls them “unnatural” and “coercive.”

Even ignoring this, Mr. Strebe continues:

In Hoppe’s example of an anarcho-capitalist society, all land is privately owned, and so freedom of movement becomes absurd. How could one individual have the untrammeled ability to traverse another person’s property? The only proper relation is one of mutual freedom of association – one property owner may decide to hang out with, say, Mexicans, while another would not. Freedom of movement becomes dependent on individual consent, which in turn (using the historical example of the monarchy) is based on calculated self-interest. This leads to another possibility: all property owners could willingly confederate and decide they will not associate with Mexicans or some other group, and freedom of movement to that group, such as it was, ceases to exist. Thus, freedom of movement as a human right is absurd in an anarcho-capitalist society because there is no freedom to traverse the unowned land.

Even though this federation could hypothetically happen under anarchism, this does not mean freedom of movement is a farce. Freedom of movement does not simply mean freedom to traverse land, it also means freedom to buy land regardless of location, or contract with other consenting individuals regardless of location. Immigration restrictions not only forbid movement across land, they also forbid freedom of contract. When I said “freedom of movement,” I was referring to a bundle of rights. Even if you do not accept that freedom to traverse land is a human right other rights would exist under anarcho-capitalism are still trampled upon by immigration restrictions.

Further, what government does by restricting immigration is not at all analogous to what happens when a private owner of land puts up no trespassing signs. Government not only restricts immigrants from entering publicly owned land (such as the area near the border, or national parks), they also restrict the immigrants from entering privately owned land, most of which they would be welcome to buy up for themselves or work on by its rightful owners, simply because that land is on the other side of a line in the sand called the border. It would be as if two people agreed to contract together on a piece of privately owned land in anarcho-capitalism, yet a third person who owned another piece of land unaffected by the contract used force to stop the contract from happening. Government is not only restricting movement to land that it properly owns, but it also restricts movement to lands privately held by its citizens.

Hoppe himself acknowledges this in his discussion of “forced exclusion:”

Now, if the government excludes a person while even one domestic resident wants to admit this very person onto his property, the result is forced exclusion (a phenomenon that does not exist under private property anarchism). Furthermore, if the government admits a person while there is not even one domestic resident who wants to have this person on his property, the result is forced integration (also non-existent under private property anarchism).

Of course, Hoppe still favors immigration restrictions because he believes it will result in more “forced integration,” a point Mr. Strebe brings up which will be addressed at length in a moment. My point: even under Hoppe’s own argument, it is incorrect to say “in a monarchy, the king owns all the land, and in a democracy, an association of elected bureaucrats holds sovereignty over all land[.]” The government doesn’t own all the land, private citizens do. Restricting citizens from voluntarily contracting with non-citizens is, in fact, restricting those citizens’ own freedom of association as well.

But even if we concede that government owns some public land, this would still not be analogous to anarcho-capitalism as Hoppe argues for it. For Hoppe, the state is illegitimate in the first place because it acquires its existence (including the ownership of public land) through an immoral monopoly on force. To bring back the example of anarchism, suppose that I claimed ownership over all the land in a region which I gained by killing the previous owners and forcing everyone else on the territory into submission with guns. Does this grant me the moral right to stop people from moving onto land I stole and do not rightfully own? If so, then we have an extremely perverse notion of property rights. Government doesn’t really “own” any property, it steals it through force.[1] Private citizens own most of the land within a geopolitical border to begin with, and the land the government does own it does not rightfully own as it only acquires land through coercion.

Even if we grant that anarcho-capitalism would end all immigration and freedom of movement, how does this carry any normative force for what policies an existing state should have? As Don Boudreaux points out, under anarcho-capitalism, one is not allowed to speak freely on private property, and hypothetically a group of property owners could form a federation and ban the expression of certain opinions. Does this mean the government is justified in restricting free speech in publicly owned areas, or in the nation at large? Of course not. Further, no one would argue that this makes “freedom of speech” a farce as a human right because there is a scenario in which it hypothetically wouldn’t exist in its pure form under anarcho-capitalism.[2]

This brings us to the point about “forced integration.” To call allowing individuals to contract with other members of a nation-state or own land in a nation state “forced integration” is really bizarre. As Don Boudreaux argues:

[L]abeling open immigration as “forced integration” is disingenuous. Such a practice is identical to labeling the First Amendment’s protection of free speech as “forced listening.” But keeping government from regulating speech is not at all the same thing as forcing people to listen. Likewise, allowing people to immigrate to America is not the same thing as forcing Americans to associate against their wills with immigrants. Under a regime of open immigration, I need not hire or dine with anyone whom I don’t wish to hire or dine with. Indeed, whenever government restricts immigration it coercively prevents me, as an American, from hiring or dining with whoever I choose to hire or dine with. An immigrant who receives no welfare payments engages only in consensual capitalist acts with those (and only those) domestic citizens who choose to deal with the immigrant. Just as trade restraints are, at bottom, restrictions on the freedoms of domestic citizens, so, too, are immigration restraints restrictions on the freedoms of domestic citizens.

As Walter Block and Anthony Gregory point out, this type of argument could justify many other statist interventions:

Hoppe’s position that keeping illegals off public property because of their supposed “invasiveness” could easily be extended to other matters, aside from free trade. Gun laws, drug laws, prostitution laws, drinking laws, smoking laws, laws against prayer—all of these things could be defended on the basis that many tax-paying property owners would not want such behavior on their own private property. Such examples are hardly without a real-world basis. Large numbers of Americans would not allow guests in their homes if those guests had machineguns or crack cocaine in their possession. The principle of the freedom to exclude and set conditions for entry onto private property simply cannot be extended to the socialized public sphere, or else all sorts of unlibertarian, illiberal policies could be as easily justified as border controls. In other words, just because an individual—or many individuals—would not want act X to occur on their property does not mean that, according to libertarian law, it can be prohibited as a general principle, even on so-called “public property.”

To me, forced integration as a concept is incoherent. Allowing someone to exist within geopolitical borders is not the same as forcing others to associate with that person. It confuses the public-private distinction, which has been fundamental to classical liberal thought throughout the entire tradition.

But even if we accept it as a legitimate argument, Mr. Strebe and Hoppe need to prove why “forced integration” is more of a problem than “forced exclusion.” They believe they have shown this under democracies because democratic politicians will be a more predominant problem in democracies because democratic leaders are more likely to admit stupid riff-raff who will vote for them rather than people who the landowners would not want to associate with. Empirically, this argument seems implausible. If this were the case, then why does the US government need to forcibly stop private citizens from engaging in labor contracts with illegal immigrants?

But even if this were the case, wouldn’t that be an argument in favor of open borders? Open borders as a policy mean that democratic politicians can make no judgments on who can and cannot enter the country, including the “bums” and “parasites” Hoppe doesn’t like get in as well as the immigrants “superior” people would want to associate with. If the citizenry truly do not want to associate with these “bad” immigrants then the immigrants will have problems finding others to contract with (meaning they won’t find a job and will be destitute), face severe social exclusion by the citizens, and will be less likely to immigrate in the first place or will go back to their country due to their misery here. Freedom of association would win out under open borders. Why would closed borders—when democratic politicians really are the only ones making decisions about who can come into the country and who cannot—result in more forced integration than open borders?

Hoppe does have somewhat of a response to this when he points out how government owned roads and non-discrimination laws will increase “forced integration.” But as Alexander Funcke argues, these arguments are based on some pretty faulty assumptions:

As he notes that migrants may “proceed on public roads and lands to every domestic resident’s doorsteps […] and to access, protected by a multitude of non-discrimination laws, […]”, etc., he need to explain why they would be a greater nuance than the current population. To argue this, he claim a strong human ethno-cultural homophilia, e.g.: “[In a residential area, the] desire for undisturbed possession—peace and quite—is best accomplished by a high degree of ethno-cultural homogenity.” This claim, hinges on two questionable assumptions. First, that the ethno-cultural distances within countries dwarfs the between-state distances. This is far from obvious, New York City and London seem ethno-culturally closer than the two U.S. cities, New York City, NY, and Albuquerque, NM. El Paso, TX, seem closer to El Ciudad Juarez, Mexico, than to Boston, MA. Secondly, as Thomas Schelling and many after him has shown, what on the surface look like strong homophily often is a product of path-dependency and at most weak preference for at least someone similar.

Even if “forced integration” were a coherent concept and more of a prevalent problem, that in itself wouldn’t refute a policy of open borders. Many immigrants choose to relocate because of conditions of poverty or oppression, to find for themselves a better life. Is it just that the state should forcibly keep non-citizens impoverished simply because some citizens feel uncomfortable existing in the same geopolitical borders as someone they do not like? It’s a pretty counter-intuitive and morally questionable argument to make.

In sum, border restrictions are not consistent with anarcho-capitalism because they restrict freedom of contract and freedom of association as well as freedom of movement, even if you deny that freedom of movement is a fundamental human right. And even if anarcho-capitalism would abolish immigration, that carries no normative force for why existing states should do so.

Ingroup vs. Outgrouping on Citizenship is Inconsistent with Anarcho-Capitalism

Mr. Strebe thinks that he can show why my reduction ad absurdum doesn’t work because it is enforced on non-citizens and not citizens:

[I]t assumes there is an equivalency between immigration and any other government policy, such as Medicare or eugenics. Without such an equivalency, Mr. Woodman’s appeal to the faulty logic of his interlocutors’ argument falls apart, as his own argument no longer possesses the balance between its two examples it relies upon for its logical and persuasive force. Here’s the problem: Medicare or eugenics are internal policies that affect the ingroup, the citizenry, only. Immigration is an external policy that affects both an outgroup, the immigrants, and the ingroup, the citizenry.

This is simply begging the question: why would such an ingroup-outgroup dynamic in regards to citizenship be morally relevant in the first place? Mr. Strebe tries to address this in his conclusion:

It should be clear that this is a non-sequitur: non-citizens do not have rights to the sovereign territory of a country, which is held either by private citizens or the public. The government does not restrict their rights when it refuses to grant them the privilege of traversing land that is publicly held for the ingroup because they had no rights to that land to begin with. Because the government is nominally beholden to the ingroup, and not to any outgroup, rights discourse concerning the outgroup is fundamentally absurd when considered in terms of Hoppe’s arguments.

I’m not sure why asking for a major premise to be justified is a non-sequitur, but Hoppe’s arguments are contradictory with what Mr. Strebe says here. Hoppe’s argument hinges on the idea that the government and its citizens are the only ones who have the right to exist within a nation. Here’s how Mr. Strebe’s argument above works, correct me if I’m wrong:

  1. If an individual or group has rightful ownership over a property, they may rightfully exclude others from trespassing on the property.
  2. The government and its citizens rightfully own the territory within a nation-state.
  3. Therefore, the government is justified in prohibiting non-citizens from entering a nation-state but not citizens.

How does 2 at all follow from Hoppe’s argument?[3] Hoppe claims that government is an illegitimate monopoly on force. If this is the case, how can the state have rightful sovereign ownership over any property in the first place? If it is the case that the state is illegitimate, and the ingroup-outgroup of citizenship and borders distinction is itself only a result of the state’s “unnatural” existence (which Hoppe himself admits), then how can you claim that only citizens as defined by an illegitimate government “have sovereign rights over the territory of a country?” It’s an argument that is inconsistent with the whole premise of anarcho-capitalism in the first place, and again confuses the private nature of property ownership with the coercive, public nature of government. If it is the case that the distinction between non-citizens and citizens is simply the result of the illegitimate, arbitrary use of government force, then the analogies to Medicare and eugenics are still valid.


I think this whole discussion is misleading because I do not believe in deontological natural rights and think Hoppe’s conception of property is untenable in the first place. Despite this, Mr. Strebe and Hoppe’s anarcho-capitalist counterfactual fails on its own terms. A state which stops its citizens from contracting with non-citizens is a fairytale in the absence of the state because the concept of “citizen” and “noncitizen” is dependent on the existence of the state in the first place. Of course, property owners could prohibit certain individuals from trespassing on their property, but that is not at all analogous with what the state does when it restricts immigration. And even if we accept everything about the anarcho-capitalist counterfactual argument, there still is no moral reason to treat non-citizens differently than citizens.

But Mr. Strebe is right about one thing: my original post did not address the problem of tribalism, which I agree is the root cause of most opposition to immigration, but that seems to be a separate issue from Hoppe’s argument and I don’t see very many libertarians making purely tribalist arguments. There are two ways to coherently argue that my reductio ad absurdum is invalid: 1) by claiming that there is some moral reason for treating citizen’s rights differently than non-citizens (tribalism) or 2) by claiming that immigration restrictions are not really violating anyone’s rights in the first place. Hoppe’s argument doesn’t make a libertarian case for tribalism: at most, it can make the case that freedom of movement isn’t really a right and so immigration restrictions are not rights-violations. I believe it fails to make this case, but that’s a separate argument from tribalism. Even if I agree that freedom of movement understood simply as the ability to traverse land isn’t a right, I can easily reply: “Sure, freedom of movement per se isn’t a right, but immigration restrictions also violate freedom of contract, so the reductio still stands.” If Mr. Strebe wants to show why the reductio ad absurdum is invalid, he will need to 1) provide an argument for why treating citizens differently than non-citizens, or tribalism, is justified or 2) show that freedom of contract isn’t really a right or that immigration restrictions do not really violate this right.

I thank Mr. Strebe for this opportunity to discuss these topics with him, and look forward to his response.

[1] It is worth noting at this point that I think that discussion over who has the “natural right” over property based on past ownership is not the best way to approach property rights, which are themselves an ever-changing result of tacit knowledge and spontaneous order, see Hayek’s discussion of property in chapter two of the Fatal Conceit. Also, I think Hoppe’s vision of anarcho-capitalism is likely not what would happen and tends too far in the direction of a perverse crytpo-feudalism. However, the point is that to say government owns property and therefore is justified in forcibly controlling who can or cannot enter a country is inconsistent with Hoppe’s own deontological arguments against government in the first place.

[2] At this point, it is worth mentioning that David Gordon argues this is not self-evidently absurd because we do not talk about “rights” when discussing the use of public land, but prudential consequences. So it is fine to restrict the “riff raff” in, for example, an airport just as it is fine to restrict movement across public roads. Even if we accept this premise, then Hoppe’s argument that we should not consider consequences and only the morality of immigration, which is how he starts his argument, is thrown out the window and we are going into the realm of consequentialist arguments, which Hoppe admits are in favor of pro-immigration even with the added caveat that economic growth isn’t the end-all-be-all of welfare because value is subjective. Hoppe’s argument against immigration will then have to hinge entirely on consequentialist cultural-based argument, such as his rhetoric about “forced integration.”

[3] I would also heavily qualify 1 with a more nuanced notion of “possession” versus “property” inspired by mutualism, but that is a discussion for another day.