Privatization—the transfer of functions and assets from the coercive “public” sector to the voluntary private sector—is a central part of the classical liberal program for radically reducing government influence in society. But not all privatization is created equal, and not everything that goes by that name would reduce the power and scope of the state. Some suggest that economic efficiency demands that the speed of transfer should take precedence over the manner of transfer, apparently on the assumption that by some “invisible hand” Coasean process assets will always end up with the “right” people. But we have good reason to believe that how privatization proceeds is important to satisfying both justice and efficiency criteria.

Is privatization even the right word for what classical liberals (henceforth liberals) want? Does it capture the essence of their objective? We shall see, but first let’s explore the virtues of getting functions and assets out of the hands of government officials.

One need not be an anarchist to appreciate that the state suffers two grave flaws in its provision of services. These have been dubbed the calculation/knowledge problem and the incentive problem.

The Calculation/Knowledge Problem

The elaboration of the knowledge problem occurred during the famous debate over the efficacy of state socialism launched in 1920 by Austrian economist Ludwig von Mises and extended in the 1930s by his student F. A. Hayek. The upshot of the critique is that outside the market—that is, the economic-political-legal environment in which the means of production are privately held and freely traded—those who make decisions to allocate scarce means among competing ends necessarily act without the indispensable ability to engage in the economic calculation that prices routinely accord market participants. Prices, Mises wrote, permit disparate resources to be expressed in terms of a common denominator—the monetary unit—enabling people to appraise and compare the profitability of various possible products and production plans.

One doesn’t engage in abstract production. Rather, one chooses to produce item X rather than item Y. But that’s just the beginning. Having decided on item X, should one use method A—with one combination of labor, raw materials, and producers’ goods—or method B—with a different combination? Mises showed that these questions cannot be rationally answered without the money prices that markets generate through the free exchange of labor services and privately owned land, raw materials, and capital goods. In the absence of prices for the means of production, even a dictator who cared nothing for what consumers wanted would be unable to calculate the most efficient way to produce all the things he thought should be produced.

Reeling from Mises’s body blows, the state socialist theorists conceded his point—even as they missed it—and revised their doctrine ostensibly to accommodate it. Yes, prices are necessary, they said, but markets are not necessary to have prices. Given a powerful enough computer, bureaucrats could use a system of simultaneous equations to impute prices to producers’ goods based on consumer demand and to achieve the conditions of general equilibrium.

At this point, Hayek grabbed the baton to argue that the “market socialists” failed to understand the true epistemological nature of prices. Much of the knowledge that motivates people’s actions is never articulated or even articulable. Thus there is no way that bureaucrats could acquire it for use in their equations. Indeed, critical “data” on which people act, far from being “given,” do not even exist before individuals are confronted with often-unanticipated alternatives. Socialist economists then regrouped and concocted schemes intended to arrive at market prices by nonmarket means—schemes Hayek demolished handily.

As Hayek wrote in his influential 1945 article, “The Use of Knowledge in Society”:

The peculiar character of the problem of a rational economic order is determined precisely by the fact that the knowledge of the circumstances of which we must make use never exists in concentrated or integrated form but solely as the dispersed bits of incomplete and frequently contradictory knowledge which all the separate individuals possess. The economic problem of society is thus not merely a problem of how to allocate “given” resources—if “given” is taken to mean given to a single mind which deliberately solves the problem set by these “data.” It is rather a problem of how to secure the best use of resources known to any of the members of society, for ends whose relative importance only these individuals know. Or, to put it briefly, it is a problem of the utilization of knowledge which is not given to anyone in its totality.

What does it mean to produce, economically speaking? It obviously does not mean that something is created out of nothing. That power is denied the human race. Production, instead, is transformation, the conversion of resources from one form to another. As praxeological beings, we logically cannot wish to convert resources from a preferred form to a non-preferred form. That is, no one can want a production process that destroys rather than creates (subjective) value. However, in a world of fallible people, a process expected to create value could end up destroying it.

We should hope, then, that the economic system would facilitate the creation of value. In a world of scarcity, devoting resources to a process that turns out to be destructive has a cost not merely in the lost resources, but the loss of whatever those resources could have produced instead.

Mises elaborated the role of prices in this context. Market-generated factor prices enable entrepreneurs to envision productive plans whose final consumer goods are anticipated to pay returns that exceed the total money cost of the inputs. That return is pure entrepreneurial profit. If things work out that way, it is a sign that value has been created.

Such profit is the result of complex arbitrage. In simple arbitrage, one buys low, then turns around and sells high. With the complex arbitrage of entrepreneurship, one buys inputs low and after a period of production, sells the transformed product high. In other words, a successful risk-taking entrepreneur intuits that inputs are undervalued with respect to how consumers appraise the final product. Profit is the entrepreneur’s reward for his correct judgment.

Losses indicate that value has been destroyed. They force producers to revise their plans to better satisfy consumers or transfer their assets to those who will. As Mises liked to point out, in a market economy, devoid of all privilege, the people as consumers ultimately determine who controls the means of production. Ironically, it is the individualist free market that delivers on the state socialist’s collectivist promise.

The lure of profit in a competitive free market induces entrepreneurs to please consumers, often in ways the consumers never imagined, and aligns personal interest with social betterment. Moreover, profits and the price system are the principal way in which individuals in what Adam Smith termed a “great society” can coordinate their disparate plans and expectations with one another, which is critical to all people’s hopes of realizing their aspirations. As Hayek emphasized, the trial and error of competition is “a discovery procedure,” enabling us to learn things we couldn’t have learned otherwise.

The implications for government provision of services should be plain. The difference between a private, profit-seeking competitive enterprise in a freed market and a monopolistic government bureaucracy is vast, as Mises explained in his book Bureaucracy. In contrast to the enterprise, the bureaucracy obtains its revenue by compulsion—that is, taxation. While an enterprise must offer its products and services to potential consumers who are free to say no, a bureaucracy looks to the taxman. Saying no is not an option for “consumers” of government services. Aside from the moral defect in this arrangement, economics tells us a bureaucracy that can compel support will lack the feedback that free consumers constantly provide competitive enterprises. In other words, government services are not subject to market pricing, leaving no way to gauge relative consumer demand. It is hardly mysterious that government services are inferior to what a competitive market would provide.

In a mixed economy bureaucracies are not totally immune from market forces. Their inputs are, after all, purchased in the marketplace. However, the revenues that fund their operations are obtained coercively and allocated to them without regard to consumer demand, so they don’t face the same constraints that competitive enterprises face. And of course not all of the government’s inputs are purchased on the market—consider those obtained through eminent domain and conscription.

The Incentive Problem

Even if we assume away the calculation/knowledge problem, the prospects of government’s satisfactory provision of services would remain bleak. Could we expect bureaucracies ably to serve their “customers”? According to the Public Choice school of political economy, no. Why not? Because there is no reason to assume that “public servants” would have the incentive to serve the public even if they knew precisely what that required.

Public Choice brought a neglected realism to the study of government by introducing the assumption that politicians and bureaucrats are motivated much like everyone else. Without dismissing the possibility that some of them sometimes intend to do good, it is nevertheless the case that they see the world, as the rest of us do, through the prism of personal interest. They are pursuing careers—striving for higher incomes, security, prestige, influence, authority, and power. Separating the personal from the public good is not as simple as it might seem. Public Choice teaches that there are simply no grounds for assuming that people in the private sector tend to be “greedy” while people in the government sector tend to be altruistic.

This, however, is an incomplete statement of the incentive problem, which can be filled out by adding the “ruling-class factor.” This is the government’s inherent tendency to serve the interests of a well-connected elite, historically most often a business elite. This is not to say that politicians and bureaucrats ignore the masses; in a democracy they must do enough to obtain and retain power. But that is not incompatible with the fact that, as the iron law of oligarchy holds, a relatively small subset of society necessarily will emerge as most influential over government policy.

Public Choice has identified one reason for this: Government programs concentrate their benefits on well-organized and relatively small interest groups, while dispersing the costs of their proposals among the mass population. The cost to any individual of any particular program is small, though the aggregate cost of all programs is huge, as is the aggregate cost to the entire society. How much would a consumer be willing to spend to oppose, say, the restrictions on sugar imports (which benefit a few domestic sugar producers), if the program costs her a few dollars per year? That assumes she is even aware of the extra cost or its cause. (Government has myriad ways to raise political transaction costs.)

The system will tend toward interventions that socialize risk and cost, create artificial scarcities (such as intellectual “property”), and privatize the resulting “excess” profits. In contrast, the free market would privatize costs and through competition progressively “socialize” benefits. (Frédéric Bastiat has much to say about this in chapter 8, “Private Property and Common Wealth,” of Economic Harmonies.)

Given the power of government to distribute other people’s money, it is inevitable in a democracy that candidates and well-organized interest groups will gravitate toward one another, resulting in an environment in which government is reasonably seen as the “executive committee of the ruling class.” (Contrary to the apparently Marxian tone of this discussion, class analysis originated with French classical-liberal economists in the early nineteenth century, something of which Marx was quite aware.)

Privatization

Thus the state cannot be counted on to provide satisfactory goods and services to the general population. So what is to be done? The answer is for government to divest itself of assets and functions. Before discussing how, we should acknowledge that some things government does should not be privatized—simply because those things shouldn’t be done by anyone. For example, if a government conscripted men and women for the military (or civilian workforce), liberals would not propose privatizing the machinery of the draft in the name of efficiency. Rather, they would propose abolishing conscription because it amounts to slavery. Similarly with many other functions governments perform, including taxation, occupational licensing (as opposed to rating and certification), administration of the “war on drugs,” and foreign military intervention.

For more than a decade, prison administration has subject to privatization. According to Corey Mason of the Sentencing Project, author of “Too Good to Be True: Private Prisons in America”:

In 2010, private prisons held 128,195 of the 1.6 million state and federal prisoners in the United States, representing eight percent of the total population. For the period 1999–2010, the number of individuals held in private prisons grew by 80 percent, compared to 18 percent for the overall prison population. While both federal and state governments increasingly relied on privatization, the federal prison system’s commitment to privatization grew much more dramatically. The number of federal prisoners held in private prisons rose from 3,828 to 33,830, an increase of 784 percent, while the number of state prisoners incarcerated privately grew by 40 percent, from 67,380 to 94,365. Today, 30 states maintain some level of privatization, with seven states housing more than a quarter of their prison populations privately.

Private prisons combine the two negatives already mentioned: Rather than ceding a service to the competitive market, prison privatization constitutes merely the contracting-out of a government monopoly. Whatever efficiencies that may be realized through the competitive bidding for a monopoly government contract, the resulting service can hardly be considered “free market.”

Moreover, many prisoners are victims of a government function no one should perform: enforcement of victimless crime laws, including limits on immigration. Indeed, Corey writes:

Public policies adopted during the 1970s and 1980s facilitated an increase in prison privatization. The War on Drugs and harsher sentencing policies, including mandatory minimum sentences, fueled a rapid expansion in the nation’s prison population. The resulting burden on the public sector led private companies to reemerge during the 1970s to operate halfway houses. They extended their reach in the 1980s by contracting with the Immigration and Naturalization Service (INS) to detain undocumented immigrants. These forms of privatization “on the ‘soft’ end of the correctional continuum” were followed by the reappearance of for-profit prison privatization.

A better way to prevent burdens on the government prison system would be to abolish victimless crime laws and reserve prison sentences for habitually violent criminals, replacing punishment with restitution whenever feasible. The alternative is to give “private” companies an incentive to support imprisonment for a wider variety of offenses, both real and victimless. As Sean Gabb, director of the Libertarian Alliance in England, writes regarding the British experience with privatization:

[It] is leading to the abolition of the distinction between public and private. Security companies, for example, are being awarded contracts to ferry defendants between prison and court, and in some cases to build and operate prisons. This has been sold to us on the—perfectly correct —grounds that it ensures better value for money. But it also involves grants of state powers of coercion to private organisations. All over the country, private companies are being given powers of surveillance and control greater than the Police used to possess.

If we confine our attention to functions that may legitimately be performed in the marketplace, the label privatization is applied to a disparate group of measures. On the one hand, it is used to describe Great Britain’s and the Czech Republic’s programs to transfer state firms to private hands in the 1980s and 1990s, respectively. On the other, when Baltimore in the 1990s contracted with Education Alternatives Inc. to run nine city schools, that also was called privatization, despite the city’s maintenance of ownership and coercive funding of the schools. It was merely contracting out a government monopoly.

The word is also used in connection with the contracting out of the New Jersey state lottery, DMV, state parks, psychiatric hospitals, and turnpike toll booths, all of which ultimately remain government responsibilities.

This should lead us to refine our idea of what liberals seek. It’s not privatization per se, but free competition through voluntary exchange, that is desirable. It matters little whether the government calls people who perform its functions public employees or private contractors. In fact, when a company becomes a monopoly government contractor, to that extent it is an arm of the state rather than a private firm. (See Kevin Carson, “‘Private’ Versus ‘Public” Sector” and “Ephemeralization: A Weapon Against Capital” and Brad Spangler, “Recognizing Faux Private Interests that Are Actually Part of the State”)

For that reason, such ersatz “privatization” devices as contracting out and the operation of charter schools won’t be discussed here. These merely blur the line between “private” and “public” sector—which is in the nature of corporatism—and undermine the case for the genuine divestiture of state-held assets.

The problem is that we don’t have a suitable word for what liberals want. Charles W. Johnson suggests that “[w]hat we advocate is the devolution of state-confiscated wealth and state-confiscated industries back to civil society… . Now, given the diversity of cases, and all of the different ways in which government might justly devolve property from State control to civil society, privatization is really too limiting a term. So instead let’s call what we want the socialization of the means of production.”

As Johnson notes, state assets can be and have been divested in a variety of ways—some clearly better than others. Perhaps most egregious was what occurred in Russia after the fall of the Soviet Union. Russia’s “privatization,” Robert W. Poole Jr. writes, featured “a botched shares-for-debt scheme … that created an instant crop of politically connected billionaires.” Among former Soviet republics, on the other hand, the Czech Republic (which had been more collectivized than its fellow Warsaw Pact members) moved quicker than others, striving to avoid corruption and to enable widespread ownership of shares in formerly state-owned enterprises. Elsewhere in Eastern Europe (and in the developing world), Poole writes, “the privatization record was mixed with many cases of less-than-transparent sale processes (to firms well connected with government officials).”

One of the earliest and largest divestiture programs took place under British Prime Minister Margaret Thatcher in the 1980s. In Thatcher’s Britain, an effort was made to maximize individual participation and ownership. Poole writes:

Thatcher succeeded in making privatization politically popular while selling off the commanding heights of the British economy: British Airways, [the] British Airports Authority, British Petroleum, British Telecom, and several million units of public housing, to name only a few examples. Thatcher’s political strategy emphasized widespread public share offerings rather than auctions to other private firms. Over the decade, this approach tripled the number of individual shareholders in Britain, giving the policy a popular base of support. [Emphasis added.]

Britain set an example for many other countries: “By the end of the 1980s, the sale of SOEs had gone global, inspired in part by the British example. Governments in France, Germany, Japan, Australia, Argentina, and Chile all sold numerous SOEs.”

But this was not an unmixed blessing, since, Poole notes, “global privatization proceeds ran in the tens of billions of dollars each year.” Skeptics about the beneficence of government power can hardly thrill to the thought of politicians acquiring tens of billions of new dollars with which to make mischief. In addition, “[i]n the case of public utilities (airports, electricity, water, etc.), privatization generally led to the creation of some form of regulatory oversight if the company remained a monopoly provider.”

Moreover, Thatcher’s privatization was not all that it seemed on the surface. As Gabb observes:

As reconstructed in the 1980s … the new statism is different. It looks like private enterprise. It makes a profit. Those in charge of it are paid vast salaries, and smugly believe they are worth every penny.

Undoubtedly, there are benefits. We have the most open and deregulated telephone network in the world. The other utilities give better value for money than they did. The internal market in the National Health Service is beginning to cap the otherwise limitless rise in welfare budgets. Applied to policing, the same policies may be about to do the same for law enforcement.

But for all its external appearance, the reality is statism. And because it makes a profit, it is more stable than the old. It is also more pervasive. Look at these privatised companies, with their boards full of retired politicians, their cosy relationships with the regulators, their quick and easy ways to get whatever privileges they want… .

Ideally, then, truly competitive privatization would be effected without cronyism or privilege. How might that be accomplished? We should first specify the less-desirable methods.

While the sale of state assets either to private bidders or in the form of stock shares to the public at large seems desirable, one can make a libertarian objection. Since government possession of those assets originated in one form of usurpation or another, the requirement that they be bought back is unjust. It may be argued that the revenue could be used to benefit the general public (say, through tax cuts), but political incentives tend to work in the other direction. Politicians will see the new revenue as an opportunity to launch new programs that offer benefits to well-organized interest groups and hence it will function as the source of votes and campaign contributions.

Better, then, that state assets be seen as existing in a state of non-ownership (unless they were expropriated from identifiable legitimate owners or their successors in interest) and opened to homesteading in the spirit of John Locke. How might that work? Government elementary and secondary schools could be turned over to the people who work in them or the students’ parents, or both groups, who would be free to decide how to run them—without tax money. A government university could become the property of its students, members of its faculty and staff, or both. Some schools might organize as joint stock companies with tradable shares, while others might become consumer or producer cooperatives. Competition would determine which forms best satisfied consumers and attracted capable producers. This approach seems feasible for a variety of state-owned assets, such as post offices, airports, Amtrak, land, and buildings.

Not every state asset will be amenable to classic Lockean homesteading—interstate highways, for instance. On the other hand, residential and commercial streets could be more easily devolved via some form of homesteading by adjacent property owners, a possibility Walter Block has discussed in The Privatization of Roads and Highways, with its voluminous references to the history of private provision of roads. Also see his “Free Market Transportation: Denationalizing the Roads” (pdf); the streets of Disney World and elsewhere are working demonstrations of private provision. Since the market for roads and highways was not permitted to evolve spontaneously, we cannot know what shape this part of the economy would have taken; hence it would be impossible to create what might have been. Thus devolution to a competitive freed market may not satisfy the criteria of justice perfectly in every respect. But that should not be a barrier removing roads and highways from the inherently inefficient and unjust political realm.

Finally, we should take note of the de facto privatization that has occurred throughout history—particularly in dispute resolution. The Law Merchant was a complex, just, efficient, and spontaneously generated body of commercial law that fostered the revival of world trade after the fall of the Roman Empire. Though later codified and nationalized by states, it is instructive that when merchants of different nationalities and cultures needed an orderly process of conflict resolution, they were able to generate it without the assistance of governments. Similarly, the prevalence of private arbitration and private security (for instance in Disney World and shopping malls) constitutes de facto “socialization” of services typically associated with government.

Radically shrinking the scope of government will require the wholesale shedding of current functions. Thus genuine privatization should play an important role in the liberal project.

Sheldon Richman makes a persuasive case for privatization, accurately noting the calculation/knowledge problem and the incentive problem as critical flaws in the government provision of services and assets that justify their ultimate devolution to the private sector.

Richman also provides a useful theoretical construct for privatization, making a sound argument that just because a public sector activity or asset can be turned over to the private sector in some fashion doesn’t necessarily mean that it gets us closer to a smaller, more constrained government; it may be an unnecessary or illegitimate function in the first place.

That said, theory and practice are different. The style of privatization that Richman argues for, which draws a clear distinction between what is and is not a legitimate government function, is a very worthy “end state” goal. However, if that goal is to be achieved, it is likely to come only through a slow unwinding of the functions and assets of government, with many incremental steps along the way from here to there.

Rather than a sweeping devolution of K–12 schools and road systems, for example, from public to private sector provision, the relatively glacial pace of government reform in practice, and deeply entrenched social and political views on the role of government (warped by a century of government sprawl started in the Progressive Era) will make devolution through privatization a long, piecemeal, and challenging process. To bridge the gap to the end state, the private sector has to demonstrate to the public that it can take on an increasing share of the load from the public sector.

Given the nature of modern politics, scaling back the size and scope of the public sector will not likely come in giant leaps, such as devolving control of K–12 schools to groups of parents and/or teachers with no ongoing state subsidy, as Richman envisions (and which represents a worthy long-term aspiration). Rather, the devolution of public functions to the private sector will (of necessity, in this author’s opinion) continue as it currently does: incrementally, through the slow adaptation and expansion of existing contracting and government procurement models, shifting ever more functions over time from public to private provision.

There are several practical constraints standing in the way of rapid and aggressive devolution of the sort implicitly envisioned in Richman’s essay. First, political and bureaucratic inertia—clinging to the status quo—dominates the public sector culture. Regardless of their intentions upon taking office, many policymakers fall prey to a sense of parochialism that grants political deference to existing public employees, making it difficult to advance even simple outsourcing contracts, much less more substantial shifts toward privatization and devolution. The public sector’s desire to retain some control over the process will, for better or worse, remain until the private sector takes on such a predominant role in the delivery of what society views as “public” services and assets that the public-sector role itself becomes seen as redundant or irrelevant.

Just launching a fairly straightforward procurement for an outsourcing contract today often requires running an internal gauntlet, despite the potential to reduce the costs of service delivery and drive better value for taxpayers’ money. Add in aggressive political campaigns by public employee unions, circus-like public meetings, media pressure, and politicians’ tendencies toward self-preservation, and it becomes even more difficult to take even baby steps toward privatization.

While some libertarians accurately note, as Richman does, that the pursuit of increased efficiency alone (through contracting out, etc.) will not ensure that government will be meaningfully shrunk, that doesn’t mean that efficiency through sensible outsourcing should not be pursued nonetheless, both on moral grounds (e.g., spending taxpayers’ money wisely) and as a necessary part of the evolution toward a more privatized end state.

Richman’s vision—only a bare bones government with the private sector delivering the vast majority of societal needs—is worthy of aspiration, but we will only get there incrementally over time. It is likely to involve devolving functions and assets to the private sector that are determined more, at least initially, by political and pragmatic expediency, than by a shared libertarian vision of a proper (and radically minimized) role for the state. This shift will probably be driven less by principle and more by immediate fiscal pressures, looming debt obligations, and a general fiscal triage as governments confront the “new normal.”

But while imperfect, this evolutionary, incremental process toward privatization is necessary for two key reasons. First, while Richman argues that “privatization devices” like contracting out “merely blur the line between ‘private’ and ‘public’ sector […] and undermine the case for the genuine divestiture of state-held assets,” this author would respectfully submit that the opposite is true. For the longer term societal shift, the private sector will need to prove itself, taking on ever more functions over time and demonstrating to the public at large that it can be trusted with an increasing share of responsibilities generally seen as “public” today.

Governments did not grow to their current size overnight, and it will similarly take generations to prune them back to a size and scope more palatable to many libertarians. The government is not just one large monopoly, but rather the aggregation of many smaller monopolies, each of which has a unique path to privatization.

There’s another practical and more immediate reason to support incremental privatization: unfair government competition with private enterprise. Governments at all levels today routinely spend tax dollars to hire in-house staff to perform a wide array of functions and services—everything from print shops and vehicle maintenance stations to park operations and various healthcare services—that are commercial in nature and, thus, could be subject to private sector competition to drive costs down and drive better quality in service delivery.

For example, let’s look to Sandy Springs, Georgia. It’s an Atlanta suburb incorporated in 2005, and it has gained national attention as a startup “contract city,” with the bulk of non-safety functions provided through private firms via competitive contracting. Long dissatisfied with Fulton County control, citizens overwhelmingly approved incorporation and opted to pursue a largely privatized service delivery model to avoid setting up a traditional—and more costly—“City Hall.” Citizens today are getting far better services for their tax dollars than when under county control, and the example has proven to many observers that there is little that a traditional city can do that cannot be successfully implemented through a privatized delivery model—a critical step toward building the case toward larger devolution.

Some libertarians could argue, correctly, that Sandy Springs doesn’t represent a radical shift to privatization. Yet Sandy Springs also did not have a realistic option to secede from Fulton County to start a “Free State”–type project, so incorporation with a primary reliance on private contracting can be seen as the next best pragmatic approach if the goal is to advance privatization.

For example, planning and zoning—which many libertarians view as illegitimate and market-distorting government functions to begin with—are now just performed by private contractors rather than public agency staff in Sandy Springs. Despite such imperfections, the Sandy Springs model nonetheless offers a powerful (albeit incomplete) example to society showing that what are thought to be “traditional” planning and zoning functions (among many others) are not inherently governmental and can be, as a starting step, handled by private contractors. Given the almost ubiquitous entrenchment of professional urban planning in local governments today, there will be many policy battles to fight—and minds to change—to reverse it. But we shouldn’t wait to take small steps toward privatization in the meantime to help sustain the larger case for radical devolution in the future.

In highways and roads, for example, this means continuing to build on the nascent movement towards the private financing of toll roads and other roadway assets in partnership with governments, as both Reason Foundation and the Cato Institute have advocated. And we should continue to build on the innovations in contracted road maintenance and other functions formerly delivered in-house by government agencies.

Similarly, in education we should not shun things like competitive contracting for non-instructional K–12 services (e.g., food, custodial, and transportation), nor should we reject as half-measures innovations like charter schools in favor of a less realistic (at least today) full devolution of K–12 schools to private actors. The success of the charter school movement is ultimately building the case for the latter, not distracting away from it.

To truly devolve the functions and assets of government, libertarians will need to continue to build the case for more radical privatization, even as they work on pragmatic, incremental reforms. The key, then, is crafting sensible procurements, contract structures, and market-based reforms that allow the public sector to steer while the private sector rows, at least until libertarians can build a broad political and social consensus that the private sector is ready to step into the captain’s chair as well.

As Sheldon Richman explains, what American politicians pass off as “privatization” is usually merely government contracting. Government outsourcing now permeates the military establishment, schools, social service agencies, prisons, and even regulatory enforcement. Hiring private contractors, as Richman points out, does not shrink the size of government or the scope of government services; if anything, it expands it by allowing the government to do ever more. The “privatization” that is merely government outsourcing creates incentives for private firms to lobby for the expansion of the relevant government services—besides the private prison operators lobbying for longer sentences and more criminalization, other entrepreneurs lobby for the state to start new social service programs that the contractors expect to run, and so on. Such contracting has the additional insidious result of skewing tangential markets in the private sector, because the large firms that get lucrative government contracts can then underprice (and eventually eliminate) their competitors in other markets, using the taxpayer dollars they receive as leverage rather than competing fairly.

Worse, outsourcing often is a bad deal for the taxpayers—one-quarter of federal contracts have no competitive bidding (thus, no competitive pricing), and many firms that lock in a government contract can later raise the price dramatically, knowing that the agency involved cannot easily switch to another contractor. The government is certainly less efficient than the private sector; but less efficient still is the government hiring the private sector to do the government’s job. True privatization is the government withdrawing from a field and leaving it entirely to the private sector.

One quibble I have with Richman’s discussion concerns the following:

Government elementary and secondary schools could be turned over to the people who work in them or the students’ parents, or both groups, who would be free to decide how to run them—without tax money. A government university could become the property of its students, members of its faculty and staff, or both. Some schools might organize as joint stock companies with tradable shares, while others might become consumer or producer cooperatives.

Devolving educational institutions to local communities is a fine idea in principle, but Richman seems unaware of the type of corporate entity typically used for such endeavors. Nonprofit or “non-stock” corporations, trusts, institutes, and benefit corporations are the usual model for such community-benefit organizations. Believing in free markets does not necessarily mean believing that all goods and services are best provided by for-profit entities. Certain useful institutions, like schools, parks, and hospitals, originated in the nonprofit (religious) sector; they often work reasonably well as projects of philanthropists or volunteer organizations.

Richman makes a useful contribution to the literature in this area by opening the discussion about how privatization should proceed—that is, the most effective way for the government to turn over its assets or activities to the private sector. We could take these points a few steps further.

Not all government assets and services are the same. It is very different to have the government withdraw from an area where it is competing with private entities, whether charities or for-profits, than to withdraw from areas where it (government) has long been the sole provider. In the former scenario, where the government is competing with existing private-sector alternatives (schools, certain social services for the poor, etc.), it is relatively easy to withdraw from the field and let new or existing private firms occupy and compete for the newly available market share. The latter situation, where the government has traditionally been a sole provider, is more complicated. Hasty privatization in these scenarios can easily result in an anticompetitive, government-created monopoly due to the lack of available competitors, the potential for obtaining an exclusive market by bribing officials, and the cultural tradition by which consumers expect a sole provider and feel uneasy about competitors.

Even where there are nonprofit and for-profit competitors with the government, as in education, we must ask the extent to which hefty government subsidies already skew the area—as with student loan programs. Under our current regime, privatizing all the state universities would merely result in more tax dollars flowing to private universities, due to the prevalence of the government loans and subsidies for education. As tempting as it might then seem to abruptly terminate all Stafford Loan programs, grants, and subsidies, there would be significant risks (i.e., market-skewing consequences) in taking drastic measures all at once. The for-profit schools are arguably more dependent on federal student loan programs (and have a higher default rate among graduates) than are the nonprofit religious schools, who receive a share of their income from private donors.

Thus, we also need to ask ourselves whether we want most of the surviving schools and universities to be religious schools. This was the situation before the government moved into the field and started competing in education. Religious movements pioneered the development of large-scale educational institutions, and the government entered the arena in the nineteenth century to provide a nonsectarian alternative. One very possible scenario after a government divests itself of educational activities would be a return to the sectarian domination of the field. Education is not, it turns out, terribly profitable without a great deal of non-tuition revenue, either from religious tithes or from taxpayer subsidies (student loan and grant programs). Religious groups will continue to have the resources and willpower to prop up their schools and colleges; it is unclear whether secular counterparts will do the same. Some will object that private, non-sectarian schools like Harvard University and the University of Phoenix thrive without religious funding, but they have long depended upon government funding as opposed to tuition that students pay out-of-pocket. It is not clear how many of these secular institutions would remain viable without the government throwing money at education. Perhaps it would be perfectly acceptable to return to sectarian-dominated education, but it is worth discussing beforehand.

Finally, it would be helpful to inquire as to the type of market failures or conditions that led to the government moving into an area in the first place—like public transportation. Was it merely demand by the poor for free stuff? Or were private firms that provided certain services (like mass transit) simply unable to survive, due to the market conditions? There is some historical evidence for this in the case of transportation. While we can imagine private firms providing public services that we now take for granted, there are situations in which an insufficient number of consumers would be willing to pay the prices for a service that the private firm would have to charge in order to stay in business. Government programs are, of course, often nothing more than redistribution—voters demanding free services— and these should be the first candidates for government contraction. These are, perhaps, the majority. Yet the cases where the government stepped in because of a genuine, chronic market failure—a useful service was for some reason unprofitable—we should proceed with more caution. These situations may be good candidates for turning the field over to nonprofits, trusts, volunteer associations, and non-stock institutions.

The key to successful government reform is getting the incentives right. True privatization makes formerly government services responsive to market signals, which are usually helpful. But incentives can be tricky.

People often apply the term “privatization” to programs that aren’t true privatization. Sometimes opponents of privatization use the term to describe failed semi-privatization efforts, such as the reform of British Rail, in order to discredit the idea of real privatization. And sometimes advocates of half-measures use the term to gain support from backers of real privatization.

During the 1980s, the government of New Zealand decided to privatize its ocean fisheries by giving every fishing boat an annual allocation, in pounds of fish, based on their historic fishing rates. After doing so, the government realized it got the incentives wrong, since every boat then had an incentive to fish its allocation whether it was sustainable or not.

So the government bought back the allocations and then gave them out again in terms of percentages of annual harvest. This gave the fishermen incentives not to overharvest, since overfishing would deplete the fishery and reduce the value of their allocations. The result is that New Zealand today has one of the few healthy ocean fisheries in the world.

British Rail’s privatization separated the infrastructure from the trains. One company, called Railtrack, was to operate and maintain the infrastructure and collect fees from other train operating companies for using the rails. This separation proved a disaster, as Railtrack neglected maintenance in order to earn as much as possible in fees from the operating companies. After several fatal accidents, the government took back the infrastructure and now operates and maintains it itself.

Meanwhile, the operating companies bid to provide passenger service on various routes. Since most passenger trains lose money, most of the bids are negative and the government ends up paying the operating companies to run the trains. Buses would be more efficient and probably wouldn’t require subsidies on many routes, but some people insist on the government providing trains. The result is high costs to taxpayers for a system that really isn’t privatized.

In the United States, many transit agencies have similarly contracted out transit service, paying private companies to operate transit buses or trains on selected routes. Denver’s Regional Transit District (RTD), for example, contracts out half of its bus routes. The operators of the half that are contracted out must pay taxes on fuel and property (which RTD is exempt from paying) and pay wages comparable to RTD (one of them is even unionized). Yet the buses that are contracted out cost half as much to operate, per vehicle mile, as buses that RTD runs itself.

Unfortunately, this savings did not translate into reduced costs to taxpayers. Instead, RTD used the savings to start building an expensive light-rail network. The anticipated costs of constructing the full, 140-mile network are about $10 billion, and after beginning construction with the savings from buses RTD convinced voters to raise taxes to complete it.

As this suggests, one problem with contracting out is that the money saved usually ends up being spent on some other, often frivolous, projects. Another problem is that it still relies on central planners to, in this case, determine the routes and frequencies of bus service, which makes it costly for taxpayers who must foot the bill for routes that can’t be supported by the market.

Most American cities forbid private companies from offering transit service, but a few do not. In Atlanta and Miami, private bus owners compete directly with the public transit agencies, often charging a little lower fare, yet managing to make a profit. This suggests that many cities could completely privatize their transit systems and no longer rely on tax subsidies that currently are used to extend service into suburban neighborhoods, whose residents rarely use transit anyway.

Atlantic City, NJ, has no public buses. Instead, the transit system consists of privately owned “jitneys” that offer 24-hour service on a variety of routes. Some routes are free because they are subsidized by local hotels. Other routes cost $2.25 a ride, with discounts for frequent riders and seniors.

A completely privatized system would schedule bus service on heavily used routes and offer non-scheduled, door-to-door shared-ride service in areas where scheduled service could not earn a profit. Such shared rides, such as SuperShuttle, are available in most cities for getting to and from airports, but most operators are not allowed to take people to other destinations.

San Juan, Puerto Rico has a private, non-scheduled bus system called públicos that competes directly with the public bus and train system. Without any subsidies, the públicos carry more people more miles each year than the subsidized buses and trains combined. All of these examples suggest that transit could be privatized.

Few countries in the world have contemplated complete privatization of roads. But many countries offer franchises to private companies to build roads and toll them for 40 or so years, after which time the road reverts to public ownership. At least one highway in the United States, the Dulles Greenway in Virginia, is completely private, and with electronic tolling, there is no technical obstacle to complete privatization of our road network.

Air traffic control is another candidate for privatization. Canada privatized its air traffic control in 1996, putting in the hands of a non-profit company whose primary goal is safety and which installed the latest air traffic control technologies. In 2010, the company won its second award from the International Air Transport Association, recognizing it as one of the world’s leading air traffic control systems. Meanwhile, the United States still uses antiquated air traffic control systems that depend on federal appropriations to continue.

If it is done wrong, as in the initial New Zealand fishery and British Rail programs, privatization can fail to solve resource problems and even make them worse. But if it is done correctly, full privatization can offer many benefits and save taxpayers money that is now being wasted supporting government bureaucracy and inefficiency.

William Lloyd Garrison, the great abolitionist, was an impatient spirit. When told that slavery would end only through baby steps in the right direction, he responded, “Gradualism in theory is perpetuity in practice.”

One need not equate chattel slavery to government services (though the income taxation that pays for those services has been likened to slavery) to see the applicability of Garrison’s maxim. To be sure, the political world moves slowly most of the time, but if those who want government shrunk radically call for gradualism, that is surely what they will get—at an even more glacial pace than if they had espoused abolition.

“Steps in the right direction” are all well and good. But the advocates of any presumed steps are obliged to demonstrate that the direction is indeed right. They usually just assert it without argument. That’s not good enough.

To my claim that “privatization” that falls short of true socialization “merely blur[s] the line between ‘private’ and ‘public’ sector … and undermine[s] the case for the genuine divestiture of state-held assets,” Leonard Gilroy responds that “the opposite is true.” But he makes no argument for this. His example of the “contract city” Sandy Springs, Georgia, does precisely what I claim: blur the line between “private” and “public.” Its residents may get services at a lower cost than they would have otherwise, but that’s not really the point of what I call socialization. As Randal O’Toole notes, it may not even save money, since politicians will always be tempted to spend any savings.

The reason I would not call contract cities a “step in the right direction” is that the lesson likely learned from them is not that government shouldn’t coercively provide services, but rather that government has a wider array of options for coercively providing services than we thought. I submit that is the wrong direction. And what lesson would we expect to learn from contracting out land-use restrictions? We really must keep our eye on the ball!

This discussion reminds me of two other hot topics in public policy. The first is medical marijuana, which some advocates of drug decriminalization believe is a “step in the right direction.” I don’t see it. I think the late Thomas Szasz was right when he said that medical marijuana, far from striking a blow for individual autonomy, merely bolsters “the therapeutic state” by conceding its power to make, in its wisdom, an exception for people it deems sick enough.

Likewise, those who say immigration restrictions should not be abolished until the welfare state is eliminated unwittingly become preservers of the welfare state by saving it from the strains that open borders might produce.

How do we end government services if politicians are able to make them appear more efficient by contracting out? And what happens to the cause when “privatization” is perceived to fail, as it did with the Baltimore schools and elsewhere?

I don’t see how we will make progress in shrinking the state if we back “reforms [that] allow the public sector to steer while the private sector rows.” At best we’ll teach the public that the market is simply an efficient means to the ends chosen by politicians. Though surely unintended, Gilroy’s strategy seems more suited to sowing confusion about what the market process really is.

I completely agree with Dru Stevenson that “believing in free markets does not necessarily mean believing that all goods and services are best provided by for-profit entities.” After all, when I wrote about socializing the schools I said, “Some schools might … become consumer or producer cooperatives.” But I don’t agree that we need to worry that real privatization might “easily result in an anticompetitive government-created monopoly”—not if government genuinely gets out of the way. There’s an easy way to overcome “the cultural tradition by which consumers expect a sole provider and feel uneasy about competitors”: Let competition freely blossom.

I also wouldn’t worry that ending government support for education will give an advantage to religious schools over secular ones. First, that should be no concern of the government, and second, it’s amazing what people can do when left to their own resources. As long as the government money keeps rolling in, the necessity to find alternative financing never arises.

Stevenson seems to think that government has come to dominate certain services like transportation because of market failure. Inspired by Public Choice, I’m more of a skeptic. Government tends to expand because rulers and their “private-sector” cronies benefit. The market-failure excuse is an ex post rationalization. At any rate, those who believe in market failure seem oblivious to the more serious and less remediable problem: government failure.

Sheldon Richman wonders whether efforts to streamline government are really worthwhile if they fall short of complete privatization. This is a legitimate concern, but the question needs to be posed carefully. Powerful political forces oppose privatization, and demands for nothing less than outright privatization may be self-defeating simply because pure privatization in one fell swoop may be impossible.

The real question for those who want to move resources from the public to the private sector is: what is the most likely political path for such a change? It is true, as Richman says, that some halfway measures may prevent eventual privatization, but other measures may open up the path for such an action.

Multiple groups typically benefit from public ownership and control. For example, public ownership of infrastructure typically benefits contractors who build the infrastructure, public employees who operate the infrastructure, politicians who get to pass out infrastructure favors, and private parties who may use the infrastructure at less-than-market cost.

Halfway measures that merely trade off one set of beneficiaries for another will not, as Richman suspects, ease the path to eventual privatization. For example, one proposal before Congress would replace Amtrak with private companies who would bid on individual Amtrak routes. Congress would determine routes, frequencies, and speeds of each train and then offer to pay companies enough of a subsidy to maintain those schedules. The proposal would also require the companies to pay employees as much as Amtrak pays them. Supposedly, the private companies would be more efficient than Amtrak, leading to an eventual savings, but they would also become lobbyists to maintain the subsidies.

Alternatively, it may be possible to plan a path that takes on one of the beneficiaries at a time without immediately threatening the others. Each of the initial steps of the path may fall short of complete privatization, yet the process would not only culminate in full privatization, but each step would make such privatization more likely.

For example, when the automobile was first developed, the main justification for public ownership of roads funded out of gasoline taxes was that the toll collections required for private ownership were both costly and delayed traffic. This same justification was used in the 1950s, when the U.S. Bureau of Public Roads advocated a federal gas tax dedicated to the Interstate Highway System.

Today, electronic tolling has reduced collection costs and eliminated delays, yet opponents of state highway privatization are legion. One of the obstacles is the states themselves, which get billions of dollars a year from the federal government—support that, under some Congressional proposals, would decline with every mile privatized. Outright privatization by the states seems impossible, yet a multi-step process could lead that way.

The first step is to devolve federal transportation funding to the states, but that in itself will be a multi-step process. The House Transportation & Infrastructure Committee is the largest committee in Congressional history because transportation has become one of the greatest sources of pork. Members of Congress who joined the committee to pass out pork are not going to give it up overnight.

Federal transportation dollars fall into three categories: formula funds, given to the states based on population and other factors; competitive grants, supposedly given to the most worthwhile projects but actually distributed largely on political criteria; and earmarks, the most obvious source of pork. Proposals to let states take over gas taxes and eliminate the federal middleman have fallen on deaf ears, but a House dominated by Tea Party Republicans forced the Senate to agree to eliminate earmarks in 2012. The next step is to eliminate the political grants. If all federal funds are formula based, pork-oriented members of Congress will lose interest, and devolution would become a likely next step.

Once federal funds are out of the picture, the states will be forced to rely on their own resources for transportation. All states fund roads out of state fuel taxes, but some play a complicated game of diverting a share of gas taxes to other uses and then spending revenues from sales, income, or other taxes on roads. This enhances legislative power because the groups that get the diverted gas taxes are grateful for the diversions while highway contractors are grateful for the use of general funds on roads.

For example, California diverts more than $2 billion a year of fuel taxes and motor vehicle registration fees to mass transit and another $400 million to general purposes. Meanwhile, the state spends $2.4 billion in general funds on roads, almost exactly the same amount that is diverted. Privatization is politically unrealistic so long as state legislators get the political benefits of playing this money game.

Rather than insist on immediate privatization, tax activists at the state level should demand that state and local governments stop spending general funds on roads, while at the same time they should oppose diversion of highway user fees to non-highway programs. Once these steps are taken, a discussion of privatization becomes more realistic.

It is interesting that Richman quotes an abolitionist to make a case against incrementalism. In the 1820s, emancipation—in which the government would buy and free the slaves—was a viable alternative to abolition that was actively supported by many slaveowners in Virginia and the Carolinas who felt the moral quandary of slavery but believed they could not afford to simply give up their investments. The emancipation movement was killed by a combination of the abolitionists, who argued that slaveowners should not be compensated for the loss of their slaves, and slaveowners in the Deep South, who argued that a government that would take some property (slaves) without compensation would be just as likely to take other property (land) without compensation. The issue was only resolved, of course, after a costly and deadly war.

In short, as Richman says, some halfway measures can themselves become obstacles to true privatization. But insistence on nothing short of complete privatization can itself become an obstacle toward that goal.

I wouldn’t oppose all the measures Randal O’Toole describes in his comment, “Making Incrementalism Work.” Prohibiting a state from diverting fuel tax revenue to general purposes and general revenue to road maintenance and construction would be fine with me. Whether it would lead to privatization of roads is another question. I suspect that special interests would find myriad ways to prevent that, but I see no harm in trying. If gradualism takes the form of tying the politicians’ hands, bring it on! So I guess you can say I don’t oppose all incrementalism, but I never said I did. See my original essay. What I said was that when “steps in the right direction” are proposed, we advocates of stigmergy want to know that the direction is indeed right.

About abolitionism and slavery: As O’Toole notes, slavery might have eventually vanished had the federal government been allowed to buy slaves’ freedom in the 1820s. I’m not sure that qualifies as incrementalism (though it would have been a compromise), since presumably all the slaves might have been freed that way at once. But we should not assume the only alternative to that program was war. (For one thing, Lincoln did not go to war to free the slaves.)

By the way, I would have happily backed one particular incremental step toward ending slavery: the internalization of enforcement costs. As Jeffrey Rogers Hummel emphasizes, “Government interventions that reinforced chattel slavery in the American South … included compulsory slave patrols, fugitive slave laws, laws against educating slaves, and legal disabilities on freed slaves… . These features socialized the enforcement costs of slave labor… .” Removing the subsidies to slaveholders would indeed have been a step in the right direction. And William Lloyd Garrison himself presumably would have approved. As he wrote in The Liberator (1831), “Urge immediate abolition as earnestly as we may, it will alas! be gradual abolition in the end. We have never said that slavery would be overthrown by a single blow; that it ought to be we shall always contend.”

I appreciate Sheldon Richman’s response that he’s not necessarily opposed to incremental privatization and that the onus should be on advocates to make the case that “the direction is indeed right.” That’s certainly a valid point, though it’s unclear from some of his other comments how surmountable the required burden of proof of what’s “right” would actually be in real life.

For example, I find it difficult to understand how the experience of Sandy Springs, Georgia and other contract cities—in which private contractors handle the vast majority of non–safety related government operations—would not be considered a “step in the right direction” toward privatization, and I could not disagree more with Richman’s view that the major lesson learned from them is likely “that government has a wider array of options for coercively providing services than we thought.”

Citizens in Sandy Springs probably see that differently. Indeed, it was a decades-long dissatisfaction with the perceived coercion imposed by Fulton County prior to incorporation (via various taxation and land use decisions, among others) that drove the creation of Sandy Springs in the first place.

But citizens weren’t looking to eliminate coercive local government outright. They simply wanted to see their tax dollars stay in the community, as opposed to being funneled elsewhere by the county. They wanted better infrastructure and more efficient services than the county was providing to them, and they wanted more local control over land use policies, as opposed to more distant county control.

Sandy Springs broke from the county to achieve this, and the contract city model they ultimately adopted was not driven by ideology, but rather a pragmatism that centered on how to best construct a city from the ground up to achieve the maximum bang for the taxpayer buck while avoiding the trappings of the traditional city model (e.g., pensions, union power, bureaucratic sprawl, etc.).

Citizens continue to voice overwhelming satisfaction with their new city, and most recognize that they are getting far more efficient and responsive public service delivery today (on about the same tax dollars) that they were when under the thumb of Fulton County prior to incorporation. They have also been able to avoid the chronic public pension liabilities so common in municipal governments today, as they adopted a 401(k) style defined contribution system from the outset for their small number of truly “public” employees. They are not on the hook for covering the retirement costs for the employees of contracted service providers, which the companies themselves are responsible for.

The success of Sandy Springs prompted citizens in nearby jurisdictions like Johns Creek and Dunwoody to subsequently break from oppressive county control and incorporate themselves, with similar contract city models that rely primarily on outsourced service delivery. In fact, these cities are not far off what Reason Foundation founder Robert Poole envisaged in his seminal privatization book, Cutting Back City Hall, which argued over thirty years ago that most of the municipal functions typically viewed as “public” could in fact be provided via contract, with a lean government that concentrated primarily on contract administration, not “traditional” in-house service delivery.

In reality, the major lesson learned from Sandy Springs for most observers that I’m familiar with—and one that goes a long way toward changing societal mental maps on the larger role of government—is that there’s very little that local governments do day-to-day that cannot be handled by the private sector. This is no small feat for a society that is generally unaware that the private sector can take on a far bigger role than just filling potholes, picking up garbage, and handling other discrete tasks. Though libertarian ideas appear to be making significant inroads into the larger society, a skeptical public will nonetheless demand proof that ideas like privatization work on a smaller scale before they become willing to tackle larger efforts. Hence, incremental reforms along the lines of Sandy Springs offer a proof of concept that work to break down societal conceptions of “traditional” government, making it easier down the line to advocate for free markets and liberty in bolder ways.

And we should never forget a critical, but often overlooked, point: It is far easier politically for policymakers to cancel or limit the scope of a contract than it is to eliminate or downsize the scope of a government bureaucracy. If Sandy Springs wanted to begin eliminating particular city services, all they would need to do is remove them from the contract terms the next time they rebid it. Sure, a private contractor could try to exert political influence to preserve the status quo, just as public employee unions do on the other side. But once a function has shifted out of the in-house bureaucracy to a private contract delivery model, the tether begins to thin considerably and is far more easily cut relative to the herculean efforts typically needed to shutter a government agency.

To dismiss Sandy Springs and other contract cities as merely “blur[ring] the line” between the public and private sectors seems shortsighted to this author. Just because we may share a long-term goal of radically downsizing government does not mean we should not embrace smaller steps in the interim that measurably improve outcomes for taxpayers today.

I have no doubt that the residents of Sandy Springs are enjoying the benefits that Leonard Gilroy describes, and I never said that “we should not embrace smaller steps in the interim that measurably improve outcomes for taxpayers today.”

Nevertheless, we’ve taken our eye off the ball. My opening essay was about truly competitive privatization—“socialization”—and radically shrinking government, not how to make the residents’ money go further, as worthwhile as that is. By the standard I set out, Sandy Springs is not relevant unless it can be judged as steps that can reasonably be expected to lead to competitive markets in what are now government services. (It might be a useful lesson in secession—which would please me no end! Secession and ever-smaller political jurisdictions are to be desired: They make voting with one’s feet cheaper and more feasible.)

Mr. Gilroy writes, “The major lesson learned from Sandy Springs for most observers that I’m familiar with—and one that goes a long way toward changing societal mental maps on the larger role of government—is that there’s very little that local governments do day-to-day that cannot be handled by the private sector.” That is incomplete. It should have read: There’s very little that local governments do day-to-day that cannot be handled by the private sector under exclusive contract to a coercive monopoly. That, I submit, is an essentially different lesson. Mr. Gilroy’s version may have some value, but it is not the same as teaching folks that government is unneeded to provide public services. I still do not see how government contracting can provide proof to “a skeptical public … that ideas like privatization work on a smaller scale” without equivocating over the word privatization and blurring the line between private and public sector.

Monopoly—not who pays the workers—is the fundamental problem. Contracting out leaves it unaddressed.

Some free-market advocates argue that any action short of complete privatization of publicly owned resources and infrastructure is a betrayal of free-market principles. My previous post agreed that some “half-way” measures may actually foreclose complete privatization because they create special interest groups that will lobby against such action. At the same time, I argued that some intermediate steps towards privatization may be necessary for complete privatization to ever take place.

In this post, I want to make the additional case that there may be some halfway measures that fall short of complete privatization without foreclosing it, and that—even if they don’t accelerate true privatization—these measures may be useful second-best solutions when there is strong opposition to outright privatization. Two such halfway measures are certain kinds of public-private partnerships and user-fee driven public agencies.

Public-private partnerships are a common if not the most common way of building new highways in Europe. The term has also been used to describe a way of financing and operating transit lines in the United States. But these are two very different institutions. For simplicity, I’ll call them the “lease model” and the “franchise model,” though these are not terms commonly used in the transport industry.

Under the lease model, the government offers to allow private companies to build a highway or other transport facility. Companies may bid on the right to build in a particular location. The winning bidder puts up private capital to pay for the facility and has the right to toll it for several decades. At the end of that time, the facility reverts to public ownership (which would probably put operations and maintenance out to bid to the same or another company). Taxpayers put up no money (though they may provide right of way), and the private investors accept all the risk that tolls may not earn them a profit.

The State of Indiana adopted this model when it offered to lease the Indiana Tollway. Several consortia of companies bid on that lease and the winner ended up paying the state $3.8 billion for a 75-year lease to collect tolls while operating, maintaining, and where demand exists, improving the tollway.

Under the franchise model, a transit agency asks private companies to bid on building and/or operating a transit line. The bids are invariably negative: the agency has to pay the company to provide the transit service. While American transit agencies have a proven track record of using franchises to save money on bus operations, in the case of new construction, the savings may be small or non-existent.

When Denver’s Regional Transit District (RTD) adopted this model to build three new rail lines, the purpose wasn’t to save money but to subvert voter intent. Voters had agreed to increase sales taxes to build new rail lines but put a limit on the amount of debt RTD could incur at one time. When (predictably) costs ballooned after the measure passed, RTD contracted out construction so that the private companies could borrow the money—money that RTD was still obligated to pay out of tax revenues but that was not on RTD’s books as debt.

Obviously, the franchise model is not efficient and even encourages waste. But I don’t see any serious free-market problems with the lease model, particularly if the contracts place as few restrictions as possible on both parties.

Another model that falls short of complete privatization yet is worth considering is that of a government agency funded exclusively out of user fees. For example, Florida and Texas allow counties to create toll road authorities. The authorities can sell bonds, build roads, and repay the bonds out of tolls. They cannot collect taxes and, if they default on a bond (which rarely if ever happens), county taxpayers are not obligated to cover the loss.

Another example is state land agencies, which manage their lands in trust for public schools or other programs. These agencies charge user fees, some of which go to agency operations and some of which go to the beneficiaries. They can even sell their land or other nonrenewable resources, in most cases putting the revenues from nonrenewable resource sales into a permanent fund for the beneficiaries. The fact that the courts treat these agencies as fiduciary trustees makes them legally obligated to operate at a profit.

When an environmental group in Washington urged the state land agency to reduce timber cutting to protect a rare species of wildlife, the agency agreed to do so—provided the environmentalists maintained returns to the beneficiaries by outbidding timber companies for the timber. The environmental group quickly raised nearly $20 million to do so.

In short, county toll agencies and state land trusts largely face the same incentives as private businesses. Ardent free-marketeers may argue that such agencies cannot be truly be insulated from politics, but so far both county toll agencies, at least, have been no more influenced by politics than private service companies. On the other hand, there are plenty of examples of private owners who are either subsidized and/or subject to a variety of political and regulatory pressures.

Neither the lease model nor the user-fee model forecloses complete privatization. Indeed, they may actually hasten privatization once these models prove that user fees are sufficient to operate well-managed infrastructure and resources.

In his recent response article, Sheldon Richman takes exception to my claim that government contracting with private firms to deliver public services offers a way to demonstrate to the public that ideas like privatization work on a smaller scale. He noted that this won’t happen “without equivocating over the word privatization and blurring the line between private and public sector.”

But for many people, a blurring of the lines is exactly what needs to happen to start shifting their belief systems on the role and necessity of government in different aspects of our lives. Richman notes the need to “[teach] folks that government is unneeded to provide public services,” and one powerful way to start teaching them is to have a private contractor show up to do a job they’re used to seeing a government employee perform.

For example, my town contracts out its entire fire department to the company Rural/Metro, a pioneer in privatized fire services. Their trucks are shiny, red, and full of water, just like a “traditional” fire department’s. Their firemen train just like their municipal counterparts do in neighboring jurisdictions. They respond to fire and EMS calls just like the government-run systems do.

The main differences I’ve discerned are that: (1) their logo—which otherwise looks much like other fire department logos—notes the name of the company underneath the name of the town, and (2) workers are covered under a private sector 401(k) plan, so our town is not on the hook for a massive future pension payout. Neither of these differences is relevant from a service delivery standpoint.

But most people who move to my town are not used to private fire services and have never heard of private fire contractors, so they tend to react as if it’s a foreign concept. Many folks have grown up to believe that firefighting and emergency medical services are inherently governmental functions, as if only government employees can be trusted to fight house fires and respond to EMS calls. Imagine their surprise and shock, then, upon discovering that their new fire department is run by a for-profit corporation. And imagine how powerful it is to realize over time that not only does this model exist, but it also works: what once may have seemed so “public” just doesn’t seem so “public” anymore.

I believe that before libertarians will be able to successfully convince the general public that fire services represent a “coercive monopoly” and something perfectly ripe for privatization, we’ll first need to convince them that there exists a legitimate private sector role in firefighting in the first place, a task that will be made difficult by the inevitable protestations made by public employee unions.

Contracting out helps to build the case for a private role. For many types of government services, more radical types of privatization are not likely to materialize tomorrow unless citizens can see that there is an existing batch of private providers demonstrating competency in delivering comparable services today. For example, if you did not have private fire services companies in existence today providing high-quality, cost-effective services to governments under contract, then most people would likely dismiss claims that the private sector can and should perform these services in the future. (They might ask, “What private sector?”) Fully private garbage collection exists today in large part due to the growth of contracted garbage collection over many decades, and now people tend to understand that this is a real business with viable private providers. This same market development and maturation process also needs to happen in areas where government almost entirely self-provides, like fire protection.

Hence, contracts along the lines of those described above can in my opinion play a critical part in deconstructing antiquated mental maps about what truly are “inherently governmental” functions and what aren’t. Like it or not, this will be required for many services traditionally conceived of as “public” in nature before true privatization can take place. Using my above example, if one can demonstrate to the uninitiated that cities can successfully contract out fire and EMS services to a for-profit corporation—even imperfectly, as a monopoly enterprise—then it’s not much further of a leap to conceive of even more aggressive privatization models and structures that could potentially eliminate the monopoly outright.