Semi-Connected Thoughts on Government Contracting and the Essential Nature of Government
by James Hanley
Earlier, here, on a post by Tod, there was a discussion about contracting out government services, with a particular focus on policing. It’s something I’ve thought about quite a bit, both because of my academic interest in political economy and public choice theory, but also because as a marginal libertarian I think government is both basically necessary and too heavily relied upon, so I frequently think about that troublesome boundary between the public and the private sector. Or rather, the indefinite in-between/overlapping area, since I become more and more convinced there is not clear discernible boundary; the two do not necessarily operate in different spheres of our lives.* Here are some generalized thoughts on the subject.
These thoughts aren’t going to change anyone’s mind about anything, but for those who are trying to puzzle out these issues, perhaps they’ll be some help to your efforts to put the pieces into place (wherever you end up landing on the issues).
To begin, it’s useful to make certain distinctions that are often glossed over. These distinctions are not generally made by opinion leaders, nor even by experts. But they should be. They’re fairly rarefied, but only by being rarely used—they’re not difficult to understand.
1. Production v. Provision
I learned this one from Nobel Prize winner Elinor Ostrom, who was the world’s leading scholar on commons problems (there, I’ve appealed to authority, so this must be true!). In her framework, to provide for something means to make sure that it’s available, by some means or other, while to produce something means to actually bring it into existence. For example, the government provides for the building of roads, but in most cases it is a private corporation that actually produces the roads. It’s not that “provision” necessarily has such a restricted meaning, but that–once again–we need terms to help us distinguish between the two different concepts.
2. Privatization vs. Contracting Out
I ran into this distinction recently,** and smacked myself on the forehead for not having thought of it myself. We normally talk about privatizing, whether we’re for or against the practice. But a truly privatized activity would be wholly moved from government hands to private hands. That’s rare. Most often we just contract out a government activity, paying a private entity with public funds to do something for the public. It’s not so much that “privatize” is an inherently inaccurate description; it’s just that we need to find words that distinguish between two different things. For example, consider the difference between a school district contracting out school bus services and the district getting out of the business of student transportation altogether, leaving it solely to private entities to provide and produce (whether that be some kind of firm (whether for or not-for profit) or just the parents themselves).
Full privatization is fundamentally a matter of values. It is based on the question of “what is it legitimate for government to do?” If you don’t think government should be the provider of something, then you’d want to privatize it, not just contract it out.
Contracting out is more about efficiency (although it’s not wholly removed from questions of values). If you think government ought to provide something, then you can move onto the second-order question of how that something ought to be produced. Presumably at this stage the concern is about how efficiently that something can be provided. Note that efficiency does not mean “on the cheap.” The government sets the standards that must be attained, and the producer, whether government agency or private firm, is responsible for achieving the standards. Competition spurs contract bidders to meet the given standards at least cost. And it allows for public agencies to bid, as well as private agencies. Here, for example, is a snippet from a brief report about Phoenix that demonstrates all three points: government setting the standards; cost savings while meeting those standards; and competitive bidding that includes both public and private entities.
[T]the city of Phoenix between 1979 and 1994 awarded 56 contracts in 13 municipal services by this process, with 34 contracts going to private contractors and 22 remaining with the city agencies, saving $30 million…
[Phoenix] initiated competition for residential solid waste collection in 1979… divid[ing] the city into three sectors and put[ting] each sector out to bid on a rotating schedule. Private firms bid against the in-house unit; the total number of bidders ranged from three to six in the six competitions that have been held. Private firms can serve no more than one of the three sectors; this statutory provision has the effect of institutionalizing and perpetuating public vs. private competition and preventing complacency and collusion by either sector. As a result, over the next 15 years the real, inflation-adjusted cost of service declined by 38 percent for all garbage collected in the city by the contractor and the city agency.
The contract specifies a fixed rate per house per month and penalties for missed collections, spillage, deviation from established routes, and other failures. …
The report is not entirely uncritical of the structure of Phoenix’s bidding process, but is, overall, favorable. Now, assuming we all want government to ensure our trash is picked up (or at least to ensure that our neighbors’ trash is picked up, so they don’t dump it on the roadside), do we really care if it’s done by an in-house government agency contractor or by a private provider? If government ensures it happens (at a desired level of quality), and ensures that we’re paying the lowest price for achieving that level of quality, then we’re golden, regardless of who the garbagemen’s employer of record is. I’m sure some people do care, and it wouldn’t be hard to find both those who want it to be the agency and those who want it to be the firm. But either one seems very illogical to me, as they both put the value judgment in the wrong place, on the secondary act of production rather than the primary act of provision. To be clear, I’m taking a position against both those who simply reject government provision of anything, and those who simply reject private production of anything government provides.
[Note: There may be a couple of objections here, about monitoring for compliance with standards and about how garbage pickup is different than policing. Please be patient with me, as I promise to address both issues.]
3. The Problem with Contracting Is Not the Profit Motive
The profit motive is often condemned, but really it is no more than a specific case of the generalized desire for gain (however a particular individual defines what gain means to them). Government employees, being human just like non-government employees, are not without the desire for gain. True monetary profit motive, barring actual embezzlement, is generally not available to them. But that only means one avenue for pursuit of gain is closed, and so they—being human—are likely to pursue others. The literature on bureaucracy (mind-numbingly dull to the average person—it takes a special type of geek to enjoy it) is filled with efforts to understand the ways in which public employees seek gains.
Perhaps the most obvious way is through shirking. For example, the union for Indiana’s highway workers negotiated a contract that required all “road work ahead” signs to be manned by two employees in a truck, whose only duties consisted of driving to the work sight, pulling off the road, making sure the sign was on, then just hanging out all day. I’m not union bashing here, just providing one example of how public employees can profit by shirking. (Nor, I want to emphasize, is shirking relegated solely to the public sector—private sector employees, like their public sector counterparts, are also human. And, hey, were I a highway worker, I’d have voted for that contract!)
Another gains-seeking technique of public employees, so it has been argued, is through the pursuit of larger budgets that allow for extra (unnecessary) trips to conferences, lavish office furnishings and the like. This undoubtedly occurs when the bureaucracy can manage it, but my observation is that only rarely do budgets allow for that, and I think most students of bureaucracy have moved beyond thinking of that as the most significant means of gains-seeking by public employees.
A third method of public-sector gains seeking that has been proposed is the pursuit of autonomy. Some argue that a standard bureaucratic response to limited budgets is to seek increased autonomy over their own activities, which allows them to better control the ways that limited budget is used and enable them to be less politically accountable for the specific policies they pursue. Consider yourself–assuming an increase in pay is not in the offering, would you appreciate more autonomy? (In fact as a private sector employee that’s one of the things I most value in my job, and the career-related issue that most depresses me is the recognition that the state and national politics of education, as well as–perhaps–institutional politics, are going to increasingly eat away at that autonomy.)
[Without going into great detail, I will also note that the profit-motive, or at least gains-motive very nearly aligned to it, may be one of the original causes of government, as suggested by Mancur Olson’s “Stationary Bandit” model and Diamond’s agricultural surplus/kleptocracy argument.]
4. The Benefit of Contracting Is Not the Profit Motive
See above. Everyone seeks gains, and the profit motive is just one form of gain, so the profit motive can’t be the actual cause of any benefits from contracting. Besides, as the example of Phoenix shows, sometimes the non-profit government agencies win the bid. The benefit really arises from the competition between would-be service providers. Monopolies are, all else being equal, worse for the general public than competitive organizations. Government monopolies in a democracy are presumably less bad than private monopolies, because even weak democratic accountability is more accountability than a private (unregulated) monopoly will face. But breaking up monopolies by forcing entities to compete against each other is generally (although not quite inevitably) an even better approach.
5. Monitoring Matters
Obviously none of you disagree about the necessity of monitoring, but it’s important to make clear that I am in agreement with everyone here. Some contracting out enthusiasts seem to forget this, while some anti-contracting out enthusiasts seem to assume it is impossibly difficult. But really it’s all about the institutional capacity of the government, and that’s a political choice (a government that saves bundles by contracting out can afford to transfer a sufficient chunk of that to compliance monitoring and still save the public money).
Does “institutional capacity” sound like a “Nirvana fallacy” argument? It’s not—most functional cities have been doing it for a long time, for various different contracted out services. A small number fail to do it, whether through ignorance or corruption. Both are fixable, although not easily, and neither gives much hope for trusting the government to do anything well—whether contracting out or doing in-house production.
Additionally, forcing competition for government services can increase oversight by adding a layer to it. In addition to the general political (democratic) oversight of public agencies that produce government-provided goods/services, we add legal (contract compliance) oversight. Politicians love to have cover-your-ass justifications for politically difficult decisions, and the law often provides a good one (“I don’t like it either, Ms. Constituent, but my hands are tied”).
6.Not Everything Is Contractable
None of this implies that everything government provides is suitable for contracting out. In a Public Administration Review article David Van Slyke makes a good argument that contracting out for social services does not work well. I’m persuaded. Or at least I’m persuaded that the specific problems he identified are a serious constraint on effective and efficient contracting out. If those conditions can be changed, then not necessarily so (and the author does not say it’s necessarily so, either), but it may be that those problems are particularly acute and difficult to solve in the case of producing social services, and they may be best left (at least as a general rule) in the hands of a public agency.
The value of identifying specific problems, as Van Slyke does, is that they are generalizable to other situations. For example he notes a lack of potential service providers, so there is no real competition in the bid process. This is an issue I took note of a few years ago when the city of Santa Clarita, California pulled its library from the County of Los Angeles Library system and contracted out library operation to a private firm. Santa Clarita is a pretty conservative community, and it has been very aggressive in running a very lean government and contracting out all the services it can (it also contracts out its police work, but to the L.A. County police, not a private firm). They’ve actually had great success with this approach, but there was only one bidder for the library contract, Library Systems & Services, Inc. (LSSI). (it’s a very new and controversial approach, so there are few providers in the game. That may not be in itself sufficient reason not to move forward with the contracting, but it provides more than sufficient reason to be very wary, and for the city to be extra careful. (Instead, Santa Clarita just wrote their bid requirements to suit LSSI. Sigh. I haven’t checked in on the issue with my Santa Clarita contacts since the contract became effective, though, so I can’t comment on how well/poorly it’s worked out.)
Another example (one I’m dealing with right now) is funding for parks and recreation activities. My city is so cash-strapped that they’re faced with having to cut out parks and rec funding.*** In response to the question of why they don’t just partner with others or contract that service out the City Manager noted that a) they already were partnering to a significant extent, both with the local YMCA and with a local church that runs lots of recreational activities; and b) even though they’d increased the direct fees for rec activities to the point where they had fewer people signing up (and poorer kids being unable to afford it), those fees were still only covering about half the cost of the activities. Those local non-profits can’t afford to foot any more of the bill, and there’s no real source of cost-savings that would lead to bidders for a contract. In this case, if government doesn’t produce the service, most of it won’t get produced (or so at least we believe–we may be forced to do the experiment for real, and see if our hypothesis is wrong).
I’ve gone on at length here for two reasons. First, I don’t want to give anybody legitimate reason for thinking that as a libertarian I’m gung-ho about privatizing every damn thing government does. Second, case-by-case analyses based on specific identifiable principles are where our general policy debates on the issue should be, because for “normal” cases, that’s the best way to distinguish cases that are suitable for contracting out and those which aren’t (“normal” is explained below).
7. Should the Monopoly on the Legitimate Use of Force Be Contracted Out?
This last section is different than those above. I’m not drawing from literature I know well here, so I’m being much more speculative.
Over on Tod’s post there was a discussion about contracting out for police services. Despite the vociferous character of the argument at points, there was mostly agreement that this isn’t a good idea. I’m in agreement with that, but I’m not entirely sure why yet. I base it on the historical examples I’ve observed, but I don’t yet have a solid (to me) theoretical explanation for those historical examples. In the absence of a theoretical explanation about why “X” necessarily occurs when condition “Y” is present, we should always entertain the possibility that there’s an alternative, confounding, variable that we haven’t yet discovered, and that may be controllable, if we develop the required knowledge for how to do so.
Policing is an example of my exception to “normal” cases; those areas where government’s use of force is part of the specific activity, an inevitable part of the active production of the service, not just its provision. In addition to policing this include such things as military engagement, detention (jails/prisons) and tax collection. Maybe there are others, but those are the ones that jump out at me. Everything else still relies on government’s coercive power (as tax collector) for provision, but doesn’t ask the private contractor to actually employ that power in producing the service. E.g., road builders don’t do the taxing that pays for the roads.
But exactly what is it that makes that private employment of government’s force unsuitable for contracting out? This ties in directly, I think, to my continuing thought about how we can distinguish between government and non-government entities in that indeterminate region I mentioned at the beginning. And, dammit, I just have to admit, with no prevarication of any sort, that I just don’t have that worked out yet. I mean, not even close. I’m still stuck at Weber’s starting point, and I haven’t had time to put in the effort to really work at the question the way I would like. But simply put, I’m dubious there’s a very simple explanation for why—or whether—some inherently governmental functions don’t contract out well, or what distinguishes government-proper from quasi-governments. I think it’s an answerable question, mind, just not a simple one.
Fortunately it’s not a pressing policy concern from my perspective. I’m quite happy to leave those particular tasks in the hands of public employees, given our current state of knowledge. But to me it’s one of the most intriguing questions in political theory and policy analysis, and I don’t think many people are asking it, bar a few folks niggling around the edges of it in their studies of self-governing institutions. Most people, including most political scientists, just assume government and study its various manifestations, rather than tackling that root-level question.
And, no, political philosophers whose work justifies government aren’t really asking that question, either. Perhaps the one who comes closest is Nozick, in Anarchy, State and Utopia, where he explains how a private protective association can morph into a state. I didn’t enjoy the book when I read it almost 20 years ago (I didn’t become a libertarian though either Nozik or Rand, despite having read significant parts of both of them prior to my shift, and I suppose that actually explains a lot about me), but I may have to skim that part again, to see if he at least provides some insights I should consider.
And I often find on questions like these that comments and questions from the League’s well-above-average-intelligence readership often provides some insights for me to chew on. (Yes, the League, including its accursed and damned-to-statist-hell liberals, have influenced my thinking.) So consider this post a two-parter–one in which I am trying to clarify some specific issues for the purpose of stimulating clearer discussion of contracting out government services, and one in which I am asking you to ruminate with me about the very nature of what makes government government.
* Max Weber, in “Politics as a Vocation,” is the source for this claim. “Sociologically, the state cannot be defined in terms of its ends. There is scarcely any task that some political association has not taken in hand, and there is no task that one could say has always been exclusive and peculiar to those associations which are designated as political ones: today the state, or historically, those associations which have been the predecessors of the modern state.”
**I forget how I stumbled across it, but it’s in this Cato article by Bruce Benson. There are things in the article I disagree with, so I’m not really praising the article itself, just the distinction. Yglesias has also used it. I’m not a big fan of his, but I’ll give credit where credit is due.
***It’s amazing how the threat of cutting parks and rec programming can unite a community. The old people (cutting the senior center!), the folks with young kids, and the young kids themselves were all giving the City Council the wicked stink eye.