Friday, March 21, 2014
Good things & bad things: competition and privatization
When I talk to students about anti-competitive behavior, I like to start with a stark illustration. There are lots of things firms might do in order to earn profit, and these things might be good or bad, broadly speaking. Profiting by producing a higher-quality product, or by producing at a lower cost than competitors: good things. Sending Fat Tony to threaten your competitors with violent death if they do not bow out of the industry: bad thing. The good things are profitable to firms because of the benefit to consumers that is generated, whereas the bad things benefit firms at the expense of everyone else. The purpose of competition policy is to preserve firms' incentives to do the good things and preclude them from doing the bad things. There are lots of complications, of course, making it difficult to sort out whether a particular strategy (e.g. a merger between two firms) is actually a good or a bad thing. But the basic idea is that when firms are doing their best to earn profit within the constraints of well-designed laws, their activities enhance consumer welfare.
The same question arises in the context of privatization. If some government-run entity, like a prison, is put into private hands, there is then someone who stands to profit if he does a good job of running the now-private entity. The hope is that a firm running a prison for profit will do good things for the sake of profits, like reducing costs, insofar as this can be done while still fulfilling the purpose of having a prison. I would guess that prisons are a target for privatization for two reasons: in addition to the general phenomenon of government agencies doing things less efficiently than private firms, incarcerating convicts is probably a relatively costly activity. Thus there would be a lot of benefit to be had by creating strong incentives for running prisons better.
A spectacularly bad thing that resulted from such privatization was the Kids for Cash scandal, wherein two judges were sending kids to juvenile detention in exchange for kickbacks from a private detention center. This is pretty old news by now, but it raises a concern that has continued to circulate: that privatization of prisons gives someone incentive to maintain a large population of prisoners. The firm in charge of the prison has to be paid for its activity, and it's easy to see why the payment would be linked to the number of prisoners. So if the firm manages to turn a profit per prisoner, then more prisoners make the firm better off, leading the firm to do whatever it can do to get more people incarcerated.
Most references I have seen to this issue, from news articles to commentary to Facebook posts, talk about this and other problems as if they are obvious indications that privatization of prisons is a terrible idea and shouldn't be happening at all, without acknowledging (or perhaps even understanding) what privatization can accomplish. It's always a danger to oversimplify an important policy issue, and in this case I see that happening on two fronts: treating the problems with privatization as insurmountable and disregarding the benefits of privatization. It's especially damaging when someone manages to state an oversimplified view in a clever way.
I'm not trying to claim that there is a knock-down argument in favor of privatization in this or any other specific context, but I would claim that any argument for or against privatization has to consider a number of factors on both sides. Otherwise it's not an argument that deserves to be taken seriously. The basic question is whether it is possible to realize at least some of the benefits of privatization while precluding most or all of the problems it could cause. Obviously it is a conflict for a judge to have any financial interest in a prison, and it's possible, at least in principle, to discover and take action on these conflicts if they arise. In practical terms, I can only guess what difficulties there might be, but this is the kind of question we need to ask. Maybe there are problems that are particularly difficult to deal with because of, say, political constraints or informational asymmetries, and if so, then that's where the argument lies. There would have to be some things we could do to mitigate negative effects. That the proprietor of a for-profit prison will always find ways to maintain a large incarcerated population, by encouraging the prosecution or even the commission of crimes, seems a bit of a stretch to me.
There are lots of examples of policies that have had perverse consequences, and this often comes down to poor implementation of the policy. If one specific law had some intended purpose but led to unintended effects, this does not necessarily mean that we should not have that sort of law at all. If we've tried privatization but it hasn't worked so well, we don't have to abandon the idea entirely. Put another way, problems caused by privatization do not automatically indicate that we should have less privatization, just as the benefits of privatization do not automatically imply that we should have more of it. The question is how to do it right. A comparison: in the wake of the financial crisis, it is simplistic to say that the problems we have experienced indicate a need for more regulation, or for less. Clearly financial regulation in the US is not accomplishing everything that we would like. We need to do something different, and the best course is likely to be some combination of creation of new laws and abandonment of old laws.
Here is another example that provides a context for asking the same kinds of questions. Most municipalities in the US require residences to pay for trash-collection services, which are provided by a government agency. An alternative to the typical scheme would be for the government to charge residents for the service but to subcontract the actual trash collection to a private firm. Another alternative would be for the government to play no role at all, leaving waste disposal entirely to the market. In the latter case, I would imagine that private disposal firms would arise, charging a fee for collection from residences, with the only real difference from the status quo being that participation is voluntary. A potential problem with this is that consumers could avoid the cost of disposal by dumping their trash in their neighbors' trash cans, or in the street or the park or the river. I'm sure that some consumers would dispose of their trash responsibly and that others would not, but I couldn't speculate about the relative proportions. If something like this is enough of a problem, then providing collection services to everyone and levying a tax to pay for it starts to look like a good idea.
A similar problem could arise if a private subcontractor does the collection: the subcontractor has an interest in disposing of the trash in the cheapest possible way, which might be to dump it in the street or the park or the river. However, it is much easier to monitor the activities of the subcontractor and enforce proper disposal than to try to do this for individual consumers. The important questions are whether we have the ability to create the right incentives for the private firm and whether it is less costly to do it this way than to have the government perform the trash collection itself (which depends on how hard it is to create the right incentives for government workers, among other things). I suspect that privatizing the service but funding it through taxation would be the best way to go, although I acknowledge that there may be considerations that I am missing.
We can't very well expect the free market to provide prisons, notwithstanding the anarcho-capitalist claim to the contrary. For-profit prisons are analogous to the publicly funded trash-collection subcontractor. The same kinds of pros and cons exist, although the specifics of these may be vastly different, and the conclusion about the most desired policy could very well be different. Perhaps the prison system is especially susceptible to corruption when it is under private rather than public auspices. Perhaps placing appropriate constraints on the stakeholders of for-profit prisons is too difficult or too costly. I.e. maybe privatization is undesirable because the bad things will inevitably outweigh the good things. But it is far from obvious that this is so.