James Surowiecki's column in the Aug. 13, 2007 New Yorker has something useful to say about the politics of privatization, in part because it's about a case that has already developed into full-scale scandal: student loans:

For decades, student-loan companies have had one of the cushiest businesses in America. [ . . . ] The federal government, for instance, guarantees the so-called Stafford loans that college students get: if a student defaults, the government will pay off almost the entire loan. On top of that, the government hands out billions of dollars in subsidies to lenders every year. In effect, lenders get a guaranteed return with very little risk.

This convoluted process is good at making student-loan companies rich -- Sallie Mae, the biggest issuer of student loans, earned $1.3 billion last year, with a return on equity that dwarfs most other companies'. But it's not very good at getting government money to students cheaply and efficiently. President Bush's 2007 budget shows, for instance, that it's four times as expensive for the government to subsidize and guarantee private loans as for it to issue those loans itself. In other words, the current system is not just corrupt. It's also inefficient. So why are we stuck with it?

In part, it's ideology, and the dominance of what you might call the privatization mystique -- the idea tha tanything the government can do, the private sector can do better. Often, this makes sense: the free market is more likely to come up with efficient ways of creating and distributing products and services than the government is. But the student-loan market isn't a free market in any meaningful sense of the term, because the government effectively determines prices, insures against losses, and subsidizes volume. In this environment, most of the competition among private companies is really just squabbling over how to split the spoils. Economists call this behavior -- when a company seeks to manipulate economic conditions rather than actually create value -- "rent-seeking." It's common in areas where the fetish for privatization has taken hold, such as the outsourcing of homeland security to private contractors and the boom in private Medicare insurance. (The private insurers are less efficient than Medicare and receive billions in subsidies from the government.) Outsourcing tasks to private companies is spposed to let government reap the benefits of the free market. But sometimes it just ends up uniting the worst of government and the worst of the private sector into one expensive mess.

Actually, I think that's a good deal more often than sometimes. In anything it chooses to do, government has four advantages over the private sector: 1) it can borrow money at lower cost because there's no risk that it will default; 2) it can insure itself less expensively against other risks; 3) it can offset income, granting itself subsidies from general tax income, in some cases completely avoiding direct charges (and therefore marketing and collection costs); 4) it can motivate workers by appealing to their sense of public duty. Given these advantages, you have to wonder how the private sector can ever compete. The main answer is politics, on both sides of the fence. Politics works to reduce the efficiency of government, to increase its costs, to de-motivate its workers. Politics lets the private sector reduce its labor costs and run its labor force more tightly. And politics puffs up ideologies that obscure what more often than not is a direct transfer of subsidies from government to the private sector. If you get rid of the political handicapping of government, in industries where there are well-defined consensus demands, government will easily out-produce the private sector, and we'll (almost) all be better off for that.

The private sector still has a role in providing goods and services where there is insufficient political will to make sure that they are provided most efficiently. But where there is a public demand -- student loans is one such case; health care is another -- the real question is how do we make public financing and orientation work. Paying rents to private interests doesn't work, no matter how lucrative it is to politically connected beneficiaries.