Faux Market News
I’ve said it before, but “privatization” of government services as such is bullshit. Or can be. And can be much worse than straightforward government provision. Yglesias explained why last month. Basically, if it’s a situation where the service has many buyers and the new private entity will have to sell to them, presume in favor of privatization. But if the new arrangement would have a private corporation selling to one buyer, the government, presume that the “prvatized” situation would suck much, much worse than good old fashioned bureaucracy.
Comment by Thoreau —
March 5, 2007 @ 6:07 am
Not to mention that when a “private contractor” runs a government program it sometimes results in double the management layers, since the private company has its people and the government has its own people watching them.
All these watchers watching with no transparency gains but lots of efficiency losses.
It’s not always that way, but it sure seems to work out that way a lot.
Comment by Marc —
March 5, 2007 @ 6:40 am
And even the best-intentioned privatization can be irretrievably hosed by inept design. See, for example, British train system privatization, which has had the effect of simultaneously raising prices, degrading capital maintenance, and reducing service efficiency. (This can be nicely compared with the Japanese private train networks, which work very well and very cheaply, though I don’t know if there is a government subsidy involved.)
Comment by John Emerson —
March 5, 2007 @ 7:18 am
And the government buyer can quit and be rehired by the privatized seller once the contract is signed.
Ufortunately, if government is not competent to offer services, it’s probably not competent to negotiate business deals either.
Buying and selling IT systems seems to be one area where the Peter Principle is especially prominent. At the beginning at least, the tech person in some organizations was just the flunky most willing to bone up in a haphazard way on something no one else wanted to bother with. This was true in the private sector too, I think.
Comment by Barry —
March 5, 2007 @ 7:31 am
Add to that cronyism and secrecy. If the contracting process is not open to the public, then corruption flourishes. The ‘private’ firm’s doings may or may not be open to the public, depending on the laws and politics. If they are not open to the public, then this is a great way of moving activities from a publicly-reviewable area to a non-reviewable area.
I have a sneaking suspicion that privatization amplifies tendencies in the government and the political situation.
If only I could prove that, I’d be picking and choosing from professorships
Comment by Fraud Guy —
March 5, 2007 @ 8:13 am
Additionally, if the service provided is one that is required by the government only (e.g. foster care, prisons, military, voting systems), then a negative feedback loop can be created as the provider attempts to politically affect the need/use of their service, and then becomes dependent on that political support and may create conditions that reinforce their chosen means of political support. Then if the political support becomes dependent on the service provider (contributions, political capital, potential scandal), they will attempt to reinforce the need for the service provider. At that point, it becomes a dance wherein each partner tries to exploit the other just short of having the other try to break the relationship.
Comment by Barry —
March 5, 2007 @ 9:55 am
Or worse, the dance cements the relationship.
Comment by Neel Krishnaswami —
March 5, 2007 @ 10:38 am
Some things, like prisons, are absolutely things that you want to be expensive and run at a loss. If they are profit centers, then you’ve created a positive incentive to incarcerate more people. That’s one of the reasons that I’m very strongly opposed to most prisoner work programs — it’s a way for people to profit on jail time.
It’s bad enough already that the prison guards’ unions lobby for more jails; there’s no sense in increasing the strength of that special interest.
Comment by Neel Krishnaswami —
March 5, 2007 @ 10:39 am
Am I getting spam filtered?
Comment by Leonard —
March 5, 2007 @ 10:44 am
If the buyers can easily do without, then yes, slam dunk. But if they can’t, and the service is a monopoly, there’s a big political problem here. The business has the incentive to raise prices to the monopoly level. That will feed back politically and cause pressure for at least heavy regulation, if not re-socialization of ownership.
In that sort of case, if there were a way for the state to really, truly, irrevocably privatize, then it would be good and it would work. This is because in the long run, the business’s monopoly profits would attract competitors. However, depending on the capital intensity of the business, that long run might take a while.
Of course, there is no way for the state to commit itself to anything, since there is no higher power than it. As a moral matter, of course, we should still attempt such privatization and be prepared to have the business stick it to us. However we should not be surprised when we do it and the political backlash is considerable.
Now, in a situation where there are multiple buyers of a service, and there is no natural monopoly — in that case, you do have a clear win from privatization, at least assuming the privatization is done right (not setting up a monopoly). However there don’t tend to be that many of these opportunities, for the obvious reason that it was the perceived monopoly aspect of many businesses that was why they were socialized in the first place.
However, there is one business currently monopolized by the state that clearly does meet the above requirements (multiple consumers, multiple suppliers). That is education.
Comment by lemuel pitkin —
March 5, 2007 @ 11:54 am
Right as far as it goes, but I would add another criterion:
Interruptions of service and/or the occasional low-quality product ahve to eb acceptable. If a minimum level of service is absolutely necessary, you’re probably better off with the government. That’s why education, despite Leonard’s other good points, is not a good candidate for privatization.
Comment by lemuel pitkin —
March 5, 2007 @ 11:56 am
John Emerson also makes a good point: privatization does not reduce the demands on gov’t personnel, who still have to manage and oversee the contractor. If you really believe government is simply not competent to handle a certain task, you need to either think about ways to improve its competence, or get it out of that line altogether.
(Another reason why Leonard is wrong about schools — gov’t is always going to have to be a provider of education, and always going to have to oversee it closely.)
Comment by Eric the .5b —
March 5, 2007 @ 12:18 pm
And when the government abjectly fails to meet that minimum of service, we’re better off with that instead of private providers because…?
Comment by lemuel pitkin —
March 5, 2007 @ 12:27 pm
Eric,
As I said in my previous post, in that case you need to either raise the level of comeptence of government or get it out of that activity altogether. I’m not saying government is omnicompetent; I’m just saying that if it’s not competent to perform an activity, it’s not going to be comeptent to hire and supervise private contractors to perform it either.
Comment by Neel Krishnaswami —
March 5, 2007 @ 12:33 pm
Lemuel, does the FDA have to provide drugs? If not, then why does the government have to provide schools?
Comment by lemuel pitkin —
March 5, 2007 @ 12:37 pm
Sorry, should have been clearer: I think we all agree that government, if nothing else, reduces teh variance of outcomes. If all you care about is average otucomes, the case for private provision is going to look betetr than if it’s also important to minimize the number of otucomes below a certain level.
Movie studio makes bad bets, goe sout of business, fine. So we ahve a free amrket in movies. Airline makes bad financial decisions, goes out of business,a gain fine. Airline makes bad safety decisions, planes crash, not fine, so we don’t ahve a free market in airline safety but instead government regulation. Markets are good at optimizing, government is good at satisficing. Which you care about depends on the case. Right?
Comment by lemuel pitkin —
March 5, 2007 @ 12:40 pm
Oh and finally, and then I’ll shut up:
Satisficing is more important than optimizing in education. A certain number of crap movies is a fine tradeoff for a certain nuber of great movies, but some kids getting a crap education is not an acceptable tradeoff for other kids getting a great education.
Comment by Leonard —
March 5, 2007 @ 1:11 pm
Lemurel, you have a theory that the state satisfices in education. But this is flatly contradicted in its inner-city school systems. Anyway, I hope you do not believe that those systems are doing good enough.
Thus, your theory is wrong.
BTW, our current society also proves that we can accept a pretty low “minimum level of service”. We have accepted it, for decades now.
Comment by Barry —
March 5, 2007 @ 1:49 pm
A comment – Stiglitz (?) was writing about IMF-required privatizations in third world countries. He stated that the bureaucrats in charge might initial b*tch about the destruction of their fiefs, but only until they realized that they’d be selling multi-billion dollar assets at a dime on the dollar. For some reason, probably rhyming with ‘Swiss bank account’, that made them happy.
Comment by mac —
March 5, 2007 @ 2:23 pm
There’s privatization, and “privatization.” The latter really started under Reagan, with the “privatization” of Landsat. What had previously been available from the government for (near) free now cost something like $100. And of course the government was te biggest buyer of Landsat images. So the taxpayers paid for the satellite, the launch, etc. Then they started paying somebody’s crony a bundle for the images as well.
Comment by Thoreau —
March 5, 2007 @ 7:03 pm
During the California energy crisis of 2001, I tried to make sense of the highly regulated but ostensibly “de-regulated” California electricity market. I observed that a straightforward public monopoly would at least make sense. We’d all know who to point fingers at, where it’s coming from, and who’s doing what. With the highly regulated “de-regulated” “market” for electricity I had no clue what was going on.
Mind you, I don’t consider it imperative that I understand everything as long as the lights stay on. But if the lights aren’t saying on, and nobody can explain why, but locales with public monopolies are getting electricity, then we should go with something simple that works. Apparently there are places where electricity is privatized and relatively deregulated (not 100% deregulated, for good or for ill, but at least the regulations are simple compared to CA) and it works, and there are public monopolies that work.
I’ll always take soemthing that works over a highly regulated and non-transparent system that doesn’t work, no matter how many ostensibly private features it has.
Trackback by Freedom Democrats —
March 5, 2007 @ 7:22 pm
Faux Privatization…
Hat tip to Jim Henley at Unqualified Offerings. Matthew Yglesias discusses when privatization works and when it doesn’t work. I can see how a government provided service, such as education or transportation, can be improved through privatization and…
Comment by Wild Pegasus —
March 5, 2007 @ 9:46 pm
but some kids getting a crap education is not an acceptable tradeoff for other kids getting a great education.
So, what happens when a school system spends $16k per student and still can’t teach half of them how to read? See Columbia, District of.
- Josh
Comment by bza —
March 6, 2007 @ 1:44 am
Lemuel, does the FDA have to provide drugs? If not, then why does the government have to provide schools?
I’m not Lemuel, but there are significant differences between the two cases. School choice is, arguably, an area where tacit and local knowledge on the part of the consumer is something that would be possible and useful to harnass. Knowledge of a drug’s effectiveness and side effects is neither tacit nor local.
Comment by Leonard —
March 6, 2007 @ 11:01 am
bza, your criteria do distinguish education and drugs. But they would seem to indicate that education should be freer from state control and regulation than drugs.
Comment by bza —
March 6, 2007 @ 12:57 pm
24: I was reponding on behalf of Lemuel, who was considering the idea that gov’t should get out of education in particular, but not be put out of business altogether.
Comment by bza —
March 6, 2007 @ 1:13 pm
At the risk of being tedious, let me try to explain more clearly. The argument concerend three levels of government involvement:
Lemuel was arguing that a purely regulatory role for the government didn’t make sense in education. Neel then countered that government has a purely regulatory role in the case of drugs. I was pointing out that there are broadly Hayekian reasons for thinking that drugs and education are different in this regard (i.e., as to the viability of governmental involvement at the merely regulatory level).
Comment by Neel Krishnaswami —
March 6, 2007 @ 4:30 pm
bza, you’re not being tedious at all– being precise in argument makes productive discussion easier. As I understood Lemuel in 11, he was arguing that, in general, if a government can’t produce a good effectively, it can’t effectively regulate it either.
I don’t think that’s true, and offered the FDA as a counter-example — I don’t think many people would argue that the government could research and produce drugs as effectively as the private sector, and yet they are also happy with the FDA’s regulation. (I think Leonard is also right that the Hayekian arguments you advance actually argue for making education less regulated than the drug market, but that’s another point.)