There is little doubt that contracting out of prison services is possible. With requirements for good information, independent monitoring and external accountability to get over the principal-agent problem it can be made to work well. If done only to reduce costs, then what you’ll get is a bad prison system that is low cost to run but has high negative externalities (the cost of recidivism is enormous and hidden, effective rehabilitation and reintegration is obvious and up front). However, if the contracting parties understand this it is a perfectly workable option.

That they can be outsourced is though an entirely different question from the normative one of whether prisons should be privately provided. There is a long standing debate on whether the state should have a monopoly on the use of coercive power. It relates not just to prisons but to armed force more generally (police and the military). Although it is quite feasible for a state to employ a mercenary army – see for the example a recent example with Sandline in our near north – there is a strong body of thought that says this is not desirable.

But there is a contiuum in relation to what constitutes the use of coercive power. It ranges from the armed forces through police, prisons, down to mall security guards. Where you draw the line through this spectrum, with state on one side and private providers on the other, is ultimately a political decision. It can only be determined through the democratic process, rather than quantitative analysis.

In governance terms, this requires any proposal to provide such service privately to be transparent and open. If it is part of a party’s platform, and voters accept that, then it is fine. If it is done as a result of private backroom lobbying in secret (as has apparently happened in some cases in the US), then the test is failed.

The public system is a monopoly, whereas the whole benefit of a private system is to make it competitive.

I doubt that’s the point of private prisons, to be honest. The benefit from privatising them is getting rid of unionised government prison workers and replacing them with non-union private workers on lower pay and benefits.

There’s also the benefit that comes with privatising anything in that people who are spending their own money tend to spend less of it than they do when they are spending someone else’s.

That said I’m always suspicious of any private business that has the government as its only client. The incentives are all wrong.

I was joking.

It’s not as though we have no basis for comparison when discussing rehabilitation versus punishment – Scandinavian prisoners have a re-offending rate of around 35%, in contrast to 50% – 60% in the US and Australia.

More importantly, the re-offending rate of prisoners after serving their first prison sentence is significantly lower. Australian prisons have a reputation as trade schools for crims rather than somewhere to get back on the straight and narrow.

Many prisons in Scandinavia are set up more like rehab centres than traditional prisons, with comfortable rooms and lots of opportunities for recreation and skills acquisition.

Imagine trying to implement that in Australia, where tabloid readers have a fit if they so much as hear that selling an ounce of pot doesn’t warrant ten years of hard labour.

http://www.unafei.or.jp/english/pdf/RS_No74/No74_06VE_Seppala2.pdf

“Since there’s no proven technology for preventing recidivism…”

Gallows, guillotine, firing squad, electric chair…

What we have officially are “Corrective Services”, and since I believe in truth in advertising, they should be paid proportionally to the number of people they can get into steady jobs, who don’t re-offend for several years.

That raises some interesting issues. The first is the issue of where service providers would get the money they need to run prisons while they wait for their outcome payments. A fashionable idea right now is ‘social impact bonds’. As the Economist explains:

Peterborough is where the proceeds of the world’s first “social-impact bond” are being spent. This instrument is not really a bond at all but behaves more like equity. In September 2010 an organisation called Social Finance raised £5m ($7.8m) from 17 investors, both individuals and charities. The money is being used to pay for a programme to help prevent ex-prisoners in Peterborough from reoffending. Reconviction rates among the prisoners recruited to the scheme will be measured against a national database of prisoners with a similar profile, and investors will get payouts from the Ministry of Justice if the Peterborough cohort does better than the rest. If all goes well, the first payouts will be made in 2013.

I see three problems with applying this model to prisons:

1. Impact measurement depends on a comparison group who are outside the program. That means you can’t apply payment-for-impact to the entire system.

2. Private sector investors will demand a premium in return for the risk of not getting paid. Since there’s no proven technology for preventing recidivism, the premium is likely to be large. It’s hard to see how government would save money.

3. Governments have other reasons for incarcerating offenders. For example – incapacitating offenders (it’s harder to commit crimes against the public if you’re in prison), deterring potential offenders, and exacting retribution.

After we’ve solved the problem of a payment-for-results model for privatised prisons, maybe we can move on to defence policy.

Instead of having politicians involved in debates over what kind of submarine to buy, perhaps we could simply contract out the defence of Australia.

How could we encourage private investment and construct a payment-for-results system? ;)

Two obvious flaws with the bread analogy:

1. Most people eat bread, so there’s strong market pressure to provide decent bread. Fewer people are concerned about the treatment of prisoners, and many think prisons should be more brutal, if anything.

2. A bakery will get more return custom if it produces better bread, but a prison will encourage recidivism if it provides poor rehabilitation, so a private prison has incentive to work AGAINST the public interest in order to maintain or increase profits.

One obvious flaw with this comment – it basically makes the same argument twice.

However, that argument is quite compelling, and I’d like to hear one against it.

Just for shits and giggles.

JC, so you don’t like the idea that someone offering a service should tell the truth about what they offer?

You need to spell out exactly what the service is that firms are bidding on and then establish agreements based on that. That was the point I was making.

You suggested that firms should only be paid on the back end, after the ex-con has been out of the clink for x number of years. We don’t exactly have that sort of expectation on the public system now.

Look, you could stick anything you want in a bid spec. You could even enter a condition that a number of the prisoners need to become CEO’s of ASX 200 before firms receive a red cent in payment. But whether you get bidders is an entirely different story. Or if you get bidders the price would be impossibly high. There’s no free lunch.

Of course the issue is mitigated if they are only paid for rehabilitation.

In the end however prison reform, like much else about good government, is rather dependent on ending the war on drugs.

You might want to check up what the WWII migrants from Europe were offered by the Australian government when they got here. Galvanised iron sheet, black tar on the roof, stuck out in inland NSW or QLD, and none of your Kevin Rudd in the attic either. They were not held as prisoners though, they were free to take long walks in the desert, I’m sure that little bit of freedom meant a lot to them.

Funny how air conditioning was once the most amazingly luxurious item, and now the only people who live without it are those poor buggers who bought a fibro hut in Western Sydney for a quarter of a million bucks, and would rather fatten up the bank’s profits instead of trying to pay the 300% markup on electricity.

What we have officially are “Corrective Services”, and since I believe in truth in advertising, they should be paid proportionally to the number of people they can get into steady jobs, who don’t re-offend for several years.

So you would expect what is an intolerably higher standard from the private sector than you currently do from the public one. Why?

And how would you compare a hypothetical bid from performing the service in-house when the conditions are onerously different.

Seriously, would you expect anyone to even contemplate placing a big on those conditions and they don’t even get to choose their own “clients”?

Naturally the convicted man should be given the right to choose whose “Corrective Services” he is to avail himself of. He is the one who has the most to lose or gain after all.

The impact of such a choice would be having firms potentially pandering to the “clients”. That would be the last thing you want to create. This isn’t the Hilton we’re talking about here.

What we have officially are “Corrective Services”, and since I believe in truth in advertising, they should be paid proportionally to the number of people they can get into steady jobs, who don’t re-offend for several years.

In terms of the difference between private and public… that should be obvious. The public system is a monopoly, whereas the whole benefit of a private system is to make it competitive. Thus, we should never have a single supplier of these “Corrective Services”, it would work very much better if we insist on always having a number of suppliers. Naturally the convicted man should be given the right to choose whose “Corrective Services” he is to avail himself of. He is the one who has the most to lose or gain after all.

For example: “I…suggested that there seems to be a lot of moralizing about this subject”. You suggested no such thing until now, of course, and my point that you’re contending is about recidivism, not moralising.

You’re welcome back to the topic whenever you’re ready.

It’s been nicely said to you by others, like Yobbo etc., that you find it impossible to follow a normal discussion. I once found you denying your own comment three times on the same thread here at Troppo even though I pointed out what you said and posted the comment number. Remember?

You must play chess the way you converse. You take the bishop and suddenly want to use it as a queen.

So, you disagree that most people in the western world eat bread,

No, sanchez, I said I think the business model for a bakery and a prison are very different and cannot really be compared so easily.

and you disagree that prisons can affect recidivism rates through their management of prisoners? Is that right?

No, I said nothing about the recidivism rate but suggested that there seems to be a lot of moralizing about this subject and it would be good to look at the hard facts in terms of outcomes through prior experience.

Sancho said:

Two obvious flaws with the bread analogy:

1. Most people eat bread, so there’s strong market pressure to provide decent bread. Fewer people are concerned about the treatment of prisoners, and many think prisons should be more brutal, if anything.

2. A bakery will get more return custom if it produces better bread, but a prison will encourage recidivism if it provides poor rehabilitation, so a private prison has incentive to work AGAINST the public interest in order to maintain or increase profits.

The above is what I was pointing out was silly. Furthermore it makes no sense at all.

The prison firms have zero involvement or influence in the judicial process prior to incarceration, so the idea they could in theory have any influence on the judicial system is flawed.

Did you bother reading the link you offered? It’s not research about the effectiveness of the US private prison business, which now houses around 3% of the US prison population thereby offering a decent sample population. It’s model based research on private vs public for a host of services, with a conclusion based on various theoretical parameters. It speaks very little about hard data results.

One consideration they discuss that has an important impact on results and may spew out one here is the level of unionization and the strength of these unions.

The link seems redundant to what is needed… and that is studies that offer some guide to how these enterprises are performing.

I’m open on the idea and have no bias either way. However it would be good to see if the contractual arrangements/objectives by and large meet expected outcome.

http://www.economics.harvard.edu/faculty/shleifer/files/proper_scope.pdf

So isn’t the obvious point then, not to award bakeries contracts run prisons?

Instead of making all these hypothetical claims, there have been private prisons around for a long time now. Are there any studies that help with the process of deciding if they work. Surely that would be a good place to start instead of worrying if bakery firms pitched a bid.

And by “work” I mean did they meet objective and conditions.

The business line is really not that much different to running hospitals.

The real problem with a private prison system is the recent shenanigans going on in the US at the moment with deficit ridden State Governments agreeing to minimum prison numbers. I don’t see that as the fault of the firms negotiating these deals, but more of problem with the states doing these repulsive deals that could cause potential conflicts down the road.

1. Most people eat bread, so there’s strong market pressure to provide decent bread. Fewer people are concerned about the treatment of prisoners, and many think prisons should be more brutal, if anything.

2. A bakery will get more return custom if it produces better bread, but a prison will encourage recidivism if it provides poor rehabilitation, so a private prison has incentive to work AGAINST the public interest in order to maintain or increase profits.