The Economic Way of Thinking

Many libertarians in the blogosphere have been up in arms lately over a decision by British officials to "charge victims of miscarriages of justice more than ?3000 for every year they spent in jail while wrongly convicted."

Perry de Havilland calls the decision "beyond belief" and the officials who made it "evil." Jim Henley thinks the story is so absurd that it must be an April Fool's Day joke. Radley Balko calls it the "Gum'mint Outrage of the Day" and says that he is speechless, with the added implication that the decision is so self-evidently wrong, it needs no further comment. Julian Sanchez writes that this decision must come "From the 'You've Got to Be S#$&ing Me' Files" and compares it to a fictional government policy of charging families of persons detained by the government for the interrogation and torture of their loved ones. Eugene Volokh, the blogosphere's Socrates, believes the decision represents "Unintentional self-parody of government overreaching" and echoes Jim Henley with a reference to an April Fool's joke.

I must admit that when I initially heard about this decision, I too thought it was a joke, and if not a joke, then a true injustice. But the more I think about it, the more reasonable the decision becomes. I think the reason why so many of us are initially appalled upon hearing that the wrongfully convicted are being charged for room and board is that we have trouble accepting the economic way of thinking.

More specifically, this seems to be a cut-and-dry case for cost-benefit analysis, which stands at the heart of economic decision making. Nearly all of us agree that people who are wrongly convicted should be justly compensated for the years they lost in prison. But how do we determine how much this compensation should be? This is not a question simple libertarian theory can answer. The answer must come from some kind of economic analysis, which measures the costs (as well as the benefits) of putting innocent people in jail, and perhaps a deterrent to prosecutors[1] to make them more careful about wrongly convicting innocent people in the future.

While I'm not going to try to do such an analysis in this blog post, both because (a) I'm lazy and (b) someone has almost certainly done it before me (which I would know if I wasn't too lazy to look it up), I will name a few of the areas I would start with were I doing a valuation for real. The most obvious, and most easily determined would be the value of lost wages. We don't need to use any fancy methods for figuring this one out. More difficult would be determining the value of one year of freedom, a year without friends and family, and so forth. But in principle, it could be done. If we can put a price on life itself (which economists do; it's about $5 million these days), we can certainly put a price on freedom.

In which case, putting free room and board on the benefits side of the ledger is not so outlandish after all. Remember that we are trying to evaluate the costs of a year of freedom, in which case any benefits of being in prison should be deducted from the costs (benefits and costs are merely the opposite of each other). Those who morally object to including the benefits as well as the costs of wrongful imprisonment are as mistaken as those who morally object to considering the possible benefits of global warming. Global warming, assuming it occurs as many claim it will, will produce costs as well as benefits. To ignore the value of either is to incorrectly determine the value of global warming as a whole. And to ignore either the benefits or the costs of wrongful imprisonment is to do the same. While many of us might feel that no amount of money can compensate someone for lost freedom, we still need to come up with some compensation technique, and this method must make sense. It must include all of the relevant variables.

fn1. Unfortunately, under our current socialized justice system, there is little if any disincentive given to government officials to prevent them from wrongfully convicting innocent people. Under a competitive, for-profit justice system, we would expect that part, if not all of the cost of compensating the wrongfully accused would come from the arbitration firm which wrongfully convicted him.

Update: Greg Goelzhauser, who writes for the Law and Economics Blog by day and Crescat Sententia by night (or is it the other way around?), takes me to task for my analysis of the this issue. Greg writes,

I don't have any knowledge of exactly how much people received in compensation, but it is almost certainly less than they deserved considering the direct costs and opportunity costs that we could value in determining payment, e.g. general happiness in pursuing leisure, procreation, spending time with loved ones living out their final years, career development, and so on. In addition, a proper cost-benefit analysis (as Micha phrases this exercise) would include compensation for some of the harmful effects resulting from sudden release, some of which were described in the excerpt above. In short, it doesn't seem reasonable even under "the economic way of thinking" to deduct cost of living expenses from the total award given these individuals. They almost certainly were not fully compensated, even assuming such a calculation could be made. The way I see it, the government's award was almost certainly low considering the resulting costs imposed by the wrongful convictions. Thus, even under "the economic way of thinking" the current billing is nonsensical and outrageous. I should add that the decision to bill may be even more nonsensical considering the certain signaling effect it will have. Of course, it's too early to measure the results of such an effect, i.e. whether the signal will impose low or high costs (if any) on the government in the long-run, but any rational calculation would probably need to weigh the costs from signaling and resulting availability effects against the general desire to collect any potential windfall pay-out.

Now, I certainly agree with Greg that the decision to deduct living expenses from compensation benefits may very well be wrong on practical grounds if the compensation is lower than it should be. But that's neither here nor there. Greg should focus his criticism on the underestimation of the costs of wrongful imprisonment, not on the inclusion of any possible benefits. The current compensation system in Britain is "nonsensical and outrageous" not because of the decision to include the benefits, but the failure to include all of the costs.

In theory, it makes perfect sense to deduct living expenses. In practice, if total costs are underestimated, it does not make sense to do this, and we might as well ignore the living expenses to account for the difference.

I'm not sure what "signaling effect" Greg is describing, so I can't respond to that part of his criticism.

Update #2: Will Baude responds:

It's more important that an economic formula like this be approximately right than it account for some randomly selected economic effect that is at best second-order. If and when the compensation levels for wrongful imprisonment reach the ballpark of $500,000-$1,000,000/year I will then be impressed by the reasoning evinced by the British Government. Until then, they should work harder on trying to get into the right range before they start picking nits.

I agree with Will that, practically speaking, it is better for a CBA valuation to be approximately right rather than accurately account for some variables but not others. The original point I was trying to make, though, is that, in theory, there is nothing outrageous about including the costs of room and board in the equation, so long as the other costs and benefits are properly accounted for. Further, although the costs of room and board are relatively small compared to the total costs to the individual of losing his freedom (and this is reflected by the fact that the amount the government wishes to charge the recently-freed is much less than the total amount the government pays in compensation), I don't see why this should be considered a "second-order effect." Perhaps I simply do not have a good understanding of the concept of second-order effects, but it seems to me that the benefits of free room and board are on the same level as the costs of losing one's freedom. And why does Will assume that the proper ballpark for compensation is $500,000 to $1,000,000 per year? Why does anyone assume that the amount the British government currently gives in compensation is too low? I accept that as a possibility, but I certainly don't see it as so blindingly obvious that it justifies immediate outrage over this issue, as so many have expressed.

One more thing. Will writes,

Focusing on second-order mis-estimations when the first-order estimations are wildly off is both bad philosophy and bad economics.

The problem with this argument is that the bloggers to whom I was responding were themselves focused on what Will considers the second-order effect of charging for room and board. None of them were outraged by the fact that the first-order effect of lost freedom was (allegedly) undercompensated. That is indeed an outrage, if it is true. But including the costs of room and board is not an outrage, and those who initially responded with scorn and derision were letting libertarian moral theory trump sound economic thinking.



Follow-ups:
Limits of the Economic Way of Thinking?
The Benefits of Incarceration

Share this

If we can put a price on

If we can put a price on life itself (which economists do; it's about $5 million these days), we can certainly put a price on freedom.

If someone offered you $5 million that can be willed to anyone you desire in exchange for your death, would you take the offer?

I have to disagree with you,

I have to disagree with you, Micha. If I understand the decision correctly, the British government is attempting to charge 3000 pounds per year in jail without at the same time compensating for lost wages and other costs born by the wrongfully accused. And while it would be less wrong if they had been taken into account simultaneously, it still seems to me that there is some gross miscarriage of justice here because the victim's decision to stay at the jail was coerced. While the state may pay 3000 a year per inmate, this seems to me a clear case of coerced purchase, every bit as unjust as taxation (perhaps even more so) that pays for programs that one wouldn't otherwise donate to.

"If someone offered you $5

"If someone offered you $5 million that can be willed to anyone you desire in exchange for your death, would you take the offer?"

I'm not sure if this is relevant to your point, but that is not what is usually meant by "value-of-life." Usually what is meant is "small-risk-value-of-life."

For example, if I say I "value my life" at $17 million, that means that, for very small risks (less than 1/10,000), I can model a risk of death as similar to a risk of losing $17 million. So, if you want me to take a one-in-a-million chance of death, you would have to pay me at least $17.

This does not scale to larger risks.

Bill is partly right, and

Bill is partly right, and for the purposes of arguing with a libertarian, this is all that matters. In public policy, however, governments must make decisions about how much money to allocate to save an additional life. Since resources are scarce, governments only have so much tax revenue available, and must decide how best to allocate these scarce resources to achieve the greatest benefits. Therefore, it helps to have some kind of metric so that we can determine where the government should direct funds.

Kian, No. Reread the article

Kian,

No. Reread the article linked at the beginning of my post. The British government does in fact compensate the wrongly accused, and this charge for room and board is fairly small compared to the overall compensation package, as we would expect.

And while it would be less wrong if they had been taken into account simultaneously, it still seems to me that there is some gross miscarriage of justice here because the victim's decision to stay at the jail was coerced. While the state may pay 3000 a year per inmate, this seems to me a clear case of coerced purchase, every bit as unjust as taxation (perhaps even more so) that pays for programs that one wouldn't otherwise donate to.

No one is denying that the original act was unjust. But the injustice has alread occurred and now the relevant question is how much should we compensate. While I'm not entirely sure if room and board should be included as a benefit, I can certainly hear such an argument. My point here is that the case is not absurd at all; the decision might be wrong, but there is no way to know that without looking at all of the relevant details of the CBA.

Also, the price per life

Also, the price per life (value of a statistical life, VSL) is around $7.6 million, last time my firm was doing CBAs for proposed EPA rules. ^_^

Ahh, well, in light of the

Ahh, well, in light of the new evidence, I accept your interpretation. Pardon me for not reading closely enough ;)

Exactly. Great post.

Exactly. Great post.

If someone offered you $5

If someone offered you $5 million that can be willed to anyone you desire in exchange for your death, would you take the offer?

The trouble is, you can't offer someone $5 million in exchange for their death - a dead man can't collect any money.

As Bill alluded to, it has do be evaluated on a probabilistic basis. Thus, if I offered you $5 million for a 10% chance of your death, you only have a 90% chance of collecting, so you would value the offer as only $4.5 million if you are risk-neutral. If it was for a 99.9% chance of you death, you have only a 0.1% chance of collecting, so you would only value the offer as $5000.

Micha: I applaud your

Micha: I applaud your Sanchezesque bravura in flouting the libertarian conventional wisdom here, but you're missing the essential factor - the evidence is that those awarding the original compensation saw the amount they set as the amount the victim would receive. Blunkett's grab is ex post facto. If he wants compensation panels and judges to take "upkeep costs" into account in making future awards, he's welcome to make that case. What would likely happen is that they would simply raise the awards. Which is why he's trying to knock down awards that have already been made.

It's crap, it is total crap, it is unfair, to quote a recently-fired basketball coach.

Micha, I totally disagree.

Micha, I totally disagree. A person wrongly imprisoned has no obligation to pay room and board to the person or institution that imprisoned him, any more than a person who is kidnapped would owe room and board to his captors.

The costs of wrongful accusations and wrongful imprisonment should be high, and none of them should be borne by the wronged party. Those costs function as a vital check on law arbitration and enforcement entities. You need look no further than the U.S. "justice" system to see how fucked up things get when compensation for the wrongly accused and imprisoned is not enforced.

Jim, If what you say is

Jim,

If what you say is true, then this is merely a case of double counting and I would agree with you the decision to charge twice for the same benefit is incorrect. But I'm not entirely sure that is true; Blunkett might just be claiming that until now, the people doing the CBA have been ignoring this factor, and should begin to take it into account.

Either way, it's certainly not an issue for libertarians to be outraged about; CBA's are notoriously more of an art than a science, especially in cases where the valuation is of something relatively abstract, like human freedom. My purpose in this post is to merely point out that reasonable people can disagree over the proper valuations of various parts of a CBA, without anyone having to express moral indignation at anyone else's valuation.

Qiwi, I certainly agree with

Qiwi,

I certainly agree with you that compensation for wrongful imprisonment in the U.S. is drastically too low. However, I do think it is important to consider any possible benefits of wrongful imprisonment in addition to the costs. Otherwise, we have no basis for claiming that the valuation of the compensation is properly determined.

Micha - Jonathan, I

Micha -

Jonathan, I understand your point, but as Bill Carone and I have already said, that is not what the value of a statistical life means. When we say that a life is worth $5 million, we don?t mean that the average person would be willing to die if he could give $5 million to his family.

I understand the point that you, Bill, and Andy are trying to make. Yet, it is irrelevant to the point I am making. I simply object to your gratuitous use of the word "we" in how valuation is performed. These figures you are quoting are prices, not values. Individuals can use their own values to judge whether the price is worth the benefits. But it is not a "we" decision.

Rather, it means , for example, that when the risk of death to construction workers is increased by working at the top of a skyscraper rather than safely on the ground, the average worker agrees to do so if and only if the additional compensation is such that the following condition holds: if the additional risk is 0.1% per year, the additional wages must be $5000 per year.

I submit that this statement is false, but that is a side-topic and one best argued on a different day.

You make a good point that

You make a good point that room and board should be included as a benefit. I think what angers people is that the cost side is not being fully acknowledged, therefore tacking on extra items in the benefit column makes the situation even more unjust.

I don't know how much people get for wrongful imprisonment - there was only brief mention in the article that Hill got a million quid. I saw no mention of awards for anyone else. Perhaps you know more about it. Are these awards being handed out randomly to only some of the wrongfully imprisoned? Or are all fully compensated for their lost wages? Do they get anything for loss of freeedom?

The charges are only good if they were being given as much as they should, or more. If they were already being less, then this moves the compensation farther from the "fair" point. So I think those decrying it have a point.

But thanks for rephrasing the debate in terms of economic efficiency instead of "i hate those gummint bastards"!

The trouble is, you can?t

The trouble is, you can?t offer someone $5 million in exchange for their death - a dead man can?t collect any money.

You guys are missing the point of my question. People leave their property to heirs in wills with quite a bit of confidence today. The exact mechanism by which this could happen is not the question, nor the degree of confidence you would have. Reputated firms could make this transfer.

My question was, if it was possible to leave the money to whatever cause you please, would you take the $5 million in exchange for your life? Micha claimed in his post that "we" could put a "value" on life. I'm asking what would he and you do in that situation?

bq. ... I do think it is

bq. ... I do think it is important to consider any possible benefits of wrongful imprisonment in addition to the costs.

The *possible* *benefits* of wrongful imprisonment? Who the hell exactly benefits from wrongful imprisonment?

How can a person unjustly imprisoned justly be held liable for the cost of holding him against his will?

Jonathan, I understand your

Jonathan, I understand your point, but as Bill Carone and I have already said, that is not what the value of a statistical life means. When we say that a life is worth $5 million, we don't mean that the average person would be willing to die if he could give $5 million to his family. Rather, it means , for example, that when the risk of death to construction workers is increased by working at the top of a skyscraper rather than safely on the ground, the average worker agrees to do so if and only if the additional compensation is such that the following condition holds: if the additional risk is 0.1% per year, the additional wages must be $5000 per year.

Qiwi, the benefit of

Qiwi, the benefit of wrongful imprisonment is that you don't have to pay for rent or food during your "stay." Of course, this doesn't mean that wrongful imprisonment as a whole is desirable or beneficial; else, we would see many more people committing crimes just to get into jail. However, it does mean that although the costs of wrongful imprisonment are many, there is a small benefit which should be included in any cost-benefit analysis.

The point is not that we should hold the wrongfully imprisoned liable for the costs of holding them against their will; the point is that we should only include the true costs of wrongful imprisonment when calculating compensation. Any benefits, however small, should be subtracted from the total costs.

Ok, that's fine. When I said

Ok, that's fine. When I said "we," I was referring to neoclassical economists, and not Austrian economists, if it is the case that Austrians reject such valuations.

I think Micha is right to

I think Micha is right to say that benefits should be considered in the entire compensation calculation. I think a fair compensation would be something like "What amount is sufficient that a reasonable person would be indifferent between choosing the payment+incarceration, or neither." And, a reasonable person would consider the benefit of the room and board in addition to the costs of lost opportunities, etc.

But...and it's a big but...

The way this seems to be handled is not as part of the compensation, but as an additional fee tacked on. It looks like an attempt to get back some of the compensation that it has grudgingly paid.

If the state doesn't charge the guilty for room and board, then it certainly has no business charging the innocent. It seems that the state accepts that the expenses of housing a prisoner include room and board and it's willing to bear that cost for the benefit of administering justice, security, etc.

Gil, You make two good

Gil,

You make two good points. In response to the first, while the government may have improper motives for acting, or at least the appearance of improper motives for acting, I don't think this changes the overall validity of the policy. Whether the costs are deducted at the time of the original valuation or later, the end result remains the same.

In response to the second, why shouldn't the state charge guilty prisoners for room and board? What sense does it make charging taxpayers, who have done nothing wrong, and not the criminal himself? In terms of feasibility, it might be easier to make prisoners pay while they are still in prison through forced labor, rather then after they are released. Then again, in terms of political feasibility, this may not be feasible at all, unfortunately.

Jonathan, One more thing I

Jonathan,

One more thing I just realized about your objection. Surely you believe that victims of wrongful imprisonment should be compensated for their loss of freedom, correct? And surely you believe that the families of victims wrongfully killed should be compensated for the loss of the victims' lives, correct? If so, how do you suggest courts go about determining this value, if not by the method of economic analysis I described?

Individuals can use their

Individuals can use their own values to judge whether the price is worth the benefits. But it is not a ?we? decision.

So is your objection to using an average estimate of how much people value their lives when calculating compensation, rather than trying to estimate a different value for each individual?

Jonathan, "Individuals can

Jonathan,

"Individuals can use their own values to judge whether the price is worth the benefits. But it is not a “we” decision."

Sorry for missing your point.

I agree that each individual should get to decide for him or herself what value to place on his or her life, depending on e.g. income, risk preference, time preference, age, lifestyle, lifetime distribution , etc.

If you delegate that decision to the government, for example, it leaves them the option to value it as low as they want.

Here are two cases I might describe as "our setting" a value-of-life. First is a case of product liability, the other is an issue with tort suits involving risks.

1.

Let's say I'm designing a product that might kill the user in an accident (a car, a lamp, a screwdriver, for example). Now, by designing it in different ways, I affect the risk and the eventual consumer price (safer means more expensive).

I might put the following notice on the product: "I have used a $20 million small-risk-value-of-life in designing this product." In other words, I assessed the risk, and increased costs whenever the cost increase would be worth it to someone with such a value-of-life.

Then each consumer could decide whether or not to use this product. If the value-of-life was approximately right, then they would buy it. If it wasn't safe enough, they would find a safer product. If it was "too safe," they might buy another product (since the "too safe" product might be too costly).

This number might also be tied to a contractual term: "If this product kills you, we agree to a $20 million payout from us instead of your suing us."

So here is a case where we "set" a value-of-life; here it is just a piece of information that helps the consumer make a decision.

2.

Another case is one where we need to figure out someone's value-of-life, and we can't necessarily trust them to tell us the truth. For example, say I inflicted a 0.1% risk of death on you without your permission. How much should I pay you as damages? It should be about 0.1% of your value-of-life.

We could use an evidentiary proceeding: imagine that, normally, people with $100,000 income, age 45 have a $10 million value-of-life (I don't know if this is really true). So we might use that as the standard.

Litigants could then provide evidence that a person does or doesn't have this value. For example, we shouldn't believe someone if he claims a $1 trillion value-of-life, since he, e.g., doesn't wear a helmet when driving, doesn't live in a concrete bunker, etc.

This way, we can figure out how much to insure ourselves for if, for example, we build a factory that inflicts, say, a one-in-a-billion chance of death on everyone in a 50 million person community. The insurance company would figure out the likely damages involved, based on the makeup of the community.

Now, we would have to set up laws about such unagreed-to risk. Maybe for very low risks (valued at less than $1), no permission is needed, and victims can sue for damages if the worst happens. For slightly higher risks, you wouldn't need permission, but would have to insure for the correct amount. Higher risks might require the victim's permission, and people could shut you down if they didn't want the risk (i.e. they could refuse any offer). Even higher risk would be on par with murder. I'm no expert on this, but I suspect something like this could be worked out in either a public or a private law system.

Again, here is a case where "we set" a value-of-life; here we are just trying to determine a revealed preference without being able to trust what the victim says.

What do you think about the above? I am much less comfortable with 2 than with 1, as you might guess.

Micha, "Bill is partly

Micha,

"Bill is partly right, and for the purposes of arguing with a libertarian, this is all that matters."

I have no idea what this means; should I ignore it?

"Since resources are scarce, governments only have so much tax revenue available, and must decide how best to allocate these scarce resources to achieve the greatest benefits. Therefore, it helps to have some kind of metric so that we can determine where the government should direct funds."

Taking off on this point ...

I can set my value for my life; this means I can decide how much _I_ should pay to protect _myself_ from risks.

I can also set my value for your life; this means I can decide how much _I_ should pay to protect _you_ from risks.

I cannot tell you how much _you_ should pay to protect _yourself_ from risks; this is about your preferences, not mine.

So, if we assume the government rightly owns the tax money it takes, then it has the perfect right to decide _its_ value for _your_ life. In other words, it can set how much it is willing to pay to reduce your risks.

It should not use this value to, e.g.,

- regulate your risky behavior for your own good, or
- figure out how much to compensate you for risks inflicted on you

since these are about _your_ value of your life, not _its_ value of your life.

So perhaps the government could provide tax incentives for safe behavior, thus paying to reduce your risks. For example, subsidizing seat belts and air bags in cars.

Unfortunately, some ideas start falling apart once you loosen the assumption about taxes being okay.

For example, taxing everyone $100, then providing a $100 tax rebate to everyone who wears their seat belt all the time isn't any different from charging $100 if you don't wear your seat belt. And this is something that depends on _your_ values, not the government's, so the government shouldn't do it.

This suggests the question: What kinds of things should the government be allowed to do to reduce our risks that doesn't amount to _de facto_ regulation of risky activities? Medical research? Pollution clean-up? Others? I haven't thought about this carefully.

Micha, I agree that the

Micha,

I agree that the guilty should pay for their upkeep, and I would say that they should be made to pay compensation to victims, too. Perhaps proper prison privatization could incorporate these things.

But, my point is that while the policy is to NOT attempt to retrieve expenses from the guilty, it seems unfair to discriminate against the innocent in this area.

Actually, perhaps if it's proper to retrieve these expenses from the guilty, then those responsible for the improper imprisonment (police, prosecutors), rather than the prisoners, should be made to pay. And they should pay out of their personal wealth, not soak the taxpayers. They should really pay the entire compensation, but that's probably not practical. Maybe they should have malpractice insurance for this.

Why should he be compensated

Why should he be compensated and "made whole" at all, Micha? You say nearly everybody agrees he should be compensated but nearly everybody thinks a lot of stupid things.

No disrespect intended but nearly everbody thinks your average Cattalarchy.net blogger is a lunatic. (And I DIG that about you people!) I don't think you want to be premising your arguments on "Nearly everybody thinks X".

Ok, that?s fine. When I said

Ok, that?s fine. When I said ?we,? I was referring to neoclassical economists, and not Austrian economists, if it is the case that Austrians reject such valuations.

Just as a clarification, my objection to your use of the word "we" had nothing to do with Austrian economists vs. neoclassical economists, but rather with the notion that "we" (plural individuals) can come up with a value of life. Thus, my question about the decision you would make given the $5 million hypothetical.

Bill, Thanks for those

Bill,

Thanks for those examples. In general, with minor semantic quibbles, I would endorse #1 easily. Yet, my point is that the vendor set a price which he believe will maximize the achievement of his ends, and consumers individually decide whether or not they value the product (and associated risks) more than the cash equivalent of the price of the product, and all other possible uses of that cash equivalent at that time and place. If they do, they make the exchange.

But this is not a collective value appraisal, just like my going to the store to buy a loaf of bread for $2 is not a collective value appraisal.

I'm still mulling over #2.

Bill, ?Bill is partly right,

Bill,

?Bill is partly right, and for the purposes of arguing with a libertarian, this is all that matters.?

I have no idea what this means; should I ignore it?

Yes, you probably should ;) My disagreement with what you said is very small, but if you?re interested, my point was this: estimating the value of life is not only useful for estimating the cost of small risks. It is also useful for governments (and perhaps insurance companies) for determining whether or not it is worth it to save an additional life, by offering kidney transplants, for example. This might not be relevant when arguing with a libertarian who does not believe the government should be making these kinds of decisions, but it is important under our current regime.

So, if we assume the government rightly owns the tax money it takes, then it has the perfect right to decide its value for your life. In other words, it can set how much it is willing to pay to reduce your risks

But remember, its not only relevant for small risks, its also relevant for deciding which medical operations the government will fund for the indigent.

John, Why should he be

John,

Why should he be compensated and ?made whole? at all, Micha? You say nearly everybody agrees he should be compensated but nearly everybody thinks a lot of stupid things.

You?re quite right that this is no reason to support a policy. I wasn?t using this as justification for anything, just a factual claim intented to establish an initial premise. In other words, I was saying that the idea of compensation is nearly universally agreed upon, even among libertarians. The real question here is not whether someone should be compensated for wrongful imprisonment, but rather, how much someone should be compensated for wrongful imprisonment. And to answer this question, we must turn to economic analysis; we cannot rely on simple libertarian moral theory to give us an answer.

If you want to discuss the justification for compensation, I?m certainly willing to do so. But that is a separate question from the one I was addressing here.

Jonathan, Just as a

Jonathan,

Just as a clarification, my objection to your use of the word ?we? had nothing to do with Austrian economists vs. neoclassical economists, but rather with the notion that ?we? (plural individuals) can come up with a value of life. Thus, my question about the decision you would make given the $5 million hypothetical.

Well, neoclassical economists try to take people?s values as a given as best they can, and when possible, they do this by observing people?s actions. If I purchase an apple for $0.50, an observer knows that I must value this apple at least more than I value $0.50. This is revealed preference.

The value of a human life can be determined in slightly similar way. If I am willing to do an action that has a 0.1% risk of death in exchange for $5000, observers can safely say that that the value I place on my own life is around $5 million.

From a large sample of such values, we can calculate an average value of human life. Although this is not quite the same as a true market price, I think it is close enough to be reliable in a great many contexts.

Well, neoclassical

Well, neoclassical economists try to take people?s values as a given as best they can, and when possible, they do this by observing people?s actions. If I purchase an apple for $0.50, an observer knows that I must value this apple at least more than I value $0.50. This is revealed preference.

Yes, Austrians also use revealed preference in their methodology. I think this is the core issue about why so many people, including myself, disagree with your claim that the prisoner benefited from the room and board. He did not reveal his preference for it. He was not given an opportunity to.

The value of a human life can be determined in slightly similar way. If I am willing to do an action that has a 0.1% risk of death in exchange for $5000, observers can safely say that that the value I place on my own life is around $5 million.

I don?t think you can say that. What I conclude from that is that you are willing to perform an action with a 0.1% risk of death for $5,000. It very well may be the case that for $10,000 you would be willing to perform an action with a 5% risk of death (i.e., not 0.2%), and for $20,000 you would be willing to perform an action with a 50% risk of death, and for $30,000 you would be willing to perform an action with a 100% risk of death.

From a large sample of such values, we can calculate an average value of human life. Although this is not quite the same as a true market price, I think it is close enough to be reliable in a great many contexts.

But markets don?t work with averages, they work at the margin.

Jonathan, Are you really

Jonathan,

Are you really willing to argue that the prisoner would not have purchased food and rent had he not been convicted, assuming he was purchasing these goods right up until conviction? Would he have decided to become a vagrant and live on the streets? Or just commit suicide? Calculating the proper amount for damages requires an estimation of costs and benefits that cannot always be easily observed through direct action. If you think it is incorrect to assume that he benefited from room and board because we cannot observe his preference directly, then you must also think that it is incorrect to assume that he was harmed by being wrongfully convicted. Perhaps that is what he prefered, and he would have framed himself for a crime had he not been convicted anyway. To be consistent, you must treat costs in the same way you treat benefits. If one is impossible to calculate, then so is the other.

And while you are correct the one?s desired value can change depending on the risk of death, the information we have available is the best thing we have to work with. We can make all of the assumptions in the world, but at the end of the day, we must make a decision about what to do, and I don?t see any method better for decision making than using the information we already have available.

Are you really willing to

Are you really willing to argue that the prisoner would not have purchased food and rent had he not been convicted, assuming he was purchasing these goods right up until conviction?

How do you know he didn?t already own his house?

Even if he did not own his house, he would have had to choose among many alternatives of what kind of living accomodations he wanted to pay for. For example, today I used $2 to buy a bagel for lunch instead of, say, a slice of pizza for $2. Forcing me to eat pizza, which I did not value today, would not have benefitted me. By subtracting room and board costs from his settlement, you are saying that he would have chose to fork over his cash to live in a prison, i.e., those particular accommodations.

Would he have decided to become a vagrant and live on the streets? Or just commit suicide?

Some people do that. Not most people, or the ?average? person, but yes, some people do that.

Calculating the proper amount for damages requires an estimation of costs and benefits that cannot always be easily observed through direct action. If you think it is incorrect to assume that he benefited from room and board because we cannot observe his preference directly, then you must also think that it is incorrect to assume that he was harmed by being wrongfully convicted. Perhaps that is what he prefered, and he would have framed himself for a crime had he not been convicted anyway. To be consistent, you must treat costs in the same way you treat benefits. If one is impossible to calculate, then so is the other.

We don?t know what his preferences would have been. But we know that he has held against his will in prison. Otherwise, prisons would be voluntary. We know that for X amount of years, he was not allowed to live his life the way he wanted to.

And while you are correct the one?s desired value can change depending on the risk of death, the information we have available is the best thing we have to work with.

Extrapolating to that degree in other decisions can have disastrous consequences. If you estimate the temperature inside a gas tank at 200 atmospheres based on the temperature at 1 atmosphere, the tank will blow up.

We can make all of the assumptions in the world, but at the end of the day, we must make a decision about what to do, and I don?t see any method better for decision making than using the information we already have available.

I largely agree with you in general, believe it or not, and this kind of situation underscores the need for a free market in arbitration. Different arbitrators would make different rulings about what the prisoner deserves. These ?benefits? would simply be a good or service sold to the buying public. They can come up with whatever economic models and/or assumptions they want to. Buyers would help determine the price of each benefit package. But the key point is that these would be goods sold, much like insurance benefits.

The difference between that system and today?s system would be that the prisoner gets to at least have some say (via subscription prior to his arrest perhaps?) in who his arbitrator would be. The main objection I have against your post is the claim that the prisoner did absolutely benefit from room and board during imprisonment. That sounds a lot like an embrace of objective value.

How do you know he didn?t

How do you know he didn?t already own his house?

If he did, then you would not include free rent as a benefit.

Even if he did not own his house, he would have had to choose among many alternatives of what kind of living accomodations he wanted to pay for. For example, today I used $2 to buy a bagel for lunch instead of, say, a slice of pizza for $2. Forcing me to eat pizza, which I did not value today, would not have benefitted me. By subtracting room and board costs from his settlement, you are saying that he would have chose to fork over his cash to live in a prison, i.e., those particular accommodations.

Then the solution is simple: charge on the low end of what most people need to keep themselves alive. I suspect prison food is much cheaper than food free people normally purchase, since it is produced in mass quantities and doesn?t aim to please.

We don?t know what his preferences would have been. But we know that he has held against his will in prison. Otherwise, prisons would be voluntary. We know that for X amount of years, he was not allowed to live his life the way he wanted to.

Correct, but that is not the question here. The question is how much he should be compensated for the damages of wrongful conviction. If not by adding up the estimated costs and subtracting the estimated benefits, how would you propose we estimate the appropriate level of compensation? This is not a question of ethics; it is a question of method.

Extrapolating to that degree in other decisions can have disastrous consequences.

It sure can. But not having any information to go by whatsoever, and simply relying on random guesses can have even more disastrous consequences. We live in a world of imperfect information and we have to do the best we can with the information available to us.

The main objection I have against your post is the claim that the prisoner did absolutely benefit from room and board during imprisonment. That sounds a lot like an embrace of objective value.

I never made such a claim, and if I did make any claim that sounded like that, it was not my intent. I?m dealing with high probabilities, not absolutes. The vast majority of people, i.e. those who were not planning on either committing suicide or framing themselves to get into prison anyway, or those who already had their room and board paid for do in fact benefit from free room and board. Although this assumption may not always be the case, it is the case in the vast majority of real life situations, which is why it is more than fair to include it in a cost-benefit analysis of appropriate compensation, unless we have evidence to the contrary. So too, any costs incurred can only be imperfectly estimated, but we still believe that the wrongfully convicted should be compensated regardless of these imperfections.

Well, only two things come

Well, only two things come to mind. First is the old expression "too clever by half" and the second is: You fucktard.