GM, Product Safety, and Death

[Occam's Banana]

GM, Product Safety, and Death
http://mises.ca/posts/blog/gm-product-safety-and-death/
Bob Murphy (05 April 2014)

The reports that GM deliberately avoided a recall for a defective ignition switch – which led to at least 13 fatalities – because of cost considerations has led to predictable denunciations of capitalist greed. Michael Moore epitomizes this reaction:

I am opposed to the death penalty, but to every rule there is usually an exception, and in this case I hope the criminals at General Motors will be arrested and made to pay for their pre-meditated decision to take human lives for a lousy ten bucks. [...]​

In this post, I do not want to discuss the particulars of the GM case. Rather, I want to explore the notion of product safety more generally. Even though Moore’s reaction probably captures how most people emotionally respond to such situations, economics offers us a more nuanced understanding.

[...]

Suppose years ago when they discovered the problem, GM had contacted all of the 2.6 million customers driving the vehicles with a faulty switch and told them, “We think 13 of you will end up dying because of the faulty switch. The problem is, we don’t know which 13 – if we did, we’d obviously fix your vehicles. Although the part itself is under $1, there is the labor involved, plus many of you will want a loaner while we fix the vehicle. All told, it would probably cost us about $50 per each of you to replace all 2.6 million switches. So we’ll give you all a choice: We can fix your switch, OR we can mail you a check for $40, and promise that if you die because of the faulty switch, we will pay your estate $1 million.”

I have no idea how many people would take that offer. I’m guessing out of 2.6 million, most would elect to get the switch fixed, but a good many would take the $40 and promise of indemnification. Certainly if we changed the numbers around – so that people got, say, a $400 check – the balance would tilt in favor of the “live with it” outcome.

The reason I have framed this hypothetical is that I’m trying to show that the underlying logic – that in situations where there is a very small chance of death – it really might make economic sense to not do a recall. In my scenario above, if a particular customer had elected to take the $40 rather than get the switch fixed, and then that person happened to die – with his estate then getting $1 million further compensation – people like Michael Moore could hardly complain about the absurdity of neglecting a repair because of cost. In such a hypothetical scenario, the victim himself (or herself) would have chosen to bear the risk in order to save on the repair cost.

Of course, one enormous difference between reality and my hypothetical scenario is that GM didn’t offer consumers this choice. [...]

In conclusion, I am NOT defending the actions of GM. Indeed, had policymakers taken my advice, GM would have gone under; I was totally opposed to the US government’s bailout of the automakers. (The joke at the time was calling it Government Motors.) Rather, I am trying to show that there is nothing evil about a company using cost/benefit analysis to guide its decisions, even when it involves defective products that can kill people. Taking Michael Moore’s logic to the extreme, companies would stop selling food (because it might be poisonous) and no one could ever fly again. The way to reform the system and lead to genuinely sensible decisions is for government to get out of the way, so that true market prices can guide cost/benefit tests.

[Full article at link: http://mises.ca/posts/blog/gm-product-safety-and-death/ ]

 

[Brian4Liberty]

Taking Michael Moore’s logic to the extreme, companies would stop selling food (because it might be poisonous) and no one could ever fly again.

Bad analogy. The risk of fatal accidents with a product is not the same as a known flaw with the basic functionality, which could result in death. The first is an unknown. The second is a negligent liability. It's known to be defective. The fact that it would have been a cheap and easy fix just makes it sound more outrageous, but the cost of the fix doesn't matter.

And in practice, people did take theses damn vehicles in for repair for this exact problem, and the known fix was not implemented, because they were covering it up! They sabotaged their own repair process.

 

[Occam's Banana]

Bad analogy. The risk of fatal accidents with a product is not the same as a known flaw with the basic functionality, which could result in death. The first is an unknown. The second is a negligent liability. It's known to be defective.

This misses the point of Murphy's analogy, which doesn't have anything to do with "known" risks versus "unknown" risks. (There is no way to incorporate unknown risks into cost-benefit analysis - precisely because they are unknown.) Murphy is talking about known risks of any kind - including (but not limited to) risks that exist because of product defects.

Murphy examined a hypothetical arrangement in which a known risk (arising from a defective part, in his fictive scenario) could be accounted for in cost-benefit analysis - and he proposes that this could be done without making such a huge & vicious botch of things as GM has done.

On the other hand, if we "tak[e] Michael Moore's logic to the extreme" (as Murphy puts it), cost-benefit analysis would NOT be allowed to have any place at all in business decsions when human lives might be involved, period - regardless of whether or not the source of a "known" risk is something like a product defect, or just the sort of risk that unavoidably comes with a potentially dangerous activity (such as flying or whatever). That was the point of Murphy's analogy (although he could have stated it better in order to make this more clear).

The fact that it would have been a cheap and easy fix just makes it sound more outrageous, but the cost of the fix doesn't matter.

The cost of the fix DOES matter - in fact, it's probably the biggest reason why GM went to such deceptive lengths to cover up the problem in the first place. They were wrong to do so and should be held liable for it - but that is an entirely separate issue from what Murphy is concerned with (which is whether there is a proper role for cost-benefit analysis even when "human lives are at stake").

And in practice, people did take theses damn vehicles in for repair for this exact problem, and the known fix was not implemented, because they were covering it up! They sabotaged their own repair process.

What has this got to do with anything Murphy said? Murphy never said anything to the contrary. As he explicitly stated, he isn't talking about what GM did in this particular case - and he certainly isn't defending how they handled it. GM lied and engaged in a systematic "cover up" - but that's NOT what Murphy is talking about. Moore wants to hitch GM's lies & cover-up with cost-benefit analysis. Murphy is trying to unhitch them. He isn't defending what GM did. Quite the contrary: he tried to show how such a situation could have and should have been handled differently - and that (contrary to Moore) cost-benefit analysis would play an important part in such an alternate scenario.

 

[donnay]

Good article. Yes, government needs to get out of the car business, indeed. But imagine my shock to find out they covered-up a problem.

 

[angelatc]

I am surprised that nobody (that I've seen yet) has mentioned Adam Smith's invisible hand.

 

[LibForestPaul]

I produce a product. I believe a part is defective and could cause harm and bodily injury. I cover the implemented fix. And...
Negligent homicide, no? Who's going to jail?

I do not want to fix the part. Yet I notify others of the problem. They have it fixed and repaired on their own dime, and send me the bill or sue for damages. They do not fix it, they die, up to them and their insurer?

 

[LibertyEagle]

This misses the point of Murphy's analogy, which doesn't have anything to do with "known" risks versus "unknown" risks. (There is no way to incorporate unknown risks into cost-benefit analysis - precisely because they are unknown.) Murphy is talking about known risks of any kind - including (but not limited to) risks that exist because of product defects.

Murphy examined a hypothetical arrangement in which a known risk (arising from a defective part, in his fictive scenario) could be accounted for in cost-benefit analysis - and he proposes that this could be done without making such a huge & vicious botch of things as GM has done.

On the other hand, if we "tak[e] Michael Moore's logic to the extreme" (as Murphy puts it), cost-benefit analysis would NOT be allowed to have any place at all in business decsions when human lives might be involved, period - regardless of whether or not the source of a "known" risk is something like a product defect, or just the sort of risk that unavoidably comes with a potentially dangerous activity (such as flying or whatever). That was the point of Murphy's analogy (although he could have stated it better in order to make this more clear).



The cost of the fix DOES matter - in fact, it's probably the biggest reason why GM went to such deceptive lengths to cover up the problem in the first place. They were wrong to do so and should be held liable for it - but that is an entirely separate issue from what Murphy is concerned with (which is whether there is a proper role for cost-benefit analysis even when "human lives are at stake").



What has this got to do with anything Murphy said? Murphy never said anything to the contrary. As he explicitly stated, he isn't talking about what GM did in this particular case - and he certainly isn't defending how they handled it. GM lied and engaged in a systematic "cover up" - but that's NOT what Murphy is talking about. Moore wants to hitch GM's lies & cover-up with cost-benefit analysis. Murphy is trying to unhitch them. He isn't defending what GM did. Quite the contrary: he tried to show how such a situation could have and should have been handled differently - and that (contrary to Moore) cost-benefit analysis would play an important part in such an alternate scenario.

Yeah, but the way that Murphy worded the whole thing it sounds too much like it's ok with him if a company knows of a defect that most assuredly would kill people, but chose not to fix it, because of cost. That doesn't exactly make us look too swift. Ya know?

It would have come out better if he had condemned them for their decision and suggested that the free market should and would reprimand them, all without government, by suing their asses into bankruptcy for their negligence.

 

[Anti Federalist]

It would have come out better if he had condemned them for their decision and suggested that the free market should and would reprimand them, all without government, by suing their asses into bankruptcy for their negligence.

Agreed...Eric Peters made the same argument, with which I agree.

http://www.ronpaulforums.com/showth...nvestigation-and-multi-million-vehicle-recall

 

[Brian4Liberty]

On the other hand, if we "tak[e] Michael Moore's logic to the extreme" (as Murphy puts it), cost-benefit analysis would NOT be allowed to have any place at all in business decsions when human lives might be involved, period - regardless of whether or not the source of a "known" risk is something like a product defect, or just the sort of risk that unavoidably comes with a potentially dangerous activity (such as flying or whatever). That was the point of Murphy's analogy (although he could have stated it better in order to make this more clear).

You seem to have missed the entire point of my post, which directly quoted his food and flying analogy. The risk of flying or eating is not the same thing as known defects in products. Selling food in general is not the same as selling food contaminated with e-coil and hoping that everyone washes and thoroughly cooks the found product before consuming, and if not, estimating that most people will only get sick and not die anyway.

The inherent risk of selling food and eating it is not the same thing as selling contaminated food.

 

[Zippyjuan]

In a totally free market- should companies be able to make and sell whatever products however they want to (good or bad) and if you feel yours is defective force you to file suit against them to try to prove they knowingly wronged you? (In this totally free market they are not required to inform anybody of any problems they become aware of). Will the market solve the problem of defects?

 

[Occam's Banana]

Yeah, but the way that Murphy worded the whole thing it sounds too much like it's ok with him if a company knows of a defect that most assuredly would kill people, but chose not to fix it, because of cost. That doesn't exactly make us look too swift. Ya know?

I really don't see how anyone can come to the conclusion that it "sounds" like Murphy is saying that it is OK for a company not to offer to fix a deadly defect. His hypothetical does NOT involve a company just choosing not to fix a defect. The option of getting a defect fixed was clearly a major & central part of the scenario he offered. He even repeated it several times ("we’d obviously fix your vehicles" - "while we fix the vehicle" - "We can fix your switch, OR ..." - etc.). I honestly don't understand how anyone can actually read his article and come away with any impression to the contrary.

It would have come out better if he had condemned them for their decision and suggested that the free market should and would reprimand them, all without government, by suing their asses into bankruptcy for their negligence.

But his purpose was NOT to address what GM actually decided to do - or what the free market should do about it after the fact. His point in citing the GM fiasco (in conjunction with Michael Moore's statement on the matter) was to use it as a springboard for his actual purpose. This is another thing he made very clear. Here are Murphy's own words again: "In this post, I do not want to discuss the particulars of the GM case. Rather, I want to explore the notion of product safety more generally." He was trying to make a case for what a company in GM's position could have (or should have) been able to do about the problem of deadly defects on the basis of cost-benefit analysis in order to avoid the need for "across the board" recalls.

 

[Occam's Banana]

You seem to have missed the entire point of my post, which directly quoted his food and flying analogy. The risk of flying or eating is not the same thing as known defects in products. Selling food in general is not the same as selling food contaminated with e-coil and hoping that everyone washes and thoroughly cooks the found product before consuming, and if not, estimating that most people will only get sick and not die anyway.

The inherent risk of selling food and eating it is not the same thing as selling contaminated food.

No, I understood the point of your post. I know exactly what you are getting at - and I do not disagree with it, as far as it goes. But I was pointing out that that was NOT the intended sense of Murphy's analogy. As I said, he could have stated it differently in order to avoid the very confusion caused by the interpretation of his analogy that you offer. I think it would have been better if he had left food out of it altogether and stuck to things that have some inherently unavoidable and (measurably) knowable risk, such as flying. (His point does apply to "defective" food for which a "fatality risk" can be known & measured - he isn't talking about generic "eating" risks associated with "selling food in general" - but the issue with respect to food is complicated by, among other things, the fact that purchasers of food are not nearly as easy to identify as purchasers of cars or airplane tickets, as well as the fact that food "defects" can more easily originate from elsewhere than the producer.)