Recently, California passed a new law requiring all homes to have Carbon Monoxide detectors . Practically, this means that if you need any work done on your home, you have to have proof of a carbon monoxide detector or you can’t get the permits. There may also be fines involved, although it’s surprisingly hard to find info on the specifics of enforcement.
Hmmm… seems a bit paternalistic. Let’s do some math and see if the benefits are so great as to justify the governments’ infringement on our choices.
According to the CDC, the death rate from carbon monoxide poisoning in California is .57 deaths per million residents . In other words for every two million people, one of them will die each year from carbon monoxide poisoning. The cost of a standard carbon monoxide detector runs between about $25 and $50. Let’s take the cheapest detector, and also assume that you only need one per home (as opposed to several in a large house). The average household size according to the census bureau is about 2 and a half. The best case estimate of the life of a carbon monoxide detector is about 10 years after which point it needs to be replaced. Doing the math, that means that the cost/benefit analysis of putting a carbon monoxide detector in each household is $2,000,000 per life saved. (and that’s a REALLY conservative estimate).
So, let’s put that in perspective. According to AIDemocracy, for less than $4 you can save a life in developing countries. (I’ve seen other estimates that you can save a life for as little as a dollar, but I think $4 is a bit more realistic and I’m trying to stay conservative with my estimates). So, the cost/benefit ratios suggest that for each life we save installing carbon monoxide detectors in every household in California, we could save 500,000 lives in the third world.
But maybe California doesn’t care about lives outside of California. Fine. We can look at what it would cost to save lives in other ways. This article provides a cost/benefit breakdown of 500 different interventions. The average cost per life saved is $42,000 – meaning that you could save 47 lives for every one you save with the carbon monoxide detectors. But, of course, that average includes a number of extremely inefficient interventions. The median medical intervention costs less than $20,000 per life saved. In other words, we could save about 100 lives for every life saved for the cost of carbon monoxide detectors if we put that money into medical care.
I’m not going to belabor you with the math here, but if we switch from deaths to hospitalizations as our unit of analysis, the numbers change a bit, but the policy remains incredibly inefficient compared to other interventions.
I’m not saying you shouldn’t buy a carbon monoxide detector if you want to. If you have the money, and it’s worth it to you to reduce that very, very small risk of carbon monoxide death, then more power to you. For me, I get a free carbon monoxide detector from my landlord, but it sits unused in my closet because I find the detector of such little value that I’d rather have access to the electrical outlet that it would prevent me from using.
It bothers me that the government would step in and force me to spend my money (and electrical outlets, if I actually had it plugged in) in such an inefficient way. Life has risks. I could spend all my resources trying to avoid those risks, but 1) I still wouldn’t be able to avoid them all and 2) that would be a pretty unpleasant life. Every individual should be able to determine what an acceptable level of risk is, and live their life accordingly. I chafe at government making that choice for me, even when the government’s choice is reasonable. This is not one of those reasonable cases.
Even if you are correct about death rates and intervention costs, you glossed over the cost of presentations at Emergency Departments and the cost of lost productivity to the economy caused by CO poisoning.
Here are some quick numbers I dug up:
- In 2005, poisonings led to $33.4 billion in medical and productivity costs
- An average of 15,200 people suffered accidental CO poisoning from 2001 through 2003
- There were 164,480 non-fatal poisonings treated in Emergency Departments in 2000
A quick bit of math using the data above suggests 9.24% of accidental poisonings treated at EDs are for CO poisoning.
Assuming that all types of poisonings incur the same cost – that would imply a total cost to the economy of 9.24% of $33.4 billion = $3.08 billion per year from CO poisoning.
Now I’ll admit these are very rough numbers, comparing different years, making assumptions etc, and it doesn’t answer the question of how much of that $3 billion in cost would be saved due to mandatory CO detectors, but it does show that the assumption that the cost of hospitalizations is unimportant, is not necessarily a correct one.
I’d be interested if you could ‘belabor’ us with your numbers.
(Disclosure – I do have a personal interest in CO detectors via this website: http://www.carbonmonoxidedetectorplacement.com)
Hey Malcom, I’m on the road and can’t reply as thoroughly as I’d like, but two quick things. First, your numbers are national, and mine for from California alone. So, that’s why the costs you’re raising seem very high in comparison to my numbers. As a quick fix, we can divide your numbers by 7, which isn’t perfect, but does get us closer to the same set of numbers.
Second, I wasn’t intending to say that hospitalizations weren’t important, but rather that the utility calculus for hospitalizations looks very similar to that for deaths. That is, for the same cost of installing CO detectors in your house, are there other things you could have done with that money that would decrease your likelihood of hospitalization even more? So, are you better off spending your $25 on a CO detector or, for example, a bike helmet? You can look at deaths (as I did) or you can look at deaths + hospitalizations. But you have to add hospitalization costs not just for CO poisoning, but for comparison safety measures as well. And when you do that, everything moves up in cost, and things move around a bit in ordinal ranking, but not really all that much. Certainly not enough to justify legally privileging CO detectors over the 400 other more cost efficient ways of reducing injury/death.
P.S. Bike helmets are also a rather inefficient safety measure for the casual cyclist. I’m not any happier about government mandates about those than I am about CO detectors.