I personally think these are the cases of over-fitting. You can come up with a higher order polynomial that fits the observed noise but it comes at the cost of losing precision for the yet to be observed data.
In case of government regulation, each regulation ends up imposing significant cost on the entire society to prevent handful of deaths which is a massive waste of resources. I am pretty sure destruction wealth also results in some human misery.
Consider USA:
Total deaths caused by unintentional injury in 2014 for children below 14 is 3857 [1]. For a population of 300M it is pretty insignificant. Even if we just look at population below the age of 14 (30M) it is merely 0.00012856666. This is just pure noise.
Sure you can come up with a regulation that imposes a very heavy cost on entire society to eliminate one cause of unintentional injury but I really wonder if it has any impact on the overall number at all.
Unfortunately I don't have more data to backup this claim but if I had to put my money on whether these rules will bring down child mortality I would bet it wont make a difference.
I also think that child mortality will go down over time even if we don't pass a lot of laws and people themselves will figure out solutions to avoid these deaths. A good data point to verify is this OSHA's impact on worksplace safety in USA.
I turns out that workplace safety was getting better even before OSHA and the giant regulation made absolutely no difference what so ever to the decline in worksplace deaths but put a significant cost on businesses. [2]
In case of government regulation, each regulation ends up imposing significant cost on the entire society to prevent handful of deaths which is a massive waste of resources. I am pretty sure destruction wealth also results in some human misery.
Consider USA:
Total deaths caused by unintentional injury in 2014 for children below 14 is 3857 [1]. For a population of 300M it is pretty insignificant. Even if we just look at population below the age of 14 (30M) it is merely 0.00012856666. This is just pure noise.
Sure you can come up with a regulation that imposes a very heavy cost on entire society to eliminate one cause of unintentional injury but I really wonder if it has any impact on the overall number at all.
Unfortunately I don't have more data to backup this claim but if I had to put my money on whether these rules will bring down child mortality I would bet it wont make a difference.
I also think that child mortality will go down over time even if we don't pass a lot of laws and people themselves will figure out solutions to avoid these deaths. A good data point to verify is this OSHA's impact on worksplace safety in USA.
I turns out that workplace safety was getting better even before OSHA and the giant regulation made absolutely no difference what so ever to the decline in worksplace deaths but put a significant cost on businesses. [2]
[1] http://www.cdc.gov/injury/images/lc-charts/leading_causes_of... [2] http://mercatus.org/sites/default/files/workplacefatalities5...