What are you talking about? The article said there were 400,000 opiate addicts in 1880 due to addicted Civil War veterans. That works out to 0.8% (population 50,000,000). By 1900, only 0.5% were addicted to either cocaine or opiates. That means there was a substantial fall in the addiction rate while drugs were still freely available. I'm not sure where you get that 1900 was a high.
Those laws delivered an almost immediate decline in addiction which lasted all the way into the 50's.
How do you know this was a causal relationship? Alcohol prohibition didn't reduce drinking for very long. By the time it was repealed, consumption was near pre-prohibition levels.
What one endorses, one will get more of - Endorse vice, get more vice. Endorse virtue, get more virtue.
Doesn't seem to work out that way when you look at addiction figures for Singapore, the US, Iran, and the Netherlands. And the addiction rate now is triple what it was in 1900, according to the DOJ.
400,000 of 50m (1880) vs. 380,000 of 76m (1900)
Several things of note...
Firstly, the 1880 census specifically states that it omits the Chinese and American Indians completely. It attempts to calculate Chinese and Indians living among whites, placing the number at some 150,000, though it is my opinion that number is grossly underestimated, especially among the Chinese.
Considering the vast establishment of Chinese industry (fishing, fish processing, wineries, laundering, road building, every form of labor, etc) in California, and the fear of the "Yellow Tide" which prompted the Chinese Exclusionary Act of '82, after which another 300,000 were still allowed to immigrate by 1900, I would consider the number to be much higher than the census claims. This is a critical population when considering addiction, especially to opiates, and the distribution thereof, particularly from the west.
Even if the numbers are correct, if Asians are not included in the addiction rates, the rates are sure to be skewed favorably.
**Note that I am not attempting to blame the entirety of opiate use/distribution upon Asians, but they certainly own the western distribution routes, and high addiction rates among Asians are undeniable.
Secondly, While the general population grew immensely (1880-1900), one must also consider the distribution capabilities, and where an addict might support his habit - This renders large swaths of the continent invalid, as timely delivery was simply not to be had, and the expense could not be paid.
To get a reasonable statistical analysis, I would imagine one would have to break down populations according to where delivery was easily available - Seacoasts, close proximity to railroads, main roads, or major rivers, and then apply that to the addiction rate... A near impossibility for me to do - Especially when considering that all records must be weighted for a nation in tremendous flux:
We are talking about the period directly involving the transcontinental railroad and the civil war. How any records could be remotely accurate would be a wonder, indeed.
And lastly comes the awareness of addiction in the medical sense, and in the public mind. We are dealing with a period of time during, or just leading up to that awareness. We are barely into the use of hypodermic needles here (mid 1800's).
It takes a while for addiction to rear it's head as a public phenomena beyond the confines of the red light district. IFAIR, the Civil War was the first broad use of morphine as medicine (actual pain relief for wounds), at least in the US.
I am much more comfortable looking at time after the knowledge of addiction is in place, and the present distribution system is reasonably accounted for. That need not include the interstate highway system, but would needfully include expanded railways, expanded road systems capable of automotive traffic, and maybe even air service.
That puts us firmly in the industrial age rather than comparing agrarian and industrial norms. It also places us past all racial disputes and into a more efficient census system, with better over-all statistical analysis.
Otherwise, one might be better served using a single coastal metropolitan area as a statistical norm, or comparing and averaging several to get the needed data.