What is your work that puts you in a position to observe statistically meaningful samples of both low and high income families?
Was this comment a response to my comment? If so, I tried to state that it was from my personal observations.
However, from a statistical point of view, it's heavily supported in survey research in psychology and sociology that income is negatively correlated with drug use. As income goes up, drug use typically goes down**. I think it would be fair to consider cigarettes and alcohol as luxuries.
**It's actually a curvilinear relationship, because once someone becomes super wealthy drug use starts to go back up again.
I have a Ph.D. in Applied Social Psychology and have worked on a few projects that measure drug use and income with survey data. Income was never a focal point of any papers because the data typically followed the normal trend and thus wasn't interesting.
When my dad was in his late 30's and early 40's he would spend about 25% of his tax home pay on alcohol, gambling and tobacco. He earned a good paycheck as a union construction worker. However, he was always broke and never saved.