There are entire fields of medicine, including psychiatry and psychology, that believe surveying people on how depressed, anxious, etc. they feel is the scientific basis of research, diagnosis, and treatment.
Your idea to just ignore what people say is just totally out there...
I don't disagree that self-reported happiness may be meaningless, but I strongly disagree with any implication that experts have some reference point to evaluate objective truth about mental states.
Psychiatric drugs are developed by surveying how patients feel. If there was some source of objective truth, based on understanding the mechanisms behind illness and medications, they'd use that.
Common sense tells you people are complex and unreliable, but that difficulty doesn't give experts a crystal ball to replace self-reporting.
You're raising a point that has no relevance to the conversation. Neither economists nor epidemiologists rely on self reporting. In the face of a large scale disaster, such as the rise in drug overdoses, self-reporting is meaningless and can be ignored. Economists have other ways of estimating the overall experience of the public, mostly looking at aggregate numbers that point back to some real experience.
Put a bit differently, if you know someone is an alcoholic, and they tell you that they are happy, you would be a complete idiot to believe them.
A poll saying that Americans are happy has a certain limited value that needs to be balanced against all of the other behaviors that Americans are currently exhibiting. If the average life span is falling because of excess "deaths of despair" that is a very large fact that you would be foolish to ignore.
"that difficulty doesn't give experts a crystal ball to replace self-reporting"
The entire profession of economics has devoted the last century to developing various aggregate numbers that give some reliable picture of the mood of the country -- reliable in the sense that investors can make actual decisions based on those numbers, with the results being good more often that bad.
Americans telling a pollster that they are happy is a fact that needs to be balanced against other facts, such as the decline in the formation of new firms, the spike in gun violence, the spike in drug overdoses, etc. Compared to people's real actions, a poll means very little.
It's amazing to me that such an obvious point should get so much push back on Hacker News.
Your idea to just ignore what people say is just totally out there...