It's an interesting paper but several things have to be kept in mind: 1. This is about cross-sectional studies. They mention a longitudinal multi-sample study

It's an interesting paper but several things have to be kept in mind: 1. This is about cross-sectional studies. They mention a longitudinal multi-sample study which came to slightly different conclusions, that peoples' happiness tends to increase in general with age. It's so important to examine this longitudinally, because there's so many things that could be going on, and it's reasonable to hypothesize that there's some kind of cohort effects going on. In could be that people in their 40s-50s are being subject to some unique combination of issues due to societal problems and patterns (e.g., that people who are reaching professional maturity are bumping up against dysfunction that the older generations are unaware of because they benefited from different socioeconomic environments in a way they don't recognize, and the younger generations aren't quite subject to the same dysfunction because they haven't reached it). But it could be teenagers in the home, becoming disillusioned with life and then learning to cope with it, hormones, who knows. 2. Strangely enough, happiness and stress aren't really totally inverses of one another. So it's possible your stress / anxiety / etc. will decrease and your happiness might also a bit. Think of it this way: it's possible to not really feel angry or fearful or depressed, but also not entirely excited or fulfilled either. Related to this are the fact that certain types of problems are definitely more likely in youth, especially those related to drugs, impulsive behavior, etc. The decline in those types of problems is much better documented and understood than these life satisfaction patterns (I actually think that a huge part of world conflicts can be accounted for by 15-25 year old males). 3. These are average curves; everyone's lives are different. I can't emphasize this enough, speaking as a researcher in this area, roughly speaking.