The income graph strongly suggests to me that increased happiness scales with the proportionate increase in income, rather than the absolute increase in raw dol

The income graph strongly suggests to me that increased happiness scales with the proportionate increase in income, rather than the absolute increase in raw dollars. That is to say, a 20% raise will increase everybody's happiness roughly equally, regardless of the absolute magnitude of the raise (eg, $10k for someone earning $50k/year and $100k for someone earning $500k/year). This probably has to do less with the utility of the money itself and more with psychological positive reinforcement ("I'm smart and successful and my boss recognizes that") as well as increasing the person's relative social status. I've read that, given the choice, people will prefer to (for example) earn $70k while their peer group earns $50k, as opposed to earning $100k while the peer group earns $150k. Income is an indicator of social status and is largely valued relative to the rest of your social cohort. Interestingly, the Japanese are said to be more efficient because their corporations tend to award social status directly to "salarymen" (sic) rather than indirectly through differential wages.