It’s tempting to assume retirement satisfaction is mostly a math problem: save enough, and the happiness follows. The actual research is more layered than that, and it’s worth being honest that wealth doesn’t simply stop mattering the way a tidy headline might suggest. A 2024 analysis by the Employee Benefit Research Institute, using data from the nationally representative Health and Retirement Study, found that 73 percent of retirees with $500,000 or more in household assets reported being very satisfied with retirement, compared with just 33 percent of those with under $25,000. That’s a real, substantial gap, and it doesn’t close entirely even among people with similar health: retirees in fair or poor health who had $500,000 or more in assets were still more than twice as likely to report high satisfaction as equally unhealthy retirees with under $25,000. Wealth keeps correlating with satisfaction well past the point where a person is simply “not poor.”
What does show a different pattern is a more specific kind of money: guaranteed, protected income, as opposed to total net worth sitting in an account. A white paper by retirement researchers Wade Pfau and Michael Finke, drawing on data from roughly 20,000 older Americans in the same Health and Retirement Study, found that satisfaction jumped sharply as guaranteed monthly income, from Social Security, a pension, or an annuity, rose toward about $3,000 a month in today’s dollars. Past that threshold, additional guaranteed income kept helping, but the size of the effect shrank noticeably, and this held true across nearly every wealth tier, including among people who were already high-net-worth. Having enough steady, reliable income to comfortably cover the basics and then some appears to matter more for peace of mind than the size of the number in a brokerage account, even though that same brokerage account balance still correlates with overall satisfaction in the broader EBRI data. The two findings aren’t contradictory so much as measuring different things: how much money exists, and how confidently a person can count on some of it showing up every month no matter what.
Once income adequacy is accounted for, the research points toward two other factors doing a lot of the remaining work: health and relationships. The EBRI analysis found health was, if anything, at least as strong a predictor of retirement satisfaction as wealth. Seventy percent of retirees who rated their own health as excellent or very good reported high satisfaction, compared with only 28 percent of those in fair or poor health, a gap similar in size to the one seen across wealth tiers. And even within the same wealth bracket, healthier retirees were consistently more satisfied than less healthy ones with identical assets.
The relationship piece comes from a different, much longer-running body of research: the Harvard Study of Adult Development, an ongoing study that has followed hundreds of participants and, in later phases, their spouses, since 1938, making it one of the longest-running studies of adult life anywhere. Its director, psychiatrist Robert Waldinger, has described one of the study’s most striking findings this way: how satisfied participants were with their close relationships at age 50 predicted their physical health decades later better than their cholesterol levels did at the same age. A related study led by Waldinger and Marc Schulz, examining older married couples day by day, found that time spent with a partner was linked to greater daily happiness, and that this link was strongly shaped by how satisfying the marriage actually was, not simply by how much time the couple spent together. Quality of connection, not just its presence, was doing the work.
None of this research proves that money is unimportant, and none of the researchers behind it make that claim. What it supports is something more specific: financial security up to a reasonably comfortable, reliably steady level appears to matter enormously, and beyond that point, the returns on additional wealth for retirement satisfaction seem to taper in a way that health and relationship quality generally don’t. It’s also worth noting that all of this research is correlational. People in better health may find it easier to maintain close relationships, and people with warmer relationships may take better care of their health, so untangling exactly which factor is doing the causing, and how much, from population-level survey data remains genuinely difficult.
The practical upshot isn’t that saving for retirement doesn’t matter. It’s that the research consistently finds a ceiling on what additional wealth alone buys, while investments in physical health and the quality of a marriage or close friendships don’t show that same ceiling in the data. A retirement plan built entirely around a number in an account, without much thought given to health habits or the people a person expects to spend that time with, is missing the two variables the research keeps identifying as doing just as much, if not more, of the real work.
