What the data actually shows
The clearest pattern is about share of income. Multiple analyses of charitable-giving data have found that lower-income households tend to donate a larger percentage of their income than higher-income households, even as the wealthy give far more in raw dollars. This 'giving as a share of income' gap is the most robust strand of evidence behind the idea that the less affluent can be proportionally more generous — though giving patterns vary a great deal by what and how people give.
The more provocative claim comes from social psychology. A widely cited series of studies by Paul Piff and colleagues reported that higher-social-class individuals behaved less generously and less ethically in various lab and field measures — for example, being less likely to yield to pedestrians or behaving more self-interestedly in economic games. This research was influential in popularising the 'higher status, lower prosociality' narrative.
Crucially, that work is contested. Some of these findings have faced difficulties on replication, later studies have produced mixed or null results, and critics have questioned the measures and interpretation. So while the share-of-income gap is reasonably well documented, the stronger claim that wealth makes people behave less generously or ethically in a general sense is not settled — it has some support and meaningful pushback.
Why this feels different from how it actually is
The 'stingy rich' story is satisfying, which is part of why it spreads faster than the evidence warrants. It fits a moral intuition and offers a comforting narrative to people with less money, so the contested lab findings get repeated as established fact while the caveats and failed replications drop away. A neat story tends to outrun a messy one.
It also feels clear-cut because the two measures that disagree are rarely held side by side. 'The poor give a bigger share of their income' and 'the rich give vastly more dollars and fund most large philanthropy' are both true, and which one you reach for shapes the whole conclusion. People tend to pick the measure that confirms what they already believe and call it settled.
And generosity is hard to observe directly, so we judge it by visible cues — a large public donation, or a small unrecorded act of help — that capture very different things. The wealthy person's giving is often formal, deferred, and tax-structured; the lower-income person's is often informal and immediate. Comparing them as if they were the same act produces a misleadingly tidy verdict.
A neat story tends to outrun a messy one — the contested lab findings get repeated as fact while the caveats drop away.
What the research says to do about it
Because this is a contested empirical question rather than a personal decision, the most useful response is mainly about how you read the claim. Treat 'rich people are less generous' as a debated finding with partial support, not an established fact — note which measure any given claim relies on (share of income, absolute dollars, or a lab task), since they genuinely point in different directions.
Be especially cautious with the lab-based 'higher status, lower ethics' results given the replication difficulties around them. The honest stance is that these studies raised a real hypothesis that has not held up cleanly, so they are best cited as 'contested' rather than as proof of how wealthy people behave.
If the underlying interest is your own giving, the research that is solid points to share of income as a more meaningful gauge of personal generosity than absolute amount — but this is a values question, not something the data can settle for you. The evidence describes group-level patterns, not a verdict on any individual.
What the research says does not help
Citing the 'stingy rich' studies as settled science does not help, because the most famous of them have faced replication challenges and mixed follow-ups. Treating a contested finding as proven is exactly the kind of overreach this site exists to push back on.
Equally, dismissing the whole question as a myth overcorrects. The share-of-income gap is reasonably well documented, and the lab research, whatever its problems, raised a real and still-open hypothesis. The honest position lives between the two confident extremes.
Drawing conclusions about any individual from these group-level, disputed averages does not help at all. Even where a population pattern exists, it says nothing reliable about whether a particular wealthy or lower-income person is generous — the variation between individuals dwarfs any class-level tendency.
Even where a population pattern holds, it cannot predict whether any particular person is generous.
What this looks like in real life
Share of income vs absolute dollars
Ask 'who gives more?' and the answer depends entirely on the measure. Lower-income households tend to donate a larger percentage of their income, while wealthier households give far more in total dollars and fund most large philanthropy. Both statements are true — which is precisely why the question feels contradictory, and why people tend to reach for whichever measure fits what they already believe.
'The stingy rich' as contested, not settled
When you meet the confident version — 'research proves rich people are less generous' — the honest move is to note which measure it leans on and to treat the lab-based findings as contested rather than proven. The Piff and colleagues studies raised a real hypothesis, but replication difficulties and mixed results mean they are best cited as an open question, not a verdict on how wealthy people behave.
Real numbers in context
The most dependable figure-based pattern is about proportion, not amount: analyses of giving data repeatedly find that lower-income households tend to donate a higher percentage of their income to charity than wealthier ones, even though wealthier households give far more in total dollars and account for the bulk of large philanthropy. Both statements are true at once, which is precisely why the question feels contradictory.
The psychological side resists clean numbers and should be hedged hard. The Piff and colleagues studies reported that higher-status participants behaved less generously or prosocially in specific tasks, but these effects were modest, context-dependent, and have faced replication difficulties. So the reliable takeaway is qualitative: there is some evidence wealth can dampen certain forms of generosity, the share-of-income gap is the strongest part of it, and the broader 'rich are less generous' claim remains genuinely disputed.