Across 34 populations, energy burned differed far less than body fat did#
Is the rise in obesity driven more by eating more or by moving less? A large 2025 study in PNAS offers one of the most direct population-level answers so far. Researchers pooled doubly labeled water measurements, the gold standard for measuring calories burned in daily life, from 4,213 adults in 34 populations across six continents. The sample ranged from hunter-gatherers and pastoralists to farmers and people in industrialized cities. After accounting for body size, total daily energy expenditure was about 6% lower in the more economically developed populations, and basal expenditure about 11% lower; activity energy expenditure showed no significant change. That difference explained roughly one-tenth of their extra body fat1.
The authors' conclusion is that higher energy intake, rather than lower expenditure, is the main driver of obesity linked to economic development. They estimate intake has been "roughly 10 times more important." Keep one limit in view: this is a cross-sectional comparison of populations. Expenditure was measured; the role of intake was inferred from what expenditure could not explain. It is a strong clue about why whole societies gain weight. It doesn't tell any one person why they did.
For an individual, the mechanics are covered in our guide to calorie deficits. This piece is about population-level causes, a different question from whether you can outrun a bad diet.
What the study measured#
Doubly labeled water works by having a person drink water tagged with harmless stable isotopes of hydrogen and oxygen, then tracking how fast each leaves the body over a period of days. The difference gives carbon dioxide production, and from that, total calories burned. It captures everything: resting metabolism, digestion, exercise, and the fidgeting and walking of ordinary life. The data came from published studies and a large international research database of these measurements.
The team placed each population on an economic scale using the UN's Human Development Index. Nonindustrialized groups (hunter-gatherers such as the Hadza of Tanzania, pastoralists, horticulturalists) were assigned ranks by the researchers based on their market integration. They then compared body fat, BMI, and energy expenditure along that scale.
| Finding | What the data showed |
|---|---|
| Body fat and BMI | Higher in more economically developed populations |
| Absolute energy expenditure | Also higher in developed populations, because bodies were larger |
| Size-adjusted total and basal expenditure | About 6% (total) and 11% (basal) lower with development, varying widely |
| Share of extra body fat explained by lower expenditure | Roughly one-tenth |
| Ultraprocessed food share (25 populations with data) | Associated with body fat percentage |
The second row surprises people. People in industrialized countries burn more calories in absolute terms, because larger bodies cost more to run. Only after adjusting for body size does a gap appear, and it is small.
An older finding, now tested at scale#
The result extends a line of research that started with small studies.
In 2012, Herman Pontzer and colleagues measured the Hadza with doubly labeled water. As expected, their physical activity level, total energy spent relative to resting energy, was higher than that of Westerners. Yet their average daily energy expenditure was no different from Westerners' once body size was accounted for2. The authors proposed that daily energy expenditure might be a physiological trait largely independent of lifestyle. That idea was later developed as the constrained energy expenditure model.
Earlier still, Westerterp and Speakman looked at doubly labeled water data from Europe and North America going back to the 1980s. They found that physical activity energy expenditure had not declined over the decades when obesity rose sharply, and in Europe had slightly increased. They also found that daily energy expenditure in Europe and North America was not significantly different from that measured in developing countries. They concluded that decreased expenditure was unlikely to have fueled the obesity epidemic3.
The 2025 study's contribution is scale. Instead of one foraging group, or trends within rich countries, it covers 34 populations across the full economic range, with body fat measured directly rather than estimated from BMI.
In absolute terms, industrialized populations burn more calories than hunter-gatherers do. Lower size-adjusted expenditure explains only a small part of the gap in body fat.
What the study cannot tell you#
The limitations deserve their own section, because this study is easy to overread.
Intake was not measured directly. The authors state that they "lack detailed dietary data for most of the populations" and "cannot establish causality." Their intake conclusion comes mostly from subtraction: expenditure explains about a tenth of the body-fat difference, so the rest most plausibly comes from the other side of the energy balance equation. Measured total expenditure plus weight change also points to greater intake in developed populations. That reasoning is sound, but it is still an inference.
The ultraprocessed food link is correlational. In the 25 populations with data, the share of calories from ultraprocessed foods tracked body fat, even after accounting for energy expenditure, age, sex, and development. The authors say plainly that the mechanisms are unclear. Ultraprocessed food may be a cause, a marker of other dietary changes, or both.
Populations are not people. A finding that separates hunter-gatherers from city dwellers doesn't mean that, within your own life, a drop in activity couldn't be why you gained weight. The study compares groups; it doesn't diagnose individuals.
Adjusted expenditure varied a lot. Within each economic tier, populations differed widely, and differences in expenditure "did not correspond closely with lifestyle." That spread is part of the finding. It isn't noise to be averaged away.
Why exercise still matters#
It would be easy to turn this study into "exercise doesn't matter." The authors reject that reading directly: "Rather than advocating for diet over exercise in public health, data from this study join an emerging consensus that both must [be] prioritized."
Physical activity has well-established benefits for the heart, metabolism, and mental health that don't depend on how many calories it burns. And the study's own design, with body-size-adjusted expenditure varying widely within economic groups, is not evidence that activity is irrelevant to weight for every person.
What the study does challenge is the idea that the obesity gap between societies exists mainly because modern people have stopped moving. By the most direct measure available, modern people burn roughly as much energy as people in foraging and farming societies. They also eat more.
What this means for your own plan#
The following is practical interpretation, not a finding of the study.
- Intake is worth attention. At the population level, intake appears to explain more of the difference; for an individual, both sides of the calorie deficit arithmetic count.
- Keep exercising for the reasons it's good at. Fitness, strength, mood, and keeping weight off all benefit from activity even when the calorie math is unimpressive.
- Daily movement still counts. Everyday activity like walking and chores is part of the total the study measured. Our explainer on NEAT covers why it varies so much between people.
FAQ#
Do people in rich countries burn fewer calories than hunter-gatherers?#
In absolute terms, no. They burn more, because their bodies are larger. After adjusting for body size, total daily energy expenditure was about 6% lower (basal about 11%) in more economically developed populations, and that difference explained only about a tenth of their extra body fat1.
Has physical activity declined since the obesity epidemic began?#
Not in the doubly labeled water data analyzed in 2008. Physical activity energy expenditure in Europe and North America had not declined since the 1980s, and in Europe it rose slightly3.
Does this study prove ultraprocessed food causes obesity?#
No. In 25 populations, the share of calories from ultraprocessed food was associated with body fat, but the study is cross-sectional and the authors say they cannot establish causality or explain the mechanism.
Sources#
- McGrosky A, Luke A, Arab L, Bedu-Addo K, et al. Energy expenditure and obesity across the economic spectrum. Proc Natl Acad Sci U S A. 2025;122(29):e2420902122.
- Pontzer H, Raichlen DA, Wood BM, Mabulla AZ, et al. Hunter-gatherer energetics and human obesity. PLoS One. 2012. doi:10.1371/journal.pone.0040503
- Westerterp KR, Speakman JR. Physical activity energy expenditure has not declined since the 1980s and matches energy expenditures of wild mammals. Int J Obes (Lond). 2008. doi:10.1038/ijo.2008.74
Source: BurnWeek — "Why obesity differs across populations: the 2025 energy-expenditure study", https://burnweek.fit/blog/is-obesity-from-eating-more-or-moving-less/. Licensed CC BY 4.0: free to quote or reuse with a link to this page.
