Yes — and the best-controlled figure is 116 calories a day#
Fiber does reduce the energy you extract from your food, the effect has been measured under conditions tight enough to trust, and it is smaller than the internet implies and stranger than the average suggests. Seventeen healthy, weight-stable adults each completed two 12-day stays in a metabolic ward, eating either a Western Diet or a Microbiome Enhancer Diet built around dietary fiber, resistant starch, large food particle size and minimal processing. The two diets were matched for metabolizable energy and total macronutrients; everything eaten, and everything excreted, was collected and measured.
On the fiber-rich diet participants lost an additional 116 ± 56 kcal per day in feces (P < 0.0001), and the share of intake that reached them as usable energy fell from 95.4 ± 0.21% to 89.5 ± 0.73% — a drop of 5.9 percentage points. Energy expenditure, hunger, satiety and food intake did not change1. That last clause is what makes the result clean: nothing else moved, so the 116 calories are an absorption effect and not an appetite effect wearing its coat.
The average is the least interesting number in the study#
Here is the finding that has not filtered out into general knowledge, and it is the reason this article is not simply a footnote to the broader question of how much of what you eat you absorb.
On the Western Diet, participants were remarkably alike: metabolizable energy spanned 94.1% to 97.0%, a range under three percentage points. On the fiber-rich diet, they scattered — 84.2% to 96.1%. In the authors' own terms, that spread amounts to 73 to 390 non-metabolized kcals per day between individuals eating the identical diet, which they call "a clinically meaningful quantitative difference"1.
Fiber does not merely lower how much energy you absorb. It removes the guarantee that your number resembles anyone else's.
That asymmetry is the mechanism showing its face. On a highly digestible diet, absorption is a property of the human small intestine, and human small intestines are much alike. Push substrate into the colon and absorption becomes a property of a microbial community, and microbial communities differ enormously between people. The study could say how much: fecal propionate and fecal biomass together explained 58% of the variance in host metabolizable energy. Someone at 84% is losing five times as much energy per day to fiber as someone at 96%, and neither of them can tell which they are.
Three exits, and one of them is made of bacteria#
The missing energy leaves by three routes, and they are usually collapsed into one.
The first is fiber's own accounting. Fiber that reaches the colon is fermented rather than digested, and anaerobic fermentation recovers far less energy than the aerobic route — which is why fiber is credited with an energy value well under 4 kcal/g, and why that value is a band rather than a constant (the arithmetic behind it sits with the macro conversion factors).
The second is collateral. The fat and protein travelling in the same meal come out less completely absorbed too, so the discount lands across the plate rather than only on its fibrous part. That has been demonstrated directly in controlled feeding work, and the absorption article works through the measurements.
The third is the one the metabolic-ward study quantified and the other two miss. Fermentable substrate grows bacteria, and bacteria are made of carbon you paid for. Microbial gene copy number — a proxy for biomass — rose significantly on the fiber-rich diet, and fecal biomass was one of the two variables explaining most of the individual variation. Some of the energy that fails to reach you has not been left undigested at all. It has been converted into a microbial population and then excreted, which is a genuinely different kind of loss from food passing through untouched.
Is it the fiber, or the microbes? A drug answers part of it#
If the microbial community is the moderator, then perturbing that community should move absorption even without changing fiber. That has been tested, with an antibiotic.
Twenty-seven inpatient volunteers with obesity and impaired glucose tolerance underwent two interventions: a randomized crossover of three days of overfeeding versus three days of underfeeding, and a randomized, double-blind, placebo-controlled course of oral vancomycin. Stool calorie loss was the primary endpoint in both. Both endpoints were met: stool calorie loss rose during underfeeding relative to overfeeding, and rose on vancomycin relative to placebo. Community structure barely shifted with the feeding manipulation but changed widely, with reduced diversity, on the antibiotic — and Akkermansia muciniphila increased under both conditions that produced greater stool calorie loss2.
Nothing here is a recommendation; oral vancomycin is a drug with real consequences and not a weight-loss tool, and the participants were a clinical population rather than healthy adults. What it establishes is causal direction. Change the microbes and the calories you absorb change, with the diet held still. That closes the loop the fiber studies leave open, and it explains why two people eating the same beans get different bills.
The label already half-knows, and cannot agree with itself#
All of which sits underneath a printed number that has never settled how to handle fiber. The gap is not subtle, and it is jurisdictional.
| Rule | What fiber is worth per gram |
|---|---|
| EU, Regulation 1169/2011 Annex XIV | 2 kcal/g, fixed |
| US, 21 CFR 101.9(c)(1)(i)(B) | 4 kcal/g — fiber sits inside total carbohydrate |
| US, 21 CFR 101.9(c)(1)(i)(C) | 0 kcal/g — non-digestible carbohydrates subtracted first |
Sources: Regulation (EU) 1169/2011, Annex XIV; 21 CFR 101.9, which permits either general-factor method.
A high-fiber bar containing 10 g of fiber can therefore be labeled at 40 calories, 20 calories or 0 calories for that fiber, all legally, depending on which side of the Atlantic it is sold on and which permitted method the manufacturer picked. None of the three is the measured answer, which lands between the first two and moves with the eater. The one thing this does not license is treating fiber grams as a calorie rebate you apply yourself — the same trap that catches people subtracting their way through a net-carb calculation, and one more reason the printed number carries a band around it.
How much of this should change what you do#
Almost none of it, and knowing why is the useful part.
The 116-calorie effect is real and it was bought with a metabolic-ward diet engineered to push substrate into the colon — not with a bowl of bran on an otherwise unchanged day. Around 5% of intake is the honest ceiling on what this mechanism delivers, it arrives with a personal multiplier you cannot measure, and it moved neither hunger nor how much people chose to eat. Nobody is going to out-absorb a surplus.
What it does justify is a small correction to how the numbers are read. A day built on legumes, intact grains and whole fruit delivers meaningfully fewer usable calories than its printed total, and a day of refined food delivers very nearly all of them — the same figure describing two different transactions. That is a reason to prefer the former, stacked on top of the much larger reasons in fiber's benefits and targets, and a reason to treat any calorie total as a range rather than a reading. Which fiber you eat matters here too, since only the fibers that resist fermentation reach the far end intact — and the ones that ferment fastest are precisely the ones building the biomass you excrete.
FAQ#
How many calories does a high-fiber diet stop you absorbing?#
About 116 kcal a day at the well-controlled extreme. In a metabolic ward, a diet designed around fiber, resistant starch and large food particles increased daily fecal energy loss by 116 ± 56 kcal versus a Western diet matched for calories and macronutrients, dropping absorbed energy from 95.4% of intake to 89.5%1. Ordinary increases in fiber deliver correspondingly less.
Do two people absorb the same calories from the same high-fiber meal?#
No, and the gap is large. Eating the identical fiber-rich diet under identical conditions, participants ranged from 84.2% to 96.1% of intake absorbed — between 73 and 390 unabsorbed calories a day, person to person. On a Western diet the same people spanned less than three percentage points1. Fecal propionate and microbial biomass explained 58% of that variance.
Does a nutrition label already count fiber's calories?#
It depends on the label. EU rules fix fiber at 2 kcal/g3, while US rules let a manufacturer either leave fiber inside total carbohydrate at 4 kcal/g or subtract non-digestible carbohydrates entirely4. The same food can carry three different legal calorie figures, and the package does not disclose which method was used.
Sources#
- Corbin KD, Carnero EA, Dirks B, et al. Host-diet-gut microbiome interactions influence human energy balance: a randomized clinical trial. Nat Commun. 2023;14:3161.
- Basolo A, Hohenadel M, Ang QY, et al. Effects of underfeeding and oral vancomycin on gut microbiome and nutrient absorption in humans. Nat Med. 2020;26(4):589-598.
- Regulation (EU) No 1169/2011 on the provision of food information to consumers, Annex XIV — conversion factors for the calculation of energy.
- 21 CFR 101.9 — Nutrition labeling of food: permitted methods for calculating caloric content. US Code of Federal Regulations (Cornell LII).
- Poutanen KS, Fiszman S, Marsaux CFM, et al. Recommendations for characterization and reporting of dietary fibers in nutrition research. Am J Clin Nutr. 2018;108(3):437-444.
