> Nonetheless, it seems improbable that too many calories in and too few calories out are the cause of modern obesity.
the post itself talks about change in caloric intake and potential decrease in metabolism.
a decrease in necessary calories doesn’t change the math, just one of the variables.
seems like it still works to simplify to calories in and calories out, just don’t assume the numbers you’re handed are right. you can always measure your changes.
Wait, if you can’t assume numbers you’re handed are right, how are you measuring changes? At some point fundamentally CICO requires some amount of counting calories??
> Even those in “light” labor roles consumed 3.2k calories daily. Examples of such jobs included toothbrush filler, rotary machine operator, and polisher. While these tasks may not be as physically undemanding as sitting at a computer, they are not much more demanding.
I find it very difficult to believe that sitting in an air-conditioned office chair is close to comparable to the conditions of even 'light' factory work in the 1950s.
Probably closer, but I still doubt it. Factory work was notoriously brutal.
Figures for calories burnt from sitting vs. standing varies between 20 and 80 cal/hour. 80cal/hr would be substanial over a workday (800cal, assuming factory work was 10 hour days in the 50s...), but 20 would not...
If we assume it's ~50cal/hr, standing alone could add up to 500cal difference for a 10 hour day, not even including additional physical exercise from the job itself.
Add to this the amount of walking that was likely substanially more common in the 50s, and it sure seems like you could chalk up a 1000+ cal difference pretty quickly.
I just read the actual study quoted just to get an idea of what they're claiming, and this further seems to indicate that 'light' labor was anything but...
> 4. The average daily calorie intake of men doing light work was about 600 Cal./day less than that of men doing heavy work. The previous failure to differentiate between intakes was thus not confirmed. The estimated calorie expenditure per minute, excluding what was require for basal metabolism, was on an average 2-2 Cal. for light work, 2-7 for light to medium work, 2.6 for medium to medium heavy work and
3-8 for heavy work.
2.2cal/min would equate to 1320cal additional burned for a 10 hour day! That's clearly far, far above typical calories used in an office job.
A 150 lb person uses 100 calories per hour at their desk, but people get up quite often to go to meetings etc. It could easily be 1150 vs 1320 calories here a meaningful difference, until you consider differences in body weight = higher calorie expenditure.
The following are examples of the grading of the job:
Light: filler (tooth brushes), rotary-machine operator, polisher.
What was a filler of tooth brushes? I assume that was just someone who made tooth brushes? Was that really a popular job? Google is totally failing me on that.
> I assume that was just someone who made tooth brushes?
Yes.
> Was that really a popular job?
No, but that's not the point. If you looked up census record occupations of the era, you wouldn't find a lot of "rotary-machine operator" or "polisher" either. It's one example of an assembly task that was part of manufacturing at the time.
Air conditioning can increase metabolic rate to keep you warm, it’s mostly heating that reduces energy expenditure and 1950’s factories had heat.
A 150lb office worker at their desk is burning ~100 calories an hour, but people only spend about 70% of their time act their desks instead moving around a fair bit over the workday. Step counters end up counting a surprising amount of activity on average.
What I think the article is missing is people doing light factory work in 1928/9 generally lacked cars. People just walked far more in their non working life.
I'd need a citation on 100cal/hr for sitting at a desk. Are you removing BMR from that calculation? That would imply people are spending 800cal/day on JUST sitting during a workday (no other movement, no BMR, no walking, etc).
If you extend this to a whole day, you would be burning 1600cal/day from sitting ... without even counting BMR which is ~1200-1600 or more.
Numbers I've seen is that sitting is much closer to 10-20cal/hr from sitting (vs. literally asleep).
I found several secondary sources for the 100/hour figure at 150lb, but no primary sources. One study had lying down while awake at 86 calories per hour so typing/using a mouse/etc would need to be 15c/h to hit 100 which seems likely. But again that's excluding time spent getting up and moving around at the office which adds up.
There was also outliers like: *But the National Institutes of Health tells us that in an hour, the average 170 lb. person burns:
134 calories sitting
206 calories standing
296 calories walking at a moderate pace
Right, and the 1950s study was measuring ~132cal/hr above BMR for their 'light' work. So, clearly, if someone burns ~100cal/hr (including BMR) for sitting in an office, then the 'light' work is much, much, much more strenous regardless of what BMR is.
Again sitting is part of the office activity, a more realistic figure is probably 110-130 per hour averaged over the day simply because walking and standing use significantly more energy.
Also, subtracting BMR doesn’t make sense from a calorie per day standpoint. A 5% increase in calorie expenditure doesn’t explain a 25-50lb increase in weight especially when that weight increases BMR.
Granting your position of 110-130/hr on average for an office job, including BMR, then that still is far lower than the quotes in the study.
BMR for men is roughly 1700cal, or 70cal/hr, so if you burn an additional 130 over BMR (what the study says for “light work”), then the “light work” was closer to 200cal/hr, or 800cal/day extra over an office job assuming a 10 hour day.
800cal/day is an enormous difference in expenditure.
Your calculations are off, but so are mine. I had initially assumed a 10 hour work day which makes a big difference in how large the gap is, but it looks like working hours dropped from 10h/day in 1890 to ~8.2 by 1929. https://eh.net/encyclopedia/hours-of-work-in-u-s-history/ Still a 6 day workweek.
Anyway, they don't actually list light in terms of calories per hour just 838/day without saying number of hours worked. Further they list BMR at 1550c/day.
So ~838 / 8.4 + (1550/24) = 164 calories per hour during work.
If office workers are at 120/h +/- 10 and they are working the same 8.2h days that's a deficit of (164 - 120) * 8.4 +/- 10 * 8.2 = 369.6 +/- (82) which is quite large but half your 800cal/day.
Though again, I had assumed a longer work day which puts the numbers a lot closer. (838/10 + 65) = 149/h vs 120/h.
They state 2.2cal/min of 'light work', and have other details on assumptions. They also state 27.9% of whole day spent on work, which would imply either 6.7 hours spent on working, or... more likely... if you include one day off, it would be 7.8 hours per day for 6 days a week. Pretty close to ~8hrs/6days/wk.
The difference between 1550 and 1700cal for BMR is only 5cal/hr, so my calculations on difference in exertion between sitting (120cal/hr) and the 'light work' of the paper is essentially the same. It's still ~195cal/hr. My assumption was 10hrs/day for 5 days... so 8 hours/6 days is nearly the same amount of total time. Makes an office job even worse in comparison.
Also, another note that is very different from today's assumptions buried in the paper: 13.1% of average day that wasn't sleep or work was spent walking, and 20.6% was spent on cycling. That's ~1h10m a day walking, and ~1h50m a day of cycling. 3 hours of walking/cycling a day alone could account for much of the difference in total calories burned.
Table 2: Classification of activities and average number of calories expended on them by the 152men (Basal metabolism is included, the allowance for it being 1.1 Cal./min
Sitting, including time spent at meal 1.6
Light games, such as darts and billiards 2.5
...
Sleeping 1.0
Light work in factories in 1950 included sitting down at a conveyor belt and removing defective items. 130 ish per hour works.
200 calories an hour * 8 hours isn't lite work by most definitions. That's someone walking constantly for the entire shift.
Again, read the paper. The whole thing. The specific text:
> 4. The average daily calorie intake of men doing light work was about 600 Cal./day less than that of men doing heavy work. The previous failure to differentiate between intakes was thus not confirmed. The estimated calorie expenditure per minute, excluding what was require for basal metabolism, was on an average 2-2 Cal. for light work, 2-7 for light to medium work, 2.6 for medium to medium heavy work and 3-8 for heavy work.
I assumed that was a mistake as light to medium is above medium to medium heavy etc, but I almost got make their numbers work though both light medium to heavy are off by 0.1.
Light: 46.5 h / week
Light to medium: 46.0 h / week
Medium to medium heavy: 46.6 h / week
Heavy: 48.8 h / week
Going back to table 7
Light 838/day
Light to medium 1063
Medium to medium heavy 1013
Heavy 1585
Light is thus 838 * 7 / 46.5 = 126 calories per hour = 2.10 /min *vs 2.2?*
Light to medium is thus 1063 * 7 / 46.0 = 161 calories per hour = 2.68 /min
Medium to medium heavy is thus 1013 * 7 / 46.6 = 152 calories per hour = 2.536 / min *vs 2.6?*
Heavy 1585 * 7 / 48.8 = 227 calories per hour = 3.79 /min
Though at this point I assume there are other math errors. Still 65 + 2.1 * 60 = 191 calories per hour for light work is more realistic than some of the other numbers you where tossing around. (191 - 120) * 8 = 568 calories per day delta.
The post mentions the "Ray Peat" community a couple times. Anyone here peripheral to that scene? The argument seems to hinge on the point "They propose that people historically had significantly faster metabolisms", which is a pretty big claim that I'd love to hear more about.
Interestingly, there's no Wikipedia article for him, just an (expectedly) nauseating one on RW.
Americans doing various "fitbit" challenges need to actually "try" to complete 10,000 steps in a day. People in 1909 would have had to try hard _not_ to complete this number of steps in a day.
90% of a mammal’s energy output is in the form of climate control, not movement.
The variable not factored in is the average ambient temperature. Americans of yesteryear shivered in winter months and sweltered in summer months.
Today, that work has been outsourced to heat pumps, gas furnaces, and refrigerated air systems.
I discovered regular workouts removed my extreme discomfort with it being cold inside. I may not enjoy it being 55 degrees, but it’s not painful like it used to be. I suspect most Americans would find the same.
Stronger by Science covered this in a podcast episode maybe a year ago or so. There does seem to be reasonably good evidence that basal metabolic rates have dropped over the last century without any conclusive idea why, though a case was presented that it has something to do with toxin/pathogen exposure - the body works less when it doesn't have to fight infection. Another case was made for indoor temperature control, which another comment brings up.
It is nowhere near enough to "explain" the rise in obesity, though. It also doesn't sharply drop out of nowhere around 1980 or so. I've been using MacroFactor now for about two years, weighing myself every morning and weighing and measuring everything I eat. Looking at the log right now, it seems I've averaged 3,348 calories a day for the past 30 days. The weight estimate for me was 172.2 lbs 30 days ago and 172.0 lbs today. Seems I'm managing to use the large amount of energy I eat, in spite of being part of a population with decreased basal metabolism. If you're not using that much energy, then also eat less.
The Internet badly needs to stop galaxy-braining this stuff. Sure, obesity is multifactorial like anything else that happens to an animal body. Potentially specific foods are less satiating in the short run than others or simply take up less space in the gut. Some are easier to eat, literally taking less time and effort and even energy, than others. Some sort of endocrine disruptor or what not maybe messes with ghrelin and leptin levels. But all energy consuming systems are still subject to basic thermodynamics. If you use less energy, then eat less energy. There may be a variety of reasons you find that hard to do, but that is nonetheless what you need to do if this matters to you.
And if you're going to propose some kind of cause specific to your region, your country, even just to humans, to me that isn't enough. I've got four cats, all of whom are over a decade old now, and I've had them their entire lives. Among the varying personality traits are food motivation. You can see it very easily. Two of them have stayed exactly the same size no matter what. They often don't finish a meal and never eat more than they need. Two are constantly begging no matter what, always hungry. One of them is so motivated by food that he has figured out how to open a refrigerator, a screwtop Talenti container, a locked cabinet, which took him six months but he still figured it out.
They're not eating sugar, heavily processed foods, not subject to marketing aimed at humans, yet we still see the same differences. Given effectively infinite access to food, some of them will still only eat what they need and some of them will eat until they puke and then eat the puke. It's an innate personality difference. The humans of 1909 did not have near infinite access to food. Humans of today do. And we're still mammals. Whatever difference exists in appetite coupling to energy usage and food motivation in cats exists in us as well. I guess I'm lucky. I only eat what I need. Many people don't. I don't think the difference is going to be solely down to microplastics in the water or whatever that somehow I magically managed to avoid. It's just a spectrum along which animals differ.
The article is earnestly written, but -- as often with blog post research -- it falls apart on faulty premises.
They rely on tables from early 20th century research, but in doing so you have to consider the history of the "calorie" itself and changes in how it was being measured and quantified in both individual ingredients and prepared food.
In their reference period, it was still all pretty much guesstimates as a thorough catalog of experimentally determined calorie counts wouldn't appear for decades.
Better research would need to go back and look for reliable diet logs (good luck) and then try to work up calorie counts using modern measures, but even that is tricky because there's reason to doubt the nutrition profiles of most ingredients/meals have held steady in the last 100+ years anyway.
So, this is a fun little "make you think" piece, but not anything you can take seriously.
28 comments
[ 4.8 ms ] story [ 115 ms ] threadthe post itself talks about change in caloric intake and potential decrease in metabolism.
a decrease in necessary calories doesn’t change the math, just one of the variables.
seems like it still works to simplify to calories in and calories out, just don’t assume the numbers you’re handed are right. you can always measure your changes.
are there more layers to this? of course.
I find it very difficult to believe that sitting in an air-conditioned office chair is close to comparable to the conditions of even 'light' factory work in the 1950s.
Figures for calories burnt from sitting vs. standing varies between 20 and 80 cal/hour. 80cal/hr would be substanial over a workday (800cal, assuming factory work was 10 hour days in the 50s...), but 20 would not...
If we assume it's ~50cal/hr, standing alone could add up to 500cal difference for a 10 hour day, not even including additional physical exercise from the job itself.
Add to this the amount of walking that was likely substanially more common in the 50s, and it sure seems like you could chalk up a 1000+ cal difference pretty quickly.
> 4. The average daily calorie intake of men doing light work was about 600 Cal./day less than that of men doing heavy work. The previous failure to differentiate between intakes was thus not confirmed. The estimated calorie expenditure per minute, excluding what was require for basal metabolism, was on an average 2-2 Cal. for light work, 2-7 for light to medium work, 2.6 for medium to medium heavy work and 3-8 for heavy work.
2.2cal/min would equate to 1320cal additional burned for a 10 hour day! That's clearly far, far above typical calories used in an office job.
Yes.
> Was that really a popular job?
No, but that's not the point. If you looked up census record occupations of the era, you wouldn't find a lot of "rotary-machine operator" or "polisher" either. It's one example of an assembly task that was part of manufacturing at the time.
A 150lb office worker at their desk is burning ~100 calories an hour, but people only spend about 70% of their time act their desks instead moving around a fair bit over the workday. Step counters end up counting a surprising amount of activity on average.
What I think the article is missing is people doing light factory work in 1928/9 generally lacked cars. People just walked far more in their non working life.
If you extend this to a whole day, you would be burning 1600cal/day from sitting ... without even counting BMR which is ~1200-1600 or more.
Numbers I've seen is that sitting is much closer to 10-20cal/hr from sitting (vs. literally asleep).
I found several secondary sources for the 100/hour figure at 150lb, but no primary sources. One study had lying down while awake at 86 calories per hour so typing/using a mouse/etc would need to be 15c/h to hit 100 which seems likely. But again that's excluding time spent getting up and moving around at the office which adds up.
There was also outliers like: *But the National Institutes of Health tells us that in an hour, the average 170 lb. person burns:
Also, subtracting BMR doesn’t make sense from a calorie per day standpoint. A 5% increase in calorie expenditure doesn’t explain a 25-50lb increase in weight especially when that weight increases BMR.
BMR for men is roughly 1700cal, or 70cal/hr, so if you burn an additional 130 over BMR (what the study says for “light work”), then the “light work” was closer to 200cal/hr, or 800cal/day extra over an office job assuming a 10 hour day.
800cal/day is an enormous difference in expenditure.
Anyway, they don't actually list light in terms of calories per hour just 838/day without saying number of hours worked. Further they list BMR at 1550c/day.
https://substackcdn.com/image/fetch/f_auto,q_auto:good,fl_pr...
edit: 1954 is 5 days but hours are up to 8.4.
So ~838 / 8.4 + (1550/24) = 164 calories per hour during work.
If office workers are at 120/h +/- 10 and they are working the same 8.2h days that's a deficit of (164 - 120) * 8.4 +/- 10 * 8.2 = 369.6 +/- (82) which is quite large but half your 800cal/day.
Though again, I had assumed a longer work day which puts the numbers a lot closer. (838/10 + 65) = 149/h vs 120/h.
They state 2.2cal/min of 'light work', and have other details on assumptions. They also state 27.9% of whole day spent on work, which would imply either 6.7 hours spent on working, or... more likely... if you include one day off, it would be 7.8 hours per day for 6 days a week. Pretty close to ~8hrs/6days/wk.
The difference between 1550 and 1700cal for BMR is only 5cal/hr, so my calculations on difference in exertion between sitting (120cal/hr) and the 'light work' of the paper is essentially the same. It's still ~195cal/hr. My assumption was 10hrs/day for 5 days... so 8 hours/6 days is nearly the same amount of total time. Makes an office job even worse in comparison.
Also, another note that is very different from today's assumptions buried in the paper: 13.1% of average day that wasn't sleep or work was spent walking, and 20.6% was spent on cycling. That's ~1h10m a day walking, and ~1h50m a day of cycling. 3 hours of walking/cycling a day alone could account for much of the difference in total calories burned.
200 calories an hour * 8 hours isn't lite work by most definitions. That's someone walking constantly for the entire shift.
> 4. The average daily calorie intake of men doing light work was about 600 Cal./day less than that of men doing heavy work. The previous failure to differentiate between intakes was thus not confirmed. The estimated calorie expenditure per minute, excluding what was require for basal metabolism, was on an average 2-2 Cal. for light work, 2-7 for light to medium work, 2.6 for medium to medium heavy work and 3-8 for heavy work.
Interestingly, there's no Wikipedia article for him, just an (expectedly) nauseating one on RW.
The variable not factored in is the average ambient temperature. Americans of yesteryear shivered in winter months and sweltered in summer months.
Today, that work has been outsourced to heat pumps, gas furnaces, and refrigerated air systems.
I discovered regular workouts removed my extreme discomfort with it being cold inside. I may not enjoy it being 55 degrees, but it’s not painful like it used to be. I suspect most Americans would find the same.
It is nowhere near enough to "explain" the rise in obesity, though. It also doesn't sharply drop out of nowhere around 1980 or so. I've been using MacroFactor now for about two years, weighing myself every morning and weighing and measuring everything I eat. Looking at the log right now, it seems I've averaged 3,348 calories a day for the past 30 days. The weight estimate for me was 172.2 lbs 30 days ago and 172.0 lbs today. Seems I'm managing to use the large amount of energy I eat, in spite of being part of a population with decreased basal metabolism. If you're not using that much energy, then also eat less.
The Internet badly needs to stop galaxy-braining this stuff. Sure, obesity is multifactorial like anything else that happens to an animal body. Potentially specific foods are less satiating in the short run than others or simply take up less space in the gut. Some are easier to eat, literally taking less time and effort and even energy, than others. Some sort of endocrine disruptor or what not maybe messes with ghrelin and leptin levels. But all energy consuming systems are still subject to basic thermodynamics. If you use less energy, then eat less energy. There may be a variety of reasons you find that hard to do, but that is nonetheless what you need to do if this matters to you.
And if you're going to propose some kind of cause specific to your region, your country, even just to humans, to me that isn't enough. I've got four cats, all of whom are over a decade old now, and I've had them their entire lives. Among the varying personality traits are food motivation. You can see it very easily. Two of them have stayed exactly the same size no matter what. They often don't finish a meal and never eat more than they need. Two are constantly begging no matter what, always hungry. One of them is so motivated by food that he has figured out how to open a refrigerator, a screwtop Talenti container, a locked cabinet, which took him six months but he still figured it out.
They're not eating sugar, heavily processed foods, not subject to marketing aimed at humans, yet we still see the same differences. Given effectively infinite access to food, some of them will still only eat what they need and some of them will eat until they puke and then eat the puke. It's an innate personality difference. The humans of 1909 did not have near infinite access to food. Humans of today do. And we're still mammals. Whatever difference exists in appetite coupling to energy usage and food motivation in cats exists in us as well. I guess I'm lucky. I only eat what I need. Many people don't. I don't think the difference is going to be solely down to microplastics in the water or whatever that somehow I magically managed to avoid. It's just a spectrum along which animals differ.
They rely on tables from early 20th century research, but in doing so you have to consider the history of the "calorie" itself and changes in how it was being measured and quantified in both individual ingredients and prepared food.
In their reference period, it was still all pretty much guesstimates as a thorough catalog of experimentally determined calorie counts wouldn't appear for decades.
https://spectrum.ieee.org/how-counting-calories-became-a-sci...
Better research would need to go back and look for reliable diet logs (good luck) and then try to work up calorie counts using modern measures, but even that is tricky because there's reason to doubt the nutrition profiles of most ingredients/meals have held steady in the last 100+ years anyway.
So, this is a fun little "make you think" piece, but not anything you can take seriously.