Editor’s note: This article is widely misquoted and misunderstood. Critics usually miss the point completely. The point is not that “Taking the car is good for the environment” (which was the title of one blog post I saw). Instead it’s this: Just like some kinds of transportation use more energy than others, some kinds of food require more energy to produce than others too, and what you eat is just as much a part of your energy footprint as how you get around. Please read the article carefully before jumping to any conclusions. Thank you.
Most people think that bicycling doesn’t use gas, but actually it does, indirectly. It takes lots of energy to produce the food for the cyclist’s effort — probably more than you thought.
Of course, we can’t just stop eating, but we can definitely choose what we eat, and here’s the kicker: meat requires much more fossil fuel to produce than vegetables and grains. How much more? About 68 times more for beef than for potatoes.1 The reason for this is simple: Cattle consume fourteen times more grain than they produce as meat. They’re food factories in reverse. So it takes a lot more energy to produce that meat (as well as more land and water). We use absolutely horrific amounts of energy to grow grain to feed to cattle. In fact, over 80% of the grain grown in this country is eaten by livestock, not people. So in short, the more meat you eat, the more gas you waste.
Dr. David Pimentel of Cornell University calculates that it takes nearly twice as much fossil energy to produce a typical American diet than a pure vegetarian diet. This works out to about an extra 200 gallons of fossil fuels per year for a meat-eater. This means that meat-eaters are “driving” an extra fourteen miles every day whether they really drive or not, when we look at how much extra fuel it takes to feed them.2
In fact, meat production is so wasteful that walking actually uses more fossil energy than driving a 35 mpg car, if you get your calories for the standard American diet.3 (On vegetarian or vegan diets, walking uses less energy than driving.)
The same is not true of bicycling vs. driving, because bicycling is more than twice as efficient as walking (calories consumed per distance traveled). Bicycling uses less fossil energy than driving even if the cyclist were eating nothing but beef.4 But to focus on this misses the point. It’s no bombshell that cycling uses less fossil energy than driving, no matter what you’re eating. What’s important is that meat-eaters use twice as much fossil energy as vegans—whether they’re bicycling or not.
What does this mean in practical terms?
It means that the amount of gas you use isn’t just related to how you get from place to place, it’s also related to what you eat. Meatless diets require half as much fuel to produce as the standard American diet. Pimentel calculated that if the entire world ate the way the U.S. does, the planet’s entire petroleum reserves would be exhausted in 13 years. The typical American could save more gas by going vegan than by giving up driving two days a week.5
Those who think the question is, “Is it better for me to drive, or (eat a wasteful diet and) walk or bike?” are missing the point. That’s like asking whether it’s better to pour oil or pesticides into the water supply. Naturally you shouldn’t do either. If you want to greatly reduce your energy and pollution footprint, one of the easiest ways is to reduce or eliminate consumption of animal foods—no matter how you get around.
Food for thought.
The article “How Virtuous is Ed Begley Jr.?” by John Tierney, referencing these calculations, appeared in the New York Times on 25 February 2008.
Buying local vs. ditching meat
Some readers have claimed that what’s important for saving energy is buying local. But the energy savings there pales compared to going vegetarian. That’s because only 4% of food’s greenhouse emissions come from transporting it to the store. (New Scientist) As the Organic Consumers put it, “It’s how food is produced, not how far it is transported, that matters most for global warming, according to new research published in ES&T.” The authors of that study say, “Shifting less than one day per week’s worth of calories from red meat and dairy products … achieves more GHG reduction than buying all locally sourced food.” (Carnegie-Mellon University) In other words, it’s far better to buy a pound of carrots shipped from far away, than to buy a pound of locally-farmed beef. Brighter Planet agrees that ditching meat is far more important than buying local.
Hummer-driving vegans use less energy than meat-eating Prius drivers?
Michael Pollan made that claim, but it’s not true. My carbon footprint calculator can help do the comparison. A 10 mpg Hummer driven the U.S. average of 1000 miles/mo. generates 13.4 tons of CO2 per year, vs. 2.7 tons for a 50 mpg Prius, for a penalty of 10.7 tons for driving the Hummer. By contrast, going vegan saves only 2.2 tons/yr. So it isn’t even close, it’s far worse to drive a Hummer than to eat meat. But this kind of comparison misses the point: Meat production really is energy-intensive and causes a tremendous amount of unnecessary pollution, it’s just that Hummers are even worse. Anyone who’s serious about reducing their carbon-, energy-, or pollution-footprint is going to cut back or eliminate meat, especially as it’s the low-hanging fruit. Cutting back or going veg is the easiest way for most people to make the biggest dent in their footprint.
|Calories Burned per Hour||130-lb. Person||180-lb. Person|
- Cyclists cover 3.67 times as much distance in the same period of time as walkers. (11 mph / 3 mph)6
- 195 walking calories x 3.67 = 715.6 walking calories to cover same distance.6
- 715.6 calories vs. 354 calories: (715.6-354)÷354 = 102%
So cycling is 117% more efficient than walking (i.e., about twice as efficient). That’s because cyclists travel 3.7 times faster than walkers, but use only about 1.8 times more calories to do so.
Some have complained about the sample speeds in the table (that they’re too fast or too slow), but it doesn’t matter. Tweaking the speed for either walking or biking (or both) doesn’t significantly change the results. If you go faster you burn calories at a faster rate, but you also travel farther for a given amount of time.
Criticism from readers
You say it takes more energy to walk somewhere than to drive your car, but this is a fairly meaningless statement since it took a whole lot of energy to create that car and the road it is driving on. — Edward Pilbrow, Electrical and Computer Engineering, University Of Canterbury, New Zealand
This completely misses the point. The point is simply to show how incredibly energy intensive meat-based diets are, by using an interesting comparison. That’s it. If I said that a stack of 220 billion quarters would reach the moon you’d be complaining that the stack would fall over after about 100 quarters or so. Even if we did consider infrastructure, someone deciding whether to walk or drive to their destination does not suddenly have to contemplate buying a vehicle and constructing a roadway to drive it on. The car has already been purchased, the roadway has already been built. Those energy costs were sunk long ago. Nobody buys a car to make a single trip.
Think of it this way: You have a choice to bike or walk somewhere, or take your car. That’s a choice that many people make every day. I’m simply comparing the fuel use for each choice at that point in time.
The point is that meat production wastes horrific amounts of energy, no matter how you slice it or spin it.
In any event, Chris Goodall, who wrote a similar article to this one, writes: “These numbers [e.g., the energy required to produce a car] are not enough to remotely affect the conclusion that car travel is less carbon intensive than walking, if the walker replaces lost energy with animal products.”
He says that making a car produces 3 tons of carbon. Over a 200,000-mile life, that’s a mere 0.033 lbs. of carbon per mile. It’s not significant.
Goodall also sums up my frustration with the reception to our articles very well: “I was extremely naive not to realise that the analysis would be perceived as an encouragement to drivers. I didn’t intend it to be read that way. My purpose was to draw attention to the carbon intensity of modern food production, particularly of meat.”
Your analysis does not include the energy that was expended to discover, extract, ship, refine, and then ship and distribute the gasoline. — Christopher E.
And you think the amount energy to produce the gasoline is somehow different if it’s put into a car than if it’s put into a tractor? It’s really simple: For X amount of gas, we know how far a car goes. For that same amount of fossil energy, we see how much food we could produce with it, then see how far the cyclist could go on that much food. We’re talking about the same amount of fossil energy either way, so if we did count it, we’d have to count it for both the car and the bike, so the conclusion would be exactly the same. Even if the amount of energy needed to produce the energy were relevant—which it is not—Chris Goodall (see above) says that you’d add only another 15% to the carbon total to account for the production of the gas. (PRI puts the energy for extraction at 5%.) And again, if you did that, you’d have to add it both sides (i.e., into both the car and the tractor), so it couldn’t be less relevant.
Your analysis is wrong because it assumes that everything else is equal, but in fact, people who drive tend to go farther distances than those who don’t.
I’m just flabbergasted at how far people will go to miss the point. Yes, the comparison assumes that all else is equal because that’s the whole point of the comparison! Once you change the variables then you’re comparing something completely different. If we want to compare the energy used by for traveling a given distance by different methods then that is what we’re comparing. We’re not trying to measure every trip that every person takes, that’s completely irrelevant. We’re simply making a single, interesting, trip vs. trip comparison. That’s it.
What about grass-fed beef?
What about it? 97% of beef raised in the U.S. is grain-fed, not grass-fed. If you’re eating beef, you’re almost certainly eating grain-fed beef. Even if you’re not it makes little difference, because it’s not possible to raise all our beef grass-fed, because there’s not nearly enough land. If we shift to grass-fed then beef gets more expensive and there’s less of it produced. There is simply not enough land to support America’s beef addiction by grazing cattle, even at higher prices.
As for raising cattle on land unsuitable for growing crops, again, there isn’t nearly enough land available. If you suggested otherwise, you simply have no idea how staggering America’s beef consumption really is.
In any event, the world’s leading researcher on energy in agriculture, Dr. David Pimental, says that even grass-fed beef uses about half the energy as grain fed, meaning it’s still more energy-intensive than growing crops for food. (American Journal of Clinical Nutrition) And Spiegel reports that 1 lb. of grass-fed beef is equivalent to driving 32 miles.
In short, if we changed the beef industry from grain-fed to grass-fed, there would be a lot less beef, and it would be a lot more expensive. For those reasons it’s not going to happen, and even if it did, it would still use much more energy than growing crops.
I’m concerned that by biking 24.4 miles roundtrip to and from work I’m not saving as much energy as I thought, since I take an extra shower when I arrive home, and I wash clothes more frequently.
You can relax, you’re still saving a ton of energy. I’m not going to do a detailed analysis, but here’s a back-of-the-envelope version:
Your 24.4-mile round trip would use about a gallon of gas in an automobile, which would cost about $3.00. Since the costs of water and energy for laundry are much lower than that, they can’t possibly use more energy than driving.
At the U.S. average cost of a $0.002/gallon, a 20-gallon shower costs $0.04.
A load of laundry uses about 35 gallons ($0.07) and 2 kWh of electricity ($0.20). At a guess, you could wash about eight trips’ of cycling clothes in a load, so that $0.27 per load would be about $0.03 per trip.
So your water and energy costs per trip are about $0.07. That tells me that the extra energy required by your biking is insignificant compared to the amount of energy you’re saving by not driving. If there were lots more energy required to provide the water for your shower and laundry, then the price of water would be a lot more. But it’s not, so you’re in the clear: The additional energy required to shower and do your laundry is trivial compared to the energy you’re saving.
Carbon Footprint Calculator. My custom calculator is one of the few that shows you how going vegetarian or vegan compares to driving.
Eating Fossil Fuels, by Dale Allen Pfeiffer
Cattle more damaging to the planet than cars. (UK’s Independent). Excerpt:
A United Nations report has identified the world’s rapidly growing herds of cattle as the greatest threat to the climate, forests and wildlife. And they are blamed for a host of other environmental crimes, from acid rain to the introduction of alien species, from producing deserts to creating dead zones in the oceans, from poisoning rivers and drinking water to destroying coral reefs.
The 400-page report by the Food and Agricultural Organisation, entitled Livestock’s Long Shadow, also surveys the damage done by sheep, chickens, pigs and goats. But in almost every case, the world’s 1.5 billion cattle are most to blame. Livestock are responsible for 18 per cent of the greenhouse gases that cause global warming, more than cars, planes and all other forms of transport put together.
Burning fuel to produce fertiliser to grow feed, to produce meat and to transport it – and clearing vegetation for grazing – produces 9 per cent of all emissions of carbon dioxide, the most common greenhouse gas. And their wind and manure emit more than one third of emissions of another, methane, which warms the world 20 times faster than carbon dioxide.
Livestock also produces more than 100 other polluting gases, including more than two-thirds of the world’s emissions of ammonia, one of the main causes of acid rain.
Ranching, the report adds, is “the major driver of deforestation” worldwide, and overgrazing is turning a fifth of all pastures and ranges into desert. Cows also soak up vast amounts of water: it takes a staggering 990 litres of water to produce one litre of milk.
In order to make the bicycle much more energy-efficient than the auto, we need to significantly lower the energy that it takes to create and transport food. That’s much easier said than done. See Food, Energy, and Society.
One way to help do this is to eat less (or no) meat that is obtained by feeding animal (become a vegetarian).
Footnotes, Calculations, Assumptions, and Sources
(1) 68 times more energy required to produce beef than potatoes.
Beef: In April 2004 Dr. David Pimentel of Cornell University shared with me an advance copy of his paper Livestock Production and Energy Use, which says that it takes 40 kilocalories (kcal) of fossil energy to produce 1 kcal of beef protein. This number updates the 35:1 ratio published in his earlier book Food, Energy and Society (1996, with Marcia Pimentel). These numbers include only production, not processing, packaging, transport, refrigeration, and the methane produced by animals and their waste. The numbers for potatoes below likewise are only for production, so we’re comparing apples to apples. Of course, beef likely uses even more energy vs. potatoes than we calculate here, considering the extra energy required for refrigeration and safety protocols. Finally, note that these figures consider all forms of fossil energy, not just gasoline. This includes fossil-fuel-based fertilizers. With that long introduction, here is the calculation for the energy required for beef production:
- 40 kcal fossil energy per 1 kcal beef protein
- 40 kcal fossil energy per 1 gram beef (1 gram of 85% lean beef has 0.25 grams of protein according to the USDA database, which is 1 kcal of protein)
- 18,160 kcal fossil energy per 1 lb. beef (40kcal x 454 grams)
- 0.585 gallons of gasoline equivalent per 1 lb. beef (18,160 kcal ÷ 31,000 kcal/gallon; see below for the 31,000 conversion factor)
- 0.516 gallons of gasoline per 1000 calories of beef (1 lb. of beef is 1135 food kilocalories, according to the USDA Food & Nutrient database)
A Japanese study concurs with the above, clocking in at 43 kcal of fossil energy per 1 gram of beef. As with Pimental, they do not include processing, packaging, refrigeration, transport, or the methane produced by animals and their waste.
- 32.3 kg of CO2e per kg of beef
- 5959 kg of CO2e per animal
- 184.5 kg per animal (5959 ÷ 32.3)
- 32.8 GJ energy consumption per animal
- 0.00018 GJ per gram (32.8 ÷ 184.5 ÷ 1000)
- 43 kcal of fossil energy per gram (0.00018 x 239,005.736138 conversion factor) Ogino, A., H. Orito, K. Shimada, and H. Hirooka. (2007). Evaluating environmental impacts of the Japanese beef cow-calf system by the life cycle assessment method. Animal Science Journal 78: 424&endash;432.
In Beyond Beef, Jeremy Rifkin, 1992, p. 225 says it takes a gallon of gasoline to produce a pound of beef. Rifkin cites as his source Alan B. Durning, “Cost of Beef for Health and Habitat,” Los Angeles Times, 21 September 1986, p. 3. I assume this old data is in error.
Note that there is some disagreement over the number of kilocalories in a gallon of gasoline. There are a few reasons for that. First of all, the kilocalorie is a measure of energy, but gasoline is not energy itself, it is a fuel that can be used to produce energy. Also, gasoline is not a static substance — the quality of gasoline varies from one batch to the next depending on the source material, processing methods, etc. Here are the competing sources I found:
- 34,800 – Woodrow Wilson Biology Institute
- 32,143 – Prof. Joe Straley and S. A. Shafer, University of Kentucky
- 31,499 – Ken DeLong
- 31,000 – HowStuffWorks.com
- ~30,000 – Dr. David Pimentel, Cornell University
- 28,807 – (derived, see below)
I derived the 28,807 figure thusly: According to the EPA there are about 113,500 BTUs in a gallon of gasoline. (Vigan Prassar says it’s 125,000, but the EPA’s data appears more credible since it contains more detail.) One kcal is equivalent to 3.97 BTUs (Google calculator), so the 113,500 BTUs in a gallon of gasoline is equivalent to 28,807 kcal.
In Dr. Pimentel’s earlier work he assigns a whopping 38,000 kcal per gallon, but he confirmed for me in a telephone conversation on April 8, 2004 that ~30,000 is a better figure.
Potatoes: On pp. 134-135 of Food, Energy and Society we see that the production of 34,384 kg of potatoes in New York required 152 litres of diesel, 272 litres of gasoline, and 47 kWh of electricity. This gives us:
- 152 litres of diesel = 40.15 gallons of diesel
- 272 litres of gas = 71.85 gallons of gas
- 47 kWh = 160,411 BTUs (1 kWh = 3413 btus) (And note that most electricity in the U.S. is produced with fossil fuels.)
- 160,411 BTUs = 1.41 gallons of gasoline (113,500 BTUs per gallon, as per EPA)
- Total energy to produce 34,384 kg of potatoes = 40.15 + 71.85 + 1.41 = 113.4 gallons
- 34,384 kg potatoes = 75,804 lbs. potatoes
- 113.4 gallons / 75,804 lbs. = 0.0015 gallons of fossil energy per lb. of potatoes
- 1 lb. of potatoes = 395 calories, as per the USDA Food & Nutrient Database
- 1000 calories ÷ 395 cals/lb = 2.53 lbs. of potatoes
- 0.0015 gallons of fossil energy per lb. of potatoes x 2.53 lbs. of potatoes = 0.003795 gallons per 1000 calories of potatoes
- When I spoke with Dr. Pimentel by telephone on April 8, 2004 to confirm my calculation above he said that I should double my result to include fossil-based fertilizers, so let’s call it 0.00759 gallons.
Beef vs. Potatoes: We thus have 0.516 gallons per 1000 calories of beef vs. 0.00759 gallons per 1000 calories of potatoes. That means that beef requires 0.516 / 0.00759 = 68 times as much fossil energy to produce as potatoes.
(2) Meat-eaters are “driving” an extra 14 miles a day, vs. not eating meat.
- Page 147 of Food, Energy and Society shows that it takes 35,000 kcal of fossil energy to produce 3500 kcal for a typical daily American diet (10 kcal energy per food kcal), while it would take only 18,000 kcal to produce a pure vegetarian diet (5.14 kcal/kcal). (While 3500 kcal seems high, the study of food production energy generally looks at the total amount of food produced divided by the population, to account for waste and uneaten food. By that measure, the U.S. actually produced 3774 kcal per person in 2002 (Diet, Energy, and Global Warming), and 3900 kcal in 2006 for an average of 2673 kcal actually consumed (U.S. Food System Factsheets).
- The 17,000 extra kcal per day of energy inputs for a non-vegetarian diet is 6,209,250 extra kcal per year.
- At 30,000 kcal per gallon of fuel that’s an extra 207 gallons per year.
- At 25 mpg that fuel could power a car for 5175 miles.
- Divided by 365.25 days in the year, that works out to 14 miles per day.
(3) Energy for walking compared to energy for driving.
Diet for a New America reported this in 1987:
“It is actually quite astounding how much energy is wasted by the standard American diet-style. Even driving many gas-guzzling luxury cars can conserve energy over walking—that is, when the calories you burn walking come from the standard American diet! (62) This is because the energy needed to produce the food you would burn in walking a given distance is greater than the energy needed to fuel your car to travel the same distance, assuming that the car gets 24 miles per gallon or better.” From Diet for a New America by John Robbins (1987, p. 375), further attributed as chapter footnote (62) to Hur, Robin and Fields, David, “How Meat Robs America of its Energy,” Vegetarian Times, April 1985.
When I run the numbers myself I find that a 24mpg car still uses more energy than walking when fueled by a standard American diet, but for a 35 mpg car, walking uses more. Below is the math for 24 mpg, with the sources for these figures coming from my carbon calculator sources page.
- Average fuel economy for U.S. passenger vehicles (20.36)
- Fuel required for car traveling 20.36 miles (1 gallon)
- Fuel required for the driver:
- Average urban traffic speed of 31 mph (How Fast is Your City?)
- Time required to drive 20.36 miles = 31 ÷ 20.36 = 1.52 hours
- Calories consumed by driver: 1.52 hours x 136 calories per hour (Exercise.com) = 207 calories
- Calories required after adjustment (302) The U.S. produces 3900 calories of food per person per day for every 2673 actually eaten, as per note #2. So we need to multiply our calories by 3900÷2673=1.46, because 207×1.46 calories are produced for every 207 calories eaten.
- Energy to produce food, typical American diet (3020 kcal) (302 x 35,000/3500 as per note #2 above)
- Energy for driver (0.1 gallons) (3020 kcal / 30,000 kcal/gallon)
- Total fuel required for driving (1.1 gallons) (1 gallon for car, 0.1 gallons for driver)
- Time to walk 20.36 miles @ 3.2 mph (6.4 hours)
- Calories required (1434) (224 calories per hour for 150-lb. person; this is 74.7 calories/mile)
- Calories required after adjustment for waste (2094) (1434 x 1.46; see “Driving” notes above for more)
- Energy to produce food, typical American diet (20,940 kcal) (2094 x 35,000/3500 as per note #2 above)
- Energy to produce food, typical American diet (0.70 gallons) (20,940 kcal / 30,000 kcal/gallon)
However, to focus on which number is larger this misses the bigger picture. What’s important to know is that the more meat you eat, the more gas you waste. Meat-eaters use about twice as much fossil energy as pure vegetarians, no matter how they get around.
By the way, the Pacific Institute calculates that a beef-eating walker produces more CO2 than a car does for the same trip, which should come as no surprise.
(4) Energy for bicycling compared to energy for driving.
The original version of this web page stated that bicycling actually uses more fossil energy than driving, if the source of the cyclist’s calories are from beef. (Yes, we know that nobody eats only beef, it was just an example to show the staggering amounts of energy required to produce beef.) I based those calculations on figures in Pimentel’s Food, Energy and Society (1996), which Dr. Pimentel has since confirmed for me are overstated. (Thanks to reader Jeremy Hubble for giving me the clue I needed to set me on the path to discovering the error in the data.) The original figure I used was 13,000 kcal of energy for 140 g of beef (93 kcal per kcal of beef protein). The new figure is 40 kcal of energy for 1 kcal of beef protein. Beef production still wastes staggering amounts of fossil fuel compared to grain and vegetable production, it’s simply not so wasteful that biking uses more gas than driving. Remember, though, meat-eaters use about twice as much fossil energy as pure vegetarians, whether they’re bicycling or not.
(5) Save more gas saved by going vegan than by giving up driving two days a week.
This is from my carbon footprint calculator, which lets you see how eating vs. driving stacks up.
An earlier version of this page quoted Beyond Beef (p. 225) by Jeremy Rifkin, which said that going vegetarian saved almost as much gas as not driving at all. Now that I run the numbers myself, I believe Rifkin’s estimate was an exaggeration.
- Exercise calories burned as per note #6. For the pedicab I assumed the average between biking 12-14.9 mph and 14-15.9 mph. I believe the 7 mph estimate is valid, as I’ve clocked numerous pedicab trips with a wristwatch GPS device.
- Energy required to produce standard American and vegan diets as per note 2. For vegetarian diets I picked a number just below the average of these two. I used 2673 calories for the average daily diet as per a note above.
- Extra carbon from methane for non-vegetarian diets. 800 kg. per person per year, as per “Diet, Energy, and Global Warming” as per note #2, p. 9. That’s 800 kg per (2700 x 365.25=) 986,175 calories consumed, or 1 kg per 1233 calories, or 2.2 lbs. per 1233 calories, or 1 lb. per 560 calories.
- Gallons of fossil energy per kcal of energy (31,000) as per note 1.
- Carbon (and equivalents) per gallon of fossil energy (20.4) as per my carbon calculator sources.
- Physicist Tom Murphy gets a similar result. I get 75-145mpg for bicycling using the default values, and he gets 70-130 mpg.
- 32.3 kg of CO2 equivalents are produced per kilogram of beef (See Japanese study in Note #1 above.)
- According to the University of Michigan, it takes seven calories of fossil fuel on average to produce one calorie of food.
- Here’s how agricultural energy consumption is broken down in the U.S.:
- 31% for the manufacture of inorganic fertilizer
- 19% for the operation of field machinery
- 16% for transportation
- 13% for irrigation
- 08% for raising livestock (not including livestock feed)
- 05% for crop drying
- 05% for pesticide production
- 08% miscellaneous From Comparison of energy inputs for inorganic fertilizer and manure based corn production, McLaughlin, N.B., et al. Canadian Agricultural Engineering, Vol. 42, No. 1, 2000.