The bicycling paradox revisited

Last week we reported on some research from UPenn professor Karl Ulrich, who also happens to the founder of TerraPass, that came to the contrarian conclusion that bicycles offer little benefit to the environment. The nub of the argument is that bicycle riders are healthier than non-cyclists, and their increased longevity places a drain on energy resources that largely cancels out the gas savings from cycling.

The post sent little ripples through the blogosphere,* jumping from these pages to Salon magazine, and thence to the New York Times and myriad other web sites.

Reactions to the research varied widely, but most were critical. It goes almost without saying that only a handful were critical of the paper’s actual arguments, which few bothered to read. Rather, they were critical of the paper’s conclusions (or, more accurately, perceived conclusions), usually on the grounds that the conclusions are icky.

I thought it would be fun to dig back into this issue, partly as an exercise in web anthropology, partly because the paper itself is interesting and deserving of consideration, and partly because many readers did raise thoughtful points.

Before I begin, I’d like to state up front and for the record that we at TerraPass strongly recommend you replace time spent in your car with time spent on your bicycle. For numerous practical reasons, doing so will almost certainly offer large benefits to the environment and to your own health.

On to the issues raised. Many critics alluded to “holes you could drive a truck through,” so let’s take a look at some of these truck-sized holes.

Hole #1: The paper assumes that moving people from cars to bikes will increase their food intake, but lard-assed Americans already eat way more than they need, so this assumption is invalid.

This actually might be a fair point. It’s not clear to me that all people will increase their food consumption when they start exercising, at least not by the full amount assumed in Ulrich’s paper. Indeed, another study has suggested that drawing down the fat stored in Americans’ bodies could provide a meaningfully large source of emission-free energy.

The problem here is that food is not the major part of the effect that causes the bicycling paradox. Increased longevity is. So this is more of a mini-hole.

Author Bio


Comments Disabled

  1. Anonymous - July 24, 2006

    on the bicycling paradox
    compared to all the other things one might do that people do to extend their lives, such as flossing teeth, colonoscopies, etc bicycle commuting has a great energy balance profile. No gas is saved flossing teeth, but lives are extended, no energy is saved getting a colonoscopy, in fact in takes plenty of energy to provide health services and access them, but lives are extended. bicycle commuting extends life but it saves energy while doing so. So if extending life is a value we pursue, bicycle commuting is an energy efficient way to accomplish that. Isn’t that what it’s all about, finding energy efficient ways to attain what we value? The problem isn’t with the conclusions but with how they’ve framed the problem. If we have a population problem, then address that as it’s own issue. If denying people good health is a way to address that problem then I suppose more war would be a good one too, and let’s use environmentally benign but very lethal weapons.

  2. Adam Stein - July 24, 2006

    I agree with what you’re saying. So does the paper. Human longevity is valued by society, and if bicycling increases longevity while also benefitting — or at least not harming — the environment, then biking is an unequivocally good thing. We should all bike more.
    Biking is also frequently held up as an important potential source of greenhouse gas reductions. That’s why it’s more interesting to frame the paper as an examination of cycling than as, say, an examination of the global warming implications of flossing.

  3. noah - July 25, 2006

    The paper makes the assumption that if you live longer you’ll use more energy. That is a false assumption. If you live longer, you may use sustainable sources of energy.
    The paper *tried* to be deliberately provocative, but instead is a tedious distraction…One more thing for red state doofuses to happily and falsely bleat.
    Thanks Karl, not.

  4. Adam Stein - July 26, 2006

    Noah –
    The paper does not make this assumption. In the model the paper uses, the longevity effects from cycling causes an increase in the population almost immediately, so there is no reason to assume a change in the carbon intensity of our energy sources. This is Hole #6, discussed above.
    I personally don’t think in reality the population bump would occur immediately. More likely it will be pushed out a few years. So you’re right that there’s a best-case scenario in which we push our energy consumption out to a time in which we’re using more renewables. This is also noted above.
    Bear in mind that there’s also a worst-case scenario in which our future energy is less clean than it is today. There is a real danger that we’ll start deriving a greater proportion of our electricity from coal if we don’t get serious about carbon reduction. So it is not valid to simply assume we’ll be using renewables in the future.
    Finally, your statement about “red state doofuses” is pretty pointless. If the reasoning of the paper is valid, it’s worthy of consideration regardless of whether we wish the conclusion were different.

  5. Nils - July 26, 2006

    I think that a healthier population would reduce energy consumption in the driving to and from the health providers. My wife and I spend a major portion of our driving travelling to appointments. It seems that a good portion of our age group (64 yrs old) do likewise. I wonder if there have been any studies on how people use their cars.

  6. Jim Crimmins - July 26, 2006

    This type of argument is pretty silly, and it will alienate all but the most hardcore anti-human greenbats. Humans are a part of “the environment” – a large part at this juncture. Their health and well-being is important in considering how “the environment” is doing. Considering “the environment” to somehow exist independent of humankind is how a lot of the dangerous, expensive and ill-considered anti-growth proposals get started, which will actually end up costing large numbers of human lives. Good examples are ideas that demonize energy consumption, which has been responsible for the most of the increase in the standard of living and life-expectancy of human kind. Energy is just the potential to do work – the ability to harness a work producing entity widely and inexpensively is one good defintion of democratic societal progress. Clearly wasting this type of resource is bad form, and externalities due to its consumption should be priced in, but to consider it somehow morally wrong to consume it is silly.

  7. Scott - July 26, 2006

    I noticed the wear and tear issue on automobiles was brought up. Two points on that: First of all building cars is a HUGE energy consumer from the staindpoints of freight, plant electricity, employee transport, etc. (for that matter so is the production of maintenance drugs consumed by those who do not exercise, but that’s another argument) so if there were a way to produce fewer cars, perform fewer oil changes, shipping fewer replacement parts, etc. we would potentially be reclaiming a lot of energy by driving them less.
    However, the flaw there is that in truth cars driven less often require equal or greater mantenace than cars driven consistantly. Drivetrain components and major structural components are still primarily made of metals which corrode more quickly when the car is not generating heat (i.e. engine running and wheels traveling). There are very few parts of a vehicle that bennefit from sitting and that problem is more pronunced as you move further from the equator.
    So to me the ideal (and I said “ideal” I understand it’s unrealistic) would be to limit the number of cars produced to begin with. One of the studies stated that cars carry 1.5 people on average which says to me that we have built over twice the number of cars we need.
    Of course I say that with two cars in my garage (and I’m single). I can’t undo the environmental damage of a car that’s already been built. My desire is to build car-sharing and ride sharing structures into the way I live so that maybe fewer people have to buy (lit. pay to manufacture) another car to begin with.

  8. Jonathan Vaas - July 26, 2006

    Hole #8 – The Substitution Effect
    [Ed. — This is the same as Hole #5 above.]
    Most people engage in some sort of physical activity. Fit people tend to be quite active. This is why millions of people spend time exercising in gyms, spend time jogging, participating in sports leagues, etc. For Americans who sit all day at their office jobs, it’s essential to engage in some of these activities to stay in shape, which is a widespread goal.
    The carbon generated from exercising in a gym accomplishes no “work” in physics terms. You start and end up in the same place. And it makes people live longer, which results in more carbon dioxide being generated over time.
    But riding a bike for transportation will have a substitution effect for other types of exercise, and riding a bike produces “work,” getting a person from one place to another, while being far cleaner than other modes of transportation.
    People who ride bicycles for transportation are far less likely to get their exercise in other, nonproductive ways. If you bike 30 minutes to and from work, there’s no need (and little time) to spend an hour in a gym.
    Since people are already staying in shape (and these fitness-oriented people are the most likely to ride bicycles), it’s better for the environment if they stay in shape on bicycles while performing the “work” of transportation, rather than doing it within the confines of a gym.
    I think this substitution effect is significant, and by itself negates the theory that biking has no positive net effects for the environment. I know I don’t go jogging or weight lifting when I’m biking long distances around town.

  9. Anonymous - July 26, 2006

    I wish Professor Ulrich had used the peer-review process on this paper and published it first in an academic journal before presenting it on the web. This process would have hopefully required him to strengthen his arguments and fill some of the holes. Bad press for him is bad press for the entire academic community that attempts to rigorously study sustainability tradeoffs.

  10. Tom - July 26, 2006

    Alternative perspective: The blogosphere is the best peer review process. Not only has a huge amount of commentary and intelligent argument been produced, but Karl (and presumably his policy audience) has the added benefit of seeing how the argument plays out in public.
    As for me, I’m on the train about to bike commute to work 😉

  11. gpc - July 26, 2006

    uh – 20 minutes on the bike machine at gym vs. 10 min. round trip bike commute? no contest…

  12. pradwastes - July 26, 2006

    I used to ride a bike a lot and came very close to being killed twice. Now I use the gym with no fear. Late in the evening I am able run further and faster than most high school students. I am 57 year old and feel wonderfull.
    It took me a long time to find a job that is only 8 miles from home and now I am able to work at home.

  13. Robert - July 28, 2006

    Your attempt to dismiss “hole 7″ was ineffective. I don’t actually believe the paper was deliberately written to bolster TerraPass, but it may well have that effect. A lot more people drive than bike commute, and if many of them have been thinking of switching to cycling for environmental reasons, this paper may dissuade them, and make them look for another answer, like TerraPass.
    I’m just sayin’.

  14. Adam Stein - July 28, 2006

    Well, there is another small problem with the conspiracy theory. Namely, the paper itself is hardly anti-cycling. In fact, it is strongly pro-cycling. People may have been generally aware that cycling is good for you, but who knew that it could literally add years to your life? Given that most people are more concerned with their waistlines than with the environment, the paper is actually a strong advertisement for cycling.
    Anyway, all of this is yesterday’s news. Keep your eye out for our upcoming research report, entitled “Hybrid Cars Give You Herpes.”

  15. Walt - August 2, 2006

    Karls article reminds me of why I no longer consider myself an enviromentalist. I don’t ride my bike to save the world. I ride my bike because I love riding my bike period. Although I did start cycling back in the 70’s with a proud ecology sticker on my water bottle, enviromentalism seems too much like a religon when the 1st priority is the earth at the expense of the individual. Does that mean I’m for reckless pollution and energy consumtion? Nope, I just live a simple life with no guilt that my personal lifestyle is destroying the planet. Heck I even have a B.O.B. bike trailer that I occasionally use to haul up to 7 radio control gliders to the local bluff and do some flying. But that’s nothing compared to some real diehards I know who ride all through the winter and have been known to haul lumber and building materials on their bikes. You’d be amazed at what a bike can do. People in many 3rd world countries realize it because many times it’s all they have.
    I do strongly encourage people who have never riden their bikes to work to just try it one day and see how difficult or how easy it is to do and if they do it for enviromental reason’s, I’m ok with that. Heck I’ll even say good morning or good afternoon when I pass you!

  16. AndyTiedye - August 4, 2006

    Looks like his exercise figures might be off by about a factor of 5.
    On page 3, he says riding a bike requires 17kJ/km.
    On page 4, he says that 4200kJ/wk corresponds to 2600km/yr of cycling. That is 50km/wk.
    This implies that each km consumes 84kJ of energy.
    If we used the figures on page 3, then these formerly
    sedentary cyclists would have to be riding 250km/wk!
    Does he think we’re all going to enter the Tour de France?
    It looks as if he may have used the food-input figure
    instead of the energy-output figure.

  17. Adam Stein - August 5, 2006

    I asked Karl about this. Here’s what he said:
    “The factor of five the reader alerts us to is about right because the metabolic efficiency of cycling is about 22%. The physiologists use total metabolic energy expenditure (4200 kJ/week) and the bicycling engineering is in terms of net energy input to the pedals, which is only 22% of that.”
    In other words, you’re right about the factor of five difference, but it’s not a bug. The factor of five is the difference between total energy burned and energy actually delivered to the pedals.

  18. justin - August 9, 2006

    One comment I have yet to see is regarding the measures of environmental impact; one point of debate that wasn’t addressed by Karl’s work. Could other measures of environmental impact other than energy expendatures also contribute to the argument in favor of cycling? I’m thinking about emissions – spilled fuel, exhaust, etc. Would more humans huffing and puffing put forth more CO2 than automobiles? What about the energy it takes to dispose of/ manufacture said transportation devices?
    Karl’s paper is far from comprehensive, but it is a great contribution to the debate.

  19. Adam Stein - August 10, 2006

    No, the paper doesn’t analyze the consequences for emissions of particulates or any of the secondary environmental effects of car ownership. On the one hand, the longevity effect really should carry over to most of the environmental consequences of participating in the modern economy, not just energy usage. On the other hand, prudence and common sense suggest that the world would be a lot better place if we replaced a portion of cars with bicycles.

  20. Thom - August 11, 2006

    I’m a bit skeptical of the longevity effects of cycling (or exercise). It is unlikely all these benefits are from exercise. Cyclists (in particular) differ from non-cyclists in all sorts of ways that are associated with longevity. These include education, lifestyle, where they live and so on.
    Karl Ullrich acknowledges this and argues “although the evidence is from observational studies, the basic findings hold even when controlling for sex, smoking, economic status, disease status, and other lifestyle factors” (which deserves credit and shows he’s been thinking carefully about it).
    My point is that it is very unlikely that the estimates of increased longevity take all the relevant factors into account. My educated hunch is that these factors are underestimated (though it remains a statistical possibility that these factors are over-estimated). Statistical controls just aren’t adequate for this task given the number of factors involved, IMO. At the very least one has to consider the potentially huge margins of error around the longevity estimates.
    Certain other impacts of cycling seem hard to gauge and probably depend on other aspects of transport policy. For example, by and large cycling generally replaces inefficient short car journeys in congested towns and cities. This could swing thing in favour of cycling, but it depends on whether reduced congestion lures people back on the roads from public transport. I suspect that promoting cycling needs to be thought of as part of an integrated transport policy. Promoting cycling could reduce congestion and (with the right policies) make public transport more attractive.

  21. Christine - August 18, 2006

    Has anyone studies the amount of CO2 expelled by humans as we exercise? I think this should be taken into account. How much would it offset the savings from driving a fuel efficient car? I love to bike, but there is no way I could bike to work. (too far plus summer sweat issues and winter weather issues).
    I also think there is a problem with the “most cyclists are already fit” logic. Most cyclists already ride their bikes. We are talking about expanding that group to include those that currently drive everywhere. Maybe some of them will use it as a substitue for their gym time, but really, most American commuters do not go to the gym as can be seen in the constant reminders the media puts out as to how fat we are.
    But really, it doesn’t matter. People won’t bike to work. It’s just not practical. My commute used to be 15 miles each way. If and when driving becomes too great a burden, improvements in public transportation and more telecommuting will fill the void.

  22. Darren - August 24, 2006

    All told a very thought provoking paper. My initial reaction was “Hogwash!” but as I think about it, transportation to and from work is a relatively small part of my anual energy budget. The current price of gas magnifies our awairnes of commuting as an energy cost.

    That said, their are a number of non-energy environmental advantages for bike riding. One I have not seen mentioned yet is that bicycles need much less infrastructure in terms of road construction and maintenance. Twenty cars trying to use a country road at the same time is a trafic jam. Twenty bikes would be a club ride.

  23. Stanton Byrd - September 21, 2006

    This is what I wrote Karl Ulrich:
    Hi Mr. Ulrich,
    I just wanted to make a comment about your paper that some say is ridiculous. Firstly, it isn’t ridiculous. But I just wanted to pose a possible alternative theory to yours.
    Your argument supposes that increased biking for commuting purposes would mean that fat people would hop on bikes, live longer, and therefore the environmental utility of their non-polluting would be offset by the environmental disutility of their living longer. BUT very few sedentary people would ever conceive of taking up bicycle commuting. It’s just too extreme. I rarely ever see fat people on bicycles going anywhere. A bicycle seat is the last thing a fat person would ever want to sit on.
    My theory is that an increase in bicycle commuting would mean (mostly), that people who are active anyway would start riding bikes instead of engaging in other useless exercise–like crunches or pushups–exercises that have no benefit other than the immediate health benefits.
    So, my theory is that increased bicycle commuting would mean that people who would be generally inclined to engage in physical activity anyway would commute by bike. A bicycle commuting revolution would only mean that people who otherwise spend $50 bucks a month to go to a gym to exercise on stationary aerobic contraptions, would instead, expend that energy for the environmentally useful purpose of going to work.
    If this is true, it would mean that fat people would continue to die–good for the environment and fewer people would be polluting–also good for the environment. Everyone wins!
    This is what Karl Ulrich wrote back:
    Yes, your logic is correct.
    Unfortunatly there aren’t very many fit people in the US. I think the
    population you refer to is less than 10 percent. So, it would be great
    to have those folks on bikes, but it would be a small win relative
    to the revolution envisioned by most cycling advocates.
    My response:
    Hello, Me again,
    I was looking around on the internet for numbers and figures and such and came across an article that appears to have run in 2002 in the Washington Post ( The article says something to the effect that 30% of Americans engage in regular physical activity and that 40% of all Americans are certifiably sedentary. I think your figure of 10%–which you suppose represents the percentage I refer to in my theory, is pessimistic. I think its probably closer to 20 or 25%. But I think even 10% would benefit the environment significantly without the spectre of our longevity offsetting said benefits. Either way, you should probably address this theory in your work to avoid being intellectually disingenuous.
    Comment on Hole # 5:
    If Ulrich is indeed arguing that getting all of America on bikes is a net loss for the environment, his argument is not profound or even useful. Since getting all of America on bikes would never happen, he is simply arguing the ins and outs of an improbable thought experiment that has no foundation in reality. What is the use in that?

  24. John - October 1, 2006

    Comment on Hole # 1.

    I started communting by bicycle this year (40 mile round trip). Food consumption went up after I lost 20 pounds. Bicycling magazine provided a calorie intake forumla that shows how many calories are required to maintain your current weight based on the number of minutes exercising per day.

    So if you are overweight, and don’t change your eating habits you lose weight. You will continue to lose weight until you body determines it needs more fuel.

    (current weight in lbs X 15) + (minutes daily exercise X 10) = total calories required to maintain current weight

  25. Christa Ansbergs - October 4, 2006

    Having now finally gone back and read the actual study (prompted by your comment in the most recent newsletter), it occurrs to me that there are a couple of other things that get left out of the analysis.

    If we’re actually talking about EVERYONE getting out of their cars and into a bike then auto traffic will be significantly reduced. This means either a) fewer traffic jams and thus increased fuel efficiency for the vehicles that are still road-bound OR b) fewer road widening projects, which use significant energy to build. Also, stores and workplaces will not need to build their parkinglots as large resulting in additional energy and resource savings.

    Also, if everyone is biking as a commuting exercise that means they must be moving closer to work or it simply wouldn’t be feasable – thus even on days when they are lazy and drive they will be driving shorter distances. I suppose they could supplement miles biked with miles on public transit but that would ALSO reduce miles driven. Personally I’ve done both of these things – I moved within biking distance of work and when I commute from farther (boyfriend’s house) I pop my bike on the light rail for about half the ride.

    I’m actually keeping a detailed spreadsheet of exactly how many miles of car driving I’ve saved by combining biking with public transit (I’d quote specific numbers but I’m at work and the spreadsheet is at home – I believe it is on the order of 700mi since the beginning of April). I then use an average gas economy for my car to calculate my monetary savings and use that as a fund from which I am allowed to buy new toys for my bike – things like a new cyclometer and a reflective vest. I haven’t actually calculated what my increased food intake is due to bicycling though. That would require taking time off biking to get a baseline consumption and I’d just as soon not do that! Plus, at around the same time as I started bike commuting I also started buying a significant chunk of my food from community supported agriculture (CSA) – a local organic farm that uses far less energy for fertilizer and transportation than conventional agriculture does.

  26. Adam Stein - October 4, 2006

    Christa —
    You are very cool. I love the idea of rewarding yourself with bike goodies based on the amount of gas you save. I had a sort of similar idea when I read Karl’s article — namely, I realized that I could easily rationalize buying that carbon fiber touring cycle I’ve been lusting after based on the economic value of the additional life-years cycling offers.
    But I like your system better, and I commend you for the obsessive devotion required to actually put it in practice.
    – Adam

  27. Tom - November 15, 2006

    The major theme I took from the paper was something along the following:
    If you think helping the environment is a priority, you shouldn’t really do anything. But when you have to go somewhere, so do on an electric scooter. Maybe. Better yet, just die early and don’t use any resources.
    In all honesty, this paper just seems to try to provoke people without posting any potential solutions in the “concluding remarks”
    In fact, just to reword the above, this paper seems to want me to choose between living a good long healthy life, or doing some good for the environment.
    Somehow, I don’t think that is the either/or the paper presents.

  28. Jake McKeegan - December 1, 2006

    Back to concern that Dr. Ulrich did not initially choose to release this paper to an peer-reviewed journal… blogospheres have their place, but the ethical first step for the release of any scientific paper is to have the research or theory mulled over and debated by those who are able to understand the science involved and can rule out conflict of interest. Once it is confirmed in that venue, then policymakers and any other Jill and Jack can take it to the public to see how it plays out. Circumventing the review process becomes dangerous, and there are plenty of unfortunate examples where this has led to a poor outcome.

    Regardless of your waiving off as untrue, this release still has the appearance of simply generating public buzz.

  29. Capra - May 7, 2007

    Karl assumes people driving a car don’t eat anything. Okay, so that may be a small part of the overall energy equation, but I’ve been bike commuting for 20 years, and I calculate that with the minimum amount of time I spend on the bike each day that means I’ve so far spent 1 1/2 years of my life biking, which is more than I’ve added to my life by his calculations. So the argument that people spend the majority of time they add biking is valid.
    Moreover (and most importantly), he does not adjust the life expectancy figures to account for the fact that while bikers are on mixed roads with cars they are breathing what is coming out of the tailpipes of those cars, which includes deathly carbon monoxide, carcinogenic dioxins and aromatic hydrocarbons, and toxic sulfur dioxide gas. I’m not convinced that biking is going to make my life longer–the contrary–but what it does is make the amount of life I will lead better.

  30. RichardL - November 25, 2007

    Additional evidence against hole #7 (Karl Ulrich’s conspiracy) is that Karl is apparently a principal in a company that makes folding commute bicycles and scooters.

  31. George Bush - April 3, 2008

    Bicycling makes people create a lot more Carbon Dioxide that advances Global Warming. If more people would sit on their butts in front of their T.V. instead of bicycling, it might keep gas from escaping form there as well.

  32. Reba Strickland - January 9, 2009

    good luck