Steven Dutch, Natural and Applied Sciences, University of Wisconsin - Green Bay
First-time Visitors: Please visit Site Map and Disclaimer. Use "Back" to return here.
Thought ExperimentOne day in 1974 or so, I was sitting in my car (actually my thesis adviser's university car) inching across the George Washington Bridge on my way to Manhattan to meet a class where I was the teaching assistant. Suddenly I asked myself "Why am I doing this?" After all, I had alternatives. A bus ran right by the Lamont Observatory where I spent most of my time and went reasonably directly to the uptown bus terminal in Manhattan. From there I could take a subway straight to Columbia University. So as mass transit goes, it was a pretty straight shot. So why was I driving? Well, for openers, the mass transit really didn't save much time, especially counting waiting time at both ends and the transfer from bus to subway. And it was impossible to do anything productive riding mass transit. Plus there was no privacy or peace and quiet, which I finally decided was the major factor for me. And people in those days worried a lot about subway muggings (realistically, on the 7th Avenue IRT in the daytime, a minor risk), but carjacking was unheard of, so there was a safety issue.
Since I have never, in the 30 years since, seen any article by advocates of mass transit - not one - that bothered to ask why people don't take mass transit despite all its supposed advantages, I thought it might be useful to explain why people prefer to drive instead of take the bus. Most advocates of mass transit dismiss drivers as selfish, short-sighted and unconcerned about the environment instead of asking whether mass transit itself is to blame for its own problems.
After this page was linked by another site, I got a number of responses that suggested a bit of clarification is in order. This page is not calling for abandonment of mass transit or extolling the virtues of the automobile. It is an attempt to lay out what mass transit is up against if it is to succeed. Pretending that the economic issues I describe can be made to go away is a guaranteed recipe for failure. They won't. Lots of people seem determined to illustrate the is/ought fallacy in action.
Also I've gotten a number of responses from people who say the factors I outline here don't apply because they spend their time on the bus or subway reading or relaxing. This amounts to an attitude all too common in environmentalism: everything will be just fine once people get enlightened and see things the way I do. But don't take my word for it - see the exchange at the end of this page. If you have access to a user-friendly mass transit system and can use the commute time productively, bully for you. I'm trying to explain why so many other people don't see it that way.
The Value of TimeApart from the cost of wages, economic planners rarely acknowledge the value of individual time, but that has absolutely no impact on the reality that people themselves do put value on their time. As John Naisbitt pointed out in Megatrends, one of the first thing people do when they acquire some affluence is begin to buy back their time. They hire out boring or unpleasant tasks like food preparation, housekeeping, child care and repairs. (Home delivery services are even enjoying a bit of a resurgence as two-earner families find themselves increasingly pressed for time.) Failure to recognize the value of time to individuals leads to unproductive results.
Nowhere is this issue clearer than in attempts to deal with the problems caused by the automobile. Critics of the automobile point out that in addition to the direct costs of the automobile like fuel, maintenance, and depreciation, there is the cost of highway construction, environmental damage, tax subsidies, defense of oil supplies, and so on – a host of “hidden costs.” For example, The International Center for Technology Assessment, in The Real Price of Gasoline, and Stephen H. Burrington in Road Kill: How Solo Driving Runs Down the Economy, both estimated the real cost of driving a car at about a dollar a mile. They estimated the cost of a bicycle at twelve cents a mile.
I live eight miles from campus. At a dollar a mile by car, it costs $16 to commute. It takes about 20 minutes each way, so figuring my salary at $25 an hour, the cost comes to about $33. Occasionally I bicycle. It takes 45 minutes each way. The cost of bicycling alone is only $2 a day, but the time cost is $37. It costs $39 a day to commute by bicycle. By mass transit, I have to walk to the bus stop, go downtown, transfer, and travel a winding route to campus. Total fare is $2.50, and counting time walking to and waiting at the bus stop at either end, it takes at least 45 minutes to make the trip by bus, bringing the total cost to around $40.
There are plenty of good reasons to encourage mass transit, but arguments about the hidden costs of the automobile fall on deaf ears because people, unconsciously or not, factor time and convenience into their decision making. The average driver knows perfectly well why she drives.
The cost of a transportation system is first of all, any flat fare. Call that F. Then there's a cost per mile (call it C) and the mileage (M). The value of your time we can call S (salary per hour), and the time it takes to travel is T. So we have Cost = F + CM + ST. Time will be mileage divided by your speed (V), so we have Cost = F + CM + SM/V = F + M(C + S/V). We can see that cost increases with mileage (obviously), high time value (every minute traveling costs more) and low speeds.
Conclusion 1: Transportation Costs Less at High Speeds. High-speed commuter rail is a great solution if there's easy access at both ends. If you have to drive five miles to a transit station only to find the commuter lot full, you may as well drive. HOV (high occupancy vehicles) and mass transit lanes on freeways are another good approach to this issue. The best features of HOV lanes for private vehicles is they offer a positive incentive to carpool (you get to pass all the solo drivers), rather than the negative penalties that are the only solution many advocates of mass transit seem capable of imagining.
Corollary: Low Speed Limits Raise the Cost of Travel. They may cut fuel consumption and costs of accidents, but the time cost rises steeply. Where I live, a nearby suburb has a four lane street with a speed limit of 25 miles an hour. It could easily be raised to 40 with no significant safety risk.
Corollary: Interruptions Raise the Cost of Travel. How much gasoline is burned daily by cars stopping and accelerating at stop signs where there is clearly no oncoming traffic, or waiting at empty intersections for traffic lights? Probably half of all stop signs could be changed to yield signs. And it should be legal to proceed through a red light if there is no oncoming traffic. Accidents would be wholly the responsibility of the driver going through the light. School buses should be required to wait for traffic to clear before turning on their signals and discharging students.
Let's assume, as critics of the automobile say, that a car costs $1 a mile and also assume a car averages 20 miles an hour in city traffic. The cost of operating a car becomes M(1 + S/20). If we assume a bicycle costs 1/8 as much per mile and goes 10 miles an hour, then the cost of riding a bicycle is M(1/8 + S/10). The extra cost of driving a car per mile is:
Cost (car M=1) - Cost (bicycle M=1) =
(1 + S/20) - (1/8 + S/10) = 7/8 - S/20.
If the cost difference is positive, bicycle is cheaper. If it's negative, a car is cheaper. When the cost difference is zero, both forms of transportation are equal. Call that the break-even point. That happens when S/20 = 7/8, or S = 17.5. If S is less than 17.5 ($17.50 an hour or $35,000 a year) then the cost is positive, otherwise it's negative; it costs more to go by bike than by car.
Conclusion 2: Slow Transportation Penalizes Affluent Customers. And these are the people most likely to have their own cars and to move further from work.
Corollary: Affluent Customers Will Not Use Mass Transit. It's not that they're selfish, or that they don't care about the environment. It's not cost-effective. The higher your salary, the more wasteful mass transit is. The only significant exception is commuter rail provided the fares offer a savings over driving and parking and the comfort and privacy allow relaxation or work en route.
Corollary: Infrequent Transit Schedules Discourage Use of Mass Transit. Duh. Or maybe not. My city is considering cutting frequency as a "cost-saving" measure.
If we assume the fare on a bus is $2, and there's no extra cost per mile, and buses average 15 miles an hour (because of stops and less direct routes), then the cost becomes 2 + S/15. The extra cost of driving is Cost (car) - Cost (bus) = M(1 + S/20) - (2 + SM/15) = M - 2 - MS/60. This is a bit harder to analyze because it's mileage-dependent. We can find the break-even point by making the cost zero and solving for S: S = 60(1 - 2/M). If M = 2, S =0; it always pays to drive because the cost of driving beats the flat fare. Regardless of how big M is, S is never greater than 60; if you earn over $120,000 a year, it always pays to drive. If M = 4, S = 30, and the break-even point is $60,000 a year. If you earn less, it pays to use mass transit.
But if the fare is $5, as it can be for long commutes, then S = 60(1 - 5/M). It never pays to take the bus for commutes less than 5 miles. For S = 30 ($60,000) a year, the break-even point is 10 miles - any longer than that and it pays to drive.
Conclusion 3: Flat Fares Discourage Use of Mass Transit for Short Commutes A fair number of cities seem to have figured this out and have free-travel zones downtown, unlimited travel passes, and similar offsets.
If traveling by car really does have high indirect costs not shared by public transportation, the case for making all mass transit free is so compelling you really have to wonder why advocates of mass transit don't propose it. Also, since a major cause of urban sprawl and congestion is the middle class moving to the suburbs, the obvious cure is to eliminate the problems that drive the middle class out. Unless there's some master plan to have buses, ambulances and fire trucks all get around on light rail, most of the indirect costs of the automobile will still plague mass transit. We can hope to lessen the dependence on petroleum, and hence ease prices and maybe reduce the defense threat. We might also hope to reduce the costs of road repair, reduce air pollution, and lessen the impact of the automobile.
There's a good reason why people who play the "hidden costs" game never factor in the value of personal time saved - it tips the balance so sharply in favor of existing technology that alternatives simply cannot compete. (Actually, when people say they "cannot" compete, they usually mean they will not compete because they don't think the rewards are great enough. Mass transit can compete against the private auto but it would require subsidies to the hated middle class and suburbs.)
One correspondent added:
An important wrinkle that I feel is missing from your analysis; Time saved in transit is added to my free time with my family, not to time at work. I value my time outside work much more than my hourly wage. That is why when my employer wants me to work more, he has to pay me time and a half. Or, when another firm wants to buy my extra hours, I charge them double to triple my hourly rate. (emphasis added) Therefore, your point is stronger than you present. The time I save by driving is extremely valuable to me. Much more than my hourly wage. I think I'm not alone.Funny how evil corporations routinely recognize the value of personal time by paying higher than normal salaries for overtime, but enlightened mass transit advocates, who care so much about people and the good of society, somehow just don't get it.
Exact ChangeIs there a single, more stupid tactic for discouraging mass transit than requiring exact change? Especially when fares change frequently enough that a new user can't find out the fare except by calling the transit company? Hopefully, rechargeable fare cards will become universal enough to remedy this problem. Systems like BART and many European systems that use vending machines for fare, of course, don't have this problem.
Fixed CostsIn addition to the per-mile indirect costs of owning a car, there are fixed costs that exist whether you drive the car or not. Chief among these is depreciation. Depreciation is not that much of an issue for people who buy used cars and drive them as long as possible, but for those who buy new cars and trade them in regularly it's a major cost. Depreciation has to be added to the cost of whatever transportation the individual uses. If the person drives, depreciation is part of the cost of driving, obviously. If the person uses mass transit, depreciation is still part of the cost of using mass transit because the person has a car sitting in the garage unused, but still declining in value. In fact, all hidden costs have to be added to the cost of mass transit - you still pay taxes to pave roads and defend oil supplies whatever you do. Only out of pocket expenses count in determining the cost-effectiveness of mass transit versus the automobile, because the indirect and "hidden" costs are still there whatever mode of transport you use.
Once someone decides to buy a car, the economic balance shifts sharply in favor of driving. The only way to shift the economic balance in favor of mass transit is to create a system where it becomes feasible for large numbers of people to give up owning a car. A few moments' thought will suffice to reveal the requirements for such a system:
- The out of pocket costs must be the same or less for public transport as for private transport. You might get away with a slight overage if public transport offers a real premium in convenience or comfort, but it had better be a clear advantage to the consumer.
- The time costs have to be comparable. This means:
- Actual travel time has to be comparable. The convoluted fractal routes that buses typically travel to access the largest possible area with the fewest routes are a guaranteed recipe for a failed mass-transit system.
- The schedule has to be frequent enough that transfers have negligible time impact. If you occasionally have to run errands en route, the transfer time factor demolishes mass transit.
- The schedule has to be frequent enough that waiting time at the trip origin has negligible time impact.
- The system has to be dense enough that transit time from the final stop to the destination has negligible time impact. Walking half a mile in the pouring rain negates anything positive mass transit has to offer (and no combination of rain protection will keep you dry in a real downpour.)
- The system has to be more dependable than a private automobile.
- Work stoppages and strikes are absolutely impermissible. I met some folks recently who saved on the outrageous hotel prices in Venice by staying in nearby Padua. Then, when it came time to catch their cruise ship, the trains were out because of a strike to protest President Bush's visit to Rome. Because, you know, people traveling from Padua to Venice are directly responsible for the war in Iraq and globalization. And labor activists wonder why unions fell out of favor in the U.S.
- The system has to have enough peak capacity to carry all passengers in reasonable comfort. Sitting down. With elbow room and a modicum of personal space.
- Routes have to be simple and absolutely fixed. Far too many systems vary routes with time of day, use the same number for different routes, omit stops or entire segments of the route at times, or change routes frequently. When I'm in a city and have a choice of rail or bus, I take rail every time, simply because you can't rip up tracks capriciously and reroute them. (I did see a city once where it happened - would you be surprised if I said it was Sofia, Bulgaria?)
- Information about the system has to be available everywhere. Every stop must have a map of the whole system with schedules and fare information, and the information must be current. Areas between stops must have frequent signs to the nearest transit stops.
- The system layout has to be predictable. Ever been in a city and walked to a major artery hoping to find a bus stop, only to find the buses don't run on that street? Instead the buses run down some residential street because the system is trying to cover the most ground with the fewest buses, or some alderman lives there and wants convenient bus transportation. And how about that system of identifying routes by the end of the line? Boy, that sure makes navigating mass transit in a strange city a breeze!
- Transportation has to be available at all times - 24/7/365. If you even occasionally find yourself going places on holidays or odd hours when transit is either unavailable or infrequent, you'll opt to get a car.
- Car pooling? If the passengers all have similar origins and destinations, it's an option. But if people need to vary their schedules, run errands en route, be out of town on business, and so on, it won't work. The lack of flexibility is probably the main impediment to car pooling.
- The system has to be absolutely safe. Law enforcement needs to be thorough enough, the penalties for crime severe enough and the judicial system hard-nosed enough that nobody would even think of committing a crime on a bus or subway. And there have to be strict rules of conduct. No, you do not have a First Amendment right to panhandle on the subway.
CargoIn New York City, someone who lives alone might be able to buy groceries every single day and tote them home. But what about someone with five kids? What about someone who needs to transport sheets of plywood or drywall, concrete blocks or sacks of fertilizer? In a few places, buses have provisions for carrying bicycles, but for the most part people who have frequent needs to haul cargo have no real alternative to the automobile. Delivery services might alleviate this problem somewhat.
GroupsWhile visiting my parents in the San Francisco Bay Area some years ago, we decided to take a trip to Fisherman's Wharf via the BART system. There were six of us altogether. We found the lot at the BART station full, so we drove in to San Francisco. Even counting bridge tolls and parking, it only cost a little more than riding BART.
When transporting a group, cars almost always beat mass transit. Mass transit systems that fail to recognize that the unit of travel is the group, not the individual, are doing more to promote automobiles than Detroit ever could.
A Visit to PhiladelphiaI don't share W. C. Fields' dark view of Philadelphia. I like the city very much. Putting it far above average for large cities is its direct rail link from the airport to downtown (that's changing as more and more cities come on line). So on a recent trip to Philadelphia, I booked a motel close to the airport to save expenses and took the train to the convention center downtown.
Both ways the train I intended to take was canceled, meaning I had to wait an extra half hour. At both ends of the trip there were fare machines out of service (although conductors will collect fares on the train). There was a bus link from the airport to my motel, and once I found the bus schedule that part of the trip worked smoothly. The buses actually were right on schedule. But it took a number of tries on the automated phone system to get the inbound schedule, and the Visitor Center downtown didn't have printed schedules. What, post the schedules at the bus stops? Are you mad? They needed those big plastic panels for advertising. At least the stops did indicate the lines that stopped there. And then there was the able bodied panhandler working the transit station downtown. All day. He hit me up coming and going, four hours apart.
On the whole, I got where I needed to go, but this anecdote illustrates all the minor indignities that mass transit advocates expect people to endure for the sake of society. And this is the state of affairs in a city with excellent mass transit. And we wonder why people prefer their cars.
Oh, and then I got the credit card bill for "long distance" calls from the airport to downtown to get schedule information. Factor that into the cost of mass transit because the information wasn't posted at bus stops or in the phone book, and the transit system didn't have a toll-free number.
Houston, You Have a ProblemFrom a Houston Chronicle story, August 7, 2008. The story details the travails of bus commuting in Houston. Spokespersons for the transit system claim:
Metro officials say they understand Jenkins and Camarillo's frustrations, but the street infrastructure has to be in place, said Jim Archer, manager of service evaluation. Roads don't go all the way through and many are incomplete, Archer said. "When you look at Airport (Blvd.), Airport would be a nice logical route except the road doesn't go all the way through," he said. "The streets make it very difficult for us to operate buses. ... This is not because of Metro."
Yet one of the people interviewed for the story takes anywhere from 80 minutes to two hours to commute, but when someone offers her a ride to work, "the commute time drops considerably — to 10 minutes." Somehow cars can cover the distance in ten minutes but buses can't, because it's "very difficult." But "This is not because of Metro."
What we got here are whining incompetents who can't, or won't, do their jobs.
Wrote one commentator:
The Irvington bus passes in front of my home. If I catch the bus to downtown, it is 1 hour 20 minute ride one-way and I have to walk 8 blocks to my office no matter what the weather. Since Metro discontinued the circulator system downtown, this adds another 20 minutes to the commute. If I drive to work, it is 25 minutes one-way and I park at the door and walk inside 20-feet. Let's see, 2 hours on the bus and walking 8 blocks in rain/cold/heat or 25 minutes and drive up to the door. Oh, let's double that for two-way-commute! ... dah ... Metro is a waste ...
ConclusionsIn New York City, it can make sense not to own a car. Parking is prohibitive, the risk of damage from on-street parking is severe, and the transit system beats driving much of the time. In Moab, Utah, fuhgeddaboutit.
In sparsely-populated areas, there simply is no practical alternative to the automobile. People who live in those places need cars to get around and haul cargo. People who need to get to places not served by mass transit also have no alternative to the automobile. So what are the possible solutions?
- Inexpensive Rental Cars. The cost of auto rental has come down to the point where it's pretty affordable, but it needs to come down still further to make it a really viable alternative to using the private auto.
- Inexpensive Taxis. These need to be considered part of the overall public transit system. Fares need to be competitive with comparable distances on mass transit, and availability needs to be great enough to avoid significant time penalties.
At off-peak times, there simply is no practical alternative to the automobile. The remedies are the same.
People who haul cargo have no practical alternative to the automobile. Remedies include inexpensive delivery services, but frequently bulk cargo purchases include small items or unanticipated on-the-spot purchases. Inexpensive shipping from the point of sale, or cheap truck rental, are additional possible remedies.
The only way to diminish reliance on the automobile is to create a mass transit system that is superior to the automobile by the standards of automobile users. In many circumstances the most effective system is the automobile and the only way to cut use of private automobiles is by supplying public automobiles, like rental cars and taxis. The sci-fi vision where you go up to a vending area, pop in a credit card, and drive off in a waiting car, needs serious consideration. Where density is high enough, the only way to cut reliance on private autos is with mass transit that is competitive with automobiles in out of pocket cost, speed, and convenience.
Attempts to promote mass transit through coercion will inevitably fail. Trying to make mass transit more competitive by raising auto registration fees, parking fees, bridge and tunnel tolls, gasoline taxes, and the like, will inevitably be seen for what it is: artificial manipulation of the marketplace to coerce drivers into using mass transit. Trying to encourage mass transit use by penalizing private auto use amounts to an open admission that mass transit cannot compete with the automobile.
Voodoo Economics won't work. I have to pay taxes to build roads and defend our oil supplies whether I drive or not, and fire trucks, ambulances, and delivery vehicles need streets to drive on. Pretending that I somehow avoid those "hidden costs" by taking the bus is beneath stupid. Telling me that 45 minutes in a crowded, lurching bus is better or a more effective use of my time than 20 minutes in my car is a couple of levels below that.
Wishful thinking won't cut it. It will do absolutely no good to say all these problems will go away if we can somehow persuade Americans to accept higher density and move back in from the suburbs. Suburbs began to sprawl back in the days of streetcars. Americans do not want to live in high density settings. Why not just accept it and plan accordingly?
Studies have repeatedly shown two things: the more transportation is available, the more people spread out. Second, commuters start to get irritable when commute times exceed half an hour. Basically, commuters move out to a distance where they feel the time cost is acceptable, and get angry when the rules change. Moral: Americans like to spread out until other individuals do not seriously impinge on their freedom of action. Deal with it.
What Gated Communities Teach UsI consider gated communities (and their cousins, the restricted covenant communities) loathsome. Whenever I hear about some homeowner embroiled in a dispute with his homeowners' association, I am torn between despising the homeowners' association for being so petty, and the homeowner for being so stupid as to live in such a place. But they are growing in popularity, and that has something to tell us, and we'd better figure out what that is. What do these communities offer?
- Safety. Covenant communities merely merge into the surrounding neighborhoods, but gated communities are walled cities. Paradoxically, concern over crime seems to get worse as society gets safer and crimes, being rarer, become more newsworthy. Nevertheless, crime is a principal reason why affluent people leave cities. So if you want to revitalize the cities, extirpate street crime (people don't triple bolt their doors against inside traders or crooked lobbyists). Not reduce, not contain, not deter, extirpate it. Eliminate from public discourse any notion that crime is ever justified.
- Decorum. Covenants don't merely regulate gross misbehavior; they manage fine details. Most of the people governed by them don't see it as intrusive to have to mow their lawns at specified intervals because they do that anyway. So people who don't share the covenanters' values may see such communities as repressive, but the covenanters themselves don't because they prefer to live that way. They want to live among people who share their standards of behavior to a high degree.
- Personal Space. Americans like to spread out and always have. But what's wrong with living in an apartment complex and having lots of park space nearby? Why does it have to be personal space? Because personal space can be controlled. Your kid can pitch a tent in the back yard or build a tree fort (not in a lot of covenant communities, though). You can sit in your back yard and not worry about twenty people with loud radios and foul mouths parking right next to you. You don't have to worry about having your favorite picnic spot taken by someone else.
Incidentally, gated communities are murder on traffic patterns because they lack through streets and therefore channel large volumes of traffic into restricted arteries.
The End of Cheap OilWhat will happen when oil hits its peak (as it is close to doing?). Will that affect the decision to drive? Possibly. But consider:
- Transit companies don't get fuel for free - they will have to raise fares to cover the extra cost. They may also cut routes and frequency to cut costs, adding to all the negatives that keep people off mass transit in the first place.
- Generally speaking, when costs go up, mass transit systems cut schedules, raise fares, and generally do everything imaginable to discourage mass transit use.
- Between the higher cost of living and higher taxes, people strapped for income will probably resist attempts to subsidize mass transit.
- There will be pressure to increase social spending to help poor people cover home heating and cooling. Taxes will go up.
- People forced to work second jobs to cover the increased cost of living will face a killer time cost. Their free time will be so diminished they will not want to spend it riding a bus. And they may well be forced to drive to get from job to job on time.
Europe Leads The WayEuropeans use mass transit far more than Americans because of the high population density, and dense and long-established transit systems. So how transit-friendly is Europe?
A Eurail Select Pass for five countries and ten days of rail travel is $748. That's $1500 for two people. I found a Volkswagen Passat (midsize) for ten days for $672. Toss in another $400 for gas and it's $1072. You do the math.
And America FollowsFrom this morning's paper, an article on traveling across America by train. Cost of a sleeper car from Portland, Oregon to New York City: $1792.90.
Drive: 3000 miles at 20 miles per gallon = 150 gallons of fuel, say $500. Three nights in good lodgings, another $500. Total: $1000. Your car will depreciate whether you drive or go by train. Of course, on a train you don't have driving fatigue and can read, watch the scenery, or chat. But then again, by driving you get to see all the scenery by day if you choose. Back in 1989 I took my family from Wisconsin to San Francisco and back by train. Even without sleepers it was a remarkably nice experience. But we found a rock bottom last-minute fare.
Fly? Boo, hiss, huge carbon footprint. Also $300 if you book in advance and catch a good fare. Plus the value of three days' time not spent traveling. Of course, if you want to see the country from the ground, that's not a factor.
Amtrak wonders why more Americans don't take the train.