Indeed, there's nothing wrong with induced demand on its own. In any other market, more demand induced by lower costs (whether those costs be monetary or in the form of commute times) would almost certainly be a good thing. The only reason it's a potential issue for roads is that road use is an externality.
Building and maintaining efficient roadways comes at a significant cost, but our current system of road construction funded primarily by income taxes means road users don't pay that cost in a manner proportional to their use of those roads. Road construction is "free" from their perspective, so there's no incentive to use alternative means of transportation even if those alternatives would be superior overall once road construction and maintenance costs were factored in.
Because of this it's hard to be sure whether the demand induced by increased supply of roadways is worth the cost in any particular instance. It could be a worthwhile increase in utility, or it could just be a waste of money.
The most "fair" way to do it would be to charge according to road wear and tear, which would look something like axle load ^2 * miles driven, but this would be hard to implement and also widely unpopular with huge swathes of the population (truck driver is the most common profession in many states)
The "purpose" of expanding lanes and building alterntae highways is to improve efficiency. For people living in a given area, reduce their time spent commuting. There is large economic cost to having large portions of your population spent 10+ of their waking time commuting to and from work each weekday.
The problem with induced demand is that yes you increase capacity and more people then move clogging up the roads until a similar equilibrum is reached as before. A much better option is both expanding the number of people that can commute by living further out, and reducing the per individual commute time by mass transit. That's the true goal.
That's still an improvement over the previous status quo. Commute times are not the only, or even necessarily the most important factor in a transportation system. Your new equilibrium transports a larger number of people than the previous equilibrium. In isolation, that's purely a good thing.
In a world where road users paid for those improvements directly in proportion to their use of the road we could just keep expanding capacity indefinitely until either all the latent demand were met, or until rising construction costs drove demand down to a level where the roads were no longer congested. Since roads are funded by income taxes though, a crucial half of that feedback loop is missing. We can't just keep expanding because there's nothing to stop road users from demanding more and more capacity even after adding that capacity becomes cost-prohibitive.
So adding roads gets half the benefits (as it doesn't gain any efficiency benefits), and comes with a ton of externalities when compared to mass transit.
Only one I can think of is pollution, but that's not even caused by roads, it's caused by burning gasoline.
Here in city (Seattle) most people drive because transit tends to be spotty and slow for most people.
So, what's the solution? Make driving worse of course. Speedbumps everywhere. Change 4-lane roads to 2-lanes. Remove parking. Lower speed limits to absurd levels. Make through streets dead ends.
Transit still mostly sucks, but now driving sucks too. Success!
I love transit and want more of it, but the transit folks realized it's hard to compete with driving, so they've just given up entirely on making transit great. It's easier to ruin driving.
Wow. You just crystalized exactly what I felt was wrong with the argument that induced demand is bad. Thanks.
The goal of transit first infrastructure is to make the majority of trips unnecessary. You shouldn't be required to own a car to participate in American society.
This means we need to rezone our residential sprawl to allow for more frequent, smaller grocery stores. We need to increase the amount of mixed zoning, increase density, decrease the insane quantity of land dedicated solely to the movement and storage of privately owned heavy machinery (automobiles) and focus on easily accessible areas of bike & bus friendly infrastructure.
The Netherlands was fully capable of transitioning from a nation of car dependent choked cities to a bike first micromobility haven in 30 years. The only thing stopping the USA from doing the same is the enormous government subsidies paid to car owners to keep the roads paved.
If the federal government stopped taking 90% of the cost of every road in the US and you had to pay gas tax to support it all do you think you'd still be driving? Do you think you'd support billion dollar bridge extensions and lane additions when it means gas is an extra $5/gallon?
The Federal Highway Trust Fund was fully funded by the gas tax and other user fees until 2008, all while a significant percentage of revenue was allocated not to roads but to mass transit. Congress has topped it up with general revenue since, but the gas tax hike required to eliminate that need would be measured in cents, not dollars.
The Netherlands is smaller than New Jersey and has twice its population - its one of the most densely populated countries. Comparing it to the 3rd largest country on earth is risible.
NYC doesn't even have good bike infrastructure.
Every major city in America could transition to bike infrastructure and the quality of life would improve across the US. We don't have to cross the great plains on a bike: We're talking micromobility here. Who cares if they're smaller? We have enormous cities choked to death with cars.
Death to cars: cars bring death. Cities are for humans, not cars.
I don't drive and I live in Ballard. Driving has always sucked in Seattle since I can remember from the late 1970s. My dad, who lived in Seattle after coming back from Vietnam said the same thing.
It's way more like the driving folks have absolutely ruined transit in almost every single city in the country.
This is obviously not the solution that anyone is proposing. You're arguing in bad faith against a strawman. The solution to bad public transit is to make public transit better.
The claim was that they put up speed bumps (and other measures) _with the intent of_ making driving worse to encourage public transit. That's obviously false. Speed bumps get put up to discourage unsafe driving.
If, when forced to drive safely, people would rather take public transit, that's kind of scary, but also a good thing I guess to get unsafe drivers off the road? However, that's not what the claim was (and in reality is unlikely to be true, though I have no data to back that up).
That does have the result of increasing trip times by cars, and thus motivating use of public transit.
Guessing intent is a fool’s game, and can’t really be true or false per se. It does antagonize automobile drivers and, from their valid but particular perspective, makes their life worse in (what would seem to them) a gratuitous fashion. Americans don’t like arbitrary and capricious as a whole.
Caution against the short-sighted pursuit of easily-quantifiable goals at the expense of actual value is not 'arguing in bad faith'.
I live in Denver, CO. Cars basically make it impossible to walk around most of the city, even in the more residential areas. Walking is an essential part of the public transit/non-car transportation experience because essentially everyone has to walk a few blocks from a bus stop, train station, bike rack, etc. to complete their trip on both ends. If walking those few blocks is unpleasant, unsafe, or impossible, people will (reasonably) prefer cars.
Unfortunately, car and pedestrian traffic are at odds in most cities. Situations that seem better for cars (turning lanes, right-on-red, faster speed limits, street parking) often make life hell for pedestrians who try to cross the road. Or make life very, very noisy for pedestrians who need to walk or live or work near those roads.
I agree wholeheartedly that we can't just make driving suck to encourage more people to walk or take public transit. But there are aspects of driving that need to be sacrificed to make public transit better. A great example: changing 4-lane roads to 2-lane roads -- if you can introduce a bike lane, bus lane, or both, those methods of transportation become significantly faster, safer, and better. Biking is basically a non-starter without lanes; busses can be so slow as to be not worth using when they get stuck in normal traffic. The same argument applies to parking removal -- instead of using an entire effective lane of traffic for parked cars, we can dedicate it to bikes or buses.
Lowering the speed limit reduces noise at street level, makes streets safer to cross for pedestrians, and allows bikes to peacefully coexist with cars in an environment where you don't need to go that fast anyway.
It would be interesting to hear what holds you back from using buses, walking, or bikes instead of your car to get around town. In Denver, the main issues I encounter are:
- bike theft
- literal crazy people shouting at me on buses/trains
- drivers who park/stop in crosswalks, or try to kill me on my bicycle
- the bus network is extremely slow to get around town
I think there's a fair argument that we should focus on solving these problems first, before we degrade car traffic. Bike theft is a really bike one in Seattle, too, iirc, and a huge blocker for folks trying to switch away from cars. But eventually you need to degrade car traffic to make public transit as good as it can be.
The DC metro deteriorated markedly, and has continued to. A lot of it is a combination of bad initial designs (lack of surplus tunnel capacity to ease maintenance) along with the aggressive, powerful, and corrupt WMATA employees union. (I was on a project to analyze WMATA's staffing issues, and within the first hour, my team identified that there was a huge incentive to understaff the maintenance/technician teams to allow existing employees to collect massive amounts of overtime. Many would simply hide and sleep during the time they claimed to be "working". Hiring more mechanics/techs was foot-dragged, because it reduced the overtime pay for the existing workers who would interview them.). 2 mechanics working normal hours cost the same as 1 mechanic pulling tons of overtime, but the gap in productivity is huge. The WMATA union doesn't care. The rudeness of the staff is pretty legendary amongst locals as well.
Anyway, all of that is a long winded and detailed way of saying that WMATA gradually became a significantly less reliable means of transportation. My brother was on a car that got stuck in a tunnel that started filling with smoke. He stopped riding. And the buses need the metro to be running well. Without that, the buses become far less reliable. It's a shit show. And it's deteriorated markedly since I last lived there.
Do you use public transit in the Denver area at all? I find myself biking to most places because the public transit routes don't really get me where I want to go, but a lot of folks I know in the area used to use the buses and light rail in the before times. Seems like it had a pretty good rep before covid.
Regarding public transit in Denver, I avoid it like the plague. If I'm by myself, I'm a lot more tolerant of it. But I can't take my kids to public places in downtown Denver anymore, including the transit. When my daughter was 4, I had her on my shoulders on Mother's Day while we walked the 16th Street Mall. As we approached the Capitol, a violent altercation occurred within 30 feet of between two chronic drug addicts. One of them had a hiking pole, and he started beating and stabbing the other one. My daughter was terrified. That's just one incident, there are far more like it.
It blows my mind how the current crop of homeless (unhoused, or whatever moronically Orweillian term has been created to signal pious, virtuous sensitivity to ingroup members) activists have pushed the utterly failed policies of San Francisco in other cities. They result is what you and I are complaining about: public spaces that are decidedly unwelcoming and unsafe to children, elderly, and women. The policies seem to do nothing but funnel money to the non-profits that employ the nutbag activists. They certainly don't accomplish anything else. It's the equivalent of the neighborhood cat lady who puts bowls of food out for strays claiming she's a wildlife rehabilitation specialist.
As a concrete example, I grew up near Seattle and regularly drove on East Lake Sammamish Parkway. This road was built and designed to efficiently carry traffic between Redmond and Issaquah at a speed of 45 miles per hour. It has smooth gentle curves, good sightlines, few driveways and intersections, etc. Sometime in the 90s or 00s people built a ton of really expensive lakefront houses between the parkway and the lake, and the new homeowners got the city to lower the speed limit to 35 (presumably to make it easier to get onto the road)
People generally drive 45 on it anyways. It is a road that practically screams "45 mph is safe" at you, and 35 feels downright glacial. If you lowered the limit to 25 people would probably still regularly do 40 on it - you need some kind of traffic calming as park of a major overhaul of the road to get speeds that are safe for pedestrians there. (And even if you could do this, most households in Sammamish travel to or through either Redmond or Issaquah anyhow, so they need some thoroughfare to do so - at best you're overloading and overstressing the other roads in the network)
We petitioned to put speed bumps in. Yes, we were hostile to the way at least some people were driving. But also note that we, the people who asked for the bumps, also drove there every day. We weren't hostile to drivers as a class. We were hostile to people trying to drive excessive speeds on suburban side streets.
The real solution is to make transit at least as good as driving (measured roughly by time to get from A-B). Not easy to do in some cities - Seattle has some unique geography to work around. But for someplace like Houston or Dallas? Making transit work shouldn't be that hard (other than the cost to build it out and getting people to agree it can work).
And now bicycling and walking suck just a little less. That is a success.
Seattle just opened light rail from Northgate to the U-District to Downtown this year. We will also have light rail from Downtown Seattle to Bellevue opening next year, and light rail to Redmond the year after that.
And the opening of light rail from Downtown to Capitol Hill to Husky Stadium a few years ago drove some pretty big changes in transit in Seattle.
For cars to be good you need lots of space for parking lots and highways. Otherwise, the car is getting you nowhere fast, and there won't be anywhere to park it when you get there. But that also means those parking lots and highways need space. In all but the densest urban metros, that space is two-dimensional, which means all that car infrastructure is spreading out all the buildings.
Transit needs the exact opposite to happen: buildings need to be close-together so that a single line can aggregate more demand, and riders have to walk less when they arrive at their destinations. This is actually how pretty much all cities used to be built, because cars didn't exist yet, so you had to give that space to pedestrian infrastructure. Not coincidentally, those are also the cities with the best transit, and the absolute worst to drive in. You can't have both cars and people sharing the same space.
I disagree. Regardless of your preferred mode of transit, look at the hours just before and after peak. Roads flow smoothly. Trains run at tight intervals and aren't too crowded. It's great for everyone.
Peak demand time will always be a clusterfuck but with enough infrastructure (ignoring petty ideological bickering about which mode should have what market share) we can probably have a system that's pretty damn decent the other 22hr of the day.
The speeds were reduced because people keep dying when getting hit by speeding cars. And this problem has gotten worse with Americans shifting to SUVs and crossovers that hit humans higher up and toss them under the wheels.
If by building a larger road through your city, you induce people to live outside of your city instead of in it, then you’ve added costs while reducing your economic activity, while creating more total wasted hours in traffic in the process.
Are the trade offs worth it? Sometimes! But induced traffic demand is not by itself a success criteria for regions: it’s only a success if it means more people are able to work and have higher productivity in a region, as opposed to just spreading out the existing workers and reducing their productivity through increased commute times.
Take DC and NoVA, typical large suburb next to a large city. If DC wants to increase economic activity, does it want to invest in a new bridge that allows more people to live in NoVA (where most of their retail/commercial activity will occur)? Or, would DC be better off spending that money on redeveloping run-down neighborhoods and adding some light rail (or other transit improvements)?
But I have no idea how much DC “needs” the vertical space for development. It isn’t nearly as dense as NYC - plenty of infill (re)development to be done, I would guess.
However, what expanding highways actually accomplishes is incentivizing people to move out of the city because they can have a commute which is longer in miles but shorter in minutes. That lasts until enough people move to the suburb that the commute is now longer in miles AND equal time or longer in minutes. You haven't made traffic any better, you've just made sprawl worse and you've increased the vehicle miles traveled to accomplish the same result of getting people where they want to be.
The message of the New Urbanists is that we should make our cities more livable and build/allow more housing units within the cities so that more people can live where their commutes are short (in terms of miles) and the increase in density will make providing transit more cost-effective.
More transportation means more trade, more services, and better life for all who live near the roads. It might be in the form of easier-to-get deliveries (Amazon goods), or new jobs that have popped up close by, or new housing developments (aka: homes that previously weren't possible due to the time of transportation, but are now possible thanks to sped up transportation times).
-----
It turns out that "individualism" is a crappy reason to do anything. The individual argument must be made because we live in a democracy, and its impossible to get the people to agree to something unless you sell them a story regarding individualism.
The purpose of adding lanes was to REDUCE DELAYS. Not support additional transportation. Your entire post hinges on an incorrect premise
Note, it can be better to add transit other than lanes of road. Even though I said add lanes, adding lanes is but one possible solution. Good transit may well be better. Figure out how to make your city serve the people who want to get around.
Well, there's an annoying edge case that must be considered as well. In some cases, "induced demand" is "stealing demand from somewhere else".
Lets say you have Town Foo and Town Bar. If you build a highway to Foo, all the additional traffic might be "stealing" traffic from Town Bar and benefiting Town Foo. Especially if people emigrate out of Town Bar for closer housing to Town Foo, you didn't really improve the lives of anyone. You just caused everyone to migrate over.
---------
Ideally, you want to build highways / roads / transportation in ways that benefits people, and causes the least inconvenience to other towns.
Again, I'm not making a value judgement here. Transit is a valid option that could be better
Don't forget that no matter how many lanes you add to a highway heading into a city, eventually that highway ends up... in the city. Too many cars in a city makes for a loud and dangerous-to-navigate environment for those who live there.
I drive 2-3 times a year from NY to South Carolina or Florida for years, always timing crossing through DC around 5AM. Traffic 15-20 years ago coming into DC extended down to Potomac Mills. When I passed though in 2019 it extended almost 60 miles, well past Fredericksburg!
More demand drives more sprawl that drives more demand for roads. Eventually metastasizes into a nightmare like LA or Long Island!
The reason "just shutting down all roads" doesn't make sense is that it doesn't solve the actual problem, which is that people want to both work in places with good jobs (traditionally often dense urban centers) while living in cheaper places that are far away and not spend significant chunks of their lives stuck in traffic. Shutting down all the roads only "solves" the traffic problem in a deliberately ridiculous sense (same as "just kill all humans").
I work for a big central business district employer. You can pay $150-250 a month to park or $75/week to take a motor coach bus from your suburban town. Those numbers drive behavior, and make for a better solution as folks who need flexibility can pay for it.
As long as your roads are saturated, they have less throughput, not more. It is kind of like a clog in your toilet: more things are there in your pipes, but not much is getting through.
Fluid analogies are crappy because fluids flow more when you add pressure and traffic doesn't.
Well, traffic can flow when pressure is added, just not in a way that is very safe.
Induced demand is a terrible name for the real issue here. Perhaps the "one more lane fallacy" would be better. The "one more lane fallacy" is not universal, it only applies in cities and other densely populated areas.
Imagine you are sitting in a suburban traffic jam, getting slowly more and more annoyed. It's easy to imagine adding marginally more road capacity would put an end to traffic jam. That thought is the 'one more lane fallacy'. That, admittedly attractive, idea is wrong because adding more capacity leaves the traffic jam exactly the same, with more lanes and more cars in them - because of induced demand.
The kicker is that in a densely populated area, there is often no realistic prospect of adding enough road capacity to end traffic jams, simply because of there is not enough space. I suspect that in most cases, road building achieves political support through the expectation of ending traffic jams. Nearly always, this expectation will not be met. This is the essence of the induced demand problem - it is not as obvious as is sometimes made out.
So, even if you are die hard driver, if you want to reduce traffic jams, you should advocate for other people to use transport methods that are more space efficient than driving. Of course, you might want to live in an area dominated by constantly congested roads, in which case, induced demand isn't a problem.
It does result in growth for an area, but quality of life stagnates. The traffic in northern Virginia/DC/Maryland is at a point where it noticeably affects the mood of a bulk of the people who live there. Spending 90 minutes each way day after day after day fucks people's heads up.
“Induced demand” in every other industry is described as “latent demand”.
This is such an oddball perspective: EDGE, 3G, LTE mobile networks all became increasingly congested as demand rose to meet supply. But nobody thought building out 5G was therefore pointless.
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8057171/
The situation with roads is not that different: building more road capacity incentivizes sprawl, leaving the entire network more congested than it was when you started.
Roads just don't scale. The daily commute is like a distributed shuffle: time to complete the shuffle scales superlinearly, probably quadratically given the limited topologies possible with roads. It's not like you can build a hypercube road network.
That depends on what you see as the goal. If the goal is reducing congestion, then induced demand means that particular goal is harder to achieve. If the goal is to get more people driving then induced demand is a clear success.
When you order goods from Amazon, that gets delivered to you. It might be a road, sea, rail, or plane, but its transportation. The more of packages ordered / delivered, the more things are happening in the city.
The more jobs being created, the more people will need to transport to-and-from work. The more homes built, the more transportation is needed. Etc. etc.
Its a crude measurement with flaws, but generally speaking, the more transportation that's happening, the bigger and better the city is functioning. People wouldn't travel unless they needed to (travel always sucks: traffic accidents, getting stuck, dealing with others on planes/trains/busses, etc. etc.). But we deal with it because without transit, we couldn't do our daily business.
Be it a meeting for work, going to school, delivering goods or other such need.
-------
Mass transit options, like Rail, get more things done with far less money. But there's a latency issue: rail can be slower for the individual... but its cheaper and more-bandwidth for the city.
This conflicts with individual options like roads: it costs a gross amount of money for an individual to buy a car / use it on the highways (plus the cost of highways themselves: rubber tires wear out faster than steel wheels on trains. Asphalt roads need replacing more often than steel rail lines. Gasoline costs much more than the electricity used to move a train). But the individual latency is such an advantage, that the individual will typically prefer car travel.
What they do have an advantage on is not having to synchronize with a schedule. If you need to travel 15 minutes by light rail, but the train only comes around once an hour, then missing the train adds an hour to your trip. It doesn't matter if the train never breaks down and has separated right-of-way - running on time is useless if you're not.
The way you work around that is by running more trains so that they can come more frequently. My rule of thumb is that if a transit line has a frequency of 15 minutes or less, I don't need to worry about the schedule because the time spent waiting for the train or bus is less than the time of the overall trip.
However, this is expensive; it only makes economic sense if you actually have that many riders that need to travel along that line. In other words, demand needs to be aggregated. The problem is that cars work the opposite way: they exist specifically to segregate demand. This occurs both directly and indirectly. The direct effect is the ability to immediately depart, which I already mentioned; but the indirect effect is the result of all the infrastructure that cars need in order to work at all. Things like wide highways and parking lots spread out where people need to go and make it far more dangerous to walk from a train or bus stop to your final destination.
In other words, cars cheat - they don't make your commute better, they make anyone who does not own a car have a worse one.
Parent comment is saying the only way to scale with higher demand of transportation in a way that feels like an actual improvement to people is public transit, because public transit scales so much better with higher numbers of people commuting.
I as a motorist could not give less of a shit if I'm stuck in traffic for 3 hours a day but the road is able to move hundreds of thousands of cars a day. I'd prefer a road that could only move 20 people a day with 0 traffic. It's the only thing I care about.
They tried this in Beijing. People would just buy second cars so they could drive on both days. Eventually they had to restrict new license plates as well.
Is there still some traffic? Yes. Is it better than it was 30 years ago, even as the roads handle way more traffic? Absolutely.
If you build a new road and it gets just as congested and it takes just as long to travel from A to B, then what did you accomplish?
The number of people who want to use a road is not the metric we're trying to optimize here. If we wanted to do that we should build as few roads as possible. Then there would be tons and tons of people who want to use each road. Hooray! Success!
More lanes generally improves throughput. However, more lanes often increases latency.
City planners may prefer this, but individuals may not.
In a metro area housing prices are generally correlated with how many minutes it takes to get to a city center. If you add highway capacity then people will choose to move further out to the suburbs. (Though that is great for property values, especially around the periphery of the commuting range.)