Quick Note: Are Freeways Safer?

Freeways are, in principle, much safer than roads with at-grade crossings. With postwar design standards, they eliminate the frictions that are responsible to a vast majority of accidents: grade crossings, left turns, opposite traffic (since they have medians by design), and so on. They also maintain higher design speeds and capacity than less safe local streets. But a more interesting question for policy purposes than “are freeways safer?” is “does the construction of freeways increase road safety?”

For some evidence that the answer is no, see PDF-page 3 of a John Adams paper from 1987 arguing for the continued primacy of Smeed’s Law. Traffic deaths per unit of vehicle distance driven had declined in both the US and UK at a rate following a multi-decade log-linear trend: 3.3% per year in the US, 4.7% in the UK. Regardless of whether Adams’ theory is correct, we can compare actual death rates to the trendline to see what happened. In the US, where the data goes farther back, the greatest period of freeway construction started in the mid-1950s and ended in about 1970; this was also a period in which traffic deaths increased, even more than the trendline based on the explosive growth in driving predicts. Of course the Interstate system also led to traffic growth on at-grade arterials, but the greatest construction growth was in freeways, and on top of this suburban sprawl meant more people would be driving on both the new freeways and the older parkways.

The Smeed’s Law explanation of this is as follows: drivers compensate for the greater safety of freeways by driving more carelessly, on both the freeways and the connecting local roads. The freeways are still safer, but the presence of any safety-improving technology will translate entirely to higher speed and capacity (i.e. drivers keep less distance than they would otherwise), and more careless driving.

There may be other explanations out there – for example, the construction of more roads will cause more dangerous vehicles to start circulating that would not otherwise. These include heavy trucks, and also cars piloted by poor drivers who would not have driven if the construction of an expansive highway had redirected development in such a way that more driving would be needed.

But in either case, what this means is that even though a freeway upgrade of a notoriously unsafe road will make it safer, it will not make the overall road network safer. To argue by analogy with congestion pricing, it is possible that the only way to bend the curve and accelerate the downward trend of vehicle deaths, beyond reducing driving, is to make it more expensive to drive unsafely. For example, insurance requirements could be raised from $25,000 to the rough insurance value of human life in the US, which is in the millions. (The same should be true of any transportation system, but buses and trains are much safer for their passengers than cars.)


  1. readerareadevelopment

    Really great post! Certainly freeways at the most basic level seem safer, but taken as a whole of what transpires is it has led to a lot of careless and mindless driving behavior.

    I agree with congestion pricing, increased insurance premiums, as well as a tax for vehicle weight. Much less damage can be done in a smaller vehicle – in many respects.

  2. Andre Lot

    This reasoning has been applied to virtually any infrastructure whose use exploded in 20th Century. And I dismiss them all as considering “bad” increased use of transportation, energy, water or what else.

    Let’s consider an extreme case: space travel. Before the 1950s, the rate of incidents and fatalities on manned space flights was zero. Currently, space travel is still a very dangerous activity that is very dangerous for astronauts (fatality rates above 3% per flight), but overall fatality rates of space travel for the general population is extremely low. If and when space travel becomes somehow more common, we’d see fatality rates from “space travel system” skyrocket among the general population.

    The general count of miles-passenger traveled in any industrialized country and most developing countries just boomed in 20th Century, regardless of mode used. From the sane and progressive practice of now putting filthy chimneys near houses anymore (living close to a factory in polluted and squalid conditions had been the tone of the Industrial Revolution) to the overall dramatic increase on national and global travel.

    So unless one is to assume that travelling is something inherently wrong to be curbed and reduced (as some environwackos propose with things like “stop thinking holidays abroad is something you should ever aspire for”), the increase in miles-passenger traveled, on itself alone, would make any transportation fatality/injury rate per unit of population higher.

    A bet if such data were available for the era of railroad spreading around US similar conclusions would be found: in 1835-40 rail fatalities per 100.000 inhabitants would be negligible. 50 years later, it would have skyrocket simply because there was much more train travel going around.

    So I dismiss this criticism of highways making the networks less safer as founded on the assumption travelling is a social evil instead of a social and economic enabler.

    • Shane Phillips

      I don’t think you’re really addressing what he’s talking about. He’s not saying that there are more car collisions, injuries, and fatalities solely because highways have increased the amount people drive, but also because of the driving behavior it encourages. There’s certainly the “more VMT = more collisions” aspect, and I think it’s worth questioning whether building homes and job centers further and further away is a smart thing to do since it requires more driving, but you can’t just throw out the whole argument because you disagree with one part.

      Specifically, it sounds like you’re saying that traffic deaths per mile driven should decline overall as building increases, but we see in the 60s that traffic deaths per mile driven actually increased. Of course when you build more and people drive more you may expect increased deaths in absolute numbers (although we should not be content with that fact, because it is not inevitable), but not the rate. Perhaps you’re confused because you think this referring to fatalities per unit of population, but this is per mile driven. So as we built more freeways, the chances of an individual dying in traffic increased even if they drove exactly the same number of miles as before.

      • Andre Lot

        Up the the early 1970s in US, and one decade later in Western Europe, we were still witnessing a process in which cars got considerably faster and young/senior driving got considerably more common – without the evolution on safety that came later from simple things like seat belts to airbags.

        I think that given freeways themselves are very safe (compared to other roads with more interference from vehicular traffic and/or external elements like pedestrians) by design, such increase in deaths per mile-driven could not be attributed to having freeways themselves. On the contrary, the alternate scenarios could be:

        (1) much more higher death rates is the traffic increase were channeled mainly to high-performance at-grade roads (think of the standard suburban strip 2×2 road with traffic lights and properties with direct access) than having traffic flow mostly on highways

        (2) design failure for streets and boulevards not meant to be fast thoroughfares

        To put the issue in yet another light, were the reasoning you presented valid, we could confirm by looking at the evolution of fatality per miles-driven rate in countries that came late to the highway building frenzy like Spain or Australia.

        My informed guess, but still a guess, is that while freeways provided a much safer outlet for road vehicles, the adaptations done in many town and city streets were dangerous, and by facing an enormous increase in traffic they made overall traffic more dangerous.

        • Alon Levy

          You’re still punting on the fact that freeway driving as a percentage of total driving went up between 1954 and 1970. Those high-performance at-grade roads carried traffic even before the Interstates (many of them were replaced by Interstates), and on top of that many cars were funneled into not-so-high-performance urban streets.

          Seat belts and airbags did not really increase safety, either, and you can trace this in two ways. First, you can look at the evolution of traffic fatalities after seat belts became mandatory: pedestrian deaths spiked, car occupant deaths stayed the same (in fact John Adams has several posts on that subject). On top of that, the cars that currently run in poor countries are modern – maybe not the latest-model, but certainly from the last 10-20 years; despite that, per-vkm deaths there are much higher than in the first world. The cross-national comparisons make it very hard to argue that there’s any safety benefit from vehicle characteristics, as opposed to road or social characteristics.

          The entry of very young and very old drivers, and more generally people of any age who are not comfortable driving, is what I’m proposing as an alternative explanation to a straight Smeed’s Law story. Orient your society around cars, and people who would otherwise have not driven will get behind the wheel. Make alternatives more convenient, and the people who drive will be disproportionately the kind who like driving and could handle a car on a freeway at 180 km/h without crashing.

          For what it’s worth, the trend in more recent decades has been for fewer American teenagers to drive, but the decline in traffic fatalities is on the same trendline as before. That said, there is an increase in the proportion of older drivers, due to the general increase in the proportion of older people, and the two trends may be significant but just balancing each other out.

        • Nathanael

          “we could confirm by looking at the evolution of fatality per miles-driven rate in countries that came late to the highway building frenzy like Spain or Australia.”

          IIRC, they actually did have noticable spikes in fatalities per miles driven when they put in their expressways.

    • Griffin

      Andre, the accident rate is measured by vehicle miles traveled. You completely miss the point.

    • Nathanael

      Andre, you’re talking nonsense, and it’s because you have failed to read Alon’s article correctly. He’s looking at deaths *per Vehicle Mile Travelled*.

      If your nonsensical arguments were true, then the *rate* of deaths would have increased with expanded airplane travel (it didn’t) or expanded space travel (it didn’t) or expanded rail travel (it didn’t).

      No, “more train travel” did NOT lead to a greater rate of fatalities per vehicle mile of rail travelled.

      But more expressways DID lead to a greater rate of fatalities per traveller and per mile of road travelled. Isn’t that interesting?

      • Steve

        I would bet on it being because the vehicles are individually piloted: most other modes of transportation use highly-qualified pilots, whereas drivers just cursory education and go.

        I wonder what the bike (accident? fatality?) VMT rate is in the Netherlands and Denmark. If the increase is indeed due primarily to the proliferation of pilots viz. other modes of transportation, then the expectation would be that bike accident and fatality rates are also higher than the baseline established by the pilot-passenger system used on airplanes, ships, trains, etc.*
        *Although the damage per accident on a bike would be greatly reduced than by car, due to the reduced max speed and hence proportionally reduced kinetic energy unleased.

  3. Alexis

    I wonder if the increased use of freeways, which causes more careless driving because they are ‘safer’, is environmentally specific, or if it carries over to more careless driving on the road network even though it doesn’t have the same safeguards?

    I don’t know how exactly this could be tested, but it’s an interesting idea.

    • ant6n

      In my town, walking near highways feels less safe, people seem be driving faster the closer they get to the highway. This is anecdotal, but as a study idea, maybe one could look at the rate of accidents relative to proximity to highways, per car-km travelled?

      • Andre Lot

        There are resources known to solve that problem, namely in terms of design, making the ramps look really like a transition zone instead of a continuation of the previous patterns.

        Walking near highways is probably less safe (or perceived as so) because there are less walking infrastructure and facilities as a result of the type of spatial occupation existing around a highway. For the same reason, walking around an airport, a big maritime terminal etc. will also look less safe though planes and ocean-going vessels never (well, very rarely) come onto streets.

    • dejv

      IMO there is “just off the freeway” effect where it takes some while before driver adjust their speed to regular street conditions.

      • Andre Lot

        There are relatively simple design feature that can greatly reduce such effect. Two examples: a roundabout just off-ramp, or a more rugged pavement that creates more noise and vibration with a lower speed which condition the behavior of many drivers.

        Small traffic islands might help accelerate the transition as well.

      • Robert Prinz

        I experience this as well, both with traffic coming off of and entering a freeway. Car drivers seem to make bad decisions quite regularly when approaching a freeway on ramp, perhaps because they are already in that high speed mindset and become much more impatient regarding even minor delays.

        As for traffic exiting an off ramp, I have experienced some of the most ridiculous behavior and am the most wary in these zones when walking or riding my bike. Better transition zones, sight lines, and infrastructure does have a positive effect, but it seems to take at least a couple of city blocks before drivers are truly able to handle 25mph traffic again.

        Maybe we should start installing metering lights as drivers come off the freeway as well, where they are forced to take a deep breath, count to ten, and watch a video of kittens playing before being allowed to proceed onto city streets.

        • Nathanael

          “Maybe we should start installing metering lights as drivers come off the freeway as well, where they are forced to take a deep breath, count to ten, and watch a video of kittens playing before being allowed to proceed onto city streets.”

          Well, something should be done. I can anecdotally verify the psychology; it’s hard to switch mindsets from “fast driving” to “slow driving” coming off a fast expressway.

    • Alon Levy

      It has to be present at least somewhere off-freeway, because Interstate-standard freeways have a lower per-VMT death rate than at-grade arterials and old parkways.

      My guess, going purely by anecdotes of which road segments I hear the most safety complaints about, is that the problem is really with major intercity roads that are not up to standard. CA 99 is a major example – it’s mostly a freeway but not a full freeway, and the few segments with grade crossings are apparently the least safe. Another example is the Tappan Zee Bridge, which is of course a freeway but has narrow lanes, since to increase capacity the state squeezes 7 lanes into a space intended for 6.

      • Nathanael

        The main issue, according to most psych studies, is *perceived road width*; this is what causes speeding. Those wide boulevards have large perceived road width *and* they have grade crossings. They are a very poor idea.

        My view is that no road with intersections should have more than one through travel lane in each direction. (You can add parking lanes and turn lanes if you need them.) If you need more capacity than that, you either need a freeway or you need a rail line. Usually you need a rail line, because freeways take up an *enormous* amount of land and can’t practically be built in urban areas. And in rural areas, how often do you need the capacity?… not often.

        • Alon Levy

          It may be true for rural areas, but in urban ones it’s not really – your average North American dense-city two-way major street has 2-3 travel lanes in each direction, because that’s what you can squeeze into a 100′ street with reasonable sidewalk width. So does your average major avenue elsewhere – and in cities with narrower side streets, like Tokyo, the arterials are sometimes even wider. Those streets do not have speeding problems if there’s enough development to cue drivers that they can’t just try going at suburban arterial speeds on those roads. The Manhattan avenues are fine, and the two-way streets are if anything too slow because of signal optimization that favors north-south traffic.

          • Steve

            Well, to take a page from Charles Marohn, we should call these suburban arterials “stroads” instead of “wide boulevards”…and, in urban conditions, work to emulate more European boulevard systems, which do allow 2-3 travel lanes in similar widths, but in a more enclosed, lower-speed format which feels significantly safer.

  4. guest

    Freeways tend to be long haul routes – so fatigue can also play a role in accidents.
    I view freeways as separating high speed long haul traffic from lower speed local traffic. Having all types of traffic on the same roads is chaotic. That’s probably why high speed railways usually have their own exclusive rights of way.

    One aspect that may not have been factored in is that I would expect freeways to reduce the number of deaths of pedestrians and cyclists (since they typically are not allowed on freeways) versus the number of deaths on arterial roads. The mere diversion of significant amounts of traffic away from the arterial roads may also reduce the pedestrain accident rate. i.e. in Vancouver, driving down any arterial from downtown to Richmond, you’ll drive through over 70 level intersections and a good number of marked crosswalks.

    • Andre Lot

      That is surely the case, the fatality rate for pedestrians in controlled-access freeways is virtually zero for obvious reasons. Pedestrians comprise around 40% of all road traffic-related fatalities in US (more in Europe, it reaches 55% in Italy) and practically all of those deaths happen in urban streets.

      • ITB

        I read pretty often of pedestrians getting killed on the freeways of NYC (at least half a dozen times a year, I’d guess). You could blame it on the “crazy” pedestrian for being in a place where pedestrians are not supposed to be, but I blame it on the pedestrian-hostile road design that makes it impossible to walk from A to B without taking a miles-long detour. If you are going to have a freeway in the middle of a city, I think you should at least provide a reasonable number of pedestrian bridges.

        • Andre Lot

          But then we go down to an argument of poor pedestrian facilities offer. It’s the same case of crazy teenagers entering a subway tunnel or lazy workers taking a shortcuts over a railway. Cities are full of interference with “straight walk lines” all the time, from natural (water bodies, swamps) to infrastructure of all types, large buildings and/or lots closed to the public etc.

          I have no sympathy for a passenger that ventures over a fence into a highway to take a shortcut, like I have no sympathy for someone who gets killed when going on the tracks to retrieve an object felt there. It is truly a case of victim’s blame.

          • Henry

            In the case of New York though, a lot of parkways replaced avenues, and even these parkways had side paths that could be used (sometimes in the beginning).

            Eventually the parkways were widened to absorb the former walkways (and in some cases the surrounding park), so the original recreational purpose of New York City parkways is no longer fulfilled.

  5. Adirondacker12800

    There may be other explanations out there – for example, the construction of more roads will cause more dangerous vehicles to start circulating that would not otherwise.

    The freeway induces demand. Instead of sitting at home because the drive into the city takes an hour and is along the stoplight laden old state highway they can be wherever in a half hour. The street that crosses the old state highway is now the primary access road to the new freeway. Since nobody takes the bus or the train on Sunday anymore the bus and the train stop running on Sunday. Which forces the few people who didn’t own cars and used the Sunday bus or train into buying a car. It’s not all one thing going on.
    And statistics can be misleading. The pedestrian death rate per 100,000 is the same in New York, New Jersey and Texas. I suspect it’s more dangerous to be a pedestrian in Texas.


    The chart makes Washington DC look downright dangerous. But I suspect even Washington DC is safer for pedestrians than Texas is – if you looked at from pedestrian miles viewpoint.

    • Alon Levy

      Charlie Komanoff has a table somewhere comparing total accident fatality rate, both pedestrian and in-car, in New York and its suburbs. The city turns out to be far safer per capita. From memory, there are 300 fatalities a year in New York, which is about one third the national average per capita, and marginally higher than the national average per registered vehicle (which is not the same as per VMT, or per car that actually drives in the city once you include suburban commuters).

    • Andre Lot

      I think this is not relevant, since having more transportation use (with appropriate infrastructure) is positive, not negative, regardless of modes used.

      Is this “induced demand, increased causalities” paradox that leaves people slipping into dangerous territory like arguing airport security measures meant to prevent hijackings are negative because more people will die due to modal shift to cars than the small numbers of people killed on planes on 9/11.

      I think whatever infrastructure that makes people travel more, especially when it doesn’t come to commute trips, is positive, because mobility (like telecommunication or use of electric appliances at home) is something inherently positive whose increased use foster a more dynamic society.

      So if new highways make people travel more and increase the radius in which they procure for leisure, churches, employment, schools, parks and shopping, they are surely fulfilling their function of increasing growth and reducing localism and geographic barriers.

      • Shane Phillips

        I disagree that having more transportation use is positive in all circumstances, but I’ll just accept it to move the conversation forward. Surely you agree that some types of transportation use (i.e., modes) are more positive than others, whether because they produce less pollution, require less infrastructure or real estate, are safer, free up money to be spent on other things, etc.

        But where I will actually bother to disagree with you is in your assertion that any infrastructure that makes people travel more is inherently positive. Taken to its logical conclusion, that would mean it would be a good thing for us to pave all dirt roads with concrete, widen all roads with the least bit of traffic delay, and build new roads anywhere someone might plausibly want to travel. Clearly this does not increase growth, or perhaps rather the “growth” that this entails amounts to having a large stock of mostly useless roads that still have nominal “value”. I’m not trying to say that freeways should be abolished, but I am saying that there is a limited amount of money in the world and we must make value judgments about how that capital is allocated. Just like if we were to build new highways, if we were to invest more in bus, rail, bicycle, or pedestrian infrastructure we would encourage people to travel more using these modes, and these modes have the added benefit of actually being good in terms of pollution, health, safety, and various other concerns.

        And just as a side note, what do you consider negative about localism? I’m as happy as the next person to take a vacation, but in what way is it actually better for my city, my state, or my country if I choose to fly to Brazil for vacation rather than travel to San Francisco, or find something fun to do in my area? This just doesn’t make sense to me unless you believe that we should treat all entities as equally deserving of our money, which is certainly nice, but quite utopian given the realities of places like Iran or North Korea and what it means to spend our money on products from those places rather than others (just to use an extreme example – there are many less severe ones I could come up with as well).

        • Nathanael

          It’s fairly obvious that more travel is bad — for an extreme example, we could all drive around in circles like the Indy 500 all the time, and that would be unhelpful.

          More *ability* to travel should it become *desirable* is good, but that’s a very different thing: we would ideally reduce the *need* to travel as well as making it more viable to travel when we need to.

      • Henry

        Mobility in general is positive, yes, but there are also costs to mobility enhancements.
        No one today would argue that Robert Moses’s plans for the LOMEX and the Mid Manhattan Expressway in New York would be positive, because then we wouldn’t have the vibrant commercial districts in SoHo and on 34th Street that we have today.

        Likewise, building six lane interstates through farmland for the sake of building six lane interstates is probably not worth the trouble. In addition, the low-density suburbs these highways generally create usually lack mobility for those without a car.

        Even replacements of things can have questionable benefits for their cost – the Kosciuszko Bridge replacement in Brooklyn will improve mobility, yes, but for some reason the bridge replacement will be eight lanes wide, when the highway on either side of the bridge is only six lanes wide. There is no political will or desire to destroy more of Brooklyn by widening the entire highway to eight lanes, so the extra lanes are unnecessary. (The replacement will have significantly reduced grades than the current bridge though, so the replacement of the bridge will have some merit.)

        As another example, the Tappan Zee Bridge replacement is essentially spending $5B+ to add another highway lane to a bridge. While yes, mobility is good, in a situation with finite resources we have to make sure we’re getting our money’s worth.

        • Nathanael

          The extra lanes thing is a highway engineer thing. They’re nuts about lanes and they want more lanes.

          Extra lanes are good for precisely nothing (they get used up by weaving movements very quickly), which is why this is so pernicious.

          • Andre Lot

            You obviously ignore basic physics and intermediate traffic engineering concepts (reinforced by your other absurd proposition that no road should have more than one lane per direction or be a freeway, but since freeways are bad there should be essentially nothing wider than a 1+1 road with shoulder).

          • Henry

            @Andre: I’m still not very sure why a road should have six lanes, then change to eight lanes for a mile, and then change back to six, especially when that mile has no exit or entrance ramps to speak of.

            The extra lane is a waste of space if it doesn’t lead to anything on either side.

          • Andre Lot

            @Henry: I was referring to Nathanael’s suggestion (on an early comment of his on this post) that no road that is not access-controlled should ever have more than one lane per direction.

          • Steve

            Perhaps this video might shed some light on the matter.

            The problem can be said to balance between two opposing issues: the clearly excessive traffic designs of modern highway engineering, too often justified by projections of more “growth” (and by “growth” we euphemistically mean sprawl), seen in e.g. this video, or in any other lived experience where stroads (what Strong Towns calls “wide boulevards”) have decimated safety, and safety for all users at that–and by extreme anti-car advocates who, like Nathanael, will advocate for undersized roads even where the benefits brought by the added traffic lane are manifestly superior to the drawbacks.

            Again, drawing from the Strong Towns body of work, we can justly say that the average American (st)road has been outfitted with one driving lane per direction too many–but that there are places in our traffic system needing improved traffic engineering. The problem then comes from the extreme mismatch between where the funds are most needed (think Philadelphia’s Roosevelt Boulevard) and where they actually go (think College Drive in the example above).

  6. Pingback: Today’s Headlines | Streetsblog Capitol Hill
  7. Pingback: In Miami, Cloaking an Anti-Cycling Proposal in Bike-Friendly Language | Streetsblog.net
  8. Jeffrey Jakucyk

    I recall when reviewing for the North Carolina drivers license test a number of years ago that their study guide mentioned the rural 2-lane roads with 55 mph speed limits were by far the most dangerous roads in the state. That always stuck with me, but it does make sense. In a state like NC those roads were laid out in the 18th or 19th century on very curvy and in many cases hilly alignments. They were never meant for such speeds, and with many small farms and rural hamlets, there’s a lot of driveways and access roads and other intersections, not to mention blind turns and hill crests, and other “fixed hazardous objects” near the side of the road. Nevertheless, compared to the interstates or other modern divided highways or even suburban arterials, they don’t have the traffic volume to show up particularly highly in crash statistics, only on a per-mile or per-user basis.

    If those rural roads are the most dangerous, the high-volume, high-speed suburban arterial must be a very close second. I’d suspect they’re the worst in most other states where the rural road network is a bit more gridded and less populated (think Indiana or Kansas). Those “stroads” have high speeds of 40-50 mph usually, lots of traffic, but none of the safety features of interstates like medians, grade separation, or limited access. They also don’t have the safety features of urban streets like parked cars, buildings near the sidewalk, street trees, or narrow lanes, all of which lead to greatly reduced speed and improved safety for vehicle occupants, pedestrians, and cyclists. It’s the worst of both worlds, with none of the benefits of either.

    • Nathanael

      The rural 2-lane roads also don’t have sidewalks, meaning that the people going at 55 mph in their cars are sharing the road with anyone who tries to walk next door.

      They’re not good things. It would be fairly desirable to restructure the rural road network to have (a) most roads would have lower speed limits, and (b) the ‘through’ roads would have sidepaths and perhaps parking lanes.

      • Beta Magellan

        There was an interesting article a couple of years ago, I believe in Slate of all places (though I can’t find it in ten seconds of Googling), that interviewed people in a part of the Dakotas where they were reverting some rural roads to unpaved as a cost-saving measure. Although such reversions to “dirt roads,” when noticed by the national press, were given as examples of how Americans can’t fund necessary infrastructure, the local response was fairly positive. Although these roads don’t have much pedestrian traffic, they get some and people actually living in the communities felt like a dirt road was friendlier.

        The roads in question were very rural—we’re not talking about state highways here—but it’s a good example how we’ve overbuilt road infrastructure in the West. The reason for reverting to unpaved roads was to keep the local government from having to raise taxes, not to fulfill the expectations of the pernicious UN Agenda 21. That de-paving it bettered the experience of the occasional rural pedestrian was a nice side effect.

        • Andre Lot

          Much ballooning there.

          Some roads have been un-paved (reverted to gravel) because there had been tons of consolidation of farm estates, and what was once a busy-ish (for rural standards) road with live produce being hauled up and down daily to the nearby markets, now is a big planted field dominated by agricultural tractors, without people actually living in the properties and with much diminished traffic.

          Then, some people overhyped these decisions (akin to closing and removing a disused rail spur that once served a now-closed mine operation or factory, or dismantling an airfield on a closed air force base) as some sort of “it’s all 1861 again” paradigm.

          Those were roads that served no wider network purposes (usually part of mile-squared rural grids with tons of parallel alternatives), situation that is extremely different from Interstates (I-80 through Nevada is mostly a long-distance long-haul artery east of Reno, for instance).

  9. Eric F

    So taking all the 18 wheelers off of the Long Island Expressway and putting them on Northern Boulevard [or insert your own local analog here] would decrease fatalities?

    • William Furr

      Before the Long Island Expressway was built, where did the 18-wheelers drive? Or did the construction of the expressway *encourage* more of them to drive into the city in lieu of smaller, safer alternatives easier to handle on local streets?

      • Henry

        The LIE was built out mostly in the 50s and 60s – 18 wheelers were probably not as common as they are today, only because shipping containers only really became widespread in the 50s.

        • Nathanael

          Indeed, if the LIE had not been built, the intermodal containers would probably have been put on the LIRR and unloaded at transshipment sites somewhere deep in Long Island.

          • Andre Lot

            But that requires TIME, even with modern transshipment yards of the latest design (the ones that have automated cranes with space for dozens of trucks fast parking alongside a “rail quay”, it still takes a lot of time to unload a long freight train, and most business operate on just-in-time basis in which the cost of time in transit or in storage is much higher than cost of actually moving a container from point A to B.

          • Adirondacker12800

            Sometimes traffic slows down so much the pedestrians on the sidewalk next to the truck are moving faster. Someone on Long Island who wants reliable just-in-time deliveries wants the Cross Harbor tunnel to be built.

  10. Pingback: UPDATED: In Miami, Cloaking an Anti-Cycling Proposal in Bike-Friendly Language | Streetsblog.net
  11. Eric F

    Another thought: have you examined the data as to what effect freeways have had on disaster evacuations? Have lives been saved in coastal communities due to the presence of relatively quick freeway egress when hurricanes threaten?

    • Adirondacker12800

      That was reason No. 27 for building the Interstates. When the air raid sirens went off we’d all hop in the car and drive away from the city and the bomb…. to be incinerated by the bomb dropped on the suburb 20 miles out but nobody brought that up…

      • Eric F

        I have used a car both for gathering supplies in the face of a pending disaster and was prepared to use one to evacuate, and was glad I had one for this purpose.

        You can make a similar point about how many lives are saved each year by people being able to obtain medicine or get to a hospital by car.

        • Richard Mlynarik

          All these years I’ve been deluded that cars would be the cause of the zombie apocalypse.

          But today I learned that, on the contrary they will be our only salvation.

          The veils have been lifted.

          Thank you, Internet!

        • Shane Phillips

          I think we’re getting off track a bit here. This isn’t an article about whether people should own cars, or whether we should build roads for cars, but rather about relative safety of freeways vs. other types of roads. I think it’s an interesting concept, but in some ways more of a thought experiment than anything, at least at this point. Highways aren’t going anywhere, so the question really is what can we do to mitigate the increased incidence of unsafe driving that accompanies highways, and also perhaps what alternatives we have to new highway-building in the future.

    • Alon Levy

      I haven’t. But that’s not usually how road builders justify things nowadays. They don’t say, “We need to upgrade Route 23590825 to a freeway to increase speed and capacity.” They instead say that the existing road is unsafe. Very rarely do you see people advocate for more roads on the grounds that they would make it easier to drive, and even in the 1940s the Interstate standards were based on the belief that grade-separated roads built to high standards could cure congestion and drastically reduce traffic accidents. Nowadays it’s even worse – you have the construction industry claiming that building more roads will actually reduce air pollution and CO2 emissions by reducing congestion.

    • Nathanael

      Freeways have been pretty much total failures for disaster evacuation; they just don’t have the capacity for it.

      Rail lines actually do have the capacity for disaster evacuation.

  12. Pingback: Car Culture: Fetishizing the Automobile | Critical Transit
  13. Greg Costikyan

    I know it’s pedantic of me, but “freeway” is a California-centric term, meaning a free-of-toll, divided highway. A “parkway” is NY-centric term, meaning a free-of-toll, divided highway from which commercial vehicle are barred. A “turnpike” is a term used throughout the north-east, meaning a tolled, divided highway, open to commercial traffic. A “throughway” is another northeastern word, meaning an untolled, divided highway, open to commercial traffic. In Britain, they have motorways; in Germany, the autobahn.

    Freeway is not a generic term applicable to all divided highways with limited entry and exit, and no at-grade crossings, since the “free” in “freeway” necessarily implies “untolled.”

    • Alon Levy

      I alternate between freeway and expressway. But the TTI uses freeway as a generic term, independently of whether the road is tolled; the advocates of toll roads at Reason and Cato do the same. The original etymology of freeway, according to The Big Roads, is town-free road – i.e. free of at-grade intersections, without regard to tolls.

    • Beta Magellan

      The terminology often varies from city to city—in part because Chicago’s high-capacity, high-speed grade-separated roadways were begun before the Federal Aid Highway Act, I’ve occasionally heard people stress that they’re called “Expressways” here.

    • Henry

      I know other cities have parkway systems (Kansas City?) but I don’t know if they’re similar in concept.

      As a New Yorker, I’ve only heard “thruway” as a reference to the New York State Thruway, which is just one very long highway across the state. “Expressway” and “highway” are more frequent terms here (which presumably means something else in California?). In my experience, I have never heard a native New Yorker use the word “freeway”.

      • Adirondacker12800

        Other cities may have a few miles of car-only limited access highways here a handful of miles there. Not a whole alternate highway system like there is in metro New York.

Leave a Reply to Richard Mlynarik Cancel reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.