Taming wide streets

Michael Lewyn's picture

Before moving to New York, I'd viewed street design through a fairly simple lens: narrow streets good, wide streets bad.  By and large, I still hold this view.  But after living here for a few months, I have learned that not all wide streets are equally bad.   The wide roads of the South are generally terrible, but New York has made some of its wide streets a bit more pedestrian-friendly.  To see why, go to Google Street View and examine three addresses: 5019 U.S. 23 in Chamblee, Georgia, 770 Eastern Parkway in Brooklyn's Crown Heights neighborhood, and 107-43 Queens Boulevard in my current Queens neighborhood of Forest Hills.

First look at the Chamblee address.  You will see a road that is seven or so lanes wide.  There is no median, so the pedestrian has to cross seven lanes at once, obviously a death-defying act.  Not surprisingly, pedestrian fatalities and injuries are fairly common on this road (known to everyone but Street View as "Buford Highway").   Other suburban Atlanta streets have medians that are so thin as to be barely usable: for example, if you go to Cobb Parkway and Galleria Parkway in Smyrna, you will see six lanes of traffic, then a median that appears to be about a foot or two wide, then four more lanes of traffic.  Would you feel safe standing on that median?

Now go to the Queens Boulevard address.  This street appears to be about twelve lanes.  But having lived here for a few months, I can say with some certainty that it is less scary to cross.  Why?  Because every three or four lanes there is a median.  So a pedestrian can walk across three lanes of one-way traffic to median A, then to median B, then finally to the other side of the street.  Moreover, a couple of the outer lanes are taken up by on-street parking, which means that the pedestrian often has to cross only one or two lanes of speeding traffic at a time rather than three.  Because each chunk of the trip is not so long, the street is not as intimidating as the Atlanta-area addresses discussed above.

Finally, go to the Eastern Parkway address.  The street itself is twelve lanes wide: three lanes, then six lanes in the middle, and three lanes.  But the northern and southern sets of lanes again have on-street parking, thus reducing the amount of car traffic that the pedestrian must contend with.   And when the pedestrian crosses the first three lanes, he or she reaches not a tiny concrete median but a "pedestrian mall" (really a median wider than most sidewalks) which is lined with greenery and street trees.  After the six-lane main stretch of Eastern Parkway, the pedestrian reaches another pedestrian mall.  Thus, our pedestrian not only does not have to cross all twelve lanes at once, but gets to walk in an fairly appealing environment if he or she just wants to be on Eastern Parkway rather than being on both ends of that street.  (On the other hand, Eastern Parkway's six-lane middle stretch might be scarier than any one part of Queens Boulevard).

In sum, both Eastern Parkway and Queens Boulevard are far more pedestrian-friendly than Buford Highway, despite being significantly wider.  Why? Because both streets use on-street parking to reduce the number of lanes a pedestrian must cross, and use medians and greenery to ensure that pedestrians don't have to cross as many lanes at a time as on Buford Highway.  In addition, Eastern Parkway uses its "pedestrian mall" to decorate the street and make it feel more welcoming than the concrete jungle of Queens Boulevard.

Michael Lewyn is an assistant professor at Touro Law Center in Long Island.



Excellent point

Excellent point, Michael. That is an obvious solution to the problems pedestrians experience with ultra wide streets.

Ultra wide streets are of course going to be less congested than they would be if they were narrower - OR population and business would be more dispersed in reaction to the congestion. Colin Clark pointed out decades ago that the "density" a city can develop to is very much constrained by the amount of road space with which it is provided. The irony is that the cities with the Smart Growth advocates favourite transit systems, often have by far the most cars per square km as well. Clark's point would be that the Transit-supporting population densities would most likely not exist without the abundant road space.

Narrow Streets and High Density

"Ultra wide streets are of course going to be less congested than they would be if they were narrower - OR population and business would be more dispersed in reaction to the congestion."

For a refutation of that claim, see my pictures of streets of Amsterdam at http://preservenet.blogspot.com/2011/06/straten-van-amsterdam-streets-of...

We could call Wodehouse's claim the "lump of traffic fallacy."

Charles Siegel

Whose fallacy?

Your example, Amsterdam, does not disprove my point. Holland has had very strict urban planning for decades, including compulsory acquisition of land for redevelopment.

Good luck trying to get compulsory acquisition of land accepted by the voting public anywhere else, so your "plans" can be made to work.

In Holland, there is almost nowhere else for population and business to disperse TO in reaction to the congestion. This is why the Dutch people have one of the world's biggest "diasporas" today.

And I still say that there is NO WAY Amsterdam could match Manhattan's densities, EVEN WITH the planners having the advantage of "compulsory acquisition", UNLESS they provide something closer to Manhattan's lane-miles and network of roads.

I also say that nations that use blunt instruments to restrict urban growth and drive their urban land prices up, are doomed to long term relative economic decline along with any nations that have actually run out of land - Malta, say. (And even then, the small island nation-state can reinvent itself as a tax haven - but this is still doom for the locally born "small producer").

Amsterdam Vs. Manhattan

"there is NO WAY Amsterdam could match Manhattan's densities, ... UNLESS they provide something closer to Manhattan's lane-miles and network of roads."

That is fine with me. I think Amsterdam is more livable than Manhattan, and I hope they keep it that way.

Charles Siegel

That is fine with me too.

That is fine with me too. I meant NO CRITICISM of Amsterdam at all, it is Manhattan I am criticising. I actually admire the willingness of the Dutch people to accept compulsory acquisition of land so as to be able to achieve better urban form. They are not radical leftists, they just have the common sense to see that this is necessary. UNLIKE most urban containment and transit-oriented-development advocates, who do not see that all they are doing is helping certain property owners rake in semi-monopoly capital gains. Have I ever quoted Anthony Downs on this subject, to you?

I think it is ironic that Manhattan is so much propounded as a kind of ideal urban model because it has the most financially sustainable transit system in the USA, when the reality is that to achieve this, it also needs to have the highest densities accompanied by the greatest amount of motor vehicles per sq km and the worst air pollution.

Fine With Both Of Us?

So, do you agree with me that it is better to build cities in a traditional urban scale, like Amsterdam, than to build high-rise cities?

The traditional-scale cities are more livable: as you say, Manhattan densities require high densities of vehicles and large areas or road space, which makes them less livable. The traditional scale cities are also completely accessible by walking, bicycling, and transit, just as Manhattan is.

Charles Siegel

I am definitely against "Manhattan-isation"

I have always been against urban "plans" that assume that so much growth in population will be accommodated within existing urban areas, that high-rise buildings will be a necessity. But I have no problem with something like Manhattan developing under free market conditions because of a rare series of historical coincidences. I also believe that the free market trend will be for such high density city centres to weaken, and I disagree with planning and subsidies that attempt to prevent this just because mass public transit is rendered more sustainable by that urban form. If you too are against this "tail wags dog" planning, i.e. start with mass public transit and then try and design employment location patterns and a whole economy around that, then we can definitely be allies.

THIS is the sort of thing that gets me really fuming:


The writer of that is either utterly ignorant about "cities", period; or he is a CBD property owner trying to boost his investment on the back of the fate of pretty well everything else in the economy.

I regard the main difference between celebrated European cities and the typical US one that is routinely sneered at by the planning classes, as "renewal" - which mostly did not happen in Europe. Many US cities slums would have gentrified of their own accord, and provided the same walkability and livability today, thanks to advances in technology, that Europe's one-time centre city slums (which were just as horrible as US ones) now do.

Amsterdam and other Dutch cities have the unique advantage that the government has long wielded the threat of compulsory acquisition of land, should property owners be seen to be getting too greedy with their asking prices. Probably the number one thing I am fighting for on this forum, is a recognition of what real estate markets and property rights and market prices actually DO to "plans" in the absence of mechanisms for compulsion and the over-riding of property rights. Anthony Downs and Patrick Troy are 2 specialist writers who got this right years ago.

Instead of causing significant changes in urban form, most "smart growth" type plans just cause the prices of property to rise, and rise MOST in the very locations that the planners wished for MORE people to live. The result is that those locations turn into exclusive ones rather than the focus of population shifts.

The London School of Economics has co-opted numerous Dutch economists (like Vermeulen, Van Ommeren, Hilbers) to assist their own very able team in analysing urban land markets and the effect of regulations, because there is apparently a good understanding of these issues in the Netherlands long since, with an academic literature in the Dutch language. Hence the Dutch acceptance of compulsory acquisition as a legitimate tool.

Capital gains and quasi monopoly rents created by government regulators "plans" should NOT be seen as part of the parcel of "property rights" at all - they are regulatory "GIVINGS", and the removal of them is NOT a "TAKING". I say this as an economic libertarian who nevertheless can see the difference between a greater evil and a lesser one. I am completely convinced by Mason Gaffney's argument in favour of land taxes as a genuine benefit-maximising growth restraining tool; Gaffney was 100% correct about "NEGATIVE" growth containment way back in 1964.


Advocates of urban growth restraint leave themselves open to charges that they are in cahoots with vested interests, by insisting on prescriptive "plans" rather than compulsory aquisition of property and/or land taxes. Even if there is no such corruption currently in existence, it is surely not wise to set the conditions so perfectly for it to come into existence (and history suggests that we would be extremely naive to assume that it does not already exist, let alone that it will not come into existence).

Maybe certain advocates need to re-assess just why they are so popular with certain people? The result of being pure but ignorant, is merely to be a useful idiot in what actually is a serious evil - one of the main causes, in fact, of the major transfer of wealth upwards and concentrated at the top of society, over the last 20 years.

Michael Lewyn's picture


Speaking as someone who probably knows less about Amsterdam than anyone else on this thread-

assuming arguendo that Amsterdam's densities are lower than those of Manhattan, is it fairer to compare Amsterdam with Manhattan or with the five boroughs?

And is there any way of measuring whether the Dutch have more of a "diaspora" than other small countries?

And is there any way of measuring whether Holland is more "congested" than comparable countries?

Amsterdam Congestion

"And is there any way of measuring whether Holland is more "congested" than comparable countries?"

I don't have quantitative measurements, but I spent three weeks walking around central Amsterdam for several hours each day, and I did not see a single traffic jam. I saw only one traffic backup, where about five cars were waiting on a one lane street for a stopped car to unload. (I did see many drivers waiting for pedestrians and driving slowly behind bicyclists.)

Despite the very small amount of street area for cars, there is no traffic congestion in Central Amsterdam (except for occasional bicycle congestion), as you can see in the pictures at the link I posted above. Central Amsterdam had far less automobile traffic and far less congestion than central Berkeley, where I live, though Berkeley has far more street area per person.

There was much more traffic and some congestion on the ring-road freeway on the edge of Amsterdam - but the congestion was not nearly as bad as on the I-80 freeway on the edge of Berkeley, which is larger than the freeways around Amsterdam.

Of course, I also saw far more bicyclists and intercity train users in the Netherlands than in Berkeley.

Michael, if you have a chance, you should visit Amsterdam. I am sure you would love it.

Charles Siegel

Michael Lewyn's picture

But does cars per square km indicate more road space?

It seems to me that "cars per square km" may be an indication not of abundant road space but of the lack of same (that is, the small number of square km). It seems to me that if a city doesn't spread out the population through certain types of road space, the number of cars per person will be smaller, the number of square km per person will be smaller and the amount of road space per person will also be smaller.

So for example, imagine two cities, New Amsterdam and Sprawltown.

New Amsterdam has 100,000 people living on two square miles (Manhattan level density). Only 5000 of them have cars. Assuming that the number of lane-miles is 5 times the number of square miles, there are 10 lane-miles of roads and streets. Thus, there is only one lane-mile per 10,000 people (pretty low) but 500 cars per lane-mile (pretty high).

Sprawltown has 100,000 people living on 100 square miles (Jacksonville, Fl. density). 70,000 of them have cars (the rest are mostly children). Again, assume the number of lane-miles is 5 times the number of square miles, so assume 500 lane-miles. If my math is correct, there is one lane-mile per 200 people (very high compared to New Amsterdam) but 140 cars per lane-mile (significantly lower than New Amsterdam).

Does it really make sense to say that road space is more abundant in New Amsterdam?

A fascinating mental exercise.

A fascinating mental exercise. To utilise your own method of illustrating by extremes, let us assume that New Amsterdam's 5000 cars for 100,000 people, is about the barest minimum number of cars that this many people can exist with. Let us also assume that 500 cars per lane mile is maximum congestion.

IF New Amsterdam only had 5 lane miles of roads instead of 10, what I am saying (from Colin Clark) is that it is far less likely that there would be 100,000 people there. OR, if New Amsterdam had 20 lane miles of road instead of 10, maybe it might have 200,000 people instead of 100,000.

I think this PRINCIPLE holds good for Manhattan. Of course there has to be a whole lot of other factors that enable this kind of density in the first place - rents in tall buildings are never cheap. "Sprawltown" almost certainly is not a centuries-old international trade and finance centre.

Yeah, but...

Yeah, but who wants to live, work or play along a wide, noisy, polluted, congested roadway, no matter how much it has been "tamed" with median islands, landscaping and buffer zones?

Narrow roads also make good economic sense. They are a smarter use of taxpayer dollars, especially during this Great Recession, and having buildings and roads built on a human scale fosters the business and personal relationships that are a result of people literally being closer together.

Related: Queens Boulevard Central Park

Check out this proposal for Queens Boulevard. http://qbcpark.com/ The project proposes to turn 8 of the 12 lanes of roadway between Grand Avenue and 64th Road into a Central Park. The park would extend over the Long Island Expressway and would accommodate various amenities and commercial and retail spaces along it between the surface and subway levels. Cross streets would tunnel through the park to permit through traffic, and bus and bike lanes would extend continuously along the park. The park would also implement green infrastructure to reduce stormwater runoff, detain and clean stormwater, and prevent Combined Sewer Overflows (CSOs) from discharging sewage and stormwater in the NYC waterways.

Prepare for the AICP* Exam

Join the thousands of students who have utilized the Planetizen AICP* Exam Preparation Class to prepare for the American Planning Association's AICP* exam.
Starting at $245

Essential Readings in Urban Planning

Planning on taking the AICP* Exam? Register for Planetizen's AICP * Exam Preparation Course to save $25.

Wear your city with style!

100% silk scarves feature detailed city maps. Choose from five cities with red or blue trim.

Brand new! Urban Grid City Collection

Each city has its own unique story. Commemorate where you came from or where you want to go.