This is 59th Street in Manhattan. Below it, spread across downtown and midtown Manhattan, reside some 684,000 New Yorkers. That’s more than the population of Vermont living in just under nine square miles or 23 square kilometers of land.

Or, that’s a population density of about 76,000 people per square mile, making Lower Manhattan singularly dense by American standards.

If this were a city by itself, it’d be the country’s most densely populated, far out-cramping current #1, Guttenberg, New Jersey, just across the river, where a fifth of its 12,000 residents are stuffed into a single building complex.

If it were its own county, its population density would almost double the nation’s second most dense: Kings County, New York. And for similarly dense cities to Lower Manhattan, you’d need to go abroad, where it would rank around Kolkata, Kathmandu, and Makati, comfortably within the world’s top ten.

But 76,000 per square mile only begins to tell the story of just how cramped this area really is. New York City also has the most office space by square footage of any city in North America, and just Manhattan by itself contains some 81% of that.

For comparison, Los Angeles, the city with the second most such space at 432 million square feet, is dwarfed by Manhattan’s 592 million.

With so much space comes a flood of commuters every morning, adding another 1.2 million or so people to the southern end of the island.

For the workday then, an area of just nine square miles has a momentary population of about two million. Two million that need to get to work, two million that need to run errands, and two million that need to get back home at the end of the day, which, for a majority, means getting back out of Manhattan.

But while this sub-city urban core that’s more populous than a handful of American states is uniquely large, uniquely mobile, and uniquely stuffed; for the past century—effectively, the age of the automobile—there’s really been no unique solution or system for managing or limiting traffic.

Aside from a prohibition on right turns on red and honking, driving in America's densest nine square miles works the same way as in the other 3.8 million square miles.

Because of this, each and every workday, some 140,000 commuters enter Manhattan by car, while about 1,000,000 people enter the commercial business district south of 60th Street by automobile—largely for only the price of the gas it takes to get there, the toll to cross the bridge, and the cost of parking.

All expenses that a rural American would scoff at, but expenses that hardly reflect the shared cost that cars create in the U.S.'s most densely populated environment—from air quality and pedestrian danger, to simple quality of life.

The result is this: New York’s #1 standing in Inrix’s 2023 congestion impact ranking, where the average driver moved at just 10 miles or 16 kilometers per hour and gave up about 101 of their hours a year to sit in traffic-related delays.

And those are stats for the whole of New York City.

Manhattan was far, far worse. While most streets and avenues are limited at 25 miles per hour, average traffic moves at a glacial 7 miles or 11 kilometers per hour.

This is not a new problem, either. Rather it’s one that’s just getting worse—in 2010 the average taxi navigated the Central Business District, or Manhattan below 60th St., at 9.1 miles per hour, and every year following through 2018, it’s slowed.

While COVID lightened the congestion for 2020 and 2021, experts are now saying that traffic hasn’t only returned, it's gotten even slower than before 2019.

Now, the exact complexion of the problem has shifted over the years.

The number of daily vehicles entering the Core Business District, for instance, has dropped, but their impact has been replaced by an ever higher share of for-hire vehicles serving ride-hailing apps and clogging the city streets.

But the very root of the problem is so remarkably simple: There are just too many cars.

Now, oftentimes in an urban context, a problem may present as simple and straightforward enough, but a solution could be anything but. That, though, is not necessarily the case here. In fact, as long as car congestion has been an issue here, one potential answer's remained equally obvious, equally straightforward, equally simple—disincentivize driving into Manhattan.

Going as far back as the 1930s, mayors have proposed tolls to cross into Manhattan over the East and Harlem Rivers to raise city funds or, in the '50s, to raise money for the subway.

In the '60s and '70s, such ideas were again returned to in an attempt first to popularize mass transit use, then to make progress toward meeting the standards of the 1970 Clean Air Act.

Further hoping to pressure individual car users, Mayor John Lindsay thought to reduce city parking in the business core, while Mayor Ed Koch in 1979 looked into banning private cars entirely.

For half a century, New York City leadership has understood that in order to clean its air, raise money, and/or support its mass transit system, it's needed to make it harder or more expensive to drive from New Jersey, or Queens, or Brooklyn, into Manhattan south of 60th.

But it has come close. It started with a speech outside the Museum of Natural History, where then Mayor Michael Bloomberg introduced the idea by name—congestion pricing. It was just one of many components to his larger PlaNYC initiative introduced on that Earth Day in 2007, but one of its boldest.

From below this line, any car entering Manhattan from 6 a.m. to 6 p.m. Monday through Friday would pay $8, while trucks would pay $21 per daily entry into the zone.

Taxis would be exempt from the fee, as would emergency vehicles 
and those with handicapped license plates, and   any vehicles skirting around the perimeter aiming 
not to enter Manhattan, but to get around it.

So, someone driving from Brooklyn to the Bronx 
by way of FDR Drive wouldn't have to pay the   fee. The tolls would be collected by E-ZPass 
readers so as not to slow down vehicles, and the toll money itself, 100% of all net revenue, 
would be invested back into city transit.

The project was bold, but between transit 
experts, bureaucrats, and elected officials, it was popular.

State Governor Eliot Spitzer, 
Representative Joseph Crowley, and even   presidential candidate Barack Obama supported the 
plan, while the US Department of Transportation supported the initiative by awarding the city 
$354 million in grants should it commit   to embracing new tech and infrastructure to 
combat urban congestion.

The MTA also strongly supported the program, as it accelerated its own 
capital planning process to align itself with the   congestion legislation.

On a hurried timeline, it 
introduced its largest ever spending program of $29.5 billion for the next five years—hoping to 
cover subway fleet expansion, the rehabilitation   of 44 subway stations, the purchase of some 2,500 
new buses, the replacement of 57 miles of mainline track, and to catch up on general maintenance 
across the transit authority's property.

And beyond bold and broadly popular 
with most politicians and bureaucrats, congestion pricing was proven—in recent years, 
Stockholm and London had experimented with   congestion pricing in their downtown cores and 
were so far experiencing major benefits.

Mass transit systems benefited from the influx of 
investment, city residents benefited from the   increase in air quality and calmer streets, and 
even those that chose to begrudgingly pay the fees benefited from the smoother traffic conditions.

But just like in London and Sweden prior to   adopting congestion pricing, the idea wasn't 
popular among all residents. At one point, polling of New York City showed that 61% of all 
respondents opposed the move, with numbers from   the other boroughs even higher, as 70% of 
Bronx-area respondents opposed the idea.

Opposition was especially vocal in areas 
like Eastern Queens and Southern Brooklyn,   where access to public transit was more difficult, 
and thus representatives argued such fees would unfairly burden their constituents.

Then across 
all of New York, there was a shared sentiment that   New Jersey commuters were getting away without 
paying their share, as fees for Jersey commuters wouldn't be meaningfully increased because the 
tolls they already paid to cross into Manhattan   would cover their congestion fee.

And then 
there was one claim across the city that's come to follow congestion pricing—that it's elitist.

Congestion pricing, some correctly claimed, would   impact the working and middle-class car commuters 
more than their upper-class counterparts.

An $8 fee for the pleasure of commuting would 
be a much larger proportional chunk of someone's   workday expenses if they work making $15 an 
hour instead of $150 an hour. Following such logic then, congestion pricing could be painted 
as a means to rid the roads of working-class   commuters—forcing only them to figure out new 
commute routes—while the city's elite were able to now drive into town with increased ease for 
a price that they could more comfortably cover.

Congestion pricing passed in the City Council, 
but once it made its way to the state assembly, Democratic lawmakers representing the 
city outside Manhattan blocked a vote,   missing the federal funding's deadline, and 
killing congestion pricing on the spot.

Ultimately, New Yorkers had proved that they 
understood this was a feasible solution and   that congestion was a real problem worth 
addressing, but the solution got hung up on issues as to whose cars, exactly, were to 
be limited, and who, exactly, would benefit   the most from such a program.

But what's easy to 
miss, what was missed by the opposition in 2008, and not made clear enough by proponents 
in 2008, was exactly how elegant and   ingenious a solution congestion pricing really is.

You see, when driving, everyone values their time differently depending on... well, a lot of things: 
where they're going, whether they're on time, why they’re driving, etc. And then these varying personal values translate to different monetary values based on differing income levels.

A person of average income driving into the city just for a visit on their day off might value their time in traffic quite low—say $10 per hour—meaning they’d theoretically be willing to pay $10 to eliminate that hour of traffic, but any cost above that and they’d rather just wait it out.

But in the car right behind them, there might be a person of the same income level who’s late to their first meeting of the morning. They might value their time quite highly—maybe $100 an hour—as the consequences of the delay from traffic are far higher.

They’d pay an enormous amount to eliminate traffic in that scenario. Theoretically, it’d be fair and good for both these drivers if the high value-of-time person paid the low value-of-time person to cut in front of them, and do the same with plenty of others in order to get to their destination faster.

In some places, this effectively happens through tolled express lanes, but that’s not practically feasible in Lower Manhattan. What is is congestion pricing. In it, the person valuing their time higher is paying, say, $15 to drive on the roads, and that $15 is effectively going to the low value-of-time person who’s been priced out of driving.

That’s because that person would get pushed onto public transit, and so if congestion charge fees go to public transit, as is typical, it’s funding improvements that make transit better, faster, or cheaper.

But this is, of course, an overly simplified reality, and it only considers the benefits to individual car users.

In practice, there are plenty of upsides to freeing up the roads even for non-drivers due to all the other scenarios where time is exceptionally valuable for a particular vehicle type.

For example, with the bus. With ten, twenty, thirty or more people, even if their individual value of time is low, the collection of all of them means one bus can have a value of time well into the hundreds per hour.

Delivery vehicles are a similar scenario—in fact, research by the Federal Highway Administration suggests that shippers value transit time between $25 and $200 per hour.

Everyone who buys anything in Manhattan effectively pays for any inefficiency in goods delivery, so it's a scenario where the costs of traffic get passed on to everyone.

And perhaps the greatest example is that of ambulances. A report by New York State Senator Brad Hoylman-Sigal and former NYC Traffic Commissioner Sam Schwartz pointed out that between 2014 and 2024, New York EMS response times for life-threatening emergencies increased by 24% from 6.09 to 7.54 minutes.

Not all of that increase can be directly attributed to traffic, but a lot can—half, in fact, according to their analysis.

Minutes matter during life-threatening emergencies. As the report points out, in cardiac arrest, a person's chances of survival decrease an average of 7-10% for every additional minute before medical intervention.

With a stroke, each additional minute is linked to the loss of an additional 1.9 million brain cells, meaning New York City traffic quite directly leads to increased disability for stroke victims.

Right now, the cost of driving in Manhattan is so low that there are plenty of people doing so even when it's barely worth it to them.

But with buses and delivery vehicles and ambulances using the same roadways and having their extremely high benefit to society diluted by the congestion caused, in part, by the low value-of-time drivers who have little disincentive to driving, the net benefit to society of Manhattan's roads is lower than it could be.

That's to say, everyone that could take a fast cross-town bus, everyone who buys goods brought in by delivery vehicles, everyone who might need an ambulance would benefit from congestion-free roads, but they can't capture that benefit now because there are all these drivers out there that only barely want to be driving.

But the logic goes further. You see, traffic is not a linear phenomenon.

That's to say, every additional vehicle does not have the same effect on how much it slows down traffic. The impact of an additional vehicle varies depending on how congested the road already is.

On a highway with a 70 mile per hour speed limit, it's expected that the roadway can have a throughput of up to 1,120 cars per hour, per lane, while still maintaining 70 mile per hour average speeds.

As the number of cars using the highway increases, their density increases, and therefore average speeds decrease—at 32 cars per mile per lane, for example, speeds slow down to just 64 miles per hour.

But even with the slower speeds, the higher density means the roadway's throughput still increases to 2,048 cars per hour.

The roadway is being used more efficiently as an asset, even as the efficiency of individual journey times gets lower.

This phenomenon continues as speeds get slower and density gets higher—up until a point. According to the analysis behind this set of numbers, on a 70 mile per hour highway, that tipping point is 53 miles per hour.

At that speed, 2,400 cars get through a lane in an hour, but it's not really ever theoretically possible to get more than 2,400 cars per hour through one of these lanes.

While the intersection-less highway is the simplest example, this same phenomenon occurs on all roads, with any congestion—there just is a theoretical maximum throughput that roads can't surpass, and overloading the road actually decreases its efficiency below this maximum throughput.

The factors that lead to this phenomenon are numerous, varied, and not universally agreed upon or understood, and even the ones that are are wildly complex, often being studied through the theories of fluid dynamics, but there are at least some examples of the nonlinear nature of traffic that make intuitive sense.

Thinking about the scenario of a long line of cars at an intersection, let's say it takes the first driver at the light a half second to react to it turning green. That's a half second of time lost by everyone in line, as it delays all of their arrival to the light.

Let's then say it takes the driver second in line another half second to react to the first driver moving.

That's now a full second of delay for everyone behind them. Now, if there are only five cars in line, this isn't a big deal, as the fifth driver only loses two and a half seconds to the cumulative inefficiencies of human reaction 
times.

If there are 20 cars, though, that’s 10   seconds. Since every car has to wait through 
that whole backup, the difference in travel   time between a lightly congested intersection 
and a heavily congested one, in this example,   is about 10 seconds—and even that’s assuming 
they all get through in the same light cycle.

Now, once again, this is only one small facet 
of the overall phenomenon. There are a myriad   of different factors that compound and decrease 
efficiency exponentially as roads get congested,   but the takeaway should be that considering 
traffic is exponential, there’s a lot of logic   in reducing vehicle numbers at least below the 
tipping point, to the point at which roadways   are most efficient.

Manhattan’s roadways are 
consistently well above that point, and congestion   pricing is the best idea for how to fix that. 
Of course, all these benefits are understood   by the people who care most about this 
issue—certainly not every New Yorker,   but what’s proved enough of them to keep the 
concept alive a decade and a half after its   failure in the state assembly.

As of the end of 
2024, congestion pricing in the central business   core of New York City is the closest it’s ever 
been to becoming a reality. The policy is set: $9   per commercial vehicle in 2025, then likely $12 in 
’28, and $15 in 2031.

The 100+ gantries required   to monitor roads and dole tolls are in place, 
too.

Now, all that’s left to do is survive the   new year and its first four days until the program 
goes live on January 5th.

But even with Governor   Kathy Hochul’s greenlight, transit experts are 
still holding their breath.

And justifiably so.  It was on the back of the spectacular, public 
meltdown of the MTA in 2017 that congestion   pricing returned into conversations over how to 
again fix the city’s crumbling public transit.

As The New York Times noted, the $500 million a 
year that congestion pricing would have generated   could’ve gone a long way to alleviating the city’s unraveling transit disaster.

Governor Andrew Cuomo, for his part, took a more future-forward stance; that congestion pricing was an idea whose time had finally come.

And yet, since its time arrived, there’ve been two state governors, two city mayors, a pandemic, a handful of lawsuits, and when the project was finally, finally, finally on the verge, a pause from Governor Hochul in early 2024.

In so many ways, congestion pricing as part of Fix NYC, initiated by Cuomo, has played out as Bloomberg’s congestion pricing as part of PlaNYC initiated a decade before it.

The more recent one was able to get the votes at the state level, but has run up against similarly heavy political opposition in different forms—from outlying towns, the state of New Jersey, and Donald Trump.

But perhaps most confidence-inspiring as to the long-term prospects of such a program is the longevity of those that 2008 and 2017 proposals drew from, the congestion pricing models in London and Stockholm that have remained in place for two decades now.

Each has now played out slightly differently, as Stockholm’s has kept total cars and congestion lower than London, but London has made real strides in turning fees into additional bike lanes and bus routes, but both remain success stories that New York’s drawn from.

And likely most instructive for the New York urbanists is the fact that congestion pricing became acceptable to Londoners polled and popular with those polled in Stockholm, popular enough to remain policy and even survive rate hikes over the years.

So with history as a guide, what’s most likely to happen is a continued, bitter fight over New York’s new tolls until the very hour they're implemented, but then, slowly, as the streets get clearer, the air gets cleaner, the noise quieter, the subways better, the ambulances faster, New York will slowly learn to love paying $9 to speed through Lower Manhattan.

Speaking of learning to love paying for something, over half a million people have done so with our sponsor, Nebula. Seriously, we know people like it because a majority of people that have ever signed up for a subscription are still subscribed, which is well over industry averages, especially for a service that’s been running for almost six years.

But I think I know why. It starts with the fact that Nebula has curated a group of thoughtful creators making unique stuff, and we all post all our normal stuff there ad and sponsorship free.

But really the core of Nebula, and what people seem to enjoy most, is the Originals.

You see, the reason Nebula is a paywalled, subscription-based streaming platform rather than an ad-supported one is that we believe that the business model is better for both creators and the audience.

We’re able to make stuff at a higher budget than what the ad-supported YouTube system can support, which has led to plenty of really great Originals.

For example: Great Cities by City Beautiful, covering the backstories of the urban planning of places like Venice and Shanghai and Paris.

There’s the brand-new Boomers by Tom Nicolas—a feature-length documentary about the consequences of the increasing power and wealth amassed by the baby boomer generation.

The Battle Over NYC Congestion Pricing

Wendover Productions is all about explaining how our world works. From travel, to economics, to geography, to marketing and more, every video will leave you with a little better understanding of our world. New videos go out every other Tuesday.

Wendover Productions

Lower Manhattan's extremely high population density compared to other American cities and even some of the densest cities globally.

684,000 residents live in under nine square miles, a density of 76,000 people per square mile.

This density would make lower Manhattan the most densely populated city in the US if it were independent.

Globally, its density is comparable to cities like Kolkata, Kathmandu, and Makati.

Manhattan contains a vast majority of New York City's office space, further increasing daytime density.

Daily commuters add 1.2 million people to lower Manhattan during the workday, bringing the total to around two million.

Manhattan's traffic congestion problem and the lack of effective solutions despite its long history.

The daily influx and outflow of people create significant traffic challenges.

Despite a century of automobile use, there's no unique system for managing Manhattan's traffic.

Thousands of commuters enter Manhattan daily by car, mainly bearing the costs of gas, tolls, and parking.

New York City ranks high in congestion, with Manhattan experiencing even worse traffic.

Traffic speeds in Manhattan average around 7 mph, significantly slower than the speed limit.

The history of congestion pricing proposals in New York City and their repeated failures.

Congestion pricing has been proposed since the 1930s for various reasons, including raising funds and reducing pollution.

Mayor Bloomberg's 2007 congestion pricing plan proposed an $8 fee for cars and $21 for trucks entering Manhattan during peak hours.

The plan initially received support from various politicians and agencies, including the US Department of Transportation and the MTA.

Public opposition, particularly from outer boroughs and those concerned about fairness and elitism, ultimately led to the plan's failure.

The economic rationale behind congestion pricing and how it addresses differing values of time.

Individuals value their time differently, and congestion pricing allows those with a higher value of time to effectively pay those with a lower value to reduce their commute time.

Revenue from congestion pricing can fund public transit improvements, benefiting those who switch from driving.

Public transit vehicles, delivery vehicles, and emergency services have a high societal value of time, making congestion reduction crucial.

The nonlinear nature of traffic congestion and how it justifies reducing vehicle numbers.

Each additional vehicle on the road doesn't have the same impact on traffic flow; congestion's impact is exponential.

Roads have a theoretical maximum throughput, and exceeding it actually decreases efficiency.

Simple examples like reaction times at intersections illustrate how delays compound with increasing congestion.

Reducing vehicle numbers below the tipping point can significantly improve overall road efficiency.

The current status and future prospects of congestion pricing in New York City.

Congestion pricing is closer than ever to implementation, with the policy set and infrastructure in place.

Despite political hurdles and delays, the program is scheduled to start in January 2025.

Successful examples from London and Stockholm provide optimism for New York's program.

It's expected that initial opposition will eventually give way to acceptance as the benefits become clear.

The Battle Over NYC Congestion Pricing

Scripts