As the discussion of equity is always tied to congestion pricing, here is a study from RAND titled Equity and Congestion Pricing [pdf]. The Summary at the beginning covers most of the findings.
> First, depending on how congestion pricing is implemented, it can be either regressive or progressive. This depends in large part on how toll revenues are used. For instance, if regions spend revenues in ways that benefit low-income individuals, congestion pricing is more likely to be progressive. However, if regions use revenues in a way that benefits all individuals equally, congestion pricing may be, overall, regressive. This is the strongest finding in the economic literature.
> Second, even when low-income and other transportation-disadvantaged groups benefit as a whole from congestion pricing, it is very likely that some individuals will still be worse off.
> Third, for all forms of congestion pricing (but more for some than for others), the distri- bution of residents and job opportunities (not to mention shopping, schools, places of worship, and other important destinations) has a large impact on the equity implications.
> Fourth, high-occupancy toll (HOT) lanes, the most common form of congestion pricing in the United States, tend to raise fewer equity concerns among motorists, since they provide drivers with an additional choice of using a set of priced lanes while allowing them to continue using parallel, free lanes if they prefer.
> Fifth, while congestion pricing has been shown to reduce emissions in general, there is scant evidence showing that congestion pricing can specifically reduce negative environmental consequences for neighborhoods disproportionately affected by emissions.
> Given the risk of negative impacts to low-income and other groups under congestion pric- ing, we looked at suggested ways to diminish these impacts. Two mechanisms are in common use: (1) revenue redistribution and (2) discounts and exemptions.
Agreed - but if you increase that to people from a household with a car, I bet the number increases. And if people are getting dropped off at work and later picked up, they’ll get dinged twice.
Unless the city was really built around transit (and not just once developed in a hub and spoke like so many cities) it’s hard to beat a car door to door. That’s what we should be focusing on fixing (things like inter borough travel in NYC). Massive congestion charges might be part of that, however.
I'd propose a somewhat automatic congestion toll: the toll for a segment of road adapts to keep (assuming only car traffic; extension makes it less trivial but by no means impossible) the road just about empty enough to let everyone drive at the design speed (AFAIK 90 mph for most of the US interstate network).
The effect is that the mechanism prevents significant latency degradation from demand.
While it's not necessarily possible to go that far over night, looking at how much of the traffic/congestion is from cars, and how many occupants they have, should allow to calculate what reduction perfect car pooling (with, say, 4 people per car) would allow, and at what speed that reduced number of cars could drive (without tailgating). Then, add some safety margin and set that reduced speed as the target for the auto-toll-pricing.
If car pooling physically can't fix the congestion, it'd take a bit for the problematic roads to get enough (mini)bus service to enable the full design speed.
WFH isn't the solution, and proper mass transit takes a while (5-10 years) to build even if you're in a hurry.
Wouldn’t the best way to do congestion pricing be to simply optimize for maximum revenue (which you then mostly redistribute either equally or progressively)?
I feel like the way to maximize revenue would just be to try either raising or lowering the price (by say 20%?) every n (5-10?) minutes, and see if your revenue goes up or down. If it goes up, continue what you’re doing, and if it goes down, reverse it?
Revenue maximisation in this case is not a great goal. The idea behind congestion pricing is to price-ration road-usage so that people make socially optimal use of roads (which in theory is achieved by setting prices at marginal social cost). Pricing should vary in line with road congestion so that people respond by altering their road usage patterns (e.g. shifting working hours so their commute is outside of peak times, make road deliveries in non-peak hours etc.)
If you go for revenue maximisation you won't be setting prices at the social welfare maximising level. Also, you'd probably want to charge higher prices for heavier vehicles as they cause more road damage per unit of distance travelled.
People can’t change what they’re doing that fast - you’d have to do it over weeks or months.
The goal is to tip the balance from “30 min drive vs 1 hour transit” to “30 min drive + 10€ vs 1 hour transit” and if the equation keeps changing people will feel trapped.
One thing that the author misses is that a dollar spent on congestion charges doesn't matter the same to everyone. A fancy lawyer probably doesn't even think about a $10 toll but a grocery cashier probably would drive an extremely inefficient route to avoid that toll.
I think we need to find a way to "charge" people in a way that matters equally to everyone. I usually advocate for things in terms of time (community service hours) because that perfectly hedges against the person's wage level. Although I suppose given their tax receipts we could just adjust the monetary charge for their observed hourly wage.
Any other ideas for adjusting "charges" to have equal importance for different people?
We don’t price much of anything like this, though? There are subsidies for low-income people, sure. But we don’t really do sliding scale payments for food, or public transit, or cars, etc etc etc.
Why does scarce road space in highly-congested (and well-connected) CBDs require a sliding scale when little else does?
It’s different than grocery stores and plumbers I think because roads are seen as a public good, and if they’re using space in a crowded city, that space was/is also a public resource. Since everyone contributes their individual sovereignty equally, there’s a sense that govt should serve everyone’s interests equal, as best it can.
> But we don’t really do sliding scale payments for food
But we do! When I was a poor student, I usually bought the bottom-shelf store-brand cheap pasta. Today, I look at a higher shelf and often get the organic high-protein stuff with fancier shapes.
It costs about the same per unit weight to make, but the latter costs much more to purchase. This is exactly a sliding scale that allows people who are more price sensitive to buy practically the same stuff except at a much lower price.
(Why do companies do this? It lets them expand their target market without getting total profits too close to the floor.)
----
This is also why we have region locks on electronics and export controls on medicine: companies are selling literally the exact same stuff with different profit margins in different regions, based on their collective disposable income.
That’s an interesting take. And probably true. The pasta that cost three times as much probably isn’t three times better. But if you really want that 5% improvement in quality, you can get it by paying three times the price.
Car registration effectively works this way in a lot of places, with age and value being a substantial factor to registration fees. The idea being that the rich guy driving a brand new Audi every couple years pays a lot more for his license plate than the person driving a 20yo Corolla.
I guess they could read the plate and charge a toll based on the car, and in no-stop tollways that already rely on plate scanners or transponders, it’d not be that hard to enforce.
Yes there is a simple way to do this that has universal applicability: users get charged equally, per use, the total revenues are uniformly refunded to everyone (not just users) and the refund is taxable income. At least, this is how it fits best with an American tax system.
Think about how this would work for a gas/carbon tax, for example.
Flattening economic inequality should probably not happen at the level of individual places where people spend money, but rather as a general principle across all of society. Most of the economic thought around capitalism assumes that actors have roughly equivalent economic power, in order to get the most benefits of effective capital allocation. E.g. If only a few, select people are allowed to launch new firms because they were born into the right amount of wealth, then we are greatly limiting our ability to improve the economy.
So instead these sorts of equalizations should happen with high progressive taxation and redistribution through something like UBI. In particular, taxing economic rents is something that economists of all political stripes agree is good. And many vast individual fortunes are built on economic rents rather than productive economic activity.
Or you take year's congestion charge revenue and split it equally across the residents.
The goal should be shifting traffic away from inefficient uses of the infrastructure's capacity.
That should provide the economics for actual ride sharing (I don't mean single-passenger taxi rides, but car pooling and busses) which would allow dynamic networks where passengers need to switch between vehicles on low-throughout links. Think a hybrid between taxi service and bus lines.
Such a network would be able to use incremental addition of fixed line public transit without today's issue of no car-grade last-mile solution for commuter rail services.
Also, public transit gets better when more people use it, due to e.g. busses needing to run more often for capacity and thus automatically/accidentally providing nice latency.
That's illogical. If there is congestion the problem is that the method of transportation is too inefficient. Smaller vehicles like bicycles are more efficient at slow speeds, during congestion speed is low to begin with. The right answer is to tax inefficient vehicles more and efficient vehicles less.
This, absolutely. The moral failing of neoliberalism and neoliberal solutions (e.g., externality pricing) is that it requires an honest belief that a $1 delta means the same thing to everyone everywhere, which is just absurd.
This is a generic problem that applies to a number of scenarios, including traffic congestion.
If there are fewer resources than there are users, a queue forms. People in the queue can be prioritised by sorting the queue by some agreed property and then admitting the people from the head of the queue as resources become available.
Sorting by the willingness and ability to pay in money is a very common property. Housing is notably discriminated by the money people can put on the table. Wealthy get the best spots, poorer people have to live further away in less desirable spots. Could this apply to traffic congestion as well? People with money drive when they want, poorer people drive when they can afford it. Money does a good job in discriminating between people in a huge number of services, possessions, and abilities.
If that sounds unfair somehow, I theoretically like the idea of a time tax. Everyone has the same amount of time each day regardless of if they make a million each day or over their whole lifetime. It's all very equal and it's up to each individual how they choose to spend those hours, whether it's waiting in traffic or doing something that is more profitable than car trips.
The way to implement this is curb down road capacity which also has the benefit of allocating less space for roads and more space for people. Useless trips get cancelled and transferable trips are moved to less congested times, and only if your trip is really important it's worth the wait in congestion. In many ways this is a more equal way of distributing access to limited resources. On the other hand, it's not very useful for anyone to sit in traffic whether you're poor or rich. Especially people whose time cost is close to zero could sit all their day in traffic if they wanted to while people whose time is very profitable couldn't do anything extra, couldn't pay anything extra or make any amends anywhere to get through their trip faster. This may be more equal among individuals but not necessarily the best use of time collectively, to the society as whole.
In communist countries short supply and high demand were resolved by queueing, to not much benefit to anyone. In capitalist countries the rich can pay their way through anything while the poor can't often pay anything, much less their way. What other methods are there for sorting the queue to a limited resource that are used in other domains of other societies?
On the other hand building large high-capacity road networks that are free to use is akin to socialism: there's supply without a distinct price which drives endless demand. Roads should be considered a scarce resource because they take space away from housing and business which are useful activies in a society unlike traffic which is just a means to access what is useful. Thus, the cost of using roads should at least mimick market pricing so that money will drive people to make educated, smart tradeoffs regarding the use of those resources and their time. And building more roads and wider roads in the first place should be considered in the light of alternative costs, not as a non-negotiable necessity.
Congestion pricing (can) increase inequality, favoring the people able/willing to pay: without it, everyone 'pays' in time delays. This is related to recent discussion on HN on scaling fines based on income or wealth:
https://news.ycombinator.com/item?id=31176458
Words are funny. I'd argue it favours those able to pay. I think "willing" is the wrong word because people aren't transporting themselves for fun, most of the time. Personal transport is a rather inflexible demand.
And those able to pay are, in our society, generally the rich.
Willing and able to pay is the most accurate description. Typically decisions to purchase are described as constrained optimization problems with the constraint typically being a budget constraint.
Principles of congestion pricing were applied rather successfully in electricity markets.
Having public congestion prices for electricity makes it easier for investor to understand what needs funding ; do we need more power plants, or more electric lines?
I whish these principles could be applied elsewhere, since ''congestion pricing'' often reflect the need for better systemic public infrastructure (aka a network or a market maintenance cost), while ''regular prices'' are often tied to private business (aka a node, or an entrepreneur).
I’m interested in the studies that will come out using pandemic traffic data and public surveys. There’s been a dramatic shift in commuting from 3 years ago and I wonder if we’ll return to the same sort of gridlock.
Some questions:
If companies broadly reject work from home policies would they also reject peak shifting policies?
Are workers interested in returning to offices and then working odd hours?
If everyone adopts hybrid schedules does it just drop white collar commuters by 40%?
If traffic stays light and downtowns are empty, which transit or highway projects are now needed or may no longer be needed?
> There’s been a dramatic shift in commuting from 3 years ago and I wonder if we’ll return to the same sort of gridlock.
AFAICT, we largely have in my area. The durable changes seems to for a very narrow slice of the workforce, which I am fortunate enough to be in, but whenever I have a (non-commuting) reason to drive in normal commute hours, it already looks pretty similar to pre-pandemic, even while the process of orgs announcing rollbacks of regular telecommuting continues.
I share your interest in these studies, and was quite hopeful to see traffic go down in my neck of the woods. Didn't happen.
> If companies broadly reject work from home policies would they also reject peak shifting policies?
This has actually been "recommended" by the French government after the first lockdown ended. Of course, I haven't actually measured this, but my subjective impression is that basically nobody applied this, not even in public transport.
During peak hour trains were still very full, despite some employers allowing partial remote working.
The issue, at least is France, is that the workday is basically 9 AM to 6 PM. Add to that an hour of commute and stores outside big cities closing around 8 PM (ie where people have long commutes), and people can't shift the work day later. They could shift it earlier, but they usually don't like getting up early, and there also seems to be, to some, some kind of stigma attached to people leaving "early".
36 comments
[ 2.7 ms ] story [ 86.3 ms ] threadhttps://www.rand.org/content/dam/rand/pubs/technical_reports...
> First, depending on how congestion pricing is implemented, it can be either regressive or progressive. This depends in large part on how toll revenues are used. For instance, if regions spend revenues in ways that benefit low-income individuals, congestion pricing is more likely to be progressive. However, if regions use revenues in a way that benefits all individuals equally, congestion pricing may be, overall, regressive. This is the strongest finding in the economic literature.
> Second, even when low-income and other transportation-disadvantaged groups benefit as a whole from congestion pricing, it is very likely that some individuals will still be worse off.
> Third, for all forms of congestion pricing (but more for some than for others), the distri- bution of residents and job opportunities (not to mention shopping, schools, places of worship, and other important destinations) has a large impact on the equity implications.
> Fourth, high-occupancy toll (HOT) lanes, the most common form of congestion pricing in the United States, tend to raise fewer equity concerns among motorists, since they provide drivers with an additional choice of using a set of priced lanes while allowing them to continue using parallel, free lanes if they prefer.
> Fifth, while congestion pricing has been shown to reduce emissions in general, there is scant evidence showing that congestion pricing can specifically reduce negative environmental consequences for neighborhoods disproportionately affected by emissions.
> Given the risk of negative impacts to low-income and other groups under congestion pric- ing, we looked at suggested ways to diminish these impacts. Two mechanisms are in common use: (1) revenue redistribution and (2) discounts and exemptions.
Unless there is some sort of exemption it effectively becomes “ban poors from cars” - which may entirely be the desired result.
If we want to help poor people we should just give them money rather than indirect subsidies that mainly favor wealthier people with cars.
Unless the city was really built around transit (and not just once developed in a hub and spoke like so many cities) it’s hard to beat a car door to door. That’s what we should be focusing on fixing (things like inter borough travel in NYC). Massive congestion charges might be part of that, however.
The effect is that the mechanism prevents significant latency degradation from demand.
While it's not necessarily possible to go that far over night, looking at how much of the traffic/congestion is from cars, and how many occupants they have, should allow to calculate what reduction perfect car pooling (with, say, 4 people per car) would allow, and at what speed that reduced number of cars could drive (without tailgating). Then, add some safety margin and set that reduced speed as the target for the auto-toll-pricing.
If car pooling physically can't fix the congestion, it'd take a bit for the problematic roads to get enough (mini)bus service to enable the full design speed.
WFH isn't the solution, and proper mass transit takes a while (5-10 years) to build even if you're in a hurry.
I feel like the way to maximize revenue would just be to try either raising or lowering the price (by say 20%?) every n (5-10?) minutes, and see if your revenue goes up or down. If it goes up, continue what you’re doing, and if it goes down, reverse it?
If you go for revenue maximisation you won't be setting prices at the social welfare maximising level. Also, you'd probably want to charge higher prices for heavier vehicles as they cause more road damage per unit of distance travelled.
This sounds really clever but I don't get it. Do you have a link to a more thorough explanation?
The goal is to tip the balance from “30 min drive vs 1 hour transit” to “30 min drive + 10€ vs 1 hour transit” and if the equation keeps changing people will feel trapped.
I think we need to find a way to "charge" people in a way that matters equally to everyone. I usually advocate for things in terms of time (community service hours) because that perfectly hedges against the person's wage level. Although I suppose given their tax receipts we could just adjust the monetary charge for their observed hourly wage.
Any other ideas for adjusting "charges" to have equal importance for different people?
Why does scarce road space in highly-congested (and well-connected) CBDs require a sliding scale when little else does?
But we do! When I was a poor student, I usually bought the bottom-shelf store-brand cheap pasta. Today, I look at a higher shelf and often get the organic high-protein stuff with fancier shapes.
It costs about the same per unit weight to make, but the latter costs much more to purchase. This is exactly a sliding scale that allows people who are more price sensitive to buy practically the same stuff except at a much lower price.
(Why do companies do this? It lets them expand their target market without getting total profits too close to the floor.)
----
This is also why we have region locks on electronics and export controls on medicine: companies are selling literally the exact same stuff with different profit margins in different regions, based on their collective disposable income.
Also it greatly increases enforcement costs so not only would it have to be better, it would have to be a lot better.
Moreover, I know a lot of very wealthy people that are very frugal and actually care more about $10 than I do.
I guess they could read the plate and charge a toll based on the car, and in no-stop tollways that already rely on plate scanners or transponders, it’d not be that hard to enforce.
Think about how this would work for a gas/carbon tax, for example.
So instead these sorts of equalizations should happen with high progressive taxation and redistribution through something like UBI. In particular, taxing economic rents is something that economists of all political stripes agree is good. And many vast individual fortunes are built on economic rents rather than productive economic activity.
The goal should be shifting traffic away from inefficient uses of the infrastructure's capacity.
That should provide the economics for actual ride sharing (I don't mean single-passenger taxi rides, but car pooling and busses) which would allow dynamic networks where passengers need to switch between vehicles on low-throughout links. Think a hybrid between taxi service and bus lines.
Such a network would be able to use incremental addition of fixed line public transit without today's issue of no car-grade last-mile solution for commuter rail services.
Also, public transit gets better when more people use it, due to e.g. busses needing to run more often for capacity and thus automatically/accidentally providing nice latency.
If there are fewer resources than there are users, a queue forms. People in the queue can be prioritised by sorting the queue by some agreed property and then admitting the people from the head of the queue as resources become available.
Sorting by the willingness and ability to pay in money is a very common property. Housing is notably discriminated by the money people can put on the table. Wealthy get the best spots, poorer people have to live further away in less desirable spots. Could this apply to traffic congestion as well? People with money drive when they want, poorer people drive when they can afford it. Money does a good job in discriminating between people in a huge number of services, possessions, and abilities.
If that sounds unfair somehow, I theoretically like the idea of a time tax. Everyone has the same amount of time each day regardless of if they make a million each day or over their whole lifetime. It's all very equal and it's up to each individual how they choose to spend those hours, whether it's waiting in traffic or doing something that is more profitable than car trips.
The way to implement this is curb down road capacity which also has the benefit of allocating less space for roads and more space for people. Useless trips get cancelled and transferable trips are moved to less congested times, and only if your trip is really important it's worth the wait in congestion. In many ways this is a more equal way of distributing access to limited resources. On the other hand, it's not very useful for anyone to sit in traffic whether you're poor or rich. Especially people whose time cost is close to zero could sit all their day in traffic if they wanted to while people whose time is very profitable couldn't do anything extra, couldn't pay anything extra or make any amends anywhere to get through their trip faster. This may be more equal among individuals but not necessarily the best use of time collectively, to the society as whole.
In communist countries short supply and high demand were resolved by queueing, to not much benefit to anyone. In capitalist countries the rich can pay their way through anything while the poor can't often pay anything, much less their way. What other methods are there for sorting the queue to a limited resource that are used in other domains of other societies?
On the other hand building large high-capacity road networks that are free to use is akin to socialism: there's supply without a distinct price which drives endless demand. Roads should be considered a scarce resource because they take space away from housing and business which are useful activies in a society unlike traffic which is just a means to access what is useful. Thus, the cost of using roads should at least mimick market pricing so that money will drive people to make educated, smart tradeoffs regarding the use of those resources and their time. And building more roads and wider roads in the first place should be considered in the light of alternative costs, not as a non-negotiable necessity.
And those able to pay are, in our society, generally the rich.
Having public congestion prices for electricity makes it easier for investor to understand what needs funding ; do we need more power plants, or more electric lines?
I whish these principles could be applied elsewhere, since ''congestion pricing'' often reflect the need for better systemic public infrastructure (aka a network or a market maintenance cost), while ''regular prices'' are often tied to private business (aka a node, or an entrepreneur).
Some questions:
If companies broadly reject work from home policies would they also reject peak shifting policies?
Are workers interested in returning to offices and then working odd hours?
If everyone adopts hybrid schedules does it just drop white collar commuters by 40%?
If traffic stays light and downtowns are empty, which transit or highway projects are now needed or may no longer be needed?
AFAICT, we largely have in my area. The durable changes seems to for a very narrow slice of the workforce, which I am fortunate enough to be in, but whenever I have a (non-commuting) reason to drive in normal commute hours, it already looks pretty similar to pre-pandemic, even while the process of orgs announcing rollbacks of regular telecommuting continues.
> If companies broadly reject work from home policies would they also reject peak shifting policies?
This has actually been "recommended" by the French government after the first lockdown ended. Of course, I haven't actually measured this, but my subjective impression is that basically nobody applied this, not even in public transport.
During peak hour trains were still very full, despite some employers allowing partial remote working.
The issue, at least is France, is that the workday is basically 9 AM to 6 PM. Add to that an hour of commute and stores outside big cities closing around 8 PM (ie where people have long commutes), and people can't shift the work day later. They could shift it earlier, but they usually don't like getting up early, and there also seems to be, to some, some kind of stigma attached to people leaving "early".