Hacker News new | past | comments | ask | show | jobs | submit login
Taxing Uber and Lyft rides is L.A's latest plan to free up congested roads (latimes.com)
172 points by prostoalex 56 days ago | hide | past | web | favorite | 314 comments



> Uber and Lyft “are using public roads, and the profit is going to their companies,” Phil Washington, Metro’s chief executive, said at a recent meeting.

Public roads are supposed to be paid for through fuel taxes, which every Uber and Lyft driver pays. California has gasoline taxes of:

* 55.22c per gallon (second only to Pennsylvania)

* 18.4c per gallon (federal)

* 2.25% sales tax

Additionally, taxing only ridesharing companies may increase incidences of drunk driving. Is another, poorly conceived, tax really a solution?

[0] https://en.wikipedia.org/wiki/Fuel_taxes_in_the_United_State...


Fuel tax revenue and other direct user fees like annual registration in California amount to less than 25% of what the state and its cities pay to maintain the roads and streets. Roads are massively subsidized from general revenues. If you are proposing a $5/gallon fuel tax then I completely agree. That would be about equivalent to what other rich countries tax motor fuel and consistent with the state's greenhouse gas emission goals.


That's misleading -- yes, in the aggregate, gas taxes cover only part of the expenses, but that's not because of personal cars. Damage on the roads -- and thus maintenance costs -- scales with the 4th power of weight per axle. So the damage is virtually all from large trucks, with sedans and SUVs as a rounding error.

If you were to apportion out the maintenance cost due to personal cars, they are overpaying.


For those wondering about the source of the "4th power of weight per axle", I found this after some short, but intense, Googling:

http://co-asphalt.com/wp-content/uploads/2015/03/TrashTruckD...

Reference 1 lead me to this: (Page I-11) https://habib00ugm.files.wordpress.com/2010/05/aashto1993.pd...

From that: "The load equivalency factor increases approximately as a function of the ratio of any given axle load to the standard 18 kip single axle load raised to the fourth power."

That is also cited here, on page 14: https://wisconsindot.gov/Documents/dmv/agri-eq-veh/ag-study/...


I am bad at math... could you help me understand. A 80,000 truck and 5,000 car drive down the road... how much more damage does the 80,000 truck do? Thank you.


If the 80k truck has 5 axles (as is typical of an 18 wheeler) and a 5k car has two, then the truck has 16k on each axle and the car has 2.5k on each axle.

Because the damage scales with the 4th power, then the damage ratio is the 4th power of the axle weight ratio:

(16k/2.5k) ^ 4 = 1677x as much damage.

Edit: Sorry, that oversimplifies it a bit. For the same vehicle, the damage scales that way, but for a fair comparison, you'd have to account for the fact that the weight is distributed over more and bigger tires on each axle. Still, that should show you how the math works.


So the damage is virtually all from large trucks, with sedans and SUVs as a rounding error.

Which is why the California tax on gasoline is 48.7¢/gallon and the California tax on diesel fuel is 67¢/gallon.[1]

Federal tax on gasoline is 18.4¢/gallon, and diesel fuel is 24.4¢/gallon.[2]

I'm not defending the trucking industry. I think more stuff should go by train instead. But truck fuel is taxed more heavily than car fuel. (For those of you in other countries, the number of cars running on diesel in the United States is about 3%.)

[1] http://www.cdtfa.ca.gov/formspubs/l561.pdf

[2] https://en.wikipedia.org/wiki/Fuel_taxes_in_the_United_State...


It’s taxed slightly more heavily. But trucks cause literally thousands of times more damage than cars.


And you do get a better MPG with Diesel so it should to be fair be taxed more.


They also require more gas per mile. It seems like it should roughly add up to paying a few more times than cars (they're a few times less fuel efficient).


Yes, the average car in the US gets about 25MPG, while the average tractor trailer is 6MPG, so about 4x the fuel per mile. Accounting for the higher tax rate, they’re paying maybe 5x the tax per mile, while causing maybe 5,000x the damage.


You could tax trucks like 1000 times more. But don't expect the butter and milk to stay the same price.


Butter and milk that don't have to travel as far by road would be cheaper. So much of our national Walmartification is possible because of roads are subsidized.


No, it would not be cheaper. It is not cheaper today and making food today is profitable (and possible) only thanks to economy of scale. Automation on this level is not here yet.

> don't have to travel as far by road

It doesn't have to. It can (and it does because it's better).


We’re paying for it either way. I’d rather the payment be apportioned by the extent to which you use products that have to be transported that way; it gets the incentives right.


The milk would cost more, but we'd save even more money on road maintenance.

Then many trucks could be adjusted to have more axles and lighter loads, saving money on both ends.

The end result is a better system.


Is this some kind of joke? If the taxes already cover 25% of the expenses then adding 1000x on top means you would tax 249 times more than you need. To make it fair you first have to give truck fuel an adequate tax (so it covers 100%) and then divide this tax by 1000 for light vehicles.


If you think it's a joke you're taking the words far too literally.

"like 1000 times more" is not a full proposal. Yes you would reduce car taxes appropriately. Yes you would figure out the actual amount to balance the budget.


Maybe eventually the cost will go down for milk because we chosen more efficient ways of transporting milk?


> The milk would cost more, but we'd save even more money on road maintenance.

This would adversely affect the poor as they drive less and still need milk.


literally anything that has any negative effect will affect the poor more because the rich have more resources to mitigate the negative effects.

so unless you have a solution with no drawbacks, there's nothing you can do to not 'affect the poors'


There are much better ways to help the poor than by subsidizing trucking.


My point is you can't just look at something and not take into account where else it may impact.

Also, we need to get much better at helping the poor.


My point is that “it will hurt the poor” is a very old and very tired response to proposals that externalities should be accounted for in the cost of goods.

Part of getting better at helping the poor is helping them directly rather than using extremely inefficient and indirect subsidies on various industries to do so.


Externalities that aren't properly accounted for (and paid for) end up causing inefficiency.


The point may be that they should not be this price to begin with.


The reduction in sales taxes would more than make up for it.


Those taxes should be proportional to the wear, otherwise it remains unbalanced.


If we are talking about maintenance costs, sure trucks should pay their share, but congestion (the reason to build additional roads) is caused primarily by passenger cars.


Congestion doesn't drive maintenance costs. It may drive infrastructure needs but that is an entirely different problem, and one which taxing conpanies such as Uber and Lyft actually help to lower.


What is more expensive, train or road maintenance? What gets better total utilization rates?


Road gets you everywhere, rail only gets you to the nearest terminal. Depending on the signalling system used rail can have a higher capacity from terminal to terminal, leaving the 'last mile' (which in reality is more likely to be the last 100 miles) to be travelled by road. Road needs more maintenance and the work itself is more labour-intensive.

Imagine a package-switching network comprising an (electrified) rail backbone with roads spreading from terminal nodes to the destination points, fully automated package store-and-forward, electric self-driving trucks.

This would cost loads of money and loads of jobs. It might also solve parts of the problem. Maybe Amazon's successor is already thinking of something like this?


> Road gets you everywhere, rail only gets you to the nearest terminal.

That is not true. Road gets you to where the road serves, just like rail. With a good enough coverage, virtually all transportation needs are met.


Exactly this. Of course, if they tax per axle and tonnage combination, it’ll just be passed on to consumers anyhow as most of trucks are distributing goods which directly or indirectly end up in the hands of consumers. However, it would be quite fair to do, I think.


In the short term it would get passed on to consumers. In the medium term supply chains would figure out how to reduce shipping by road. In the longer term (in conjunction with other changes) we might be able to get larger-scale urban development on a more sustainable track.

If we could properly price externalities (not just road wear but also ground water contamination, CO2 emissions etc.) there are many parts of our economy which could be made dramatically more efficient and less harmful without that much overall change to human quality of life.


This would devastate the economy.

For example: Exxon mobil profits in 2017 were 19.7 billion. Fukushima cleanup costs exceed 180 billion.

There is no way to do what you are proposing without massive price increases in electricity, gasoline and virtually every other good. I wouldn't be surprised if when you priced all of this in you discovered that people couldn't really afford electricity anymore.

Also this economy would never be able to have exports because the price of producing goods would just be insane.

The truth is we can't afford the externalities without an economic implosion that would realistically end with an extreme political group taking control in the ensuing chaos and undoing everything you are proposing.


Mitigating climate change is essentially going to cost all of the direct profit ever made by all the companies in the fossil fuel discovery, extraction, refining, and retailing businesses.

Think 85 trillion USD, real value. That's about 1/3rd of the total current wealth of the entire world, by my back of the envelope calculations.


> For example: Exxon mobil profits in 2017 were 19.7 billion. Fukushima cleanup costs exceed 180 billion.

So you could cleanup one of the worst "spills" in history with a mere 10 years of a single company's profits?

That sounds perfectly OK.


Even better, it would have been possible to avoid that spill with that money and you'd probably have enough left over for a healthy profit afterwards.


The thing about externalities is: somebody pays, it's just not the person who caused them. Redirecting that cost back to the source would be a good thing. There would then be incentives to do the most efficient thing instead of the most wasteful thing. This would actually make everybody richer.

How much of our wealth is being wasted this way? I'm guessing that it's a lot.


Sure, but it might also have some other nice side-benefits, like favouring railroad expansion. Having transport by lorry being artificially cheaper is non-ideal for lots of reasons.


Actually, of trucks were taxed on a ton/axle basis instead of ton+axle basis, the net road wear would be dramatically reduced. Think 6-axle, tripple-tire trailers.


It's not that simple. HN readers could absorb the increase, but the poor can't.


That's not really the view from a city budget. Cities don't have that many truck routes and their costs are dominated by maintaining streets where virtually all the traffic is light vehicles. For the state it is a bigger problem because almost all state route are truck routes.


Are you sure about that? I see garbage trucks, delivery trucks (package, furniture, appliance etc.), lawn service trucks, salt trucks, snow plowing trucks etc day-in, day-out.[1] And to do their job, they probably need traverse 95% of the paved roadways on a regular basis. Granted, you don't typically see 18-wheelers tooling around most neighborhoods[2] but there is quite a bit of truck traffic through even moderately populated urban and suburban areas every day.

[1] If your argument is that these are all 'light' vehicles, my city disagrees with you. They recently cited road damage from the variety of garbage hauling services as a reason to limit our options. Not that I entirely buy that as the only or even main reason they did so... local politics and all that.

[2] However they have been known to use secondary streets when they shouldn't.


We're talking about California. There are no major cities in California in possession of snow plows and salt trucks. A garbage truck goes down the street once per week, yes. The overwhelming majority of traffic on most city streets is cars and light trucks. The main cause of damage to most city streets is the mere passage of time, and rain.


We are talking about California. What rain? :-P


Thankfully CA is now finally pretty well drought free, after getting pretty decent rain the past couple months. https://www.sacbee.com/latest-news/article226462260.html

Drought map https://droughtmonitor.unl.edu/CurrentMap/StateDroughtMonito...

Reservoirs are doing well http://cdec.water.ca.gov/reportapp/javareports?name=rescond....

Of course, this is an El Niño year, so we’ll see whether we return to severe drought in the coming years.


The salt and plowing trucks are part of the maintenance of the road needed for the cars. You defeat your own point here.


Not at all. I was making the point that there are lots of trucks on most roads all the time, not that they aren't necessary.

Also, while salt makes the road drive-able in the winter, it actually does long term damage to the road (increases the number of freeze/thaw cycles.) Similar with plows: they increase the rate at which the road surface deteriorates. While they both are necessary to keep roads operable for vehicles in the winter, they are the opposite of maintenance longer term as far as the road surface is concerned.


Even then, the same phenomenon means that the dominant source of damage to those streets is the delivery vehicles; smaller cars still cause a disproportionately small fraction of the damage and are overpaying via taxes. Ubers are not externalizing here.


How does this account for all the upkeep on the residential roads where a large truck almost never travels. Sure, large trucks may account for the vast majority of wear on roads where they travel, but I imagine the majority of the roads in a city rarely see large trucks, and still need regular upkeep.

Without some sources to back up your claims, I'm not sure if it's accounting for that or not. Can you supply your sources so we can look into exactly what that statistic means?


Without much better data the best you can say is that Ubers are benefitting from externalities at a rate lower than other, heavier vehicles. It's still likely they aren't paying their fair share.


A vehicle twice as heavy per axle is doing 16x the damage. Unless the taxes on delivery vehicles scale like that, a sedan is subsidizing the costs.


A road that's never driven on suffers damage from the elements, and the passage of time.

A truck does more damage then a car? Great. But there's an upfront road maintenance (not to mention construction) cost that you have to pay, regardless of who uses the road. Gas taxes do not come close to capturing that.


This is why the Netherlands also has weight-based vehicle taxes (on top of the gas tax and many other taxes).


As do the states. Commercial trucks pay a lot more road use tax than passenger cars. Whether it's proportional to the damage they cause I don't know.


There are counter examples like the federal and some state (e.g. California) tax breaks for trucks over 6,000 pounds, providing relief to all of those orthodontists who needed a Hummer to get to the office.


I'm waiting for people to buy the actual military surplus Hummers for all those Whole Foods that are on top of mountains.


Damage isn't the only thing involved in people using the roads. Roads are a public good (in contrast to railroad tracks which are a private good).

The state effectively charges for the use of roads both to pay for maintenance and to make sure that the road remain usable to a reasonable cross-section of people. This is just as landlords charge for the use of land even when that land isn't consumed.


Here in northern Sweden we have a lot of wear on the roads. After a few years you can clearly see where most drive because there is a deep depression after the wheels. If you drive a truck in the track you will notice that you don't fit, the distance is slightly smaller than the trucks wheel distance. It does however fit exactly for the cars.

We do however all have spikes on the cars in winter, I'm guessting this helps a lot. The trucks usually does not have spikes. There is a lot of trucks too, hauling wood out of the forests.


Worth noting that the damage to the air, increase in noise pollution, and increased costs of health care spending from injuries, lung disease, etc. from both is substantial.


The more technical term for this is "fatigue damage". I read years ago that the fatigue damage inflicted by a truck loaded to the legal weight limit (and it's often exceeded) is 9,000 times that of a car.

The damage done by cars is usually from snow tires.


And add to that pollution caused by large trucks. Some are literally hard to follow behind due to the cloud of poison they are fuming all over.

California's crusade on sedans to cut down on pollution is a joke.


You are joking right? Just because the pollution from cars is not visible, doesn't mean it doesn't exist. And there are a 1000x more cars than trucks.


No need to use the 1000x number, there's actually information about this out there, at least when it comes to greenhouse gasses. In 2016:

https://nepis.epa.gov/Exe/ZyPDF.cgi?Dockey=P100USI5.pdf

Medium and Heavy duty trucks: 425 Tg CO2

Light-Duty Vehicles (passenger cars + SUVs, pickup trucks, etc): 1,106 Tg CO2


Are there studies for how these measure up in practice? Also curious about wear per weight characteristics between concrete and asphalt.


Damage on the roads -- and thus maintenance costs -- scales with the 4th power of weight per axle.

[citation needed]


Except there are a 1000 times more cars on the road than trucks, and we actually need trucks. We don't need cars, at all.


Fuel tax revenue and other direct user fees like annual registration in California amount to less than 25% of what the state and its cities pay to maintain the roads and streets. Roads are massively subsidized from general revenues.

Which is why I always laugh when I hear people complaining about evil subsidized high-speed rail in America. That's just talking points from the airline and trucking industries, which are already far more heavily subsidized than rail.


The problem with the US and high-speed rail isn't on-going subsidization per se, it's the comically enormous cost to initially build it out. Major infrastructure is wildly expensive to build in much of the US. To do national high-speed rail, you'd have to figure that out. The only solution I've ever seen that might work, would be a belligerent Federal Government that just brutalizes anyone that gets in the way, threatening uncooperative states via funding, aggressively using eminent domain to take land, bulldozing through environment reviews, etc. You'd have to use the full power of the Federal Government to rapidly get through obstacles and it would piss off everyone in the process.

There's only one major state where you can build cost-effective high-speed rail in the US - Texas. They're going to do what California couldn't and at ~1/4th the cost per km of rail. 20-30 years from now Texas will probably have high-speed rail between all of its major cities.

https://www.houstonchronicle.com/neighborhood/spring/news/ar...


Florida already did - gobrightline.com. Private rail linking west palm beach with Miami, expansion plans to Orlando.


If the usage fees aren't sufficient, then that's true for all drivers, whether it's Uber, or an eighteen wheeler, or a personal driver. If you get those usage fees right, then the argument that "specific group X profits from using public roads" goes away.


Do you not have a road licence tax for all vehicles with trucks and HGV's charged a lot more (as they do the moast damage) as we do in the UK.


Those roads are what allow stuff to be delivered. How do you ship an air conditioning unit on a bike path? The point is that the roads serve commerce and a tax on roads is really a tax on everything. Roads aren’t exactly massively subsidized — the commerce they generate results in taxes. Tax revenue generated per mile driven is far greater than any supposed subsidy. We could make an argument that schools are massively subsidized — the parents of students attending don’t pay the actual cost of attendance however, the tax revenues generated by an educated workforce are considered a net win compared to the cost of the “subsidy.” My kids don’t go to public school and some people don’t drive on roads, but we all benefit from roads and other people’s kids being educated and the economic gain vastly outweighs the subsidy. Not everyone uses the roads, but everyone benefits from them — thus a direct user-fee really serves no purpose, it’s just an additional tax on everyone. Since practically every single product touches a road at some point, taxing the roads more just makes everything more expensive.

You are also incorrect suggesting that $5 per gallon is “about equivalent to what other rich countries tax motor fuel.” In France, it’s $2.77 per gallon. In the UK, it’s $2.83 per gallon. Netherlands is $3.5 per gallon. Sweden is $3.51. In Australia, it’s $1.12 per gallon, New Zealand is $1.58. None of those countries are even close to $5 per gallon.


Taxing vehicle use reduces vehicle use. That's quite important if you want to do something about global warming. Gasoline in Berlin costs $5.75 but I can still order stuff from Amazon.

As for transporting air conditioning units, trailers and cargo bikes are a thing. You can move a refrigerator by bike

https://www.bikesatwork.com/blog/moving-a-refrigerator-by-bi...

Electric assist bicycles can transport even more cargo.


"You are also incorrect suggesting that $5 per gallon is “about equivalent to what other rich countries tax motor fuel.” In France, it’s $2.77 per gallon. In the UK, it’s $2.83 per gallon. Netherlands is $3.5 per gallon. Sweden is $3.51. In Australia, it’s $1.12 per gallon, New Zealand is $1.58. None of those countries are even close to $5 per gallon."

This is incorrect, in the UK at least. The best price I get is about £1.21 per liter. Which is $1.61 per liter Or $7.31 per Gallon.

So we are well over $5 a gallon. Don't forget the USA is pretty much the only country to measure liquid volume in gallons, we all use liters - 1L = 4.54609 Gallons


I think you have liters and gallons backwards, unless the British gallon had shrank significantly with the Brexit.


umm, nope British Gallon = 4.546090L US Gallon = 3.785412 https://en.wikipedia.org/wiki/Gallon

Though I had forgotten that the US used a different one. So my maths is off above. Thanks for pointing that out!

Using the US Gallon: £1.21 per liter == $1.61 per liter == $6.09 per Gallon.


I meant that "1L = 4.54609 Gallons " seems to be off a bit.


> How do you ship an air conditioning unit on a bike path?

Use a trailer? While not possible for everything, we could easily do with 1/10000 of the cars we have.


You can always rent a van for the couple hours you need it [1]

I still regret having bought an expensive percussion drill to install a bathroom vanity, when I could have just rented one from the local hardware store chain. [2]

[1] https://www.greenwheels.com/nl/prive/jouw-autos

[2] https://www.praxis.nl/service/gereedschap-verhuur


Am Dutch, just looked out the window at what is considered a gas station with cheap prices (not next to a highway) and gas is €1.559/l, diesel is €1.319/l.

This comes to $6.72 per US gallon for gas. More than half of this price is VAT and duty. On top of this we pay road taxes based on the weight and fuel type of your vehicle, about €50/month for a small (1000kg) gas powered car.


Gas in Switzerland is currently about $9 USD per gallon (CHF 2.27 per liter), chiefly in taxes.


Correct, if the price in the US is $3 a gallon and the price in Europe is $8-9 a gallon then there must be around 5-6 of additional taxes in Europe.


What's your average distance driven per year? In the US it's 13,476 miles (21k km).


So because you drive longer, which of course produce more Co2, you should have a lower price? l/100km or mpg is not really the most important, what count is l or g per year. Last time gas prices where high sales of SUVs and trucks in US did go down, as soon as the gas price did go down sales of cars with terrible mpg ratio did go up


In Finland, the price is currently about $6 per gallon. Average car drives about 10500 miles per year & there's about 0.49 cars per person.


$5 per gallon is roughly in line with the tax in Europe.

Compare the UK and the USA, conveniently adjacent in the table. Both have refineries, so the untaxed price should be similar, but the UK taxed price is $4 higher.

https://en.wikipedia.org/wiki/Gasoline_and_diesel_usage_and_...


Keep in mind there is also 20-25% VAT on those European prices. Which is on the total cost including the cost of the fuel and any duties.

It definitely adds up to around £1 a litre in the UK of tax (59p duty plus 30p VAT) depending on exact wholesale prices. This works out at around $5 per (US) gallon.


This seems to be a recurring problem. We have all these subsidies and taxes that penalize or prioritize certain kinds of actions instead of charging everything at its true cost. This means the government is forcing a certain kind of solution to the problem rather than charging the problem makers the cost of their problem.

A better solution would be to charge all drivers the exact cost to run their cars including all road maintenance for the roads they use and the cost to the health system and environment that their cars create.

Then provide multiple options such as rail and bike lanes. People will naturally gravitate to the best solution.

I often see complaints about how this will raise the cost of products like food because they need a lot of road and fuel resources. But the thing is you already pay for this cost through other means. By charging the true cost the price of an apple may go up but the price in your tax goes down. And this way the seller of the apple is now incentivized to source the apple locally and reduce the transport costs because its no longer paid for by everyone.


This is very hard to implement; How do you amortize startup costs (building the roads, buying new paving trucks) vs maintenance costs (raw cement, human, cost @ year 1, cost @ year 10, cost to replace the system after X years).

Do you penalize users/tax payers who have a lower use and less scaled systems? (e.g. should it cost more to mail my grandmother a letter because her post office is less at scale than mine?)

Do you increase the costs if some of your users leave (and risk even more attrition)?

This has been investigated in depth with schools and local businesses. Gov gives them a sweetheart deal with the hope that they contribute back and as the deal get's less good (and Gov starts ramping taxes to deal with paying infrastructure costs) businesses leave. For me the natural conclusion is the government has to provide some balancing (rain day funds, progressive taxes, ...)


Many Uber/Lyft/taxis cars are fuel sipping Toyota Priuses. Fuel taxes break down as fuel economies reach a certain point. Yes, it is not just Uber drivers using Priuses, but it is a substantial problem that has increased road usage without the same corresponding increases in fuel tax revenues.

Of course electric presents even bigger problems that many states now have a surcharge on them during registration. It is obvious that fuel taxes are going to be outdated in 10 years or so as EVs become more popular. Real-time tolling has been suggested as an alternative to this problem in Europe, but that probably wouldn't fly in the states.


If efficient cars are a problem, tax efficient cars, not Uber. But given CA's environmental goals, this would be counterproductive.


As others have mentioned, taxing efficient cars doesn't seem to make much sense. Maybe tax actual miles traveled instead? That would solve the problem of increasing efficiency and also decrease the value of loitering and cruising in-between pick-ups.


I'm real curious as to how taxing miles travelled would work, especially if you travel between states. Would you have to keep track of how many miles you drove in State A and State B, and how would you pay State B if you lived in State A? Since we'd need to replace the federal fuel tax as well, would the federal government require states to hold yearly odometer inspections, or would you have to fill out a form on your yearly taxes stating how far you drove? I supposed that these questions could be solved by utilizing GPS, but I'm not sure I like the idea of the government tracking the movements of my vehicle.


In Norway the use of GPS has been discussed for a few years, and as far I understand the newest solutions being discussed have been approved by the data privacy watchdog, which generally tends to be very negative to government surveillance.

There are many options there, at the cost of making audits/complaints harder, such as e.g. only keeping GPS traces for a very short time (or not at all), and gradually overwrite more and more details, such as coordinates, reducing time precision, and finally consolidate to amounts owed. Those settings could be left under the control of the car owner as well, or let you dump a signed record of the precise data if you as the owner suspect inaccuracies and want to collect evidence, and then allow you to wipe the details from the device and only leave aggregate numbers.

The point is if the device is tamper-proof enough, you don't need it to record your movements, you just need it to monitor your movements and location with sufficient precision to decide what amounts should be added to running totals of tax.

Of course there is a risk some governments will decide they'd really like more detailed records.


There are experiments with this sort of system in my state in the US. In our case, the car owner signs up with a private firm that does the data collection and only reports the total distance driven to the government, not the actual locations.

I would assume that there is some law which requires the detailed data to be deleted, but I'm not sure about that.


I'm real curious as to how taxing miles travelled would work, especially if you travel between states.

Trucks have been doing it for years with transponders in the cabs. That's how when a semi fills up with fuel in one state and ends its trip in another state the state in between still gets some fuel tax to pay for road maintenance.


That is on the table in Europe with real-time tolling based on miles driven.


It would make sense to use a differentiated sales tax for car for hire services. That'd be proportional with usage. It would also tax people who don't have a car and externalize car taxes and maintainance costs by using these services. EU countries already do ths for things like food, books, accomodation and restaurant services. Usually the sales tax is smaller for these sectors.


If you want usage fees to be proportionate, I think you'd need to tax according to the maintenance costs caused by the vehicle per mile (eighteen wheelers cause vastly more road wear than cars). That's really all you need, except perhaps for some fixed cost per mile for the size of the vehicle, to tax congestion.


You’d also want to discourage congestion with use taxes, as another commons tragedy.


Wait, how would taxing miles traveled decrease the value of loitering between pick-ups?


> If efficient cars are a problem, tax efficient cars

California already does that. (ZEVs pay a supplemental $100 annual fee.)


Yes, I mentioned that is already beginning to happen.

> But given CA's environmental goals, this would be counterproductive.

That is irrelevant, we are reaching a point where taxation on road miles driven vs. fueled used becomes unavoidable. What will wind up happening is that efficient and clean energy vehicles will still need to be taxed to account for usage, while less efficient vehicles will be taxed even more (to discourage use).


Why should less efficient vehicles be taxed more? If they are already paying more for fuel, doesn’t that already “tax” them more? It seems like such a tax is regressive — rich people can buy new efficient cars more easily than poor people.


> Why should less efficient vehicles be taxed more?

Internalizing the environmental externalities of operating them.

> If they are already paying more for fuel, doesn’t that already “tax” them more?

Fuel taxes are a mechanism for the additional (that is, beyond road usage, in the context of the grandparent post) taxes on inefficient vehicles.

> It seems like such a tax is regressive — rich people can buy new efficient cars more easily than poor people.

That's actually a reason you might want a method other than exclusively fuel taxes to implement the additional taxation (which may be via, say, a means tested tax subsidy for purchase of more efficient cars, rather than directly another extra tax on inefficient ones), because fuel taxes are operations taxes but don't directly effect purchases (they may be taken into account, but the poor who feel they need a car will often buy what they can afford up front even if it is not fully rational because real people tend to fall short of rationality in some predictable ways, and discounting deferred costs is a big one.)


> Why should less efficient vehicles be taxed more?

Taxation serves two purposes: (a) to fund roads, and (b) to discourage society damaging behavior. For (a), we need to tax all vehicles, for (b), we need to tax some vehicles more than others.

> It seems like such a tax is regressive — rich people can buy new efficient cars more easily than poor people.

True. Perhaps the efficiency tax can be built into the initial new car purchase while the use tax is ongoing (those old cars will eventually fall apart or fail an emissions test anyways and be taken off the road). Anyways, these things can be worked out.


Many countries (Norway would be an example) have special taxes on cars that takes into account a variety of metrics around pollution, weight and engine capacity, as well as price. Overall car taxes in Norway are among the highest in the world, to the extent that it massively skews which car models people buy.

Here's the Norwegian tax authorities form for calculating taxes. While not all the descriptive text is available in English it seems most of the form is:

https://www.skatteetaten.no/en/person/duties/cars-and-other-...


We’ve had a federal “gas guzzler tax” (or new car sales) since 1978. (USA)


Because you are also emitting more co2. From CA perspective it might make more sense.


Efficient cars aren't the problem.

The problem is that Uber/Lyft drivers "cruise" when the demand is slack rather than park.

So, for certain areas and times (generally about a hour before when that area is about to go from slack to busy), your street becomes a never-ending stream of ridesharing cars--generally doing stupid traffic maneuvers when they finally get summoned.


> The problem is that Uber/Lyft drivers "cruise" when the demand is slack rather than park

Isn't the whole point that the apps limit cruising time? I recall a study that claimdd the average traditional taxi cruises for half an hour on average between each trip. Are there plans to tax traditional cabs by an amount commensurate with their cruising time? If not, this legislation seems more like protectionism for traditional cabs.


> Are there plans to tax traditional cabs by an amount commensurate with their cruising time?

Traditional taxis already pay a steep annual franchise fee to the city (even after it was cut down on recent years, probably in part because less-taxed rideshares reduced the level of franchise fee that was supportable).


Franchise fees also benefits the taxi companies. It constrains the supply to taxis, thus shielding the taxi companies from potential competition. They're paying for protection from competition.


> Franchise fees also benefits the taxi companies.

In the presence of a substitute service not subject to them, they do not.

Actually, since there seems to be both a non-automatic approval process and a franchise fee, they don't really even without such competition; the approval process shields them from competition, the franchise fee is a pure cost. (Of course, there is a linkage between the two, but as long as the process is in place, incumbent taxi firms benefit from minimizing the fee.)


Do you want EV adoption or fuel revenue? Choose one.


Efficient cars cost the city less money.

This is because efficient cars are usually smaller, and cause much less damage to the roads.

It is a good thing for society that they are using Priuses. It makes road maintenance cheaper.

Why should we be trying to lose money, by discouraging behavior that helps the city out? IE, fuel efficient, smaller cars.


Hybrid cars like the Prius actually weigh more because of the weight of the batteries. Tesla cars other than the roadster are also huge and heavy.


2019 models curb weight, base model data from Google:

Corolla: 2,840 lbs

Prius L: 3,075 lbs

Camry: 3,241 to 3,572 lbs

Model 3: 3,686 lbs

Model S: 4,769 lbs


Upvote for providing data, instead of guessing! We should encourage more of this behavior.


> Public roads are supposed to be paid for through fuel taxes,

No, they aren't, at least in California; fuel excise taxes (and diesel, but not gasoline, sales tax) are a part of road funding, but other sources, including general revenue (largely, sales taxes and personal and corporate income tax), are used.

EDIT: Not that taxing ride-sharing is likely to reduce congestion. It plausibly could increase it, if it induces residents of mostly walkable communities (which, sure, LA has fewer of than it should) who might otherwise rely on ride-sharing for occasional car needs to instead keep a personal vehicle, which then tends to encourage additional, less-necessary driving.


NYC streets are worse because of 250,000+ (obviously, not all at the same time) Uber/Lyft/Via/etc. vehicles being on the streets. I don’t care if there is demand, they have made congestion worse, period. I think all municipalities should treat all “private”, “shared”, whatever you want to call them, car services just like taxis, and require them to be painted a special color. You know not to travel too closely behind a taxi because they could stop or swerve at any moment. Unless you look at the very fine print on the license plate (and this is only in cities that require these cars to register), you do you know it is a car for hire, and to not follow too closely, unless you want to be blocked while they pick up a fare. How is it that these upstarts can “disrupt” an entire industry, and have a significantly lower barrier to entry than the incumbent? I say, require them all to be painted pink!


Fuel taxes do not come remotely close to funding the roads.


Fuel is still extremely cheap in the US compared to more modern countries. It does not cover proper infrastructure, as can be witnessed in the US anywhere.


> Uber and Lyft “are using public roads, and the profit is going to their companies,”

Using this logic, they should be taxing all companies that use public roads (food delivery, package delivery, service companies), instead they are specifically cherry picking Uber and Lyft to target. These sort of anti-business policies are causing companies and people in masses to flee California and New York. Perhaps that is the goal.


> Public roads are supposed to be paid for through fuel taxes, which every Uber and Lyft driver pays. California has gasoline taxes of:

Even in California, collected gas taxes do not remotely approach the cost of construction and repair of roadways.


This is a very bad solution, especially as share rides decrease the number of cars when people use uber pool


If you believe that Uber and Lyft are decreasing the number of cars on the road, then taxing them to "free up congested roads" is misguided, as it will have exactly the opposite effect.


Not sure on the accuracy of this report but there are some indications that Uber, etc increase the number of cars on the road: https://www.businessinsider.com.au/uber-lyft-creating-traffi...

I think the implication is that people with cars keep using them and people without cars use Uber instead of alternative transport.


Talk to anyone with a reasonable public transit system about how ridesharing is clearing up the roads (spoiler alert: it's not, they're decreasing public transit ridership and contributing to a constantly worsening congestion issue).


Perhaps public transit systems could look into what ride-sharing is doing, what elements of it are so appealing to users, and if there’s any similar actions they could offer to make public transit more appealing.


It seems pretty obvious. Uber is more convenient and it comes directly to your home. People will pay more for that, which is why it probably shouldn’t be so cheap as to make public transit easy to dismiss, especially when that pricing structure encourages more cars on the road with only 1-2 people in them and the negatives that go along with it (more carbon usage, more cars idling in traffic, more road damage, more pollution to inner city dwellers, more danger to pedestrians and cyclists, etc)


So, rather than examining if there are ways to make public transport more convenient, appealing, and competitive, the only solution is to make private transport less appealing?


I didn't say that, but maybe it should be a combination of both. Which is why some cities are partnering with Uber/Lyft to do the last mile part (to the home), but using public transit for the bulk of it.

I do think you really have to question an economic model where Uber/Lyft have clobbered cabs using an unsustainable VC-funded model that will eventually run out and substantially increase cost to the end-user.


I think they're likely profitable in the cities where they've achieved scale. Specifically, I suspect that Uber (and Uber drivers) in the Boston/Cambridge market (where I live) are making a net profit overall.

Drivers who drive full time can afford to pay for the car expenses (variable and fixed). Drivers who drive part time are able to cover the variable costs (and they already had their fixed costs anyway). Uber is probably not losing money in Boston. If all three of those things are true, Uber is sustainable in Boston.


The problem is that ride sharing isn't providing a service with a realistic and sustainable price. In Boston I can get a 15 minute Uber pool trip door-to-door for $3.

They're undercutting public transportation and traditional taxi services that need to operate with a realistic budget that isn't fueled by billions of VC dollars.


If there are two people in the Pool on average, $6 for a 15 minute ride feels like it could be a sustainable price to me. When Pool first launched in Boston/Cambridge, I used to almost always be solo. Now, I would say that my average occupancy is 1.75 to 2 on Pool rides. It's also to the point that Pool rides seem to be cheap when they would result in Pooling and almost the same price as UberX when they'd more likely be solo.

If they're undercutting public transport and doing so at a sustainable price, I view that as healthy competition. If they're undercutting solely based on VC subsidies, that will inevitably end. In the Boston market, I think they're likely profitable on an overall basis, just as grocers are overall (very slightly) profitable, even when they sell loss-leaders and give other subsidies to consumers.


You're not talking a lot into account, especially that drivers are freelancers who own their own vehicles. That $6 per 15 minutes turns into $24/hr (if you're literally non-stop taking fares, which is impossible).

Then you have maintenance, wear, healthcare, retirement, etc... Just accounting for fuel and maintenance these drivers are making minimum wage without any benefits, even if they drive full-time.

The so-called gig economy is bullshit that erodes just about everything it touches, from workers rights to public services.


That's also about the worst case for the driver. There are many Ubers I take that are over $2/mile or $60+/hr.


Agreed. This sounds like a tax that will generate money for the govt and nothing else - it’s not like suddenly people are going to stop going places when the price goes up a tiny amount since there are no viable alternatives to driving in LA, either rideshare or private car.


Exactly. I could understand if the tax were tailored to the ostensible goals:

- No Uber taxes outside of rush hour, since they're not contributing to congestion then.

- Money is specially allocated to mass transit projects and subsidies for Uber pools/Lyft lines at rush hour where they are carrying multiple passengers.

But of course, that's not the proposal, it's just another golden goose.


The money from this tax would be dedicated to mass transit projects.

It would be ridiculous to use the tax to subsidize Uber as they don't pay the tax, the riders do.


> since there are no viable alternatives to driving in LA, either rideshare or private car.

The article seems to imply otherwise, that tons of money has been spent on improving public transit.

Not agreeing with you or the article either way, just pointing that out.


Seems like the article is going out of its way to make public transit sound attractive with phrasing like "Metro’s sprawling bus network". Anecdotally, nobody I know in LA considers public transit as a realistic option and they routinely make jokes about how bad it is.


Anecdotally, everyone I know considers LA mass transit to be a viable option. This ranges from professionals like doctors and lawyers to the homeless.

I know a number of transplants from NYC, Chicago, and DC who ditched their cars after they moved to LA because it's possible to get everywhere worth going by Metro.


LA Public Transit is fine if you have a very specific set of routes (WeHo/Hollywood<->SFV, Pasadena<->SFV, Most anywhere<->downtown, etc); but it's terrible for one off/ad hoc routes or uncommon commutes.


> Public roads are supposed to be paid for through fuel taxes

Some States do, California does not. The underlying issue is that California roads can infamously cost an order of magnitude more to build and maintain per mile than the same road segment in other States (e.g. interstate highways). It is possible to pay for roads with fuel and use taxes, in some other States it is a constitutional requirement, but the extreme cost of roads in California make it difficult to fund the DoT that way and there is no evidence that California is going to curtail the obvious waste anytime soon.


>The underlying issue is that California roads can infamously cost an order of magnitude more to build and maintain per mile than the same road segment in other States

Why is that? It's not like they have harsh winters like New England or anything.


In my opinion, all taxes besides a land value tax are poor solutions. rent is the final resting place of capital generated by labor and land combined. Taxing unused land higher than used land forces landowners to sell the land to someone who could use the land productively enough to afford the tax bill.

All of taxes deter productivity from increasing. It is better to tax the land date trees grow on, than to tax the output of the trees themselves.


Lots of ways around that just demolish any buildings stick some goats on to graze and call it a farm "get oorf moi land revenue - waves shotgun"


They really said that? Your local pizza place also uses public roads, and they get the profit.

What kind of stupid.........sigh.

The point of having public roads is that it benefits the city by increasing economic output.

Did he forget that there are PEOPLE in those Uber and Lyft cars? Those same people he is supposed to serve? Or does he think the cars are driving around for no purpose making profit the whole time?


Politicians don't serve the people, they serve themselves and seek reelection. Sometimes it happens to be the case that seeking reelection motivates serving the people, but that is nothing more than a happy coincidence. Sometimes a dictator earnestly wants to help his people; that too is a happy coincidence. The genuinely altruistic politician in a democracy is no more common than a genuinely altruistic dictator.


> Did he forget that there are PEOPLE in those Uber and Lyft cars? Those same people he is supposed to serve?

Many of them are tourists or business visitors, not the people local government is supposed to serve. Local government fees designed to milk those people (which incidentally also hit residents) are not at all unusual.


Something else to consider is that traffic was bad way before ride sharing companies started popping up. Secondly, additional taxes will ultimately end up being paid for by people who do not own a vehicle. A better way to help alleviate traffic conditions is to raise gas taxes across the board or maybe just using the tax dollars more efficiently to build public transportation like trains on top of the existing infrastructure that cars typically use. I don't know why California is wasting time on obviously bad legislation. Why do they feel the need to consider the crazy option?


Public roads are supposed to be paid for through fuel taxes, which every Uber and Lyft driver pays

Do Uber and Lyft pay those drivers' fuel costs?


Not to mention Uber and Lyft drivers are wholly responsible for vehicle fees, maintenance, registration, their cellular data plans, wear and tear. California is out of control with it’s liberal tax policies. I spent the first twenty years of my life here and every time I return back I start to remember why I left.


Is "drunk driving" the new "the children"?


Gotta tax drunk driving.


eeeeeeeeeeeehhhhhhhhhhhhhhhh

Uber undercuts both public transit and traditional taxi services with prices that are unsustainable and can't provide a living wage for their drivers.

This was OK initially because taxis in many places were benefiting from localized monopolies and also screwing many drivers, but in the long run it means Uber kills all the competition and then just jacks prices back up anyway.

I don't dig the drunk driving fear mongering either. If I hired the homeless $3 a trip to carry drunk people home that doesn't mean my exploitative program should suddenly be free of municipal adjustment.


Thank you for bringing this up. Uber and Lyft are already taxed. Taxing them more doesn't solve anything.


Agreed. Also consider that using taxation as a punitive tool is both immoral and ignores the fact that the government doesn't deserve the resulting revenue.


Also, they plan to tax ride sharing (which people have shown demand for) and use the money to improve public transit infrastructure (which people have shown distaste for). Why, why, aren't taxes like this directed to improve the infrastructure for the services that are being taxed? Elon Musk is on the right track - build out more roads, under the city, to improve capacity and reduce congestion. Don't burn the money on a service with ridership numbers that are in freefall.

Imagine if a private company substantially raised the price of its premier product (streaming movies, for example) to build tons of infrastructure for the product that nobody wants (DVD distribution hubs, for example) and neglected to build out the infrastructure required for the product in demand. It'd be gone overnight. This is absolute insanity.

The real solution:

- Tax all road users - most easily accomplished through license/registration taxes (IE, you pay an annual fee for a sticker that gives you permission to drive your car/bus/truck in the taxed area) and fuel taxes - By statute, require the money to be spent to improve/expand the road infrastructure - Where pollution is a problem, (optionally) spend a portion of the tax to build out EV infrastructure, and encourage EV use by charging reduced fees to EV users - Subsidize use of ride-sharing type services for people who would normally use public transit services

This would provide efficient, clean, point-to-point transit services for ALL users. Economically disadvantaged folks could finally stop paying the hidden 'poor people transit tax' (increased transit times).


This doesn’t fundamentally change the supply and demand problem in LA, which is a city built entirely for cars and nowhere near the road capacity for everyone to be able to make all the car trips they would like to.

For cities that want people to be able to get around AND for the roads to be relatively uncontested, the only reasonable solution is to lower demand by raising the cost of driving - ie. significant congestion pricing (tolls) across the metro - and invest the proceeds into a vast increase in more space efficient transit options. Typically this means express bus and rail service, although the number of people who can cycle efficiently on a dedicated bike way is pretty impressive, and with the scooter revolution there are a lot of other personal mobility choices for people who are less physically able.

Traffic congestion is a problem of geometry and public will - with public will being the primary impediment to progress.


>the only reasonable solution is to lower demand by raising the cost of driving

You can also reduce the need to travel by increasing density, which LA is attempting in certain respects.


How so? There was this reddit comment where someone listed all the ridiculous parking requirements they impose that torpedos apartment projects, regardless of how close to downtown or transit they are.

[1] https://np.reddit.com/r/LosAngeles/comments/6lvwh4/im_an_arc...


Every bit of new construction seems to be mixed use commercial, retail, and housing with large underground parking structures. He's right that its "luxury" condos but its still a lot denser than the traditional two story apartment buildings.


Building luxury condos and apartments is fine. Rich people will live in them instead of outbidding poorer people for the older apartments.

People who oppose luxury apartment construction need to explain how they will prevent rich people from competing with poor people over existing housing and causing the prices to go up.


- Rich people will probably not lower their standards enough to live in what we plebs do, thus move somewhere else, reducing demand - Doesn't luxury living space pretty much by definition require more space? In other words, you could likely house more poor people there instead, reducing demand for housing. Even if the same rich people move in as well

Yes, those kind of policies seem to be on the more reasonable spectrum to me, though I'd prefer having incentives instead of hard requirements. Of cause prices for luxury space would go up, but this doesn't seem like much of a problem for most residents.

You seem to see new luxury apartment as only option over currently existing housing. That might be the only realistic option when owners want to maximize profits, but for society it might be better to optimize other metrics, like people living there.


i wish all the ugly parking went underground. a lot of new construction in LA is mixed use but most of the parking is above ground in podium style structures because it’s cheaper.

but once you dig the hole, it’s like building the same building but starting 30 feet (or whatever) lower in elevation (yes, that requires reinforcing the ground, more environmental review, etc., but still...)


Ironically, use of Uber and Lyft decrease the need for parking.


If you raise the cost of driving without a vastly improved public transit option, you're just putting in more regressive taxes on the middle class.

dmitrygr 56 days ago [flagged]

  > public will being the primary
  > impediment to progress.
What an extremely arrogant statement. If the public will does not agree with your world view, perhaps the latter needs adjustment, and not the former, like you imply?


I’m not saying public will is currently “wrong,” I’m saying we shouldn’t kid ourselves about why we’re in the mess we’re in.

Most of American city-dwellers want to “have their cake and eat it too,” that is, they want to live in a major metro AND be able to easily drive and park everywhere they want to go.

Those two things (major population and easy driving) are mutually exclusive, because car oriented development doesn’t scale beyond rural / suburban size very well.

But so far I haven’t seen any public leaders or communities acknowledge that and start working on alternatives. Instead they play games like blaming Uber and Lyft and pretending that an extra tax on them will make any real difference, when in reality it’s political theater.


The American public is unbelievably misinformed. Public opposition to an idea indicates only how much cable TV the public has been watching recently and says little about the merits of the idea. Just as an example of such a thing, the median survey respondent believes that foreign aid is the largest component of federal spending, and the overwhelming majority are against increasing it. However when those same people are informed that foreign aid is among the smallest parts of federal spending and that the USA is by far last among rich countries in per-capita and per-GDP aid those same people in the same survey at the same moment largely change their minds.


"The American public is unbelievably misinformed."

This may be true without implying any defense of something that is contrary to the public will.

The public can be totally awful and that doesn't mean people who have contempt for the public are any good.


I don’t see how saying that “public will is an impediment to solving traffic congestion” indicates contempt for the public. It seems like a straight-forward factual statement, and not a particularly controversial one.

After we acknowledge that “public will” is an impediment, we can then probe whether the problem is really so bad after all, what “public will” really means, whether it is self-contradictory, how strongly entrenched those interests/ideas are, whether there are ways of reframing the conversation to significantly sway public perception and action, and so on.

It might turn out that everyone loves griping about traffic congestion but isn’t really too bothered. Or that “public will” is just reflexive opposition to misunderstood alternatives conditioned by decades of deliberate propaganda. Or that “public will” is very weakly held and malleable if an alternative proposal does a charismatic marketing blitz. Or that the disparate powerful interest groups involved are in direct opposition and political progress working towards compromise is at a total impasse and impossible. Or ...


Op said that "public will is an impediment to progress". That is much more arrogant statement than the one you claim to defend: "public will is an impediment to solving traffic congestion"


The context made it abundantly clear that this meant “... impediment to progress [with respect to traffic congestion]”.

Not just that the public is generically “anti-progress”.


It isn't arrogant, just another "tragedy of the commons" that we've experienced many times before.

Roads are expensive and we don't have room to build anymore anyways. Without an effective way to increase capacity, something obviously has to give.


Public will used to hold some very shocking positions in he past.


Popularity does not imply correctness.


in a democracy, popularity implies policy. literally the definition


That's the problem with democracy that ignores the existence of divine authority, or at least natural law.


Internment camps were popular. That made them no less an abomination.


> is to lower demand by raising the cost of driving - ie. significant congestion pricing (tolls) across the metro

All that does is increase the cost of living.

Either people need to drive or they don't. People don't drive just for the fun of it anymore.

And if you succeed and do manage to reduce travel you also managed to reduce economic activity.

If you goal is to get people to switch, then you have to do it on the other end: Make the alternatives better than driving.


> Either people need to drive or they don't. People don't drive just for the fun of it anymore.

This isn't true. Some people live in places where it's possible to reduce their driving quite a lot, but when they decide to do that, it isn't as easy as simply choosing not to drive. Once they're motivated, they start to figure out opportunities to drive less. They start to figure out which other forms of transit are practical in their area, and they figure them out one by one. They start to figure out how they can adjust their habits to reduce driving. (Fewer, larger trips to the grocery store, for example.)

There are a lot of degrees of "not driving." You might bike to work in perfect weather at one level of motivation, and bike to work in the rain at a different level of motivation. One level of motivation might prompt you to make fewer, larger trips to the grocery store; at another level of motivation, when you run out of milk unexpectedly, you might walk five blocks to a corner store of driving ten minutes to the supermarket.

The only perspective from which it seems like a clean either/or is for people who haven't yet found a reason to try. People living in the same neighborhood with the same economic means can and do drive drastically different amounts depending on how motivated they are.


There are no practical means of transit in LA. Sure, there are light rail here and there but it doesn’t get you most places you want to go.

Charging tolls does not fix anything. All that does is enable those with money to have a slightly less stressful commute and add an empty lane to an otherwise crowded highway.


At some point there will be an incentive to move closer to where you drive to the most often. For a lot of people this would be their job.

Unfortunate that the sprawl is so heavily subsidized. In a world where it wasn't, people would make different choices about where to live and work.


It’s not so simple. Many couples don’t work in the same place. You could have the wife working in city A. And the husband working in city B 30 miles away. Moving to where they work only works for one and not the other. In other cases they have kids and the school districts in large cities are absolute shit. In that case you move for the school district so that your kids can have better chances in the same rat race you are in.


Which is why I included the second part of my statement, which you did not quote: we would need to use the proceeds to dramatically increase the quality and quantity of alternatives to driving so that people had a realistic choice.

I totally agree with you that people are driving mostly because they have to, not for fun. However, retrofitting car-oriented development to be viable for non-car users is exceptionally difficult.

I can’t imagine any way of accomplishing that, unless we begin by acknowledging that cars won’t scale up any farther and that we have to redirect our efforts into things that will.


Public transit has to happen first.

If you introduce congestion charges without also providing a functional public transit alternative you'll be voted out of office. Plain and simple.

Congestion charges worked in London, for example, because public transit is amazing there and people could stop driving without decreasing their quality of life. Yet it was still a source of much controversy in London at the time. What are people supposed to do in LA -- walk everywhere?


> we would need to use the proceeds to dramatically increase the quality and quantity of alternatives to driving so that people had a realistic choice.

I agree, but public transit projects take years to complete, so in the interim you'd have very expensive driving and worse congestion than before (due to construction), which sounds like hell on earth. Would it be feasible to invest the costs up front and collect the payout later?


Sevilla, Spain is the poster-child of rapidly building out a bike network[0], and with the political will, that could be replicated elsewhere (especially somewhere flat and sunny like LA). Another easy way to improve transit is paint dedicated bus lanes for BRT lines. Not as sexy as dropping a few mil on light rail, but quick, cheap, and rapidly scalable.

[0]: https://usa.streetsblog.org/2018/05/07/six-secrets-from-the-...


> which you did not quote

I did not quote, but I did address "Make the alternatives better than driving."

You have things completely backwards, first you want to make everyone miserable, then hopefully they'll fix things.

> retrofitting car-oriented development to be viable for non-car users is exceptionally difficult.

It's actually not about retrofitting, it's because cars are just that good. Nothing else comes close.

> unless we begin by acknowledging that cars won’t scale up any farther

They'll scale just fine if we space out our cities a bit more.

It's actually pretty self balancing, when we run out of road space, people spread out a bit more. So there's no scaling issues, there's plenty of space in the world.

I mean if your goal is to pack people in like livestock then it doesn't scale, but that's not something I want.

> and that we have to redirect our efforts into things that will.

Such as? So far nothing exists. And I'm not just saying that. I tried it. I exclusively used public transportation for 1.5 months, in a city that is world renounced for probably the best public transportation of anywhere. (There was literally nowhere you couldn't go with public transportation.)

It was horrible. I feel really bad for anyone who has to use public transportation.

That's the reality.

If you want change, change that first. You won't need to tax cars off the road, they'll leave on their own.


The congestion caused by Uber and Lyft in SF is bad. It's not because of the number of cars on the road, although that's certainly something, it's because of how those cars behave.

Ubers and Lyfts pull over at unpredictable spots (wherever the app says to) and then just throw on their flashers. It doesn't matter if they're blocking a lane in rush hour they just stop.

I will never for my life understand why an Uber or Lyft driver thinks it's any better to double park in an active lane of traffic than to temporarily pull out and block a driveway or a red zone. Both are illegal, but at least one of those doesn't block traffic and endanger other drivers/bikers.

Almost every day I see an Uber driver double-parked within 50' of somewhere they could have pulled over safely.

If the cops would give out a few more tickets for this, the behavior would get better and so would the traffic. And then we wouldn't have to tax the road use. Taxing the road use is crap, everyone should be able to drive their vehicle on public roads as much as they want as long as they follow traffic laws.


All the delivery services park illegally as well. For a long time I assumed they had some sort of exemption because it was so common. They just eat the tickets as cost of doing business.


I suppose tickets should have exponential increase and decay built-in - this way, collecting a bunch of tickets in a row wouldn't be edible for companies, but intermittent tickets (corresponding to base rate of traffic violations) would cost no more than before.


Linear is more predictable and easier to manage, mentally, so I don't think that will change any time soon.


Exponential is linear in log scale, so maybe let's just say that a ticket will cost you N zeros, where N is equal to number of tickets you had in the past month.


That’s not a bad way of phrasing it


I think more often they just don't get the tickets. I've had Uber and Lyft drivers complain to me about tickets they got and it seems like they were paying out of pocket for them and it was a very huge and noticeable expense.

Pretty sure one $200 ticket can wipe out 2-4 days of hard driving profits.


I'm a LA resident. People in this city like to talk about congestion as though they see it as a problem, but they don't actually see it as a problem.

Only people in the valley see it as a problem because there are limited choke points into and out of LA proper which keep people from getting to work.

LA's limited transportation infrastructure prevents DTLA's homeless from getting to Beverly Hills, Long Beach's gang-related activity from getting to Brentwood, etc.

If LA wanted to fix their transportation problem then they could build trains parallel to the 405 and 110 and solve the problem, but they don't want to fix the problem because the problem allows them (us?) to keep the city segregated by income.


LA is building a rail parallel to the 405, and the Silver Express (bus) has run down the 110 for more than a decade. LA is also building 2 more rail lines in the SFV. The reason they're not already built is that the SFV opposed these lines 3 decades ago when they were originally proposed and only recently dropped its opposition.

Also, Buses run from downtown to Beverly Hills, and getting from Long Beach to Brentwood can be done entirely by public transportation.


-- Edit --

Sorry, I see now the above comments are more a refutation than a statement of, "Look how good the transportation system is." I'll leave these though just to make my general point about how horrible the options are.

-- END EDIT --

Just pulled up google maps for Long Beach to Brentwood:

Public Transit Option: 3 hours

Car: 39 minutes (29.9 miles)

Just pulled up Downtown to Beverly Hills

Public Transit Option: 1h30

Car: 30 minutes


To be fair, that was at 7:30pm, the tail end of traffic and after the busses switch to a reduced schedule. I'd be curious to see those results at 5:30pm.

I suspect the car would still be faster though.


Plenty of people still need to get around at 8PM at night. So, I'd say a bodega worker in Brentwood going home to their less expensive apartment in Long Beach would prefer to be able to commute home at 10PM without it taking 3 hours :)


> LA's limited transportation infrastructure prevents DTLA's homeless from getting to Beverly Hills, Long Beach's gang-related activity from getting to Brentwood, etc.

Gangsters have cars and don't rely on public transportation. DTLA homeless can and do use the MTA bus that goes literally from downtown to Beverly Hills.

Care to clarify your comment?


No, your perception of LA's public transportation is completely out of touch with reality.

Subway and train systems are fundamentally different modes of transit than bus systems. They are more accessible, run on regular schedules, and they are anonymous.

The urban poor in other cities utilize them disproportionately which is why Beverly Hills shuts down every hint of a conversation of adding public transit connection beyond commuter buses.

It's too bad if you're offended by what I said, but LA has more nimby's per capita than anywhere I have lived which is exactly why we do not have a public transportation system.


For the specific example you gave: does an irregular schedule really create an obstacle for a homeless person who wants to make a trip from DTLA to Beverly Hills? What does it mean "more accessible"? What do you mean by "anonymous"?

What makes you think I'm offended? I just didn't understand your argument (and still don't), and I wanted to see what you meant.

And what does it mean "my perception of LA's public transportation is out of touch"? I only said two sentences, one about gangsters typically having cars, and the other about homeless people often taking MTA bus from DTLA to Beverly Hills. Which of those was incorrect?

Edit: I can think of a reason why rich people may not like subways going to their neighborhoods. Slow and unreliable buses are a problem not for gangsters and homeless, but for the hard-working middle-class and poor people who need to commute to work - even if they have cars, traffic makes it very tough, and subway would have made it really nice. It's quite possible that rich people don't want too many middle-class and poor people getting jobs in their neighborhood. I have no idea if it's true, but at least it sounds plausible.


My argument is that the west side has had many opportunities to alleviate our transportation problem and has consistently decided that we would rather deal with hour long traffic jams than even incremental amounts of change.

People have been trying for years to do things like add safe bike lanes, rebuild the streetcar system, and build a subway system. Angelinos have been far more concerned with the temporary traffic disruptions, slight inconveniences, and potential demographic shifts of each.

My entire neighborhood threw a fit when someone tried to install a roundabout instead of a 4 way stop to reduce congestion.

Nobody would have been negatively impacted for more than a few months, but the people here just don't care about collective benefit, at all.

And that's before we start discussing the people who have deliberately induced greater congestion by adding speed bumps to side streets in an effort to protect property value.

LA chose to have this problem. It's not something that just happened on its own.


And to your point, they absolutely could solve the chokepoint issues with eminent domain, and adding additional arteries, but they choose not to.

For people who think a congestion tax will improve traffic, it won't. Congestion taxes work if they can shift behaviors to alternate modes of transportation. If there aren't any alternatives, it's just a levy.


> and adding additional arteries

Time and time again cities learn this does nothing. https://en.wikipedia.org/wiki/Induced_demand


Induced demand doesn’t say that nothing happened. It says that demand just fills up the new supply.

There are no alternatives to a road system in LA for most residents. If the roads were choked and there was an underutilized subway that I could use, I’d totally get the argument of taxing the road system or capping capacity, so as to push people toward more efficient methods. But, it’s not feesible to build out a subway system in this city that is expansive, so we need to work with what we have- Cars.

Make LA the autonomous car capital of the world. Let’s be the best and first at linking cars together in convoys and shuttling them fast down our massive freeway network.

Let’s build tunnels under the mountain passes to add more connections into Hollywood so people have better access than winding canyon roads.

Let’s build expansive subway networks through south central and downtown to bring cost effective transportation to lower income neighborhoods.

Let’s do a ton. But let’s also build a bunch of roads.


> Let’s build tunnels under the mountain passes to add more connections into Hollywood so people have better access than winding canyon roads.

This is an interesting idea. I just measured it out, and assuming you could build a straight tunnel to bypass the canyon at the 405, it would be 5.3 miles.

To bypass Laurel Canyon would be a 3.0 mile tunnel.

Coldwater bypass would be 4.1 miles.

So the question is, do we the technology to build tunnels that long and deep (all of those are about 1000 feet). And is it a good idea to build a 1000 foot deep tunnel in an area known for earthquakes?


I'm not an engineer of that sort, but I'd imagine if we can build subways in earthquake-prone zones like SF, LA, and Tokyo, we can build safe tunnels under mountains in an earthquake zone.


So removing existing roads will lessen congestion, right?


Congestion taxes can also shift road use to other times. Imagine many people on the roads are making relatively short trips (eg. a trip that takes 20 mins in traffic instead of 10 mins) that generate a lot of lane changes and contribute to congestion. A congestion tax could encourage them to drive in non-peak times, use the bus, or not make the trip.


The city is already built this way and people already made decisions on where to work and live based on facts on the ground. Unless congestion charges are levies only on newcomers, it has no legitimacy. Try these social experiments on a new city.


> The city is already built this way and people already made decisions on where to work and live based on facts on the ground.

This would indicate to me that they are already OK with the level of congestion they currently have. But that's not what they say. Some kind of disconnect.


You consent to public policy changes by living in society. In particular, you chose to live in a city with lawmaking bodies that are expected to enact policies to adapt to changing times. If you think the pros of some change outweigh the cons, make the argument, but don’t claim that public policy writ large is illegitimate.


I can't tell if this is a humorist take on the issue or not because its so completely off base. If you were really a knowledgeable LA resident you would realize that the Valley actually pays for most of LA's expenses (which is why the vote to make the Valley a separate city decades ago failed), but also more importantly realize that there is a lot more to LA's traffic situation than just the 405 and 110 chokepoints.

The simple fact of the matter is that LA has a serious housing and urban planning crisis with no real solution. The city has already been investing heavily in public transportation for years now with multiple heavy rail lines been under construction (purple and crenshaw) and some already opened (gold) - but the fact of the matter is that until a public transportation system that can replicate the freeway system - meaning a system that can replicate both the breadth and depth of the system (Think a 'good' Euro rail system spread across a 30-60 mile radius) - LA will always be a congested mess - unless a paradigm shift in transportation occurs (hyperloop, Uber Elevate, etc.)


I spent a lot of my childhood in Beverly Hills and still have friends in the area: what you're saying has historically been absolutely true, but in the last 5 years, my perception has been that the seemingly all-the-time-and-everywhere traffic has gotten to the point where even people living in the nice parts of West LA are fed up and opening up to a functional transit system.

This is obviously somewhat anecdotal, though; it's possible that for some reason my sample is skewed enough that it's different from the population's views in general.


if those rails were built, those ghettos would be gentrified so quick.


> Metro’s rough estimates suggest a 20-cent fee on each trip could bring in $401 million over a decade, while a $2.75 fee could raise $5.5 billion.

Hmm, $401 million / 20 cents * $2.75 = $5.51 billion.

Is there some "law" that public transportation officials never account for the fact that increased fares or taxes will reduce demand?

Maybe most people don't care, but I find it hard to take seriously anyone who doesn't believe in price elasticity of demand. Perhaps they are ignorant of economics, or perhaps they are so well off that that $2.75 isn't worth counting. Either way…


In this case, reduced demand is desirable because Uber and Lyft drivers are a large part of the renewed traffic problems in LA.


A lot of people's thinking doesn't go beyond stage one. We should all be asking ourselves "then what will happen?" many times before advocating for economic policies, especially tax increases.


I expect this to backfire. More people will buy cars if hiring out their commute becomes more expensive. Once they have bought a car, they'll use it even for perfectly walkable errands, just like everyone else.

"Metro also is considering a fee on bicycles, electric scooters, and other devices"

Ditto.


Your comment actually made me read the article. That is a direct quotation, and pushes the whole thing over the edge in my opinion from "very poorly thought-out plan" to "just barely pretending"


You tax something, you get less of it.


NYC is doing this as well and it makes no sense to me. What is congestion? Is it “people who own cars getting stuck in traffic”? To me it seems like all this benefits is wealthy people who want to drive comfortably to/around their neighborhood.


Not only that, if you look at plates, a good portion of the cars around Manhattan besides taxis and hire cars (T-plates) have out-of-state plates. Why should residents pay hefty city/state taxes AND pay more to use the roads (through the new Uber/Lyft tax) so that there is less congestion for people who live out of state and drive in? What behavior is the city trying to incentivize?


Congestion pricing is paid at the point of entry, so the non local cars also pay the fee. In fact, it's easier to collect from them.


But in the article LA residents pay increased Uber and Lyft fares. If this policy were in NYC, the gp is suggesting NYC residents get hit with an additional fine that people who drive from out of state would not.


Well, it doesn't matter where that uber came from.

Reducing congestion isn't just about making it easier to drive, it makes is safer as well.


Exactly. In fact, NYC does have a similar policy (but they delayed implementation so I’m not sure if it’s active yet.)


I’m not talking about proposed congestion pricing (which hasn’t passed in NYC yet AFAIK), I’m talking about a recent fee that only applies to Uber/Lyft-type rides.


This is fffing absurd. I am a hard core liberal but people on my side are not really smart when it comes to these types of decisions.


I feel this is a completely misjudged solution. The reason so many people use Lyft & Uber is because of lack of public transport. This seems like city finding another way to tax people instead of helping its citizens.


Pretty sure Los Angeles had bad traffic way before Uber and Lyft arrived.

I distinctly remember staring out of the Hilton at Universal City in 2006, on my first visit to California, with a group of Canadians marveling that there could be a traffic jam at 10PM at night.


I don't think they're trying to argue that Uber and Lyft created LA traffic but I would agree that their arrival put more cars on the roads. There's a huge market of people who are willing to pay a little extra to avoid the hassle of LA public transportation and considering that, in general, people Uber to busy places (LAX, downtown, etc.), it's bringing more cars onto the already clogged roads that wouldn't be there otherwise.


But is it really cannibalizing public transportation? Or is it cannibalizing private car driving? I assume somebody has studied this.


I got stuck in a traffic jam on the 405 near LAX at around 1:00 AM on my second-to-last day in LA last month. I'd stayed late at work to sober up after my farewell happy hour. I don't know if it was an accident or construction or what that caused it, since I didn't see anything.


Also of note... don't most people use Uber Pool to save a few $?

Just a guess but if the route planning models are good then an uberpool with 2 passengers + driver qualifies for 3+ carpools and will be much faster than a car that doesn't qualify for carpool with a single passenger and driver.


Comparing rideshare to public transit, this tells me there's clearly a need for easier, faster transit even with higher pricing.

Solution? Try to ruin it.

Instead, maybe we should embrace it and provide incentives for ridesharing vs driving alone? E.g. give uber a lane on the highway for 2+ passengers and tax only single rider transits. Give a few lyft ride coupons with every monthly bus pass. Provide benefits to using busses (which are a terrible experience) by occasional free lyft rides ...


Can we stop with the "rideshare" junk yet? Uber and Lyft are computer-dispatched taxi services. The overwhelming majority of their trips involve the driver going to someplace they have no need of going, with either zero or one other person in the car.


I have no idea what you are really trying to say. How do you define need? All my Lyft trips have either been to/from an auto mechanic or the airport. Is that "need"? In principle, I could have gotten friends to drive me, given enough incentive. Or I could have parked at the airport, or ridden the bus, or gone to a mechanic that made it convenient to wait for work to be done.


Gp is trying to say that nobody is really "sharing a ride" that they would be driving anyway, as in hitchhiking with compensation. It's virtually all professional drivers going where the client goes exclusively because the client goes there. Hence the term "ridesharing" is misleading. A very successful piece of clever viral marketing, or maybe a fossil of an early strategy.

It's a semantic point, but a valid one. Words matter.


We don’t call them taxis in California because in most California cities they’re different because of:

* no access to red lanes

* only app-based payment

* unique pickup/dropoff locations at airports and other such places

If we called them computer-dispatched taxi we’d confuse that with Flywheel-style stuff which can use red lanes, take cash, and use taxi locations. We could call them Taxi Type 2 but that’s super confusing.

“TNCs” is just an awful name so we use something that sounds intuitive “ridesharing”, since we know that pooling is possible on these services. It’s just a name. No one has proposed an alternative simple enough.


In Transport for London terminology they are called "Private Hire" companies. Separately regulated, you can't flag them down on the street, and no access to taxi lanes.


In the UK we've had this semantic split for decades and it's simply "private hire" vs "taxi".


Well, yeah, that was obvious. But I was focusing on a different semantic point, which apparently you have no interest in discussing.


My impression is that by "the driver going to someplace they have no need of going", they meant "drivers going where the client goes exclusively because the client goes there", contrasting with the hitchhiking-like scenario implied by the word "ridesharing". Possibly, you interpreted that as questioning the client's real need of going?


> The overwhelming majority of their trips involve the driver going to someplace they have no need of going, with either zero or one other person in the car.

Can you cite a source? At least where I am (Seattle) Uber Pool and Express are quite popular - and I rarely see hard numbers.


You say "even with higher pricing", but are also saying that higher pricing (due to taxes) would ruin the services?

Uber and Lyft fares are unsustainably low right now, and at their current prices, they depress public transit interest and ridership. But what happens in 5 years when these ride sharing companies need to turn a profit? Self driving cars is potentially a solution, I suppose, but will it be ready at that scale by then?

Not only that, but in the meantime, they may be causing more environmental harm than good: https://www.citylab.com/transportation/2018/04/how-uber-and-...


I think the popularity of Uber and Lyft is a direct result of the lack of adequate city transit solutions. If big city transit was robust and at least at the level of NYC or more people would likely opt for it and the big cities could plan their real estate around it better.


Funnily enough, I am one of those people-my partner and I have one car between us (in LA), and I rely pretty heavily on Uber pool (or the Lyft equivalent) to get around on a day to day basis.

However, it feels unsustainable. I can't imagine the fares remaining this low, and because of people like me, transit ridership is declining. I don't think these rideshare companies can sustain such low prices indefinitely, and my guess is it's delaying real improvements to transit infrastructure.

That's just my hunch though.


Not really. In transit poor cities people own cars and drive their own cars. In transit rich cities you can feasibly live without a car and use rideshare to fill in the occasional gap. Transit and rideshare are complementary. Only rich people and businesses use rideshare as chauffeur rather than necessary transportation.


Not necessarily. When on vacation I used rideshare everywhere I went in Texas as I didnt rent a car. When I lived in Texas I had my own car because ride share isnt an option I could use regularly. In NYC I never needed ride share at all. I used it once in a blue moon. NYC grew in population but number of transit riders has remained steady or declined since 2015. Even the NYT did an article on 8/2018 linking Uber to the decline in transit usage.


Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact

Search: