General MBTA Topics (Multi Modal, Budget, MassDOT)

[odurandina]

https://www.bostonglobe.com/busines...-study-says/m81MOB3tNBlaW19e2zBAMN/story.html

The MAPC survey suggests that 15 percent of Uber and Lyft rides add cars to the streets during the morning and evening rushes, and their cars are noticeably prevalent in downtown Boston. Last fall, the council gave tablets to 10 ride-hail drivers to survey passengers during trips. The month-long query found that 42 percent of passengers would have otherwise used public transit, while another 17 percent would have walked, biked, or not traveled at all.

“Even adding a small amount of additional congestion during rush hour has a very significant impact,” Draisen said.

The survey was the first of its scale in the region. Separately, a survey by the Massachusetts Bay Transportation Authority showed that 30 percent of riders sometimes take Uber or Lyft instead of public transit.

 

[tysmith95]

^10 drivers is not a large sample size. If they are smart Uber drivers, they often reject UberPool requests during certain times. This could explain why such a large percentage of respondents were UberX passengers.

 

[KentXie]

For those who are curious:

https://d3044s2alrsxog.cloudfront.n...ng-docs/2018/february/2018-02-05-fmcb-dgm.pdf

- MHSL returned to full service yesterday.

- MBTA is investing around $8M in the PCC fleet rehab in 2019.

- Draft options report for the long-term planning study is due this spring, so watch for it and comment.

I don't think this is an Arborway situation.

I swear a saw a PCC car being transported above the Masspike on Saturday morning. Maybe that's related?

 

[JumboBuc]

Of course Uber/Lyft/etc increase traffic. Anything that lowers to cost of travel-by-private-car--which ride services do--is going to increase the quantity travel-by-private-car. And that's going to increase traffic. It's not rocket science...

What I struggle with on all of these sort of findings is the "so what?". Just because ride sharing services lead to more traffic doesn't mean that they're "bad," or that they need to be stopped. Plenty of things in our society lead to more traffic. Anything that encourages people to leave their houses for any reason, really, increases traffic. So what? Ride sharing services provide clear, inarguable benefits for their riders. An analysis of their cost with respect to increased traffic needs to be weighed against these benefits. And I know of no good reason why people driving themselves in their own vehicles have any more of a right to the road than others riding in Ubers/Lyfts.

If the powers that be want people to substitute Uber/Lyft rides for public transit rides they should make public transit rides better. There are plenty of times when I could take the T somewhere, but I take Uber/Lyft because it's way faster and easier. I frequently travel between Cambridge/Somerville and Fenway/Longwood/Brookline. Some of these trips need to happen around 6 am. That trip on the T requires multiple transfers and takes about an hour. On the weekends or at odd hours it takes even longer (especially when you have weekend bustitutions). In an Uber/Lyft, it typically costs $5 - $15 (Pool/Line) and takes about 20-25 min. Of course I take Uber/Lyft 90+% of the time I'm making these trips, even though I theoretically could take the T and in a world without Uber and Lyft where I had to make these trips I likely would take the T. I greatly benefit by using Uber/Lyft in these situations. There's nothing wrong with that, and Uber/Lyft should be celebrated for filling in the gaps in T service. Criticizing this Uber/Lyft trip because I'm "causing traffic" is ridiculous.

 

[CSTH]

I agree with everything you’ve said.

What id add however is that I think one of the sticking points is that the marginal private cost of many car trips is lower than the marginal social cost of those car trips. This is true of both conventional and ride sharing trips. What makes uber in particular especially problematic is that at the moment that marginal private cost is also heavily subsidized by organized capital, which is doing so in order to starve other modes of resources so that uber will be increasingly attractive in relative terms - but not absolute terms - over time.

So in my view the issue isn’t so much that you personally are causing traffic - it’s the incentives that induce you to take an action that transfers costs from one actor to another in a way that overtime will make things less good for everyone than they might be otherwise.

In other words, uber inc. is paying for a portion of each ride in order to keep some of those rides (the marginal ones) from going to left, or the t, or a bike instead. For now. Eventually they want to recapture that subsidy by charging us all more onceall the taxis are gone and the t falls apart because we don’t invest in it because we have uber now.

(Like small donuts and 5$ coffee at Dunkies, but for getting around.)

The problem is the distorted incentives, not you!

 

[tysmith95]

If we believe that there is a large social cost to driving then the city should implement policies to discourage it like congestion pricing or higher gas taxes. I don't think using Uber/Lyft as a scapegoat is the right policy. I don't think the former method of capping the number of for hire cars (taxis) on the road was the right idea. It really only helped medallion owners and speculators, but it hurt riders which had to pay higher fares and drivers which had to pay large costs to rent a medallion. Plus you have the side effect of there not being enough cars during the most in demand times (eg bar close). Also I'm willing to bet that it's much easier for someone in places like Dorchester or Roxbury to request a car then it was before ride-share became a thing.

The amount of Ubers on the road has not surpassed the amount of private cars on the road. And I don't see how an Uber is worse then a private car. It creates jobs after all.

I think the biggest effect that Uber has had is reducing the demand for parking in downtown areas. I see that as a good thing, as redeveloping parking lots is good for the city.

One thing the city should look at is finding more legal places for Ubers/Lyfts to pull over to pick up and drop off passenger. They need to work with Uber/Lyft to direct passengers to only request cars in this spot. As it is now, drivers often have to block traffic for a short time to pick up and drop off people. Of course many drivers are idiots and don't care about the people around them, but the way the app works it can be hard to avoid double parking for a quick second.

Uber is implementing a new system called express pool, where people walk to pick up points for a more efficient route. It's good in theory, but I swear every single time the computer directs pick up points to be in the middle of intersections. The middle of intersections where it's dangerous to stop and wait for passengers. I wish cities and Uber could work together to create special ride-share spaces to make this process more efficient.

 

[cadetcarl]

If only cities could have their own express pool, only maybe with even larger vehicles for more efficiency! Too far-fetched.

Ridesharing services are just as good and not-good as private cars, and as such should be disincentivized the same way. But I'd add that there's an inherent conflict of interest which, when acted upon by cities, negates all the advantages of these services.

If increased traffic is bad (and it still is bad--in our haste to exonerate Uber let's not forget about congestion and pollution and wear on infrastructure), then all the tools cities use to suppress it will hurt riders and drivers alike. Congestion pricing will pass through to the rider, or the already squeezed and barely profitable drivers will have to eat it, but Uber and Lyft never will. Ditto taxing these services for operating at any scale that would make an impact.

 

[CSTH]

^ I'm with you on most of this too...

...but if you've spent a lot of time on the transit part of aB (and other similar discussions) you must know that at some point it's just not possible to get people to or from a big event in a reasonable amount of time, in cars, without a big congested mess.

At the end of a Celtics game, there will never be a way to get a meaningful number of fans into cars close to the Garden in a short amount of time. Even if there are four passengers in each car.

The constraints are geometric. Because you still need something like 2 feet of curb for 60+ seconds for every person who wants to depart in a car. And if the bar is closing,
count on like 2 minutes per person instead. (but yes I recognize that was only one of the several points you were making).

 

[Joel N. Weber II]

What I struggle with on all of these sort of findings is the "so what?". Just because ride sharing services lead to more traffic doesn't mean that they're "bad," or that they need to be stopped. Plenty of things in our society lead to more traffic. Anything that encourages people to leave their houses for any reason, really, increases traffic. So what? Ride sharing services provide clear, inarguable benefits for their riders. An analysis of their cost with respect to increased traffic needs to be weighed against these benefits. And I know of no good reason why people driving themselves in their own vehicles have any more of a right to the road than others riding in Ubers/Lyfts.

I believe in downtown Boston there are limits on how much parking is allowed to be built, along with rules saying that residential parking isn't allowed to be rented out to employees of nearby businesses during the day, precisely because parking has traditionally been used as the place to have some approximation of a congestion charge. It might be reasonable to have some sort of a per-vehicle congestion fee on Uber/Lyft type vehicles in downtown during peak travel times, perhaps with some sort of exception for people who have disabilities that make them unable to use mass transit.

If the powers that be want people to substitute Uber/Lyft rides for public transit rides they should make public transit rides better. There are plenty of times when I could take the T somewhere, but I take Uber/Lyft because it's way faster and easier. I frequently travel between Cambridge/Somerville and Fenway/Longwood/Brookline. Some of these trips need to happen around 6 am. That trip on the T requires multiple transfers and takes about an hour. On the weekends or at odd hours it takes even longer (especially when you have weekend bustitutions). In an Uber/Lyft, it typically costs $5 - $15 (Pool/Line) and takes about 20-25 min. Of course I take Uber/Lyft 90+% of the time I'm making these trips, even though I theoretically could take the T and in a world without Uber and Lyft where I had to make these trips I likely would take the T. I greatly benefit by using Uber/Lyft in these situations. There's nothing wrong with that, and Uber/Lyft should be celebrated for filling in the gaps in T service. Criticizing this Uber/Lyft trip because I'm "causing traffic" is ridiculous.

In the off peak times if we don't have enough demand to justify running 40' vehicles, something like Lyft is indeed great.

I do wonder if the Chevy Bolt and Tesla Model 3 are good enough, cheap enough, and starting to become available enough that we could start to require electrification of services like Lyft.

 

[Joel N. Weber II]

Uber is implementing a new system called express pool, where people walk to pick up points for a more efficient route. It's good in theory, but I swear every single time the computer directs pick up points to be in the middle of intersections. The middle of intersections where it's dangerous to stop and wait for passengers. I wish cities and Uber could work together to create special ride-share spaces to make this process more efficient.

Would loading zones not specifically designated for ride hailing services potentially be good enough?

 

[tysmith95]

Would loading zones not specifically designated for ride hailing services potentially be good enough?

Probably, although you might have issues with the loading zones being filled with delivery trucks.

I think it would be helpful to have Uber specific spots in places like South Station or Back Bay Station. Those areas have basically turned into shitshows during busy times. There shouldn't be any metered parking next to those places. It's mostly Uber/Lyft drivers, but you also have people picking up family at South Station.

One advantage of Uber is that you wouldn't have drivers waiting for long at the pick up zones. It's not like taxi drivers which sit at the stands and wait for a long period of time until they get a passenger. Ubers tend to be used more efficiently.

Another pro to Uber is that the drivers can pick up everywhere. Before every time a Cambridge taxi got a trip outside of Cambridge (which i'm sure happens more then half of the time) they would have to drive back to Cambridge to get there next fare. Now Uber/Lyft drivers can take a passenger from say Cambridge to Southie, and then pick up their next passenger in Southie. Used right it's more efficient.

I've always wondered if Boston and surrounding inner suburbs could merge their taxi forces to better compete with Uber/Lyft, although it's probably too late for that. There is no logical reason why Cambridge/Somerville/Medford/Malden have different taxi forces other then outdated governmental bureaucracy.

^ I'm with you on most of this too...

...but if you've spent a lot of time on the transit part of aB (and other similar discussions) you must know that at some point it's just not possible to get people to or from a big event in a reasonable amount of time, in cars, without a big congested mess.

At the end of a Celtics game, there will never be a way to get a meaningful number of fans into cars close to the Garden in a short amount of time. Even if there are four passengers in each car.

The constraints are geometric. Because you still need something like 2 feet of curb for 60+ seconds for every person who wants to depart in a car. And if the bar is closing,
count on like 2 minutes per person instead. (but yes I recognize that was only one of the several points you were making).

Luckily the majority of big Garden or Fenway events do not end during rush hour so it only hurts traffic in a small area. No matter what the city does those are going to be shitshows. 20 thousand (at the garden) or 37 thousand (at Fenway) people leaving at the same time will always create problems. Although in the Gardens case the construction on causeway plus the bike lanes (I know I know) has made that area much tougher on game nights for drivers.

 

[bakgwailo]

I agree with everything you’ve said.

What id add however is that I think one of the sticking points is that the marginal private cost of many car trips is lower than the marginal social cost of those car trips. This is true of both conventional and ride sharing trips. What makes uber in particular especially problematic is that at the moment that marginal private cost is also heavily subsidized by organized capital, which is doing so in order to starve other modes of resources so that uber will be increasingly attractive in relative terms - but not absolute terms - over time.

I think this is the biggest point - Uber/etc are cheap now, but that is because they are intentionally undercutting the market fueled by large VC injections and don't care about running at a loss - this can't last forever, and eventually something will have to happen for them to get profitable. The big gamble having been self driving cars (so they could fire all of their human drivers), but, that bet seems like it won't pan out quite as they anticipated.

 

[tysmith95]

I think this is the biggest point - Uber/etc are cheap now, but that is because they are intentionally undercutting the market fueled by large VC injections and don't care about running at a loss - this can't last forever, and eventually something will have to happen for them to get profitable. The big gamble having been self driving cars (so they could fire all of their human drivers), but, that bet seems like it won't pan out quite as they anticipated.

I think the biggest reason they're running a loss is because of all of their expansion attempts, legal fees, and self driving investments. If they were a leaner company and only focused on ride dispatch they could easily run a profit. They do take 25-30% of drivers fares, I don't see how that isn't profitable.

It's not like they are loosing money on each ride.

 

[sm89]

I think the biggest reason they're running a loss is because of all of their expansion attempts, legal fees, and self driving investments. If they were a leaner company and only focused on ride dispatch they could easily run a profit. They do take 25-30% of drivers fares, I don't see how that isn't profitable.

It's not like they are loosing money on each ride.

I paid $10.32 in San Francisco for an express pool (that wasn't matched with anyone) from the Castro to SFO.

I had the 40% off promo that I've gotten every single week for the last 3 months.
There was a $5 toll that was included in the fare already.
The driver was paid by the minute and by distance (unaffected by promos and tolls)

Uber loses money on lots of rides.

 

[tysmith95]

I just know from the 25k or so that I made from doing Uber part time uber charged passengers 34k.

I think what Uber is doing is subsdizeing pool and overcharging UberX. Their only costs really are insurance, credit card fees, app development, and overhead. I don't understand why they're losing money.

 

[bakgwailo]

Not to mention the driver promos and guaranteed rates to entice drivers at certain hours. As for cost - I think you are vastly underestimating the cost of software engineers, r&d, servers, hosting/isp costs, PCI compliance, etc - not even getting into r&d like self driving cars, which, they got into a bit of trouble over for poaching Google's tech for it. Add on marketing, support (both consumer and driver), etc, it all adds up quite a bit. Again - they float it and are OK as of now because of the vast amount of VC they are receiving to stay afloat while enticing both drivers and consumers to use their services.

 

[tysmith95]

Not to mention the driver promos and guaranteed rates to entice drivers at certain hours. As for cost - I think you are vastly underestimating the cost of software engineers, r&d, servers, hosting/isp costs, PCI compliance, etc - not even getting into r&d like self driving cars, which, they got into a bit of trouble over for poaching Google's tech for it. Add on marketing, support (both consumer and driver), etc, it all adds up quite a bit. Again - they float it and are OK as of now because of the vast amount of VC they are receiving to stay afloat while enticing both drivers and consumers to use their services.

I'm including the promos, anyway I would think that the immense size of Uber would lower the cost of developing the app. Economies of scale with regards to most of the development aspect. Google is able to develop an advanced mapping app and they give it out for free. With support they use low cost overseas labor, i'm sure it's not too expensive. Self driving cars and R and D, yes that's expensive. Marketing too.

 

[bakgwailo]

I'm including the promos, anyway I would think that the immense size of Uber would lower the cost of developing the app. Economies of scale with regards to most of the development aspect. Google is able to develop an advanced mapping app and they give it out for free. With support they use low cost overseas labor, i'm sure it's not too expensive. Self driving cars and R and D, yes that's expensive. Marketing too.

Lol. Google doesn't give maps away for free - everything everyone does on it goes to more data for better advertising, which is what they make money off of (same for search and other 'free' products). Also, economics of scale reducing the price of the app? What? Your looking at anywhere from 300-500k+ cost to the company per developer, and I would guess that they are on the higher end of that given talent competition and the complex routing algorithms (among other things) that they need to support. If anything the larger a company's engineering team grows its costs go up even more as one loses quite a bit of nimbleness vs the wild wild west startup phase.

 

[tysmith95]

Lol. Google doesn't give maps away for free - everything everyone does on it goes to more data for better advertising, which is what they make money off of (same for search and other 'free' products). Also, economics of scale reducing the price of the app? What? Your looking at anywhere from 300-500k+ cost to the company per developer, and I would guess that they are on the higher end of that given talent competition and the complex routing algorithms (among other things) that they need to support. If anything the larger a company's engineering team grows its costs go up even more as one loses quite a bit of nimbleness vs the wild wild west startup phase.

I'm saying that the costs of developing a ride-share app with millions of users won't be much more expensive then developing a ride-share app for thousands of users. Sure the server costs are higher but the development costs shouldn't be much different.

Of course I wonder if Google Maps/Waze is a way for Alphabet to get data to help out with Waymo. Google Maps/Waze are both far ahead of Ubers navigation, so I have more confidence in Waymo releasing self driving cars first. We'll see though.

 

[bakgwailo]

I'm saying that the costs of developing a ride-share app with millions of users won't be much more expensive then developing a ride-share app for thousands of users. Sure the server costs are higher but the development costs shouldn't be much different.

That is... completely not true.

Of course I wonder if Google Maps/Waze is a way for Alphabet to get data to help out with Waymo. Google Maps/Waze are both far ahead of Ubers navigation, so I have more confidence in Waymo releasing self driving cars first. We'll see though.

Considering Uber is in court right now with Waymo/Google/etc for stealing trade secrets to accurate their own self driving car, I would say thats probably accurate.