Friday, November 22, 2019

Ari and Miles rate Boston Magazine's "40 Ideas"

It’s transportation week here in Boston, apparently, what with the Globe Spotlight series dropping and Boston Magazine publishing their list of 40 ideas for transportation.

I started reading the 40 ideas piece and with each one had a reaction. Mostly “yeah,” “mm-hmm,” or “what, are you out of your goddamn mind?” Then I realized: I should rate these. On a scale of 1 to 10. On my blog. Because who wouldn’t want to read one person’s ranting about other people’s ideas, rated on a 1 to 10 scale.

Well, a lot of people. But what if … two people each rated the ideas, on a 1 to 10 scale. That’s a little more interesting. Certainly more fun to read. And when I think of rating transportation on a 1 to 10 scale, I think of one person: Miles “Miles on the MBTA” Taylor (well, now that he’s in undergrad in Philly, he’s going by “Miles in Transit” but we know where his loyalties lie.)

That’s right: guest co-blogging this post is the one and only Miles on the MBTA.

GET EXCITED. Set your snark detector to high.

A few notes before we dive in. First, these are listed in the order they appeared in the Boston Magazine story (on several different pages), not based on our rankings (so you don’t have to jump around, note that the magazine has the ideas grouped by mode, and I'll link to each page). I set some ground rules: points for (or against) creativity, feasibility, is it a “big” idea, and not “staying in your lane,” i.e. if someone works for an organization which advocates for x, and their proposal is to do x, it’s not as exciting. Also, it's worth noting that Miles wrote his ratings and responses without looking at mine.

So, we’re off. Thanks, Miles! And somehow this is under 4000 words (I should stop typing!).

Group 1: Cars

1. Congestion modeling as an investment (John Fish)

Ari: 9. Good idea, coming from a business leader. Good framing. One question: what do we do with the money? Does it go all towards transit? What if it was just put in an account and divided by the number of people in the state and returned as a tax rebate a la Canada’s carbon tax? Is this a good idea? How do we implement?
Miles: 8. Congestion pricing is awesome and Boston needs to do it. My one qualm: it’s really tough to find alternate, non-driving ways of commuting from low-income gateway cities like Brockton or Lawrence when the Commuter Rail is so inadequate. There could be equity issues here.

2. Tax garage parking (Ari!)

Ari: 10. Perfect. Do it. (But actually, a good idea, lots of precedent elsewhere, and, yeah, I’m going outside my usual transit “lane.”)
Miles: 10. Ew, Ari Ofsevit wrote this one—I can’t stand that guy. But seriously, this is a no-brainer, and every other major city has done it already.

3. Be less of a Masshole (Kenny Young)

Ari: 4. Fine idea, but doesn’t really do much. Sure, calming down is good. Looking at photos on your phone, however … demerits for that.
Miles: 3. Um, okay. Seems pretty fluffy to me. And aside from the phone use while driving (which doesn’t seem like a great way to “save transportation”), the bit about “singing at the top of your lungs” perpetuates the fact that it’s just you and your car—i.e. the geometry problem that got us into this mess in the first place. I like his title, though: “Personal driver, Kenny’s car.” That’s pretty funny.

4. Make drivers suffer (Sue O’Connell)

Ari: 9. Yes! Pedestrianization! Make it hard to drive! Although free transit is fraught.
Miles: 8. Pretty much a grander version of number 1, and it sounds amazing. I love the idea of having a city as painful to drive in as Amsterdam. O’Connell even addresses the potential equity problems with higher fees for having cars. It’s a lot, though, and pretty much a utopian vision with the current administration.

5. Late night transit (Garrett Harker)

Ari: 8. Make Boston more of a late night city? Great. Transit for late night employees? Great. Doing this through TNCs? Not sure what happens when they have to make money someday.
Miles: 6. Yeah, it would be nice to live in a city that doesn’t go to bed at 8 PM. Using a more vibrant nightlife to fuel 24-hour transit is an interesting idea, but I think the latter has to come before the former. And while it’s neat that he partnered with Lyft to get his workers home cheaply, early-morning MBTA buses get too packed to do something similar for them.

6. Autonomous cars for all (Lee Pelton)

Ari: 1. Can I go to zero? Can I go negative? Give me a break. We can all ride in an autonomous car driving around the city because there’s so much room on our roads. Seriously, Emerson? Has this guy ever looked out the window? Autonomous cars are cars. Cars in the city are bad. Please. I’ll keep him at 1 because at least he admits this is decades away.
Miles: 1. A Celtics game at TD Garden lets out. You have 20,000 people trying to get home. As the line of 20,000 individuals vehicles gets backed up along Causeway Street (and Commercial Street, and Atlantic Ave, etc. etc. etc.) waiting to pick everyone up, those autonomous cars suddenly don’t look so efficient, do they? What a shame.

7. Smarter tolls (Chris Dempsey)

Ari: 8. Good idea, should be implemented, but points off for staying too much in your lane. In other words, we know what to do? Now, how do we do it?
Miles: 9. It’s very much inside the box, but using existing infrastructure to cheaply instate congestion pricing is a great idea. The Tobin Bridge is a good place to try it, but the plan should come with a bus lane for the 111—goodness knows it needs it.

8. Fix potholes (Ernie Boch, Jr)

Ari: 2. If you don’t like potholes, move to Arizona or somewhere that doesn’t have freeze thaw cycles. Or start paying more for the roads, you Trump-loving sycophant.
Miles: 2. Are the potholes ruining your beautiful expensive cars? I’m playing the world’s smallest violin for you. Not a 1 because buses drive on the roads too, so at least they would benefit too.

9. Get cars off the road (Kristen Eck)

Ari: 10. Simple, and coming from the traffic reporter. More people without more roads = more traffic.
Miles: 6. It’s a good mindset to have, but the implementation is really vague and doesn’t do anything to fix the alternatives to driving. Sure, we can subsidize MBTA passes for employees, but if the Red Line breaks down for the 15th time that week, how valuable will those passes look?

10. Worcester-to-Boston (Mary Connaughton)

Ari: 8. Fixing the Turnpike outside of Allston is a good idea. But fixing the Worcester Line would do a lot more. People on Twitter think this could be done with toll revenue. Both, and?

Miles: 6. Kind of a mishmash of ideas here, some good, some bad. Improving Worcester Line service is a no-brainer, but keep the added service going past the time of construction, unlike what this entry seems to suggest. Fixing the monstrosity known as Newton Corner, fantastic. Increasing parking at Commuter Rail stations...fine at some of the more middle-of-nowhere ones, I guess. But building I-90 at grade just to make the project go faster? Hope you like diesel fumes!

11. Fix signage (Scott Ferson)

Ari. 2. This guy doesn’t seem to like the MUTCD. He rails against control cities. And he doesn’t understand that the exit number sign already indicates whether an exit is a left-hand exit or not. Not sure what this would do for congestion even if it made him feel better.
Miles: 3. I mean … whatever, man. Massachusetts has its share of crazy signs, but a lot of this comes down to the insane interchanges rather than problems with the signage itself. I’m not against the idea, per se, but its inclusion just feels like filler.

12. Give commercial vehicles more space (Ana Cristina Fragoso)

Ari: 5. Sort of a fine idea, I guess. More and better loading zones would make sense. Moving some stuff to rail? Fine. Ferries? Meh. More important are probably local loading zones, and getting people to stop ordering stupid shit from Amazon Prime and getting single rolls of toilet paper delivered by cars. We need a bit less Veruca Salt ("Don’t care how, I want it now!") syndrome and a bit more walking half a mile to the corner store.
Miles: 5. Points for being a unique consideration (it’s not something I’ve ever thought about). I like the idea of focusing on other ways of getting freight into the city, although it is going to be a tough sell to dedicate entire roads to commercial vehicles.

13. End distracted driving (The Rev Laura Everett)

Ari: 10. Yes! Put your goddamn phone away. Not out of your hand. Not a glowing rectangle next to your face. No. Away-away. In the glove box until you get to your destination. Look at the directions before you start driving, like we used to do before we had phones. If you don’t know where you are going, you shouldn’t be on the road.
Miles: 8. Yeah, okay, let’s do it! This is by no means groundbreaking, but any way to eliminate pedestrian deaths is a good thing.

Group 2: Buses

14: BRT (Julia Wallerce)

Ari: 8. BRT is great, but a) not going to solve all of our problems, and b) very much in Julia’s lane. But, yeah, put in some goddamn bus lanes already.
Miles: 8. BRT is a fantastic way to improve capacity in Boston’s overcrowded bus system. While this is a rosy look (how many Boston streets are wide enough for center-running BRT?), it’s still a great system to strive for.

15. Better interior bus design (Zuleny Gonzales)

Ari: 10. I was on the 70 and an already late-and-full bus was delayed several minutes while the driver helped a mobility-impaired person safely board. Plus the driver had to deploy the ramp by hand because it was broken. Making buses work better for people with mobility issues makes buses work better for everyone (goes for trains, too).
Miles: 10. Absolutely needed. If we can make riding the bus easier for everyone, especially those who are disabled, then not only will more of those riders take the bus, but speeds will be increased—I can’t be the only one who internally groans when someone in a wheelchair gets on and I know we’ll have to endure the two-minute strap-in process. It’s not the person’s fault, it’s the vehicle design.

16. Make way for buses (Jarrett Walker)

Ari: 8. Bus lanes and bus lane enforcement. Not a bad idea at all. Getting the police to care, though …
Miles: 10. I’m not a Jarrett Walker fanboy, I swear! Seriously, though, bus lanes are sorely needed in this city, along with proper enforcement. It’s definitely inside the box, and many bus lane projects are already happening, but there’s no denying that if we want to fix Boston’s bus problem, bus lanes are objective number 1.

17. Blue Hill Ave buses (Quincy Miller)

Ari: 10. Blue Hill Ave is the top corridor in the state for BRT. Tear out the concrete medians, put in BRT, and plant some trees.
Miles: 10. This is one of the only corridors where the treatment in number 16 would work. Plus, the 28 is the MBTA’s highest-ridership bus route—what better place for a BRT line? Redundant to the others, but I can’t argue with true BRT on Blue Hill Ave.

Group 3: Rail

18. Electrification (Adam Gaffin)

Ari: 10. Adam gets it.
Miles: 10. Everyone knows this is sorely needed. It’s been talked about to death already, and it seems like the only people holding back are those up at the top. Let’s do it.

19. Build the Grand Junction (C.A. Webb)

Ari: 10. The Grand Junction is the best ROI for a rapid transit “crosstown” link. North Station-"Cambridge Crossing"–East Cambridge–Kendall–Cambridgeport–Allston. It doesn’t sound like a busy transit line today, but with ongoing development, that order of stations may soon read something more like North Station–Kendall–Kendall–Kendall–Kendall–Harvard (Allston). With the Allston project, the time to do this is now.
Miles: 8. More transit to Kendall Square is really important, and the Grand Junction is the only existing rail asset there besides the Red Line. I’ve never been a giant fan of the whole “run Worcester trains via the Grand Junction to North Station” thing, though—you end up with a reverse-branch that could potentially limit the number of trains serving South Station. Instead, how about a frequent shuttle from Allston to North Station?

20.  Gondola!!!1! (Roger Berkowitz)

Ari: 1. Bus lane!!!1! (Or if you want a subway, put trains in the Silver Line tunnel. Remember, the gondola doesn’t work just because you say words.)
Miles: 1. THE SEAPORT GONDOLA IS BACK!!!!!! Look, we have a "subway tunnel service" to the Seaport—it’s the Silver Line, and yes, it is insanely overcrowded at rush hour. Convert it to light rail and suddenly you’re going to see a lot more capacity for that "subway tunnel service." And reopen the Northern Ave Bridge for car traffic? Ever heard of induced demand?

21. Big rail (Kathryn Carlson)

Ari: 10. Yes, regional rail! And let’s remember, those cost estimates are … high. But $30 billion (less, though, because this includes the cost of buying rolling stock, which the T has to do no matter what propulsion system is used) over the next 10 years is $600 per person per year inside of 495, much of which you could get from the feds, to completely transform the regional transportation system. It’s really not that much.
Miles: 10. Another one that’s been talked to death, and pretty much every transit advocate agrees this needs to be done. Carlson even brings up the hefty cost and defends it.

22. Better PR (Max Grinnell)

Ari: 4. I mean, better PR would be nice, but it won’t make the trains run on time.
Miles: 2. I don’t think Max Grinnell reads the MBTA’s Twitter replies. When they’re stuck in a line of trains trapped behind a broken signal, people don’t feel any better by looking at cutesy PSAs. I guess it’s a nice thought, but it’s nothing more than that: a nice thought.

23. Widett transportation hub (Tom Tinlin)

Ari: 2. Oh, it’s Track 61 again. Like the gondola, it also still doesn’t work just because your mouthparts move.
Miles: 2. Points for originality, but I’m not quite sure what the goal is here. "A Widett Circle station could let commuter-rail passengers from the southeast get to Back Bay or Ruggles without going all the way to South Station." What? How? Track 61, I guess? And Tinlin talks about connecting the Red Line and Commuter Rail here. So there’d be a new Red Line subway station? With a 700-foot connection from Dot Ave to a new Commuter Rail station? This had better coincide with a redevelopment of the area, or else you’re going to end up with a lot of wasted money.

Group 4: Bikes

24: Franklin Park Bike Hub (Elijah Evans)

Ari: 7. This would be great, having Franklin Park be more usable by bicycles (and overall, some of the paths there are overgrown, it’s kind of weird). And I do like the idea of making roads nearby more bike-friendly. But it’s not going to solve the regional transportation issue, even if it helps some.
Miles: 7. This seems like mostly a PR campaign to make local residents more aware of the ease of biking through Franklin Park, which is not a bad thing. Importantly, though, it would have to come with better bike infrastructure on surrounding roads—the park itself isn’t really gonna get you anywhere.

25. Bike the last mile (David Montague)

Ari: 6. A fine idea, although congestion isn’t always the last mile. This is also very much in the lane for David Montague, who owns a folding bike company.
Miles: 4. A novel concept for sure. But think about a cold winter day: you’re in your warm car, and you come to the “park-and-bike” lot—do you really want to get out of your private vehicle and get on your freezing bike through the pounding snow instead? Probably not. This isn’t a problem with a typical transit-based park-and-ride system.

26. Share the road (Jenny Johnson)

Ari: 9. Give bicycles better facilities and they’ll be more law abiding? Yup. Go to Europe and see how design begets ridership? Sure.
Miles: 10. Johnson’s anecdote about bikers waiting at the light in Vienna really drives home the idea that people will follow the rules when they’re fair. In Boston, bikers get so little compared to cars—of course they’re going to rebel and run the light. Give bikers the safe corridors they need, and you’ll see a lot more harmony on our roads.

27. Helmets for Hubway (Joanne Chang)

Ari: 3. I mean, fine, but helmets don’t keep people from riding BlueBikes. In fact, these upright, Dutch-style, stable bikes are probably the ones with the least need for helmets. This does basically nothing. (I’m surprisingly libertarian on bike helmets.)
Miles: 3. No one wears helmets in Amsterdam, because bike fatalities are ridiculously low, because the streets are designed for them. While helmets may make you feel safer here, they’re not doing anything to actually solve the transportation problem for bikers.

28. Connect the paths (Juliet Eldred)

Ari: 10. NUMTOT! But even if it wasn’t NUMTOT-in-chief @drooliet, it’s also a great idea. The real last mile issue with bikes (see above) is that the last mile of a bike trip is often unsafe. So people drive. Let’s cover the city with a web of bike lanes, instead of traffic lanes.
Miles: 10. As a proud NUMTOT, it is so cool to see that Juliet made the cut here. And her idea makes so much sense: why do we have these fancy separated bike lanes that connect to lines of paint on the road? We can’t wave a magic wand and instantaneously get separated bike lanes on every street in the city, but we can at least try to connect the ones we have.

Group 5: Other

29. Podcars (Bill James)

Ari: 1. No.
Miles: 1. Aahgahgaghhhghhh.

30. Road diets (Joe Curtatone)

Ari: 9. Yes. Now, can you call over to Marty and get Boston to stop shitting all over this?
Miles: 10. Putting aside the fantastic points, ideas, and examples put forth here, it’s amazing to see that this was written by an elected official, the mayor of Somerville. It gives me hope that there are people on the top who know what’s going on and want to fix it.

31. Flying cars! (Gwen Lighter)

Ari: 1. Also no. Not as crazy as podcars because they aren’t low-capacity elevated railroads, but has anyone talked to the FAA about this?
Miles: 1. What? Is this a joke? Am I reading an Onion article?

32. Fare equity (Tanisha Sullivan)

Ari: 8. Fare equity is great. Implementation is, well, not as easy. Price discrimination allows us to do things like keeping certain services from being even more overrun when they are already near capacity. Without this, we get a tragedy of the commons with poor service and long wait times like, well, like the roads at rush hour. (Off-peak pricing, now, there's an idea!)
Miles: 7. This is a good, moderate look at how to improve access for people. Free MBTA fares for students and seniors is more realistic than making the whole system free, while encouraging flexible work hours would ease rush hour traffic a bit. Nothing groundbreaking here, but some fine ideas.

33. Ferries (Alice Brown)

Ari: 5. Ferries are useful for a few demand corridors, but not really a game-changer regionally. And this is very in-her-lane for the head of Boston Harbor Now.
Miles: 4. Utilizing our water assets and creating new ferry routes is good in theory, but who’s taking a ferry from East Boston to Long Wharf when the Blue Line exists? Maybe the people in the expensive luxury homes near the water, but not much else. East Boston to the Seaport is better, but even then, bus lanes in the Ted Williams for the SL3 would help here, too. I just don’t want Boston to end up with a repeat of the NYC Ferry http://secondavenuesagas.com/2019/10/06/its-time-to-stop-subsidizing-the-nyc-ferry-fares/.

34. Think outside the box (Jim Canales)

Ari: 7. Let’s be a world class city with outside the box ideas is fine, but vague.
Miles: 7. This is vague as heck, but the general call to [waves hand] think outside the box is solid.

35. Dockless scooters! (Lama Bou Mjahed)

Ari: 2. Small-wheeled death traps which are expensive and still lose money. Pass.
Miles: 2. This is listed under "Big Ideas"? Dockless bikes and scooters are already being implemented around the country, and even in the Boston region! It might take a few (a few) cars off the road, but this isn’t the big solution we’re looking for. And Bluebikes already has great coverage that’s expanding.

36. Robotaxis (Karl Iagnemma)

Ari: 1. Hooray, more cars! Using technology we don’t have! Magic!
Miles: 2. Okay … I guess it’s not hyping these things as a full-on takeover, but rather a low-cost last-mile solution that’s apparently being implemented in Vegas already. One point added for being realistic.

37. Teamwork (Patrick Sullivan)

Ari: 7. Getting people together is great. But as the Globe points out, if they are all subsidizing driving, it doesn’t help that much.
Miles: 6. Working with companies to open up new transportation options...okay, I can roll with that. But the example of the North Station/Seaport Ferry shows the flip side of it: it’s great for members of participating companies, but for members of the public, the ferry is near-unknown, and you can only buy tickets online in advance. These partnerships need to ensure that the public can benefit as well, not just the private companies involved.

38. Listen to users (Ambar Johnson)

Ari: 8. Not necessarily a big idea, but having a better way to have public feedback with the agency and cities and towns, especially outside of public meetings attended by old white people, would be good.
Miles: 10. I mean … it’s probably the least revolutionary idea on here, but it’s also one of the most obvious. The users know best, and the piece summarizes it well: "Listen and trust their judgement, and just implement it."

39. Invest in Gateway Cities (Tracey Corley)

Ari: 9. Yes. Gateway cities are an opportunity to build more housing near transit, and build more walkable, livable areas people want to live. Millennials don’t necessarily want to be in Boston (or Cambridge, or Somerville) but to live in a walkable, urban community with amenities and good transit. Gateway Cities provide this opportunity.
Miles: 8. The idea of gateway cities being self-contained, you know, cities that are disconnected from Boston is such an alien idea, so this gets major points for being outside of the box. And mention RTAs and you’ve got me hooked—the transformation of gateway cities needs to be accompanied by improvements to their bus systems, which vary wildly in quality. Regional Rail isn’t mentioned at all despite being a key factor in gateway city revitalization, though.

40. Mayor of 128 (Amy Dain)

Ari: 6. Fine, but reducing car use along 128 will be quite difficult unless we nuke all the development along 128 and rebuild it very densely around transit stations along 128.
Miles: 7. Sure, I guess. It would be good to have someone looking to the future of a corridor that is really screwed up right now. Smarter future development will at least prevent more traffic from using 128. As a big final proposal though, this is a little underwhelming.

Friday, November 1, 2019

What's up with the taxi queue at Logan?

Early Monday morning, on Twitter, Cambridge City Councilor Jan Devereux posted this photo of the taxicab line at Logan Airport:
Then she went to check out the Uber/Lyft line and it was no better:
By 1:30, others reported that the cab line was an hour long.
What's going on here? Where are all the taxis? Where are all the Ubers? Why are people arriving at the airport after midnight then waiting an hour for a ride home? What can we do about this?

It's not really a simple answer. Please, though, follow me down a rabbit hole.

Logan airport is not a hub airport (okay, not really: Delta and especially JetBlue are doing their best to create a hub, but it still is mostly an origin-destination airport). This is a matter of geography: because it is tucked away at the northeast corner of the country, the airport mostly handles passengers flying in one direction, south or west, either on direct flights, or to other domestic hubs (although international travel has increased markedly in recent years, and JetBlue even has a bank of later domestic flights to accommodate connecting passengers).

Boston's flights can be grouped in to three main types: short haul flights to non-hub cities, short-haul flights to hub cities, and long hauls, both domestic and international. Only the first group operates without geographic constraints which dictate that flights can only arrive and depart at certain times of day. Non-hub short-haul flights, mostly on JetBlue and Delta (to places like AUS, BUF, RDU, MKE) are spread more evenly throughout the day, because they don't have to make connections at hubs. Hub-based flights within the eastern half of the United States (say, as far as MIA, DFW and MSP). These flights account for many of the early departures, but few leave Boston after early evening. They don't arrive in Boston until mid-morning, but are the bulk of the last flights arriving later in the evening. Because of these travel patterns, Logan has a lot of early and late flights: in the morning, people want to get to a connecting hub for the first "bank" of connecting flights, and in the evening, flights have to leave the connecting hub after the last bank has occurred.

(Not all airlines utilize banking at hubs—Southwest most notably runs continuous hubs—and there are pluses and minuses to each method, about which I won't go into too much depth here. But basically, banking decreases fleet utilization and increases congestion at hub airports since flights arrive and depart all at the same time, followed, in some cases, by periods of relatively low flight activity, but passengers have much shorter connection times at hubs since connections are coordinated. It's actually something like a pulse system for buses, except that airports have finite numbers of vehicles which can arrive and depart at any given time, and longer dwell times. As airlines have consolidated, hubs have grown and seen more frequent banks such that they are now closer to continuous operation, especially at large airports like Atlanta and Chicago. See how this is a rabbit hole? Also, when I say people want to connect to the first bank, I mostly mean airlines, no rational human being wants to be on a flight departing at 5:15 a.m.)

Then there are longer-haul flights. Transcons have to deal with time changes, cycle times, and the fact that flights generally don't depart or arrive between 1 a.m. and 5 a.m. local time. So they arrive from the West Coast either as redeyes between 5 and 10 a.m., and then turn to depart between 6 and noon, or as day flights, arriving in the afternoon or evening. Thus, in addition to the overnight lull, there are basically no transcons which arrive in or depart Boston during the midday. International flights (mostly TATLs) generally are redeye flights going out, leaving Boston in the evening, and return during the middle of the day, arriving in the afternoon. All told, Boston has demand throughout the day, but particularly high demand for flights arriving later in the evening.

For instance, O'Hare and Atlanta, the two busiest airports in the country (by number of aircraft movements), have 29 and 24 flights scheduled to arrive after 11 p.m., respectively. Boston has 39. After 11 p.m., Boston is basically the busiest passenger airport in the country (about tied with LAX), and possibly the world (since in many countries, airports have noise regulations which limit late night flights). Yet at the same time, there are few flights departing Boston. The airport has only 14 scheduled departures after 11 p.m., and only six of these are domestic flights (all are JetBlue E190s, so they're small planes). There is basically zero demand for passengers to get to the airport late at night. This creates a demand imbalance for ground transportation: there's a lot of demand to leave the airport at midnight, but almost no one who wants to go there.

Here are some charts of the approximate number of airline seats arriving and leaving Boston. I adjusted for the typical arrival time at the airport (60 minutes for domestic, 120 for international) and assumed it would take 30 minutes for the average international passenger to clear customs.



(Note, the seat numbers for arrivals and departures don't exactly match because each day at Logan is not identical and this was a snapshot of a day. This is total seats available, not total passengers, and is also a rough estimate based on plane sizes at different times of day, but should show general trends well.)

So, it's clear that there is a good deal more demand to get to the airport in the morning, which doesn't even out until around 9 a.m., various points of imbalance during the day, and then, starting around 7 p.m., significantly more demand to leave the airport.

A couple of personal anecdotes can illustrate this. The first illustrates the imabalce in the morning. Back before the Big Dig was completed (and, in fact, pre-Silver Line, and back when the airport had half the traffic it has today), when I was going to college, getting to the airport was, perhaps, worse than today. It involved both the elevated Central Artery and then the Sumner/Callahan tunnel complex: a trip to or from the airport to downtown could take an hour (which is, of course, not much different than it is now). For several years, however, the Ted Williams Tunnel had been open to commercial vehicles only. The connecting highways were not yet complete, so this was a way to keep the local streets from getting overrun by people trying avoid the congestion.

From Newton, where I grew up, my father came up with a solution, especially for early morning departures. He would drive me to South Station. Rather than risk the airport traffic, I'd get in a cab for the usually $10 or $12 trip under the harbor in the tunnel only cabs could use. Early one morning, I got in a cab at South Station and told the driver I was going to the airport. He quoted the fare: "$20" and didn't turn on the meter. What I should have done is say "I'll pay you whatever the meter quotes at the end of the ride, so it's in your best interest to turn it on now," but I was 18 and hadn't quite figured that out, so at the airport, I paid him $20, noted his medallion number, and immediately reported it.

A month later my dad got a check for $20 from the Boston Police (which oversee taxi medallions), along with a note that the taxi driver had been given a stern talking to that he was never to refuse to turn on the meter for a trip within the city.

But I understand why the driver was reticent to take the fare. He would get to the airport and have two (bad) decisions. One would be to go to the taxi pool and wait in line for an fare back to the city: a long line, because there is much less demand going in to the city at 7 a.m. than there are taxis arriving at the airport. The other would be go cross back downtown without a passenger, but still incur the cost of driving, as well as the tunnel toll (which was one-way inbound at that point), with no passenger to pay it. Still, because of redeye arrivals from the West Coast in the morning, there is some traffic for cabs that do make it to the airport going back, although they have to cycle through the cab pool (or the Uber/Lyft pool) before their next fare. Even now, if you take a cab or a ride-hail vehicle to the airport in the morning, the driver is probably not particularly happy taking the fare.

The second example was an extreme example of the late-night issue. I was flying back from SLC and the flight was delayed several hours. Originally scheduled in at 11:30 (plenty of time to catch the Silver Line downtown and take the T home), the plane was more than three hours late, and didn't arrive in Boston until about 2:30. The airport was empty. Yet a 757 had arrived with 200 passengers, and no one was making the trip at that point to pick us up. So we all converged on the cab stand, but there wasn't a cab in sight. Immediately, people started self-pooling: it was clear that if we all took our own cabs, the line wouldn't clear for hours. "Who's going to Brookline?" "I'm going to JP, that's close!" "I'm going to Concord." "I'm going to Lexington, let's share." Every few minutes, a stray cab would roll up, three overtired strangers would pile in, and the line would get slightly shorter. I found a cab pretty quickly, but imagine those at the back of the line may have seen the sun rise.

Finally, here's a picture of the departures level at Terminal A at 8 p.m. this past week.


Delta's last domestic flight leaves at 7:45 (to MCO) and the last hub-based flight leaves at 6:56 (for ATL). Two international flights leave around 8:30. At this time, it's a ghost town. Note: if you're picking someone up at the airport in the evening, plan to meet them on the upper level, and note that Terminal A is a great place to wait without having State Police harass you. (Terminal B was pretty quiet, too.)

Here are scheduled arrivals and departures, by carrier, at different times of day:


Note: not all regional flights appear to be in this sample, for instance, there are no Cape Air flights shown after mid-evening, when Cape Air flies several late flights out of Boston, but these flights are minimal as far as number of arriving passengers is concerned.

Notice how departures peak in the morning, then lull in the midday, and then have a secondary peak in the evening, before domestic departures (except for JetBlue) tail off quickly after 7 p.m. International carrier flights are clustered arriving in the afternoon and leaving in the evening. But there is a clear imbalance for flights arriving and departing the airport.

In any case, this has been a problem for years, and it's a structural issue pertinent to Logan Airport based on the airport's geography on both a macro and micro scale. On a macro scale, the geography of the airport at the corner of the country means that, late in the evening, flights feed into it but don't feed out. On a micro scale, the airport's geography encourages taxi/app-ride/ride-hail use (I'll call these taxis, for simplicity). The constrained location means that parking costs are high, because demand for parking outstrips supply. The proximity to areas with high trip generation (downtown, and high density areas nearby) means that taxi costs are often significantly lower than a day's parking cost ($38) in the garage. The combination of these factors push many people to use taxis.

Much of the day, taxi supply roughly matches demand, and there is a minimal delay for these services. But this breaks down at the beginning and end of the day, especially in the evening. Once again, Logan's geography comes into play. While the airport is close to the city geographically, it is expensive to get to for a taxi driver. When there is high demand for fares back to the city at 1 a.m.—especially once buses have, for the most part, stopped running—there is negligible demand to get to the airport, or even East Boston in general, so to pick up any fare would require the driver to deadhead to the airport.

Unless a driver happens to pick up a stray fare to East Boston, this requires a driver to travel several miles, and to pay the cost of the tunnel toll. Once at the airport, there is no promise that the trip home will be lucrative enough to cover these costs. They might get a $50 fare to a far-flung suburb. But it might only be a $15 fare to a downtown hotel or, worse, to Revere or Winthrop, meaning a driver would then have to drive back home—likely through the tunnel—and foot the bill for the toll both ways. Moreover, this is the end of the day for most drivers: except on a Friday or Saturday night, there is only so much demand for rides after 1 a.m. For many drivers, the potential upside of getting a decent fare from the airport doesn't make up for the potential downsides, especially when the alternative is shutting off the app (or taximeter), driving home, and going to bed an hour earlier. There is a high disincentive to be in the last group of taxis at the airport: a driver might get one of the last passengers, but if not, there is not likely a job for several hours when the first redeye flights start to trickle in around 4 a.m. There is little incentive for taxicab drivers to go to the airport during this dwindling time, so demand is only met by drivers already in East Boston who need a fare back to the city.

This is not an easy issue to solve. It also shows why Uber and Lyft are basically just taxis: they are subject to the same supply and demand issues that affect the rest of the market. Alas, they're providing the same service: a chauffeured ride from Point A to Point B. And the airport is pretty much the only place that cabs still have a foothold, partially because matching passengers to specific vehicles is quite inefficient with large groups, like you might find at an airport. This has been a recent point of contention at LAX, which consolidated its taxi and ride-hail facilities away from the terminals (as Logan is planning). It hasn't gone particularly well to start, but I would venture to guess that there are similar supply and demand issues at LAX. (As several people have pointed out, the buses there are running much more smoothly, too.)

Of course, at LA, there is a temporal aspect to the complaints about #LAXit. Most of the issues are in the evening. Los Angeles has more balanced operations, with plenty of departures in the evening, both redeyes to the East Coast and transcontinentals, mostly TPACs but some TATLs as well. There is probably both more demand for cabs in the evening because LA is on the opposite side of the country as Boston, and also because most of the transcon redeyes leave before 11 p.m., and most of the later flights are international, which are fed more by connecting travelers and by passengers arriving at the airport much earlier. Thus, for the actual demand for taxis, there is a similar, if less pronounced, demand compared with Boston.

I scraped Twitter for #LAXit from the first few days and it seems clear that the issue is mostly in the evening.


But back to Boston. Here's what the Logan cab pool Twitter feed (because of course Logan's taxis have a Twitter feed) looked like last Sunday. There was a cab shortage by 8 p.m., and the late night arrival issue was foreseen by 10 p.m. It wasn't a surprise. It is a market issue.


Is this feed a bot? Certainly not! Note the wrong months, typos, misspellings, random numbers of hyphens and use of quotations. The one thing that seems constant is the use of the word "need" when the airport has a cab shortage (with various qualifiers like all-caps or exclamation marks). I'm pretty sure it's a guy standing by the taxi pool furiously typing tweets into the Twitter machine. So, I decided to scrape the feed (about 60,000 tweets), and code each tweet by whether it included the word "need" or not, to get a very rough estimate of the frequency of taxi delays at Logan.

From the charts of arrivals and departures above, we would assume that the airport would generally need cabs mid-afternoon, as well as in the evening. If we chart the arrival and departure relationship and the Twitter feed's need for cabs together, and we shift the need for cabs back 1:15 (probably due to actual behavior of arriving passengers), voila, they match pretty darned well.


A few other notes on the need for cabs:

  • There is more demand on Sundays (36% of Tweets include the word "need"), followed by Mondays, Thursdays and Fridays (28%), with the least demand on Saturdays (20%)
  • The cab shortage is generally higher in the summer than the winter (and highest in May, June, September and October, while lowest in December, January and February), and it has been particularly high this year. In fact, the only higher demand for cabs at the airport was in February of 2015. I wonder why.


TL;DR: it is definitely getting harder to get a cab at the airport, especially during certain times of the year and at certain times of day. And when flights are delayed, this is exacerbated. Departing passengers generally still get to the airport on time, but arriving passengers get in later, meaning the cab shortage is even more acute. Which leads to situations like the one which occurred earlier this week at the start of this post, which may have taken you nearly that long to read

So what can be done about it? Well, Uber/Lyft could use their surge features to increase the cost of a trip to the point where it would make economic sense for drivers to come to the airport at this time of day. But that might double or triple the cost of a ride, so while it is a very market-based solution to the problem, it is not consumer-friendly. It (and taxi starters) could do a better job of pooling rides, and moving the Ubers and Lyfts to the same site might make pooling easier there. A very low-tech idea might be to figure out where vehicles are going and, at high-demand times, put people into lines based on regions to help them self-pool. This might nibble around the edges on the demand side, but it doesn't help supply. It turns out that Lyft and Uber are not a magic panacea to mobility: they are subject to the same supply and demand issues as the rest of the world. And if the answer is pricing, it's not a great answer, especially since prices might have to go up significantly to provide enough supply or suppress demand enough to satisfy market equilibrium, especially after midnight when there are really no other options available.

One solution, I think, is that rather than trying to bring supply to the demand, we should move the demand to where there is more supply: get people, en masse, downtown. This requires a magical invention: the bus. While the supply of hire cars is not limitless (there are only about 1800 taxicabs in Boston, and probably far fewer are active today, and many more Ubers and Lyfts, but a finite number), it is significantly less constrained on the other side of the Harbor. Instead of trying to entice drivers to come to the airport, we could instead move riders to where are there are more cars. Most riders are going north, west or south of the airport, and need a ride through Downtown Boston anyway. The bus might not be that appealing to a traveler at 1 a.m., but neither is a 45 minute taxi ride. This basically takes what LA is doing, and extends it a couple of miles.

The MBTA runs the Silver Line until 1:15 a.m., which could easily bring passengers to South Station, where catching a taxi or ride hail vehicle is easier than at the airport. Massport and the T, however, do a poor job of advertising these late buses, both with signage telling passengers the hours as well as real-time information about the buses' whereabouts (important especially if you are unsure if the last bus has left). But these are easy issues to solve. The bus real-time data is available: I have no issue pulling it up on mbtainfo.com, for instance.

Not helpful, especially if you're trying to figure out if the last bus has left, or even deciding if it's worth waiting in a cab line versus waiting for the bus.

And Massport could put up static signage:

NO CABS? NO PROBLEM!
GET A CAB ACROSS THE HARBOR

MBTA BUS TO SOUTH STATION

SKIP THE LINE — SAVE MONEY

BUSES EVERY 8-15 MIN UNTIL 1:15 AM
ADDITIONAL BUS AT 2:30 AM

And as I've written before, with some minor schedule tweaks, the T could use the Silver Line 3 returns from Chelsea to supplement this service even later, and keep one bus in service to make an extra round trip to provide service until 2:30 when, on most nights, the planes have all landed. Alternatively, or in addition, Massport could continue its Back Bay Logan Express bus later in the evening and into the early morning (or even 24/7), providing late night trips to the taxicab-rich parts of Boston when there are cab shortages at the airport (and perhaps even direct service to large hotels in the Back Bay).

Even more, the Logan Taxi Twitter feed often includes this kind of Tweet:
By 10:00 most evenings, someone at Logan knows how many late flights are coming in. So, conceivably, the bus driver could be held on duty to make extra trips in the cases of delays. Massport, which already helps to subsidize the fares for the Silver Line, could shoulder the rather minimal cost of the extra trips. And passengers arriving at midnight would no longer have to face an hour-long wait for a cab, when there are many more options across the harbor.

The market has never provided enough cabs at Logan when they are needed, and at certain times of day, lines of cabs are the rule, not the exception. Assuming the market will take care of this has never worked, and it is unlikely that it ever will. If Massport worked with the MBTA, however, it could pilot a project to move people downtown and create a secondary taxi queue there, where drivers would be much more willing to go for a fare, because even if the demand had dried up, they wouldn't be out the tunnel tolls, mileage and time to show up at the airport.