I felt like going for a run today, but not a long run because it’s still cold outside and I haven’t been running in a while. I decided to run to my gym, do a short workout, and then run back home. I only live about 1.4 miles from my gym, so it’s about a 10-minute run. On the way out of the gym though, I found myself thinking, “I’m pretty tired. If I had brought my wallet, I might be willing to take the bus home.” This was immediately followed by a second thought, “It’s not worth $2 to go 1.4 miles anyway.”
This got me thinking about transit’s role for short trips on the way home:
- Why should a 1.4 mile trip cost the same as a trip across the whole city?
- Does anyone with a choice actually use transit for these short trips?
- How far would I have to drive to spend $2 of gas (or $1.75 in Columbus)?
- Should transit be priced differently?
For most people in Columbus, a car is a reality of life. I previously noted that the 2008 Census American Community Survey found 97.5% of workers in Franklin County have at least one vehicle available. So the cost of vehicle ownership and insurance are sunk costs for most people. Variable costs are what matter when deciding to make a trip by car or transit. These include gas and some percentage of maintenance, which amounts to 19 cents per mile according to the IRS. At that rate, you would have to travel 9.2 miles ($1.75 / $0.19 per mile) on a bus before it’s cheaper than taking your car (obviously results will vary based on your car’s MPG and the price of gas).
However, to go 9.2 miles on a bus could take a while, 44.5 minutes at the average COTA bus speed of 12.4 MPH in 2009. That is likely to be a lot longer than a car trip if you have that option available. So basically, most transit service (around the country, not just in Columbus) is too expensive for short trips and too slow for long trips. So why would people take transit? I could only think of a few reasons:
- No car available – Whether by choice or affordability, there is that 2.5% of workers who don’t have a car. This is remarkably similar to the 2.2% transit users reported by Kiplinger here, but I haven’t fact-checked that number.
- Monthly pass – If you have a pass or a Buck-ID, there is no extra per trip cost to ride transit, so you may be more likely to take it for short trips.
- Parking – If a destination has limited or expensive parking, that can change the time and cost equations for the transit versus driving comparison. In Columbus, that means transit is basically useful for going downtown or to campus, but most of those campus-based trips are using Buck-IDs anyway.
- Work en route – Some people would rather read or get some kind of work done on transit that they can’t do while driving.
- Partying – Transit makes a great designated driver if you’re going out drinking.
So what can transit providers do to make their service more appealing? There are lots of possibilities. They could and should try to speed it up to make it more competitive with cars. People are willing to sacrifice some time riding transit because they can do other stuff like read, but not too much time. They could try to sell more monthly passes, or create special arrangements with employers just like they do with students. They could give away a month or more of passes to people who agree to give up their car. They could offer free wi-fi.
I think one major potential solution though is a change in the pricing structure. Some kind of distance based pricing would probably be best in theory, but a per mile system is impractical given the typical fare collection system on a local bus service. A zone system could help some, but unless there are a lot of zones, it probably wouldn’t make a difference for most short trips. It also introduces new problems like added system complexity (for passengers and drivers) and still have high pricing for short trips that just happen to cross a zone boundary.
I’m wondering if reducing or eliminating fares on certain routes or at certain times of day would encourage people to take transit for short trips that are otherwise not worth the cost. I’m guessing, and this is just a guess, that most crosstown routes have extra capacity, or maybe most off-peak trips, or weekend trips. You could encourage more people to use transit at low-demand times or on low-demand routes by charging a different fare. This would encourage transit trips that aren’t to the CBD, where parking economics make transit more appealing.
The downside of such an idea is of course that COTA may be giving up revenue by charging less (or nothing at all). Consider though that fares only pay for 17% of operating funds anyway, or $13.8 million in 2009. I don’t have any numbers, but I would guess that since the current system is mostly designed to serve downtown, that most of those fares are monthly passes, student fees, or single trips to downtown. It may not be that big a financial hit to operate some free routes or time periods.
I would suggest phasing something like this in. Try free Sunday trips, or free late-night trips, or free crosstown routes, or some other low productivity route. Measure how it impacts ridership. Determine if it’s worth expanding such a policy or rolling it back. My ultimate goal is to provide as many trips per service hour as possible. I think this may be one of the most practical ways to boost ridership in a place with high automobile-ownership rates. What do you think of this kind of idea?
Great post, John, and a good look at some of the numbers.
The one thing that jumped out at me as surprising was this: “97.5% of workers in Franklin County have at least one vehicle available.”
That sounds incredibly high. Does that mean that a family household of two working adults and two 17 year old teenagers with part time jobs and only 1 household car is included as four works all with “one vehicle available” in that figure? If so, that sounds like it has the potential to be skewed and misleading.
I agree that pricing fares per distance sounds like a good idea, but I have to wonder how much confusion it would add to the system and what logistical issues would arise. I just returned from a trip to San Francisco last week and the BART is priced based on distance. I rode one stop over for $1.75, and also rode from Downtown to the Airport for $8.10. The tickets are purchased in advanced from an ATM-like system in the station and read at the turnstyles at your entry and exit points.
I’m not sure if that translates well to a bus system for entry/exit or if there is a better system out there for that.
I think that also has the potential to leave a lot of people confused on pricing for trips. If you’re going somewhere that you’re unfamiliar with, you don’t know what it will cost you in advance, which can be stressful for many people. Do I have exact change? Should I just drive and not worry about it?
I think your assumptions are correct. The title of the table is “WORKERS 16 YEARS AND OVER IN HOUSEHOLDS,” so I think it does consider the number of vehicles available in the entire household. This could indeed overstate vehicle availability. For example, I might be said to have one vehicle available, but in reality my wife needs it to get to work, so I can’t drive to work.
Nevertheless, the data still show that only 3.5% of workers in households with one vehicle available take transit. Less than 1% of workers in households with two vehicles available take transit.
Yeah, those are pretty poor numbers for workforce-related transit. I wonder how things change if definition is opened up to non-workers (students, stay-at-home-parents, unemployed, etc).
John,
You may already be familiar with it, but you raise a couple of issues that have also been explored at Urbanophile.com:
- http://www.urbanophile.com/2010/03/05/replay-small-cities-should-have-fareless-transit/
and
- http://www.urbanophile.com/2011/02/25/replay-transit-ridership-framework/
Great discussion in this post… we have similar problems re: the bus system here in Ann Arbor, even though we are much smaller than Columbus.
Yes, I’ve read Aaron’s posts on those. Thanks for linking them here though. He did a much better job articulating the case than I did.
I wonder if this is a good cause for considering a free circulator route for Downtown or the near-downtown areas in Columbus. Not only could it encourage ridership through those short trips of less than 1 or 2 miles, but could also incentivize urban living by providing a free transit solution for those willing to invest in homes near the free lines (though there would need to be some way to ensure that the lines could be long-term investments).
I actually think trips from the inner neighborhoods to downtown could be COTA’s best market segment right now. Auto ownership is relatively low in the inner neighborhoods, parking downtown costs something (although not terribly expensive), they’re served by multiple routes with decent frequencies, and the trips won’t be excessively long travel times compared to cars.
So I don’t think I would give those trips away for free, at least not during peak hours when overcrowding could become a problem. I do think a free downtown fare zone would be great, but I probably wouldn’t extend it beyond the innerbelt. That’s just my gut feeling though.
The ‘proof of payment’ system used in Germany and some of Canada could make possible zones of lesser payment but not random hopping for fee. Taking the plunge with a monthly pass instead of auto, maintenance, and insurance payments, solves that hypothetical dilema very nicely. If there’s any fee restructuring to be done in COTA, some texts would likely suggest a mid-day and weekend fare reduction, to encourage off peak customers, along with an increase of express fares, to compensate for lower density. I wouldn’t be in favor of free service either unless it be a true loss leader like a downtown shuttle extending no further than offices and parking garages as a free introductory offer.
An average of 12.4 MPH? No wonder when I bike pass a bus it never catches up. Maybe it’s just me, but I think that I shouldn’t be able to easily bike faster vs. riding a bus. For me, the short trip problem is if I want to take the #2 up High, get a transfer that’s good for an hour to run an errand and then get back on the next #2, they won’t honor it simply because it’s the same route: you *have* to use it for a different route in order for it to be redeemable. You would think paying for the transfer would be enough, but not with COTA. When I did attempt to take the bus from around OSU to the southern end of the Short North I found that I’d end up walking down there without seeing a bus (#s 2, 8, & 5)until *maybe* when I was already nearing my destination.
The free low-ridership time frame sounds great, but once again we are talking about COTA. Any time you suggest an improvement they’ll just respond with “we’ll look into it”, which translates as, “Who the f!ck do you think you are to even question us!? Just for that, we’re going to add that onto the checklist of things we’ll never do. What are you gonna do anyway? Ride the subway?”
I *was* one of those people who wanted to support COTA by being a semi-regular customer. Now unless the weather demands it, I refuse to give COTA my hard-earned money for being horribly late when I really needed a ride home on numerous occasions (how is the bus out near 270 when it’s supposed to be Downtown right now!?) and for having CPD escort me off the bus for being short of the exact fare the one time during the 2 years I had ridden COTA on and off. The one thing COTA does well is sour your experience with mass transit, regardless of how close/far away your destination is and on that they succeeded. COTA is far and away worse than any mass transit system I’ve used in the US or abroad.
Anyone’s whim to take the bus up a short distance is usually quashed anyway because they realize they don’t have exact change for a $1.75 (how much do you estimate they are losing from this alone?) and the long distances are still ridiculously slow because of the comical distribution of bus stops every block or two on some urban routes; by the time the bus approaches the 20-25MPH it’s already slowing down to stop at the next stop. Now that I think about it, for the most popular routes like the #2 and #1, the 12.4 MPH figure is probably skewed too high by express routes that have much fewer stops. COTA, and mass transit in general, is a lost cause in Cowlumbus.
Congrats on biking faster than a bus that makes frequent stops for passengers. How many people can you carry on your bike?
Walker:
I totally agree with your thought of providing a free circulator route downtown. However, they did have that COTA Connect(?) circulator in the early 2000s which flopped.
I think you need to do something in between a regular bus and a light rail. You need a light-bus concept that allows a dedicated lane for the bus.
Besides significantly speeding up times, it would provide one of the important advantages of light rail-visibility. I think having a route painted on a street and well-marked stops provide easy visual cues to riders. “Oh yeah, that free bus goes right by that restaurant I’m going to.” Denver’s MallRide does this extremely well, and gets something like 14 million riders a year.
I did some quick calcs for the downtown plan update to install a temporary circulator route, (which I submitted at the last minute, admittedly very late in the game). I thought the concept of “painting” a temporary light rail route on a street would generate demand, but obviously is not feasible. I thought about using a “light-bus” concept, which you could do as a temporary trial.
The route I used was one lane north on High Street and one lane south on 3rd Street through downtown running in a continuous loop. I figured using temporary bollards, traffic paint and signage, plus labor.
The conclusion-you could create a temporary dedicated bus circulator route for about $500,000. Not cheap, but a heck of a lot cheaper than $100 million for a light rail, and it could abe a “proving” device for high-frequency transit.
There are many other issues to address, but I think this is something the City and COTA should seriously explore, especially now that they are discussing moving buses off of High Street.
Some of your branding and visibility ideas sound a lot like BRT, which COTA is studying:
http://xingcolumbus.wordpress.com/2011/01/10/cleveland-avenue-brt/
As for high frequency transit, I think Columbus already has it. I guess it depend on the definition of high frequency. See here:
http://xingcolumbus.wordpress.com/2010/10/06/cota-frequent-network-map/
The BRT is dead, because the funds got axed by either the State or Feds, I can’t remember which.
This is sort of a BRT-lite concept. It’s more of a BRT circulator, very much like the Denver Mallbus. The difference would be that I think you could try it out on a temporary basis by just using some road paint, plastic bollards and a little engineering consideration to make sure that you don’t severely impact current traffic patterns. The idea is to see if you can overcome what in my mind are two of the most significant hurdles:
-Cost vs. frequency
-The reluctance of those that aren’t used to riding buses to ride them
I don’t think the old COTA Link overcame the second obstacle. I never rode it and I ride buses regularly.
I think my proposal would help on both counts:
-Cost-it would be free or a quarter, making it much more available to people. The temporary lane reservation would allow buses to avoid getting clogged in traffic without a heavy investment in infrastructure-at least on a trial basis.
-The clear visual branding of the bus, the stops AND most critically the route would help people get comfortable with it.
I think the route “branding” is really a critical aspect. If you’re downtown and walk or drive several blocks for lunch, it’s easy to ignore buses. But if you see a bus that looks different, has a fixed route and passes by you a couple of times each way, suddenly you are trying it out, and then suddenly you can’t live without it.
Does anyone know the route, frequency, and service hours of that earlier COTA downtown circulator?
I found the schedules. The COTA Link (#14) ran from High & Goodale to Sycamore & Front every 6-7 minutes from approximately 11:00 AM to 2:00 PM. Fares were $0.25.
It looks like it started in 1997 and lasted until at least late 2003, but was gone by early 2005.
Several people want to see it return:
http://www.columbusunderground.com/forums/topic/bring-back-cota-link
Thanks John. Good that it went by the Arena District, Convention Center, Hyatt and Westin. Did it bump out to pick up Columbus State Community College (via E Long), the Columbus Museum of Art, Capital University Law School, Columbus Metropolitan Library (per Grant/Washington), and Grant Medical Center, Frankiln University, Greyhound Terminal (via E Town)? For downtown tourists business that would include the Drury Inn, Red Roof Inn, the Renaissance, the Holiday Inn and eateries.
I believe it was just on High Street with a turnaround at the south end using Front in the Brewery District.
It would be hard for one route to serve all those destinations, especially if the goal is to create a simple-easy to use route with high frequency and short travel times.
21 minutes if you drop the Brewery District as Google Maps driving suggests. So 30 minutes for those requested stops? I dunno.
I think it would be fun if you drew a proposed route map. Then we can find the total mileage and divide by about 10 MPH for the speed of a bus in the downtown area.
That slow, eh? It’s 3.1 miles roundtrip if you go out to Cleveland Ave on E Nauten, south on Cleveland Ave/Grant, then east by Town and so on, but 2.3 miles if you were to turn south at 4th Street counterclockwise to the same E Mound for a southern boundary. However, if that’s the speed of a downtown bus, and the roundtrip distance between N High St @ E Nationwide and the Brewery District is 1.5 mi X 2, then a 6-7 minute headway would’ve required 2 busses.
And then of course there’s the general problem with loop routes that take people too far out of their way to be attractive for both the departure and return trips. I prefer lines to loops when possible.
If we look to Indygo, they have a 2.8 mi downtown red line circulator which intersects quite a few local routes (for bus hopping and fudge factor), and hits more than a few major institutions. I agree that a loop is better than a simple back and forth. Maybe a 4th Street to Mound loop @ 2.1 mi would work. Hypothetical as this all is for 25 cents.
That’s the opposite of what I said. Loops can cause problems. Depends on the size of the loop though. If it’s juts a one-way pair of streets that could be okay.
My mistake. It should go counter clockwise east from S High on one way E Main, north on one way 4th, and return west on one way E Spring St to N High. Sound any better?
Maybe. I still get concerned about the practicality of the loop. If you’re a southbound passenger on High, that works great. But if you’re a northbound passenger on High, you not only have to travel out of your way on the bus, you have to walk an extra two blocks west from wherever it drops you off on 4th. It becomes more time-effective to just walk. Since most transit trips are round-trips, this creates a problem.
High and 4th are close enough that it might work though. There are two other things that concern me with that particular proposal. First, it misses some major destinations like the Short North, the Convention Center, the North Market, and the Arena District (not too far a walk from Spring though). The other thing is frequency requirements. If you’re at Spring & High and you want to go south to Main & High, that is a 15 minute walk. Since the bus travel time is about 5 minutes, you would be willing to wait no longer than 10 minutes for a bus. That sholuld be doable, but it could become a problem for shorter trips, say from Broad & High to Main & High. Even worse would be from Broad & High to 4th & Spring, since it’s only a 10-minute walk and it would probably be about a 10-minute bus ride.