Jack Turek

In my last post, I looked at overall ridership trends for VIA Metropolitan Transit and how they’ve changed over the last two decades. VIA ridership took major hits during the Great Recession of 2008-09 and the COVID-19 pandemic, and ridership hasn’t recovered despite an increasing population. I detailed some of the factors that could be depressing ridership of VIA (development patterns, funding) but in this analysis I’m going to look “under the hood”, so to speak, at VIA’s General Transit Feed Specification (GTFS) data.

GTFS is a common format for public transit schedules and geographic information. Public transit agencies publish their transit data in a series of text files which model a specific aspect of transit information – stops, routes, trips and other schedule data. This data can be used to create trip planners, publish time tables, and create performance dashboards, something many transit agencies make publicly available.

Using VIA’s GTFS data, I can get a better understanding of how the system actually operates and how effective it is at moving San Antonians from Point A to Point B. Let’s define two important terms, first.

The first is headway: this is defined as the time interval between two consecutive vehicles passing a particular point (i.e. a bus stop) on a specific route. Shorter headways means more frequent service. A 10-minute headway implies that, on average, a passenger would have to wait only about 10 minutes for the next us if they arrived at the stop at some random time.

Next, is frequency: frequency is the number of vehicles (in this case, buses) that pass a specific point (a bus stop) on a route within a given time period. Higher frequencies mean that more vehicles are passing a bus stop within a certain time frame. For example, if VIA operates with 15-minute headways (do they?), then the frequency is 4 buses per hour.

VIA publishes their GTFS data here, so let’s dive in!

Let’s start by calculating some headways for VIA. I’m going to be using the tidytransit R package to analyze VIA’s GTFS data, starting with stop frequencies between 6:00am and 10:00am on weekdays. This particular time window was chosen as it’s a likely commuting time for people using VIA to get to work.

The bus with the shortest headways between 6:00am and 10:00am is the Prímo Zarzamora (bus route 103), which runs every 12 minutes. The 103 runs North-South along Zarzamora St and connects the Madla Transit Center and Crossroads Park & Ride. The next fastest routes are VIA’s other two Prímo routes, the 100 and 102. Although slightly less frequent, those buses run every 15 minutes. These Prímo buses are sort of a Bus Rapid Transit-lite network mainly because there’s fewer stops on the routes and the buses get transit signal priority over cars.

All of the Prímo routes are in the top 5 for ridership, and have experienced an increase in passenger trips when comparing June 2022 and June 2023. 12-15 minute headways for these Prímo buses is good, and certainly contribute to higher ridership numbers. But these are only 3 routes out of the entire VIA system. What are headways like for the 552, 20, and other routes with less ridership?

Between 6:00am and 6:00pm, the 552 Loop 410/WW White Skip averages headways just over 15 minute headways, comparable to the Prímo routes. Looking at the route map below, I would classify this bus as another kind of “express” bus. It travels almost the entirety of Loop 410 and connects to several other VIA transit centers. It would take nearly 2 hours to ride this route from end-to-end.

The 20 New Braunfels (Frequent) route also has average headways of 15 minutes. This route goes from Brooks City Base on the south side before cutting west to link up at VIA’s Centro Plaza transit hub downtown.

So the other 2 routes in VIA’s system that have the highest ridership also average 15 minute headways. Prímo’s only advantage then is signal priority, as I don’t consider fewer stops to necessarily be an advantage.

The other three routes I want to look at are ones that I’ve personally used – the 2 (Blanco Frequent), 3 (San Pedro Skip), and 4 (San Pedro Frequent). Here’s where we can start to see headway issues within VIA’s system. The average headway for the Blanco Frequent (Route 2) is 29 minutes over the course of 12 hours during a weekday. To put it bluntly, that is not good. Remember the definition of headway above – the average time I would have to wait for the 2 bus if I showed up to that bus stop at some random time is nearly 30 minutes. The San Pedro Skip (Route 3) and San Pedro Frequent (Route 4) are marginally better, but all three routes have headways that are longer than what it would take me to drive to most destinations I’m trying to get to.

Granted, those headways are for a 12-hour period. If we only look at average headways for morning and evening commuting times, there’s little change. In fact, both the 2 and 3 routes are a couple minutes worse during those times. On average, a prospective rider of the the Blanco Frequent, San Pedro Skip, or San Pedro Frequent would need to wait 20-30 minutes for one of those buses to arrive.

Because of the extensive freeway and stroad system in San Antonio, anyone with access to a car can likely get to wherever they want within 20-30 minutes. Even if a household is spending most of their income on rent and transportation-related costs, driving 20-30 minutes to the store is much more attractive than potentially waiting 20-30 minutes for a bus to show up, ride the bus to the store, do their shopping and then endure another 20-30 minute wait for the ride home.

What might be the most frustrating thing about these headways on the 2, 3, and 4 routes is how they change as a bus drives along the route. The 2 bus starts at its southern-most point around 20 minute headways. About the final third of the 2 route is right around 40-50 minute headways. The San Pedro Skip route (middle graph) has varied headways throughout the whole route while the San Pedro Frequent route has a similar headway distribution as the 2 – starts “fast” but gets slower. I think VIA should remove “Frequent” from the names of these bus routes.

There’s plenty more routes in VIA’s network than the Prímo routes and the 2, 3, and 4 routes. Judging by the average headways of routes that VIA classifies as “frequent”, however, I’m willing to go out on a limb and say that the rest of the bus network is likely as slow or slower than these routes. Keep in mind, too, that the 2, 3, and 4 routes all traverse through San Antonio’s downtown, one of the denser areas and an area that should have better bus service.

If you want to see how high (or not, depending on your point of view) VIA is aiming when it comes to improvements to their bus network’s service, check out their Keep SA Moving plan. Their first stated goal is that all routes will have frequencies/headways of 30 minutes or better during weekday peak times. This means that there are VIA routes where the headways must be significantly worse than 30 minutes if an average wait time for a rider is going to be reduced a half hour. At some unspecified date in the future, VIA wants to have 15% of routes at 10-12 minute headways, a significant improvement to be sure. Take a look at the pie charts from their plan below – I’m not colorblind but it sure is hard for me to differentiate between the red palettes they’re using. I’ll be generous and go on the assumption that in the future (TBD), VIA wants to get to a point where 85% of their routes have headways between 15-30 minutes. I guess that’s better than 90% of routes having headways that long. Way to shoot for the moon, VIA.

Some might ask why bus riders don’t just look up the bus schedule before going to their stop? While it’s a valid question, I don’t think constantly checking a bus schedule is convenient for users. First, not every bus rider may have a smart phone and even if they did, the app riders need to use to check the status of their bus can be confusing to decipher. Second, I believe that the mark of an effective transit system is one which checking a bus schedule isn’t necessary. If you’re transferring between buses and routes then absolutely check the schedule. But, ideally, you should be able to show up at your stop and have the confidence that the bus will arrive within 10-12 minutes of you getting there.

San Antonio residents absolutely deserve better public transit. I know the pandemic really hit public transit agencies hard (see my last post for just how hard a hit VIA took). But as San Antonio continues to grow, changes in development patterns that encourage alternative modes of transportation need to be implemented. San Antonio was one of the cities that joined the Paris Climate agreement after Trump pulled the U.S. out of it. San Antonio’s own Climate Action & Adaptation Plan (CAAP) states that 38% of emissions came from the transportation sector (34% from private transportation). There’s plenty of strategies mentioned in the CAAP for reducing transportation emissions, such as enabling mobility as a service using “big data solutions”, but like so many other plans San Antonio has created there’s little said about how the city actually plans to implement these strategies.

Using phrases like “smart city” and “big data solutions” sounds all technocratic and cool, but unless San Antonio can actually deliver on this 7 year old plan (and the delivery of any such strategies will need to happen fast if we don’t want another summer of 80+ days over 100 degrees), residents will continue to be saddled with an inadequate public transit system and few other mobility options other than using their personal vehicles.

I grew up consistently using public transportation, but as I’ve gotten older and lived in less transit-rich parts of the country, I’ve found myself resistant to hopping on the bus in San Antonio to get somewhere I want to go. Some of this resistance is that taking the bus to a destination like the office where I work would take approximately 1 hour and 10 minutes, while driving would take 15 minutes. Another reason is sheer laziness – taking the bus to a restaurant may take 20-25 minutes but driving will “only” take 10-15 minutes. A third aspect working against transit’s favor is its ease of use – I don’t feel confident that I can walk to one of several bus tops within 5 minutes of my house and only have to wait a few minutes before the bus I want shows up.

I’m aware that the above very much sounds like “woe-is-me” complaining. I have options (privilege, actually) – I could drive my car, take the bus, ride my bike to a closer restaurant, or (and this the healthier and bigger money saver option) I could just eat at home. I have the luxury of choosing to ride the bus. I don’t have to, like so many other people who take the bus to work, to shop, to visit friends and family, to get to doctor appointments, or to many other places.

Here’s the part that gets me every time – when I take the bus, I’m really glad I did. I didn’t have to drive around looking for parking, I wasn’t worried about other maniacs on the road, and I could zone out or read on the bus.

VIA Metropolitan Transit, San Antonio’s public transportation provider, has always had to battle for funding. Only in 2020 did San Antonio voters approve a 1/8 cent sales tax to give VIA more funding, although it won’t go into effect until 2026. VIA also receives the least amount of sales tax dollars that other transit agencies in Texas cities receive, such as Austin, Houston and Dallas, which receive a full 1 cent of available sales tax. With increased funding starting in the next 3 years, VIA should be able to increase route frequencies, make capital improvements, and enhance ridership experience.

A big project will be VIA’s Advanced Rapid Transit (ART) North-South bus route, going from the airport to Southtown. I’m not sure exactly what’s “advanced” about it since it’s really just a bus-rapid transit line, with dedicated lanes and level all-door boarding, which is fairly common throughout the rest of the world. Although the North-South line is scheduled to be completed first, VIA also has plans to complete an East-West route, which could be completed faster than originally anticipated if VIA gets more funding.

I’m definitely looking forward to getting more mass-transit capacity in San Antonio, though I do have questions about what exactly VIA wants to get out of these ART routes. As I gather more data and do some more research, I’ll come back with another post on that.

For now, I just want to take a simple look at VIA’s ridership over the last 20 or so years. To get ridership data, I used the National Transit Database, which contains tons of data on public transit agencies all across the country. I’m using a complete monthly ridership time series dataset, which contains all unlinked passenger trips (UPT) per month. The American Public Transportation Association defines UPTs as the number of passengers who board public transit vehicles regardless of how many vehicles they use to travel from their origin to their destination. So if I take the bus from my house to my favorite ramen restaurant, and I take two buses because I have to transfer, that counts as two trips.

The graph below shows VIA’s monthly UPTs from January 2002 through June 2023. UPTs peaked after 2008, around the Great Recession, before falling off again as the recession worsened. Ridership steadily climbed upwards around 2010-2011, but never reached pre-2009 levels. Since 2012, UPTs have generally declined before the precipitous drop-off in ridership at the onset of the pandemic. Although ridership is starting to increase, there’s still a long way to go before VIA reaches ridership levels seen over a decade ago.

Breaking out VIA’s monthly UPTs by year shows that there have always been some ups and downs throughout the year. Ridership looks to fall somewhat in the summer months from 2002 to 2007 before increasing in the early fall before dropping again, but in 2008 there’s a definite upward trend all year until right around October, when ridership dipped to its lowest point. Ridership remains steady in 2009 until a drop, again around the last quarter of the calendar year. We see similar ups and downs throughout the 2010s, though overall UPTs are generally lower than in the decade before. VIA’s ridership remains mostly between 3-4 million UPTs until the pandemic hits in March 2020. Since then, UPTs have remained relatively flat at around 2 million trips.

To get a more normalized view of VIA ridership in relation to San Antonio’s population growth, I calculated UPTs per capita, or per person, annually. To get population data, I used the 2000 decennial census and applied that to years 2002-2008. From 2009-2020, I used the ACS 5-Year Survey population data. It’s not a perfect way to do it, since I’m applying one decennial census’ population value across an 8 year period, but I think it will still show a clear picture of VIA’s ridership in relation to population.

The graph below shows VIA’s yearly UPTs per capita from 2002-2019 and the San Antonio-New Braunfels MSA population estimates from the census (I used the decennial census for 2002-2008 and the ACS 5-Year estimate for the remaining years). ¹ Since the population estimates come from the SA-NB MSA, the population shown below is more than the total number of people in San Antonio proper. UPTs per capita increased steadily from 2002 through 2008, when we see a drop in ridership during the Great Recession. Rides per person peaked at 4.5 in 2008. Still, it shows a sharp decline in yearly UPTs per capita from 2008 onward despite steadily increasing population. VIA never seems to have fully recovered from the Great Recession back in 2008-09, which does not bode well for VIA’s potential recovery from the pandemic.

I’m not too familiar with the history of VIA, so I’m unsure about how many new routes or how much new capacity was added from 2002 onwards. Adding new routes could potentially increase ridership, as well as improving the frequency of a route (how long between each bus arrives at a stop). The drop-offs in ridership near the end of the year are interesting, though, and I’m not quite sure what to make of them.

I suspect, however, that a major factor effecting VIA’s ridership is population growth. San Antonio is growing rapidly, with nearly 20,000 people between July 2021 and July 2022. All these new residents need places to live and places to work. Lots of new developments are springing up outside of Loops 410 & 1604, many of them single-family homes or large apartment complexes. The median home price of new housing construction in San Antonio the last several years is close to $300,000. New homes that are built outside of Loops 410 and 1604 may not be serviced as well by VIA as are homes that are within Loop 410 and close to the downtown core. Additionally, San Antonio’s downtown vacancy rate of commercial and office real estate is nearly 40%, and many large employers, such as banking giant USAA, have left their downtown offices in favor of consolidated large campuses outside of Loop 410. This likely forces workers to look for residences outside of San Antonio’s freeway loops, where homes may generally be cheaper.

Almost every census tract within San Antonio has experienced some amount of population growth. Except for a small census tract right in the middle of San Antonio, much of San Antonio’s population growth is occurring outside of the city center. Some of these population estimates may not be 100% accurate due to changing tract boundaries or the addition of new ones, but it’s clear that tracts far outside of the core city center are increasing at a faster rate than tracts closer to downtown. Much of San Antonio is zoned for single-family housing and building these types of dwellings is easier to do at the edges of the city, where land is more available and fairly cheap.

Besides a growing population, there are other factors that are likely effecting VIA’s ridership. Important factors like route frequency, route speed, and connections all play a role in how VIA is utilized. I’m going to spend some time in the coming months diving into VIA’s Google Transit Feed Specification (GTFS) data, which will allow me to map VIA’s routes through San Antonio and examine route frequency and speed in more detail. With GTFS data, I should be able to exactly see where VIA is most efficient and where VIA’s service could use improvement. But in a city of 1.7 million people and counting, the lack of viable and frequent public transit will continue to cause headaches for residents trying to get around the city.