Micro-Transit: A Better Use for TDD

A couple of weeks back, Patrick Richmond wrote a mailbag letter in the Webster-Kirkwood Times that I think made some great points:

First, I think he’s spot on that the only real way of combatting congestion and a perceived shortage of parking in a growing Kirkwood is to get some people to switch from driving a car to using a mode of transport that takes up significantly less road and parking space. It doesn’t have to be everyone, and it doesn’t have to be every trip, but to make progress, we do need to encourage some people to walk, bike, or take the bus at least some of the time. The image below shows the basic spatial math at work:

There are a few ways to achieve this modal shift using both carrots and sticks. On the stick front, we could make tremendous progress by simply charging people a fair price to store their cars in Downtown Kirkwood. We currently give away one of our most valuable assets (valuable Downtown Kirkwood land in the form of parking spaces) away for free. That mismatch between value and price leads people to drive and use up more parking than they otherwise need to. Here’s a comment I got from reader Natalie Strickler earlier this year when I made my pitch for implementing paid parking:

I live within walking distance of basically anywhere in Downtown Kirkwood. A lot of times I walk, but sometimes, because it’s easy or convenient (but mostly because it’s free), I drive. I’d be really interested to see how many of those parking spots that are taken at certain times of day and night are by people who actually live within walking distance, and you have to pay to park, you might actually walk more!

But we also have to make sure that we’re using carrots to make sure that people actually do have legitimate no-driving options. One of those carrots is to make walking and bicycling in Kirkwood easier and safer. We’ve made significant progress on this goal in recent years, but there’s still a lot of ways we can improve, including significantly improving pedestrian safety on South Clay, and South Kirkwood Rd, and allowing people to live along Grant’s Trail so that they can directly access Downtown via mixed-use path.

But even with those changes, not everyone in the community lives within an easy walk or bike of Downtown Kirkwood. I think that’s where the carrot Mr. Richmond suggests, our own dedicated bus service, comes in. I’ve been thinking a lot about how we’d actually pull this off, and here’s what I’ve come up with.

What Makes for a Successful Route?

The only real alteration that I’d make to Richmond’s original idea is that I think we really want to really focus our efforts on places where taking the bus actually makes sense rather than trying to serve all parts of town. That’s because a transit route requires two primary things to be successful: density, and connectivity.

Density

Most people in Kirkwood own cars and cars, in most cases, cars are more convenient than public transit (especially as long as we continue to subsidize it via free parking). That means that convincing people to switch from driving to taking the bus will only work in a few limited situations.

First, you need the bus stop to be pretty close to where you live. Typically your car is parked in your driveway, so if the bus is much further away than that, you won’t take it. That means the bus service needs to be concentrated in the densest parts of town where people live right on top of it.

The second problem is that your car takes you to almost exactly where you want to go while a bus usually can only drop you off kind of close to that destination since everyone on it has a slightly different final destination in mind. The only time busses can match cars are in situations where parking at that destination is scarce (like Downtown Kirkwood) and you’re going to have to park a ways away or where it’s expensive (like at the train station).

Unfortunately the only place in Kirkwood that meets these criteria is Downtown. And a bus that just circles Downtown Kirkwood isn’t super useful since Downtown Kirkwood is pretty small and usually it’s just easier/faster to walk.

So then we have to think about where else, outside of Kirkwood, the incentives to take the bus are actually well aligned. And I think the most immediate answers are the urban clusters of Webster: Old Webster and Old Orchard.

Connectivity

The other big factor in how useful a given transit route is (i.e. how much ridership it gets) is the strength of its connections. Transit ridership loosely follows Metcalfe’s Law: the value of a network is proportional to the square of the number of nodes. In English, that means that the whole is more than the sum of its parts.

For us, that means that once our bus is driving the 15 minutes to get to Old Orchard, we probably want it to go three minutes to the Sunnen MetroLink Station too. Building a carless connection between MetroLink and Kirkwood’s Amtrak Station drastically increases the potential use cases for such a service, both for Kirkwood/Webster residents headed elsewhere in the region and for the rest of the region visiting Kirkwood. Such an extension has the added benefit of getting riders fairly close to Downtown Maplewood, another node of density where bus service might be somewhat useful.

Okay, enough preamble, let’s look at some maps.

Route, Frequency, & Scheduling

You can check out the full interactive version of the map here (complete with Metro’s existing Kirkwood bus routes, the MetroLink lines, and a couple of alternative routes you can toggle in between), but I’m worried it’s a little overwhelming unless you’re a die-hard transit fan, so for now I’m just going to show you static pictures.

Here’s my rough outline of the route (color coded by the municipality that section of the route is in):

According to Google Maps, the full route from Kirkwood Amtrak Station to Sunnen Station is an 18 minute drive. MetroLink’s Sunnen Station gets trains every 20 minutes (at an impressive 98% on-time performance rate), so in order to make the connection to the MetroLink a useful one, you want to make sure you have a bus at every single one of those MetroLink arrivals every 20 minutes. If the bus is one minute late, riders would have to wait for 20 minutes for the next one and I’m pretty confident relying on the Kirkwood-Webster bus is a mistake they wouldn’t make twice. That high of stakes for the bus arriving on time means two things.

First it means we’d need at least two busses operating at all times. When one bus arrives at Sunnen, the other bus needs to be leaving Kirkwood Station and headed towards Sunnen to make sure it’s there for the next train’s arrival.

The second thing that it means is that because the 18 minutes is the time between Kirkwood Station and Sunnen if you were just to drive it straight through without any stops whereas a bus has to continuously pull over to let passengers on and off, you’d also have to figure out ways to speed the travel time for busses up to something like ~15 minute travel times make sure you’re hitting those 20 minute frequencies once you include stops (aka dwell times).

In order to to do that, you’d have to consider things like: 1) giving busses a dedicated lane so they don’t get caught up in traffic, 2) avoiding railroad crossings to ensure a coal train doesn’t throw the whole schedule off, 3) making sure bus stops are constructed and positioned to reduce dwell times as much as possible.

Is that enough to get you to 20 minute headways? Maybe. Running three busses ~15 minutes apart with five minutes for stops/delays would be a safer bet to hit the 20 minute frequencies we’re looking for, but obviously three busses cost more to acquire, fuel, and operate than two.

Road Redesign

The good news is that the proposed route is well suited for the sorts of improvements I mentioned above. That’s especially true of the 2-mile middle portion of the route which runs along Lockwood between Downtown Webster and Downtown Kirkwood, which, at ~80ft wide, has a ton of underutilized capacity:

If we used that extra space more efficiently, here are some potential reconfigurations that would fit within the space available that would yield significant improvements for walkers, bikers, and bus-riders alike:

Then, once East Lockwood turns into West Lockwood at Mary Queen of Peace, the route I’m proposing follows the route of the #56 Metro bus route, which runs from the Shrewsbury MetroLink stop all the way to Meramec Community College. That means that any bus-oriented improvements to this stretch would double as improvements to existing Metro bus service, making the expected return on investment even greater and potentially opening up additional funding opportunities should Metro want to get in on the action.

Busses

Beyond the engineering costs for any such roadway improvements, the biggest fixed costs for rolling out bus service is the busses themselves. In his Webster-Kirkwood Times piece Richmond mentioned the 29′ Gillig busses, which I think would work fine, but he also mentions the possibility of using old Metro busses, which I think make even more sense, especially if we could get our hands on some of the short models they already uses for their call-a-ride services (and which you occasionally see rolling through Kirkwood):

These busses: 1) should be widely available in the St. Louis area making sourcing and servicing them much more affordable, 2) have a capacity of 13 riders which means that drivers wouldn’t need a Commercial Driver’s License (significantly increasing the available pool of operators while reducing salary costs) and 3) are already operated by Metro, making any future handover of operations to the regional transit agency much more feasible.

To ensure consistent operations, we’d probably need 3-4 of these busses. Two in operation, and 1-2 ready to go incase of mechanical issues/to seamlessly rotate drivers in and out throughout the day.

Funding

So how do you pay for this thing? Well, we’re looking at two primary funding sources, federal grants and sales tax funding. Let’s take a closer look:

Grants

The grants (typically structured as 80% federal funding with a 20% local match) would be used for the large one-time planning and capital expenses (like the purchasing of busses and the reconfiguration of the road). Obviously the 80% is spoken for, but the 20% local match would probably best be sourced via something like metered parking in Downtown Kirkwood/Webster/Old Orchard. The idea here is that you want people to bear more of the social costs of driving (pollution, occupying otherwise valuable public space) so you charge them to park, but you also want to give them other realistic options for visiting the places they used to drive to (taking the bus). And because paid parking would encourage at least some people not to drive, it would have the added benefit of freeing up parking spots for people who really do need to drive (large families, those buying large plants at the farmer’s market etc).

These grants could come from the Surface Transportation Program (STP) or Transportation Alternatives Program (TAP) grants administered by the East-West Gateway that we used to fund the Grant’s Trail, or we could also utilize the Safer Streets for All (SS4A) implementation grants from the DOT that we qualify for by virtue of our completed Vision Zero Action Plan. While on the latter of these Kirkwood is the only entity eligible to serve as the primary applicant, the other communities involved (Webster, Glendale, Maplewood, Oakland) would still contribute to the local match.

Sales Tax (TDD)

The other available funding mechanism is sales tax revenue derived from a Transportation Development District (TDD), like the one Kirkwood is asking voters to approve in November. While the grants above would cover capital expenses, the TDD sales tax revenue would be used for long term operating expenses like driver salaries, insurance, and fuel. Let’s take a look at the proposed TDD, and see how it could be adjusted to fund something like bus service.

SBD – TDD

If approved, Kirkwood’s city-wide TDD is expected to bring in ~$3m per year. 5% of that total would go to the Downtown Kirkwood Special Business District, yielding approximately $150-160k per year. Because the vast majority of the benefits of my proposed bus service would accrue to the Downtown area, I think it’s fair that such a project be primarily funded by this subset of the TDD rather than the Kirkwood TDD as a whole. The good news is that Webster has three SBDs of its own, two of which are Old Webster and Old Orchard. These two areas are obviously smaller than Downtown Kirkwood, so let’s call it another $150k between them. That would bring us to $300k per year in SBD TDD revenue.

Now we’d need a bit more than that to fully subsidize the program so then we’d need the other communities serviced along the route to contribute something. Combined, maybe Glendale, Oakland, and Maplewood get us another $200k a year, bringing us to $500k. Is that enough to cover operating expenses? Maybe you’d need the cities of Kirkwood and Webster to chip in a little extra, but I think it’s roughly in the ballpark.

How to Proceed

Admittedly, we’re a long way off from anything like this happening, but there’s still plenty of room for us to make progress in the interim. At the end of his letter, Patrick Richmond suggests that anybody who supports the idea of micro-transit in Kirkwood should raise the idea to Kirkwood City Council (emails available here). I agree with that assessment, but I’d encourage residents of Webster (or Glendale, Oakland, or Maplewood) do the same. If you know anyone in those communities, please feel free to pass this piece along.

The other productive step you can take is to vote against the TDD question on the ballot this November. TDD is a useful tool, but we’ll waste it if we roll it out without any sort of plan as to how we can most effectively use the funds. Until such a plan emerges, it’s best we keep our powder dry.