Last August I was a guest on Disruptive Conversations, a podcast by Keita Demming. After that interview, Keita offered me the honor of turning the tables and interviewing him for his final episode of Season 1. That episode was released yesterday; we’ll be running our version later this week. Keita has done some fascinating interviews and I recommend subscribing.
In the show where I interviewed him, Keita discussed the difference between systems that are simple, complicated or complex. I found his explanation very easy to grasp and so I’ll provide it here:
There are three types of problems. There are simple problems, like a recipe for baking a cake. There are complicated problems like launching a rocket or organizing a conference with 10,000 people. With enough time and resources, you can figure it out. A complex problem is more like raising a child. As a parent changes their response, a child changes their response. The child changes their behavior and now the parent changes their response. That is a continuous, dynamic situation.
I later asked Keita about mistaking complex problems for simple or complicated problems and he commented that this was the essence of systems problems that he experiences everywhere.
I’m on my local planning commission and last week we had a presentation by the local school district. This is part of an ongoing conversation that is too large to get into here, but I’d like to focus on one aspect because it stuck with me and I think is representative of deep misunderstandings policy people tend to have about the systems they manage.
The New School on the Edge of Town
Back in 2005, the school district opened a new school – Forestview – in the neighboring city of Baxter. The move including closing one Depression-era neighborhood school (Washington) and repurposing another (Mississippi Horizons) and followed the closing of two other Depression-era schools (Franklin and Whittier). All of the schools that were closed or repurposed were neighborhood schools in the core of the city.
Here’s how the local newspaper reported on the opening of the new school:
Forestview opened with few glitches on Jan. 4. Heavy morning traffic created long lines of vehicles leading to the school that morning. School officials believe many parents wanted to drop off their children on their first day, which led to the traffic congestion. Within a week, traffic problems had been solved as parents found their best routes to the school or had their children ride the bus or carpool with other students.
Only, the traffic problems were not solved. In fact, the traffic problems at the pickup and drop off area have been the greatest complaint the school district has received about Forestview. As reported in the local paper last May:
The two most common suggested improvements for the school include an improved pickup and dropoff area for students and a more secure entrance.
Steve Lund, director of business services for the district, touched on both improvements during a presentation to the 15-20 community members in attendance. Parents are familiar with the traffic issues at the school after school lets out, he said, and there's ways to make the traffic flow more efficiently.
"I know I did it for four years and it tested my patience," Lund said.
The aspect of last week’s presentation that stuck with me was the vigorous response on this particular issue. I don’t have the exact quote from district officials, but the gist of it was the traffic problem will be fixed. Not, we’re going to work on it. Not, we’re going to make improvements. No, it was Joe Namath guaranteeing the Super Bowl victory: This. Will. Be. Fixed.
Is the traffic backup at Forestview a simple problem, a complicated problem or a complex problem?
The way district officials speak about it, they seem to view it as a simple or, perhaps, a complicated problem. More parents want to pick up their kids than we anticipated, ergo, we must build more capacity to handle them. As Keita suggested, let’s just devote enough time and resources and the problem will be solved.
The concern I have right away is that this kind of lets us off the hook intellectually as to why we have the problem in the first place. I think this is a complex system – action, reaction, new action, new reaction – and I’m not convinced that whatever we do now will fix the problem, at least as we’ve defined it (taking too long to pick up your kid).
What Went Wrong?
Let’s first ponder why the very smart people who designed this facility got it wrong in the first place. How did they so dramatically underestimate the number of kids who would choose to get picked up and dropped off?
Look at the map below. My school district is HUGE – I live in a small town in a rural area and we cover a lot of space – but most of the students live in the cities of Brainerd or Baxter. In the map, I’ve circled the new Forestview school in yellow (it is in Baxter) and the four schools that were closed or repurposed in this process in light blue (they are all in Brainerd).
While it is true that we bus kids in – and many are driven in – from far and wide, looking at that map, is it possible that the action of putting a school on cheap land far out of town may create this reaction of parents choosing to take their kids to school via car?
I can attest for my family. We live near that uppermost closed school circled in light blue. Prior to Forestview, our kids could easily bike or walk to school throughout their entire education. They were eligible for busing during grade school as we live just a little over a mile away (the cutoff distance) and the bus ride — when we used that option — was short. Our kids would have gotten on 15-20 minutes before school started and gotten off about the same interval afterward.
For our kids to ride the bus now, it’s a much longer time commitment. In fact, last year my youngest (who was still in elementary school in our neighborhood) would get on the bus and ride all the way out to Forestview – a ride that took her literally right past the front door of our house. The bus would pick up more kids at Forestview and then drive back into town to drop them off. She was on the bus almost 50 minutes. Needless to say, we rarely used this option.
So when the schools were located in town, all the kids in the hinterland were either bused or dropped off (a choice their parents knew when they chose where to live). Many of the Brainerd kids were bused or driven as well, but there were other options for them. Some involved getting there at different times. I knew if I dropped my youngest off at 8:10 it was going to take longer than if I were there at 8:00. During pickup, there was so much parking space on the blocks around the school that I would park, walk to the door and meet her and then walk back to the car. It was never an issue and there was never any real congestion.
Now that Forestview is in the hinterland, all the kids are bused or dropped off. And more are being dropped off, at least partially because the bus rides are so long. Note, this hasn’t reduced busing costs – the buses have to drive the routes whether the kids are getting on or not – so this has just added other costs for everyone else (not to mention made us really fragile to gas price increases).
So would a redesign of the drop off lane solve the problem? I think there are really good reasons to believe that it wouldn’t. While the action/reaction of different families is hard to predict, I’ll use our family as a sample. I hate picking up the kids – it’s a total nightmare waste of my time to sit in the long line and pick them up like they are a Big Mac – so, when they don’t have an activity after school that I need to bring them to, they ride the bus.
They hate it. I’m not really keen on it either – there's a lot of misbehaving kids, as well as just a ton of wasted time – but twice a week during dance/piano season is not bad. Still, if it was easier to pick them up, if it didn’t require a maddening 20-30 minutes of sitting in a line of cars, I’d probably pick them up.
And now we’re right back to where we started: action and reaction. There are parents that must pick up their kids – like me on dance days – and there are others who balance the costs and benefits and choose to pick up their kids for a variety of reasons. Reduce the cost of time for picking up your kid and more people will do it until the congestion returns and the costs get high again — action and reaction in a complex system.
Addressing the Problem
Short of an expensive redesign not guaranteed to work, what are some of the things the school district could do to actually address this problem?
The first thing would be to shorten bus rides for those who are choosing to drive. Make the cost of time lower for riding the bus and fewer parents will find it worth spending their time to pick up their kids. This will only work with the pickup optional locations – mostly those in Brainerd. Those in the hinterland won’t be affected either way.
A second option would be to allow kids to bus to alternative destinations. Right now, we only allow kids one destination. For my family, that one destination has to be home. That means on dance days, they either bus home and are late for dance or I pick them up. In a school with 1,800 students, there are about a hundred that go to dance. That’s probably about 60 to 70 of the cars in line. And that’s just dance; where else do the other kids need to go that they aren't allowed to bus to?
Having a more sophisticated busing approach, one that gives families options day-to-day, would require a little more administrative headache but would save a lot of money in engineering and constructing a new pickup lane (that is not going to solve the problem anyway). The system right now feels like it’s designed for ease of administration. That seems more than silly when we’re paying to drive half-full buses past places where kids need to go.
A third option would be to redeploy the security guard that stands out there watching the line back up to doing something like traffic control. The greatest congestion at the pickup circle happens when people park at the top of the circle closest to the school (the shortest distance for their kid to walk) instead of pulling forward to the end of the circle. If people pulled all the way ahead, the entire circumference of the circle would be available for simultaneous loading/unloading. If the security guard would simply motion people to move ahead if they were idling with empty space in front of them, that would work wonders. It’s maddening this doesn’t happen already.
A fourth option would be to deal with the exit from the pickup lane back onto Knollwood Drive. Cars get backed up here waiting for a gap in traffic on Knollwood, which is designed for fast speeds with wide lanes and lots of clear space. Slowing Knollwood with a traffic circle at the pickup lane exit – a simple traffic calming device – would allow traffic to continue to flow out of the pickup lane, even during peak times, freeing up space for others.
I could go on, but it doesn’t matter. It’s very unlikely that any of these things would ever happen, even if we all agreed they were great ideas. That is true because our school districts are set up very much like our city governments: Hierarchy, silos and all the incentives to never work incrementally.
Shortening bus rides for families where that would reduce driving would require an intimate knowledge of those preferences, and we don’t take the time to communicate with our families in that way; it's just not efficient (our buzzword). It might also require us to reexamine whether Forestview was a good idea (in general, it wasn't) and that's not really a conversation we want to have.
Providing flexible drop off locations for the bus would require a rethinking of the logistics of busing and, from the district’s standpoint, there’s no real incentive to spend that effort (not to mention take on a decent amount of risk if you screw it up even once). The security guard doesn’t work for the transportation department in the district; nobody’s going to tell him to be a traffic cop, too. A roundabout would require city approval and, since it’s a state aid street, some type of variance from state officials (good luck with that). If we could get approval, the funding would have to come out of some other budget – maybe art supplies or maintaining the roof – and that would just be hard.
It’s the path of least resistance to tell parents (voters) that, if they approve a huge bond referendum to close another neighborhood school, demolish yet another and build two new schools outside of the core neighborhoods, that we can take care of that pickup and drop off lane at Forestview in the process.
Our school district is run by very smart people working in a very dumb system. We – the state, the district, the city, parents, voters – must stop thinking of that system as simple or even complicated because it is, in fact, complex. Action, reaction, reaction to the reaction, and so on.
The only way to successfully improve complex systems is incrementally. Investing tens of millions all at once every 25 to 30 years is a recipe for making a lot of big mistakes. I love my school district but I really hate this approach.
(Top photo from City of Baxter)