How We Made “The Water Drain”
Tracing water from Lake Michigan to residents’ taps, and investigating why some pay much more
The Chicago Tribune’s recent investigation, The Water Drain, took a sweeping look at how Lake Michigan’s water reaches 163 municipalities around the Chicago area. To piece together the bigger picture of water usage and how much people pay, the Tribune team used a variety of data sources, including their own survey. They found wide disparities in what residents were paying for water, with the poorest communities paying the most. Source spoke with Cecilia Reyes at the Chicago Tribune this week, to learn more about how The Water Drain got made. —Eds.
It Starts with the Lake
Source: How did this project get started? This is a huge initiative and covers so much. Was water something you’d been following for a while?
Cecilia Reyes: The Tribune had done previous reports on Lake Michigan having to do with water quality and a little bit of water pricing, When Chicago passed rate hikes in 2012 to fund infrastructure repair, it was definitely on people’s minds. But we wanted to try to take a couple of steps back and consider (or try to consider) as big of a picture as we could. So we settled on trying to understand how Lake Michigan water got to people’s taps both in Chicago and the surrounding area. And when we narrowed our view on that, we found that a lot of those places were managing their own water systems—meaning that they weren’t getting services through private companies.
So we took a deep dive on how those worked. And we started with what seemed to us was a fairly prominent question, “How much do I pay for the water? Is it the same thing that other people pay for their water?” Once we picked our area of interest and narrowed it down, we started finding data sources. We cared about infrastructure, supply chain, water pricing, water conservation. And just how we were taking advantage (or not) of a great resource—Lake Michigan.
Source: How did you get all this data, and how did you start making sense of all of it?
CR: We were using various data sources. One of them was the LMO–2 forms—water audits that municipalities have to fill out if they get an allocation from Lake Michigan water. If they’re allowed by the state to use that water, they have to report how much they’re using, how much they’re losing, what the cost of those losses are. It changed in more recent years but they also used to have to report how old their infrastructure was. Those forms are required by the state to be turned into the Illinois Department of Natural Resources. A colleague had already FOIAed for those documents. They were PDFs. We knew that was available so we decided to FOIA for 2014, which had information on municipalities’ infrastructure as of that year, and we FOIAed for the most recent forms which were 2016, at the time. That was mostly focused on age of infrastructure, water loss, and the cost associated with that water loss.
We hand-transcribed all of [the PDFs]. They weren’t available in any other format. (We tried.) This was work mostly done by a colleague who left halfway through the project, Angela Caputo. She went through each of those forms and entered them into a spreadsheet and confirmed them twice.
The other data sources…one was the Census, which we brought in at the end. The other came from the Illinois State Water survey. They had put together a map of sellers and buyers in Northeast Illinois for 2012. When we started the project, we wanted to answer another natural question, “Is what I’m paying for my water related to the steps that it takes for me to get the water? If it has to get from point A to B to C to D, how does that affect either what I pay or how my water is managed?” The water survey had tried to collect information for 2012 by asking municipalities who they got their water from. So we had those links there. And then the water pricing information, the rate information.
We also did our own survey. We started with methodology that came from a survey done by The University of North Carolina and CMAP. So UNC, the environmental financial center, had done these types of surveys not just in Illinois but in Wisconsin and other places. They were asking how much municipalities were charging for their water. They asked not just about the water rate but any fees associated with water consumption and water service. So we used their methodology to release our own survey of all the municipalities that get water from Lake Michigan with publicly owned water services. And what we asked every town was their rate information for 2013 and the most recent year. We wanted to do a comparison between 2013 and 2017. We also asked them for any fees that were related to water consumption or water service. And we also asked them for their billing practices. Some places would bill every two months or quarterly and would use different units. We standardized water consumption at 5,000 gallons because that was the standard that had first been published by UNC when they had done their first survey of water consumption, and we wanted to have a baseline to compare price A to price B. We did all our calculations at 5,000 gallons. We excluded sewers and wastewater and always asked the financial director at that village to confirm our calculations. So we put together our own dataset for 163 total municipalities.
From Data to Feature
Source: What was the timeline for this project like, and what kinds of roles were on the team?
CR: It took about eight months…. Angela [Caputo] helped with the data gathering and data reporting. I also helped with both of those when she left. We were all reporters for the story. We all went out, interviewed people in the industry and on the street. The way that we separated the work at least reporting-wise was, “we want to hit all of these to-dos for the week” and then we would divide up the work. I did the statistics. The rest of the work, figuring out if we had enough information on this topic, or enough experts, we divided between the three of us.
Source: How did the lookup interactive come about, where you can find your town?
CR: When we were starting to wind down and think about other things that were going to go with the story, we wanted to make the data available to people and provide them with a place where they could see all of it together. A lot of what we’ve been hearing people say they appreciated is that it was mixing a lot of data sources. For the first time they were able to see water loss, cost, age of infrastructure, pricing, consumption and demographics. So when we were thinking about the lookup, we wanted to prioritize making it easy for someone to get that big picture. That was something that I built. It used some libraries for the drop-down menu… it’s not the first time that a lookup has been done, and lookups are very popular for our audience, especially when we do reports for different places. It was fairly straightforward to build, I used lodash for a little bit of the cleaning… I didn’t want to use jquery at all; I tried to make it as minimal as possible.
Source: How did the other graphics come about?
CR: Chad Yoder on the graphics team helped build those. We wanted to have a scatterplot that was inspired by a concept of affordability which we do talk about in the story a little bit. The EPA developed guidelines in the late 1990s on how much of a burden water utilities could be on a person’s household income. So they determined that if you were spending more than 2.5 percent of your income on paying for water, it was not affordable. Since then people have tried to start with this way of looking at it and finessed it a little more. But the main idea is that we wanted to show what it meant for a water bill to be high when your income is not very high to begin with. For a lot of the comparisons we make in the story, we’re looking at the raw total, the bill. But in context, it really matters whether you’re paying $85 a month with a yearly income of less than $30,000 than if your income is close to $60,000. That’s where that scatterplot was coming from. It was true that across the board, non-white communities would have lower incomes, so we wanted to try to get that across.
The big, big chart with the map was based on the network map of connections that the state water survey had done, with updated links and also with the amount of water that was imported to or exported from each community—that information was coming from the LMO forms. There’s another version that’s more like a subway map, and you can see all the stops that water will go down. We included a graphic on the water basin in the area. That was something we had done for previous stories and wanted to include for context.
We also had a 3-D map of different water uses for Lake Michigan, built and reported by Jemal Brinson and Kori Rumore.
Source: What were some of the harder challenges that you ran into?
CR: Cleaning the data. Making sure that we were—because a lot of these datasets were either made by us or hand-entered—that we were double-checking and triple checking all of our numbers and more than that. Also, streamlining how we all contributed to the data gathering. When it came to the surveying of these 163 towns, that was work that we broke up amongst ourselves. So we wanted to make sure we were asking the same things, entering data in a way that wasn’t going to compromise the rest of it or not make sense to us later. Upfront, that was a big part of it. We stayed very much in touch all the time. What we found also is that all these municipalities did things a little bit differently and we needed to standardize as much as we could so that we could eventually compare one against the other.
What Caught On
Source: What has the response been like, from readers and others?
CR: We’ve gotten a very wide variety of responses from “look at my town” to “explore privatization” to “explore different ways to fund infrastructure” to “what about other utilities?” and “what about other states?” Even though this is a very data-heavy project, i think it resonated with people because it was a very straightforward project, focused on something that everyone uses, and everyone has an opinion on it. That’s something we didn’t want to lose sight of. Behind this water bill is a town that has to fund its infrastructure and deal with water loss, and deal with a system in which you’re passing along rate hikes. There’s no direct state oversight on that. Once it was published we were all very surprised at how well-received it was and how much it made sense to people. That’s a testament to Ted and Patrick who were the main writers. But also we were so entrenched in water for a long time that we almost didn’t think about how, this is important to most people and this is something that they hadn’t really seen a report on before.
…We’re not done with water.
Source: Any advice for news teams that wanted to do something similar?
CR: The most interesting thing, and what has resonated with people the most, is that you could ask your neighbor in the suburb next to you how much they pay, but they don’t really ask people who live forty miles away from you how much they pay. Even though the way that those rates are set is similar, there was no sense of “I’m aware that I pay too much.” There was that sense individually, like “I can’t keep up with my bills” but there was very little understanding of the whole picture, the average or the median water bill in the area. I think that for us it was very straightforward to start there because if we had wanted to answer the question, “How does the lake get to my tap? And why do I pay what I pay?” There wasn’t already an answer to that. Start with the very simple questions and those lead to other, trickier questions. At the end of the day, people will remember and hold on to what happened when you were trying to answer the simple thing.
Source: Right, people forget to ask those simple questions because they become part of the background of your every day.
CR: Right, people just think how much you pay for basic things is just, life. It’s just what you have to bear. It taught us a lot when we were working on it. I think it was successful because it was very immediate to most people, and it was surprising.
Source: Have their been any proposed changes to the water system, as a result of this investigation?
CR: It was difficult for us to even predict what a solution could be because there were so many factors in getting [the water system] to be the way it is now. It’s all very local—when we talked to residents, they’d say something like, “I complained to my village manager…” We’re trying to keep an eye on anything that could be introduced as far as wanting to change how the system works. The Cook County Board voted yesterday to hold a hearing on the disparities. Here’s our story on that.