Charles Waldheim on the Future of the American City

In this special episode, host and Harvard Kennedy School Professor Stephen Goldsmith and Harvard University Graduate School of Design Professor Charles Waldheim co-host a podcast crossover. Pr. Waldheim, host of the Future of the American City podcast and Director of the Office for Urbanization, speaks with Pr. Goldsmith about neighborhood regeneration, community-driven adaptation, and how to incorporate data into established decision-making routines. 

Listen here, or wherever you get your podcasts. The following is a transcript of their conversation.

Charles Waldheim:

Hi, I am Charles Waldheim. We're here today with Professor Stephen Goldsmith of the Data-Smart City Solutions Program in the Bloomberg Center for Cities at Harvard University. Stephen joins us today to share with us his podcast Data-Smart City Pod. Stephen, welcome.

Stephen Goldsmith:

Thank you. It's nice to exchange ideas and podcasts with you.

Charles Waldheim:

Only separated by a campus. I've really enjoyed coming to learn about your pod and about your larger program. I know that you've been focused for many years in your research on the role of data in decision-making and governance. There's some overlap between that and our mission in the Office for Urbanization and our Future of the American City pod. So hopefully we'll be able to compare notes here.

I just want to start by asking you what are the most interesting horizons for data and decision-making in municipal governance? Let's begin by talking about municipalities in this US, for example. What are the most interesting projects or initiatives that you see arising these days?

Stephen Goldsmith:

Well, let's start with why I'd rather be you than me just for a second. Data's okay, but data only makes a difference if it helps with the livability of a city or a neighborhood or a community, right? So how do we make for better cities? And that's what you specialize in. I think if I had to do it again, I'd rather be an architect and designer than just a talker about cities, so I'm glad we're exchanging ideas.

And to some extent, and maybe this is an opportunity for you to tell our listeners a little bit about yourself, Charles, but to some extent, just to answer your question and kind of flip it back on you, we're trying to look at how the digitization of cities allows for more livable places through better city services, better design processes, better thinking of a neighborhood as a community and not, we often use the phrase in our work that cities are organized vertically, but people live horizontally. They live at 10th and Main, they don't live in the parks department or the street department or the whatever department.

So many of our projects today say, "How can we get data, IoT data, analytics from transactional data and the like to help cities organize how they invest, how they maintain, and how they encourage communities?" And it feels like that is a cousin of what you do so well in your work in the Graduate School of Design.

Charles Waldheim:

Well, we like to think that you and your folks at the Kennedy School actually know things. We're invited to intervene in cities as they're airborne, so there's a slightly different division of labor. But it is true, I think we share an interest in urbanity, what makes for good environments for people, what makes for good outcomes for people even before they're thought of as citizens or consumers.

One topic I know that we share an interest in that maybe we could open up has to do with the role of AI and generative AI in helping cities make decisions faster or help understand more fully the desires and needs of their citizenry. Are there examples in the United States of programs that are using AI that you think are estimable or that we should keep track of?

Stephen Goldsmith:

Well, I think there's a couple ways to respond to your question. One that, again, is close to your work and one that may be a little more into ours.

So we've had a substantial amount of effort in thinking about how layering of data, think about mapping, spatial analytics, sense of place, open up insights about causation of problems or opportunities for solutions, whether you're thinking about emergency management or public health or how to design your urban park. To the extent that we can layer the data and analyze the data, we see things we wouldn't see, and that also serves as a platform for multi-agency or cross-sector collaboration because when you focus on a place, then you can have these insights.

We have work today we're dealing with funded by Robert Wood Johnson Foundation on how data will identify pollutants that are causing public health problems that are harming people in the community, right? There's a fair amount of work, as you know better than anyone, highlighted by Raj Chetty and others that suggest that the social infrastructure of a city, of a neighborhood will affect the opportunities of people in that neighborhood. So what are the greenways? What are the parks? What are the bike lanes? What are the sidewalks? How do people live together?

So many of the things we're looking at use generative AI and AI to better understand a place by allowing people, meaning officials and planners, to more quickly analyze what previously were too large of amount, too much data, to inquire of that data, to inquire that data with the community and to iterate it in a way that leads to better decision making.

Charles Waldheim:

I'm thinking of the work of Herbert Simon these days. Simon was the political economist, political theorist, won the Nobel Prize for Satisfacement, not exactly behavioral economics, but heading in that direction, the kind of limitations of the optimal decision maker in a market. I'm thinking about that and his argument that more and more data didn't always correlate in his research to better decision making. That's obviously something you've spent a career, a lifetime thinking about. Help us understand how at certain points, a quantity of data, a volume of data transforms into something that's much more qualitative.

Stephen Goldsmith:

So I think it's an interesting question, and I would definitely say the perfect can be the enemy of the good. So here's a really mundane example.

So a few years ago, when Bloomberg Philanthropies funded an effort to catalyze the use of more data in cities, the city of Chicago was an early awardee of a substantial grant. We went out to help them a little bit with their talented CIO to create more of a data analytics center effort. And she was having difficulty getting the operating agencies to pay attention because they had day-to-day problems, they weren't interested in data exercises.

So then she said to them, "Okay, name a problem that if I could find the data, I can help you understand how to address that problem more effectively." And the first problem she suggested was rats. So the issue was could we predict where the infestations are going to occur before they occurred? And can we identify the causes of the infestation in a way that allows us to intervene?

Now, that's not quite as the same and fancy stuff that like Maurice Cox of your organization has done in designing places in Chicago. But it is an example, I think, where a city's tendency to do things by routines can be improved through some uses of layered data which identify and help you address causation.

You've done so many creative things with data helping regenerate a neighborhood or design a neighborhood park or retail corner. How does your team think about the use of data as part of the planning exercise?

Charles Waldheim:

Well, as you suggested in your answer just now, it has to do with beginning with an interesting or a good question. Sometimes I've been in contexts where there's an initial desire to have all the data, and that usually doesn't lead to very productive outcomes in my experience because it's most often beginning with, as you say, a problem or a concern or something that's meaningful to people, that they're emotionally connected to. And also beginning with an insightful question where the application of data, empirical observation, testing these things can be useful.

So I lead a research group called the Office for Urbanization at the design school here at Harvard, and our mandate is to help folks think about change in the built environment. We work with NGOs, foundations, governments. And we've worked in the US and around the world on a range of different topics, but almost always we find a way to identify existing data.

And in my experience, most of these challenges really that we confront are not really met with a paucity of data. Almost everywhere we work, there's quite a lot of information. And mostly, our advice is trying to align the information with the decision-making apparatus, the policies, procedures. As you said about mapping, you reminded me of when Shaun Donovan was appointed HUD secretary in the Obama administration, he's a graduate of the design school. And he said that when people make a housing decision, they're also making an education decision. They're also making an employment decision. These things are all connected, right?

So I like this formulation that you have, Stephen, of the horizontality that we occupy. We live horizontally in ways that are geospatially connected. And I think the geospatial is another area where obviously we would have quite a lot of overlap.

Stephen Goldsmith:

You've commented a little bit in the past about The High Line in New York and we're funded by Knight Foundation and they had us convene groups like The High Line, similar but not quite so dramatic in DC as well. If you wanted to think about if a city wanted to copy that experience, what data should it look at? How does it involve the community in the planning exercise so that the planning exercise isn't just superficial, but it's meaningful, right? What data would they present? How would they visualize it? How would they capture the feedback?

Charles Waldheim:

It's a great question, Stephen. Thanks for that. We found in one of our portfolios, we have three primary portfolios of research in the Office for Urbanization, one of which is obsolete or redundant transportation infrastructure. So I happen to be curiously an expert in abandoned airports for reasons that we can go into, but abandoned or redundant elevated highways or rail lines. These are, of course, infrastructures that have been made redundant by economic and social change, this happens around the world. And we've been involved in a number of projects. We were engaged in Buenos Aires, working with the Argentine government, Inter-American Development Bank, and World Bank on the development of an elevated High Line style park on a former elevated highway.

In many instances, we begin by trying to talk people out of the idea because simply copying an idea, no matter how good it was, in context may not be necessarily the right solution. And perhaps the most important thing to say is that in different cultures, we do a lot of work in Latin America, a lot of work in East Asia, in many cultures around the world, we find that there are much more fully developed political economies or forms of communication between civil society and elected and appointed officials.

When we work with an elected or appointed official in most of Latin America, they will be introduced with the number of days they have left in office and how many projects that they have to build, and they will deliver to us a fairly robust form of civil discourse. We'll meet with the community leaders, that'll be well organized.

Of course, in the United States, things are a little bit different than that. The first slide, the first image I was shown as a student of architecture in the 1980s was the demolition of the Pruitt-Igoe Housing Project in St. Louis. The first thing I was taught was the sins of our grandparents and the failure of the modern project, and we've internalized that as designers. Somehow we were responsible for the failure of that social project.

So in that context, The High Line, as you mentioned, is a wonderful example of a community meeting. As you may know, the Giuliani administration and its planning department saw The High Line as abandoned, redundant kind of freight infrastructure to be an impediment to development of Chelsea and the west side of Manhattan. And in fact, it was at a community meeting. Now, that neighborhood, let's say it's a special kind of community, but it was through a community meeting process in which the Friends of The High Line was formed.

The second piece of that was, of course, the Bloomberg administration's understanding of the ability to partner with them and that public-private partnership. So in that context, the kind of specific information that I look for is how well organized is the community? What levels of trust might there be between elected and appointed officials and citizenry?

The other thing I would say is that The High Line is a good example of in each context that I see innovative projects happening around the world, they don't always go through the traditional channels. More often than not, I'm surprised by them going through a different pattern or a different process, let's say. And often, it's the development of that workflow, that critical pathway that becomes maybe as interesting as the built realm itself.

Stephen Goldsmith:

Never met an expert in the obsolete before. This is quite a credential.

Charles Waldheim:

Yeah, it causes my family to wonder where we go on vacation occasionally but...

Stephen Goldsmith:

One of the things that interested me about some of the work in design school, we've had Maurice come over and talk to our cities a couple times, is this complicated issue of when and where do you inject public capital to try to catalyze the rebirth of a community? Because there's, in many urban cities, particularly Chicago, but others, there's a lot of areas of opportunity where disinvestment or redlining over the years has caused a price.

As part of the planning exercise, how do you think about that? What data would you get that would help you figure out, okay, these three places have potential? And then honestly, though you probably not say it out loud, these three places don't have enough discretionary spending? Because with the mayors we bring together and the chiefs of staff, they're all highly motivated, but they only have so much public capital to invest. So if you're trying to invest it to create a more livable place, what criteria, what would you look at in terms of data to draw a conclusion?

Charles Waldheim:

Well, you mentioned Maurice Cox, our colleague at the design school who we've had on the Future of the American City podcast series when he was director of planning in the city of Chicago. And I can tell you what he told me in that conversation, which is in a city where so many of the public resources and so much of the private philanthropy had been devoted to a relatively small area on the lakefront at the center of the city, his goal in that role was to try to distribute public projects and public funds across various entities across every zip code or every district within the city where there was a train stop. Wherever the L stopped, he wanted to be present and see some visible evidence of some kind of public project.

And that kind of push-pull of the centralization, the desire to build a Millennium Park in Chicago and have these kind of crown jewels on the lakefront as cultural assets, and then the redistribution of that back toward the neighborhood. So that's a long question in American cities, as you know.

But among the data that he was looking at, the things that I would look at would be how far do kids have to walk to get to a playground or a park space? What is the relative distribution of the maintenance of those park spaces? Do people have access to recreation? You've touched on questions of public health, air quality, heat index. These are, of course, thoughtful as well. And I would link that also to the positive outcomes that we know people have when they live in proximity to recreational amenities and parks and open space among them, the most fundamental.

As you look at municipalities in the United States, particularly if you think of Chicago or New York or others, the way in which they organize their maintenance and their budgeting for those kinds of public resources are also highly uneven. So understanding that each opportunity is not going to be made available for public resources as public resources are necessarily limited, identifying the places where a modest amount of increment can in fact leverage some bigger, and that will depend ultimately upon private capital as well.

Stephen Goldsmith:

I know we don't have much time left, but I want to compliment myself for a second. I'm taking total credit for the work of another one of your professors, Toni Griffith. I was the chair of the Anacostia Redevelopment Commission in Washington DC. Toni was chief planner, architect, and one of the top officials. Southwest waterfront, area around the Navy yard, area around the Washington Nationals baseball Stadium, wonderfully successful, almost amazing in what they've done to the city.

How do you think about whether when enough is too much, right? When the gentrification is too extreme, right? You can't say there's not going to be any. Is there data you look at? Is that a planning exercise that we get that wrong at the beginning? Were we too successful in those efforts? How should I think about that?

Charles Waldheim:

Well, I appreciate you acknowledging my colleague, Toni Griffin, who's an extraordinary planner and architect. Has been involved in work in a range of leading US cities. Has been involved on our end at the design school, helping to organize the Mayors' Institute on City Design where they invite mayors and directors of planning to bring her, bring our colleagues their biggest, hairiest problems, right? That's one way in which we can be of service.

The question of gentrification, of course, it goes hand in glove with our political economy and the ways in which we leverage private capital to develop cities. I think most of the people that I've talked to in that context suggest that, in fact, we don't do nearly enough to understand the limits of gentrification. And that I'll go back to my comments earlier about the political economy and relative level of trust. I think I see very few examples, unfortunately, a fairly robust trust-based relationships between communities and their elected and appointed leaders around questions of the built environment. I think there's a long history of that in which most elected and appointed officials come in and want to do good for people as they can and find themselves in a context where they don't necessarily have the mechanisms or the venues for that kind of conversation.

My friends that are engaged in the kind of work tell me that this requires years and years and decades of ongoing continuity. And at the same time, we have political frameworks in which people are elected and appointed for four or five-year terms. So that kind of relationship and the timing of capital, for example, we were invited through the Office of Urbanization to do work in South Florida. The mayor and city councilors in Miami Beach invited us down to help them think about they had water in the streets on a clear blue day, clear blue skies, so-called nuisance flooding in the context of South Florida.

The mayor was elected there on a YouTube advertisement of him kayaking down the street full of water on a clear blue... And he was arguing, "This is not a left-right, not a red-blue issue, this is simply business. We can't operate the city this way." And as they were engaging in pumping water back into Biscayne Bay and elevating roads, what they encountered was that most of the legislative, most of the juridical, most of the legal frameworks coming out of the '70s and '80s were really fighting the last war, frankly. They made most of the regular adaptation efforts almost impossible.

So we were brought in there as advisors, and what we found there was that, in fact, individuals that were being impacted by this set of activities wanted more engagement, they wanted more opportunities for conversation and input. And that really, at the level of the structure of governance, most municipalities don't really have those structures in place.

Stephen Goldsmith:

Yes. Just maybe a final comment. We're looking more and more at how data visualizations can enhance community engagement and lead to more participatory and interactive design process. We have it in infrastructure, we have it in health and a couple other areas. This, for us, has proven to be more difficult than I expected. It requires a set of skills that maybe are in your shop, which is the visualization skills. It requires some software to allow interactivity, whether you want to call it augmented reality, whatever, so that you can see different scenarios. It requires a way to capture engagement and ideas from people who aren't present in a physical meeting, and then translate that into iterative policy. And we look forward to maybe your help in trying to help us figure that out because we're not seeing a lot of great models. Are you?

Charles Waldheim:

Well, that's central to our preoccupations these days. We've been, for the past five years, collaborating on the North Shore here in Massachusetts with folks on Cape Ann, that's Gloucester, Manchester, Essex, Rockport. They called us in January of 2020 when they found that they were not really getting the kind of attention that other parts of Massachusetts were getting in terms of adaptation to an already changing climate. So that's maybe a good example to illustrate some of the ways in which we use visualization.

That research is heavily data intensive, and most often we're borrowing, loaning that data from people in the sciences, whether they be hurricane modelers or geoengineering corporations or a range of different disciplines. So we don't really acquire original data, we're not going into the field to measure things, but we're partnering with people both on campus here, at MIT who have deep, deep bodies of knowledge in these areas.

And in those contexts, what we found was our work was primarily scenario based planning. It was if this, then that. And we did a project with them looking at what we call a scenario zero. Most often when we work with folks, you get off a plane with a big H on your sweater and congratulations, we're here to help. And somehow you're dealing with folks who think that we're, you and I are agents of change.

And I try to diffuse that immediately by saying that, "We're not here to make decisions on behalf of this local community. Ultimately, what we can do, what we are quite good at is looking at the steady state conditions. So a scenario zero in our parlance would be just assume continuity. The world doesn't work that way. Things change all the time, whether it be climate patterns or demographics or birth rates or economies, but we just try to model out what will happen, Ms. Mayor or Mr. City Councilor, over the course of the next 5, 10, 15 years."

And one exercise we engaged in was modeling a modest category two or category three tropical storm or hurricane. That's something that brings data from a range of different disciplines and professions on campus, but fits into our skill set in the design school and our ability to visualize and quantify what kinds of impacts might be statistically available because we have very good data on the history of storms, and we can then project forward and model, in scenario planning, a way of thinking.

And what this helps people to do is maybe to get out of the existing binaries that they're used to. It's the developer versus the community or the mayor and city council against this local business, to get them past that by thinking about a future condition that they could have input on. So as you know, scenario planning is quite good at revealing, not predicting the future exactly, but revealing these secondary and tertiary non-obvious connections. That we went into this exercise and we didn't realize, oh, this is leading now to a very robust emergency preparedness and evacuation planning exercise on Cape Ann. That was not something that we started with, but led the community to in that work. So yeah, we'll look forward to collaborations with you and your folks in the Bloomberg Center when you have an opportunity.

Stephen Goldsmith:

Charles, I've enjoyed my time with you. And maybe as I continue it, I'll be able to apply to your program and learn how to be like you when I grow up.

Charles Waldheim:

We'll be happy to host you.

Stephen Goldsmith:

Thanks for your time. I appreciate it, and hope we can have more exchanges.

Charles Waldheim:

My pleasure.