#1: Mapping Wicked Problems — The Lack of Affordable Housing in Pittsburgh, PA
A reflection on our approach to mapping the lack of affordable housing in Pittsburgh, PA.
By Team Emergence: Tomar Pierson-Brown, Joe Nangle, Janice Lyu, & Bingjie Sheng.
It started off as a relatively straightforward proposition. Assess the number of individuals in Pittsburgh needing affordable housing, identify the number of affordable units available in Pittsburgh, then subtract. The scope of the problem of a lack of affordable housing in Pittsburgh should be easily quantifiable.
But wait — doesn’t every person in Pittsburgh need housing they can afford? It’s really a question of who can afford what; how much of what housing is available is affordable for the most people. Then again, just because a housing unit is available, doesn’t mean it’s accessible, practical or safe. What is accessible, practical, or safe depends both on the individual and the demographic groups that individual identifies with.
Defining the scope of a problem like the lack of affordable housing is challenging because doing so raises many questions, each of which raises further questions. One factor is tied to several other factors, rooted in several other problems. Issues that present with this level of complexity, that have endured over broad spans of time, and which are rooted in other macro- and micro-issues, can be considered “wicked problems”. Wicked problems are also referred to as systems problems because they possess many of the same intractable and self-reinforcing characteristics as institutional systems.
Coined by design theorists Rittel and Weber, “wicked problems” are difficult to resolve because they emerge from, or are the product of, multiple systemic and institutional structures.
On the surface, the wicked problem of a lack of affordable housing in Pittsburgh is a product of the relationship between governmental regulatory systems that create and enforce zoning laws and the economic systems that incentivize housing construction and mortgage lending. The institutions of our state and federal government and our economy perform in response to overarching social systems, such as racism and classism.
Team Emergence, comprised of participants in the Transition Design course at Carnegie Mellon University, has been tasked with using the Transition Design approach to unpack and address the wicked, systemic problem of a lack of affordable housing in Pittsburgh. In this post, the first in a series of five, we will discuss the anatomy and dynamics of our wicked problem, as depicted in our systems map below. We will also discuss the challenges of systems mapping as an approach to depicting complexity. Finally, we will summarize the relationships highlighted by our map and the insights those relationships fostered.
The Challenges of Surfacing Problem Structure and Dynamics
One key element of wicked problems is their interconnectedness. Systems mapping aims to visualize these complex interconnections, and in so doing, make it possible to understand effective points of intervention. By organizing our map around 5 categories, often abbreviated STEEP (social, technology/infrastructure, environmental, economic, and policy/legal), we purposefully examined the problem from several perspectives.
But when mapping a wicked problem, does one start with cause or effect? How objective or subjective can the process be?
After dividing research assignments between the STEEP issues, we examined each category in a standalone discussion. Taken in isolation, there appeared to be a strong case that a lack of political will to push for more affordable housing was at the center of this wicked issue. It seemed that many of the effects spiraled out from this singular cause. Transition Design, however, encourages mapping problems as a necessary precursor to framing wicked problems, so we endeavored to put our preconceived conclusions on hold in order to be as open to learning from the process as possible.
As we merged the isolated topic-specific research into an iterative series of combined maps, it became clear that pinning responsibility to any one element of the system, or even a single category, was shortsighted.
The process of mapping a wicked problem challenged our notion of cause and effect. In many instances, what appeared to be a cause in one category could easily be recast as an effect of another.
For example, neither Pittsburgh nor the state of Pennsylvania has passed legislation to mandate a minimum wage above the Federal minimum of $7.25 per hour. Stagnant wages jumped out as a policy-related cause of our wicked problem. It would be easy to conclude that housing is unaffordable in Pittsburgh because wages have not kept pace with other market factors. But at the same time, efforts to raise the minimum wage would fail to meaningfully improve access to housing. Although Pennsylvania Governor Tom Wolf has indicated support for an increase to $12 in 2021 and $15 by 2027, these stated objectives would not guarantee a living wage, even if the 2027 target were enacted immediately. If the minimum wage jumped immediately to $15/hour, only workers or families with no children would earn a living wage. From this perspective, stagnant wages can be viewed as an effect; perhaps of the rise in neoliberal politics. Elsewhere on our map, we identified political disinvestment from “big government” as a cause of inflated housing prices.
The process of surfacing wicked problem structure and dynamics was challenging, given that the distinction between cause and effect was so often a matter of perspective.
The Anatomy or Structure of the Problem
Our team was given a two week span to map the wicked problem of lack of affordable housing in Pittsburgh. We conducted a STEEP (Social, Technological, Economical, Environmental, and Political) analysis to get a detailed overview of what external factors may be contributing to the affordable housing issue in Pittsburgh. During the first week, we each took on a specific issue domain and individually conducted research. At the end of the week, we transcribed the data we gathered onto a post-it and displayed it on a Miro board.
We shared our findings in roundtable format and began to generate key findings that summarize specific learnings or collection of data. We began mapping these key findings to post-its on our Miro board. With key findings visualized, we were able to begin drawing connections to issues within the categories.
As we started this exercise, we began to identify issues that were more directly related vs. sub-issues that stemmed from these direct issues. Therefore, we next distinguished between the direct and indirect issues with colored and outlined boxes (colored for direct issues and outlined boxes for sub issues). Once we made all the connections within a category, we also began to make connections between different categories using fine vs. dotted lines for distinction. We discussed as a group and agreed on the relationship between issues when making inter-category connections. We also distinguished which issues were effects vs. causes but quickly came to find that several issues behaved in a feedback loop where issues we saw as effects circled back and became causes of lack of affordable housing.
After we made these inter- and intra-connections, we saw how convoluted our maps had become. It struck us as a great visual for any audience doubting how complicated and intertwined these wicked problems are. But the intertwinement and criss-crossing lines made it difficult to follow key relationships.
So we spent time trying to make a cleaner map where direct issues circled the center, while sub-issues branched out. This way, connections made within the circle showed direct to direct inter-category issues while direct to indirect inter-category issues were connected with lines outside the circle.
Summary: The Dynamics of our Wicked Problem
Creating our systems map led us to several takeaways. First, wicked problems must be analyzed contextually and systematically. Taking a look at our final draft map for the first time, one is likely shocked by its complexity. Indeed, we thought the same thing when we first tried to connect all the issues. The connections within and across the problem sub-categories kept reminding us of the multiple levels of systemic complexity.
The second lesson reinforced is simply: don’t jump to conclusions. When we first discovered that political issues were closely related to our wicked problem, we were tempted to put the politics issue category at the center of the map. But his approach would have limited our ability to explore more problems under other issue categories. In the end, based on Terry’s feedback and our own reflection, we decided not to draw conclusions during the map-making process. Instead, we kept pushing ourselves to suspend judgment and to let the final mapping outcome show us the plurality of conclusions. Jumping to premature conclusions is a sign of oversimplifying the wicked problem.
Further, the way we curate the information on a wicked problem map reflects our way of thinking, even as the activity itself shapes our thinking about the wicked problem. This level of metacognitive awareness was both one of the most unexpected yet enriching aspects of creating our map. Documenting the seemingly inextricable chaos of complexity might have caused confusion or fogginess in our understanding of the interrelated issues. Rather, as we progressed toward a circular depiction of direct issues haloed by interconnected subissues, our fluency in the issues and how they link to one another improved. The evolving order of the maps reflected the increasing order of our understanding.
Further, the demands of creating a bounded representation of an expansive problem impacted how we understood the complexity. By the end of the exercise we could each distinguish direct issues that drove other direct issues (the legacy of redlining as it relates to the value that bodies of different race has on property values, in figure 3, above), from direct issues that had consequences that reinforced other “subissues” (i.e. affordable housing that is far from public transportation forces increased reliance on individual vehicles, in turn contributing to poor regional air quality, exacerbating health conditions, particularly in low-income households who lack access to quality health care, in figure 3, above).
Mapping the system dynamics revealed opportunities for change as well as the challenges of implementing solutions. The causal dynamics depicted in our map underscore that many existing policies and practices are short-sighted. For example, for the same number of residents, a well-designed multi-story affordable housing building is much more energy efficient than many single-family houses. In private construction, however, regulatory authorities have little control over issues such as construction pollution, waste of resources, etc. These problems lead to global warming and environmental pollution that further affects people’s health. Without wisely targeted intervention, these feedback loops will exacerbate the existing wicked problems and create new ones.
For example, while there may be cause to celebrate major tech & innovation companies coming to Pittsburgh and bringing new opportunities, this change is not free of side effects. With more high-income jobs created, rising real estate and land prices are making housing even less affordable, worsening the affordability of housing in Pittsburgh.
Finally, the impact of the COVID pandemic on this wicked problem cannot be ignored. COVID-19 adds an additional layer of complexity to the overall system. Its unknown boundaries may make past experience inapplicable to the current situation. Economic and public health pressures may leave the government with less money and communities with less energy to invest in supplying affordable housing developments.
Ultimately, iterating upon several versions of the map allowed us to not only explore broadly, but also understand the complexity of the system that has created a lack of affordable housing in Pittsburgh.
In our next assignment, we’ll begin to introduce specific stakeholders to the mix, bringing in a new level of specificity, and bringing us closer to understanding how to intervene in this dynamic system.
