Talking about Antifragility (Pt. 1)

This week, we've invited Strong Towns members to respond to a series of questions on Nassim Taleb's book, Antifragile. You should really read the book (it's a big inspiration for Strong Towns thinking), but if you haven't, you'll still find it easy to jump in on these topics and conversations, based on the second section of Antifragile.

The following is Strong Towns member, Marielle Brown's answer to the following question:

When considering a project or policy approach, local decision makers often want to know who else has done it and what happened there. Taleb suggests that we should look less at past examples -- good or bad -- and more at the potential for harm if we are wrong. For cities considering capital investments, how would we practically apply Taleb’s approach?

The first portion of her answer will be published today and the second, later this week.


Less is more; move the discourse to (anti)fragility.
— Nassim Taleb

The way we plan and make decisions about where and how to build is broken. The history of city planning is largely a story of meddling and overreaction creating ever more fragile cities by reducing any apparent volatility while increasing debt, building out a system that is not financially productive, and ruining the neighborhoods of our most disadvantaged residents. Even worse, planners and engineers have codified all of these disastrous decisions into zoning codes, engineering standards, and planning processes that have stopped cities from experimenting with solutions and learning how to address the serious problems we have created.

The following essay is a thought experiment on what a process that values antifragility might look like. The first part looks at what a practical application might be for cities already considering capital improvement projects. The second part (to be published later this week) looks at how we might redesign planning processes so that large capital improvement projects are seen as a last resort, not an expected part of an annual budget.

Nassim Nicholas Taleb’s concept of the Fourth Quadrant gives us a map for situating what kind of information and questions are relevant when it comes to capital improvement projects. Taleb divides payoffs into "Simple," where the question is “yes or no?” and "Complex," where the question is “how much?” He also distinguishes between Distribution 1, or Mediocristan, where outcomes vary slightly but constantly, and Distribution 2, or Extremistan, where outcomes vary by large amounts, but not regularly.

Table 1: The Four Quadrants (adapated from Silent Risk)

The pseudo-rigor of traffic projections makes it clear that our engineers, planners, and politicians think that capital improvement projects live in the First Quadrant where decisions are clear cut and statistics are well-suited. However, capital improvement projects fall along a continuum of complexity and cost, involving both Second Quadrant and Fourth Quadrant concerns.

Installing a curb ramp will have some uncertainty, and some chance of being destroyed by a black swan event, but a city has installed hundreds of these and will be able to give a relatively small range of potential costs. We can also be pretty certain that the project will achieve its purpose of helping people with wheelchairs, strollers, and carts access the sidewalk. On the other end of the continuum, the range of what mega-projects cost and the likelihood that they will accomplish their purpose falls squarely in the Fourth Quadrant, where it is more important to figure out how to minimize harm.

CenturyLink Field under construction in Seattle (image from Seattle municipal archives)

Key Questions for Engineers and Planners

A city considering a capital improvement project should consider both the Second Quadrant questions that involve estimating project cost and lifespan and the Fourth Quadrant questions about risk and harm, weighing the fourth quadrant questions more heavily as the project increases in scale and complexity. All cost estimates should be rounded and stated as a range to show the inherent uncertainty; as projects get bigger, larger ranges should be used in cost estimates and the numbers should be rounded to show less precision. An initial set of questions for the project engineers and planners should include:

  1. What is the total project cost?

  2. What are annual maintenance costs over a 40 year period? Costs should be displayed in a graph, not as an average, to illustrate that maintenance costs are often lumpy (i.e. repaving a road every 7 years) and tend to rise as infrastructure ages.

  3. What is the cost of replacing the infrastructure after 40 years?

  4. How will the project be paid for? If the project is financed through debt, what are the assumptions behind the future revenue growth? What happens if the revenues are a small fraction of what is projected?

  5. How might this alter the tax base over the next 40 years? For example, if a new road is removing homes and businesses, how will the city make up lost property tax? Could the new road degrade property values?

  6. How strong is the evidence for thinking the project will achieve its goals? Who produced the research on these types of projects? What are the critiques of these types of projects?

  7. What is the smallest incremental step for this project? Can it be tested temporarily?

  8. What will happen if the project utterly fails in its goals or is destroyed by an unforeseen event?

  9. Will the city be stronger or weaker with this project during the next depression or natural disaster?

Key Questions for Residents

The next questions start in the Second Quadrant and move into the Fourth Quadrant, and need to be answered by residents, with a special emphasis on the views of the least advantaged groups in the city, critics, busybodies, and practitioners in different fields. Some questions are deliberately repeated as the planners and engineers, and the public will likely have very different answers:

  1. How could the resources for the project be used to meet a different need the city has?

  2. Who will benefit from this project? Who will be harmed? Does the project increase or decrease disparity in opportunity and quality of life?

  3. What is the smallest incremental step for this project? Can it be tested temporarily?

  4. How could this project make things worse in the city? What other systems are we tampering with? What are the foreseen consequences? What is the worst case scenario?

  5. What will happen if the project utterly fails in its goals or is destroyed by an unforeseen event?

  6. Will the city be stronger or weaker with this project during the next depression or natural disaster?

If decision makers take the answers to these questions seriously and consider the risks they are taking, cities could avoid the worst projects. Not making very large mistakes is a huge step forward, but we need to nudge our cities into the third quadrant, where they are antifragile to Black Swans.

The second part of the essay will propose a different process, where capital improvement projects are considered the last resort, not the first option.

Read all our Antifragile coverage.


Related stories


About the author

Marielle Brown, AICP, is passionate about helping communities create lovable places that work for people of all ages and abilities. She lives in St. Louis, where she works as the Director of Policy and Strategy at Trailnet, a bicycling and walking advocacy organization. In addition to her advocacy and planning work, she has first-hand experience with multi-modal transportation planning around the world through her experiences living in Beijing, Hiroshima, Paris, and Seoul. Marielle received her Master's in Urban and Regional Planning from Portland State University.