What Strong Towns Really Says About Infrastructure Spending

A couple of months ago, a Substack named Arpitrage — written by Arpit Gupta — presented a column called “Contra Strong Towns” that got a few people in our circles worked up and asking for a response. Sadly, Contra Strong Towns had nothing to do with the classic video game nor was it a thoughtful academic critique of our ideas, so I’ve kind of let it sit.

“Contra” is Latin for “against,” which is kind of what the article was: motivated reasoning against an avatar the author called Strong Towns. Even the first section, which was supposed to be steelmanning Strong Towns ideas (presenting them in their strongest form for later refutation), was merely a superficial statement of areas of perceived agreement.

Either way, there were a few assertions on Arpitrage that I have been asked to weigh in on publicly, so I’m going to do that here. Let me first, however, quote what I think is the main contention of the author: “Suburbs are not Ponzi schemes, they are not inherently fiscally unsustainable, and in fact cities are the ones host to a lot of cost bloat issues.”

The “Ponzi” Scheme

I wouldn’t spend much time on this section, but it signaled to me throughout that the author wasn’t seriously engaging with our ideas. Whether it is a focus on “suburban sprawl” — which isn’t a term we use here — or the comparisons of density — again, specifically not our jam — the language used told me that, at best, the author’s priors were an impediment to hearing what we are saying.

This clouded understanding leads the author to use the strawman in excess. For example, I don’t recognize the presentation of my supposed thesis. Arbitrage wrote this: “The argument is that suburban sprawl literally does not pay for itself when considered on a lifecycle basis: that the deferred maintenance and lifecycle costs of suburban developments are higher than the benefits, such that bankruptcy is inevitable.”

If I were to restate this in a way I have actually asserted, I would say this: The argument is that the Suburban Experiment (an urban, suburban, and rural phenomenon) encourages cities to pursue short-term cash advantages and discount accrued long-term liabilities, leading to higher taxes, a reduction of services, deferred maintenance of critical infrastructure and, ultimately, municipal insolvency.

The Suburban Experiment (not suburban sprawl) is an approach predicated on large leaps in development followed by stasis, the opposite of a complex and adaptive system (this is literally Chapter 1 of my first book). Deferred maintenance is a consequence, not a cause. Local governments rarely declare bankruptcy — we’re not predicting that, and that isn’t what “insolvency” means. Cities' insolvency just makes them increasingly desperate, self-serving and predatory, the opposite of what they should be.

I’ll admit, it is also a bit triggering to me when academics talk about costs and benefits. It is another area where we’re generally not talking about the same things. It was a throwaway line in the article, so I’ll only note this as another likely divergence in our ways of viewing the world.

Advocates Exaggerate Infrastructure Spending

This is actually the most substantive part of this article, although it is still a weak version of an old argument. I’m going to do my own steelmanning because it is important to understand what is being asserted, which can be seductive.

Steelmanned Argument: "Cities typically spend a small amount of their budget on infrastructure. Most of what they spend money on can be categorized as services, including police, fire, park maintenance, elections, etc… If cities were in a real pinch, they could double or triple infrastructure spending and it would still only be a minority of the budget. Thus, the claim that cities are insolvent because of deferred infrastructure spending is wildly inflated."

Even if we agree that cities are not budgeting enough to maintain their core infrastructure, the pertinent question is this: How big is that gap? The answer is that nobody knows because cities don’t track and report this information. In fact, cities bizarrely account for infrastructure as a depreciating asset instead of what it really is: an infinitely recurring future liability. We sometimes get glimpses of how bad the gap is — like in Lafayette (Louisiana), Houston, and Silicon Valley — but there is no official accounting.

To overcome that knowledge gap, Arbitrage relies on a comparison of a one-year budget for Minneapolis and one of its second-ring suburbs. This is a bit like looking outside the window, seeing that it's raining and predicting that the climate is in for an extended wet cycle. City budgets vary wildly from year to year.

Even so, just looking at the cash budget is leaving out a lot, especially in terms of infrastructure. Most of a city's infrastructure spending is found in its capital budget. Things like interest on debt, shifting salaries, inter-fund transfers and many more line items obscure where the money is coming and going.

These are technical refutations of the argument in the column, but the core issue here is that the Arpitrage site displays a misunderstanding of the difference between infrastructure spending and all the other spending that the author cites.

Put plainly, infrastructure is investment spending and the other services are consumption spending.

When a city makes an investment in a new road, pipe or water tower, that investment needs a financial return. It needs to generate enough tax base and revenue to not only cover its ongoing costs, but also generate enough return to pay for all of the other consumption spending. Where else does that consumption spending come from?

Relatedly, nobody expects fire departments, police services or election officials to create a return on investment. There is no investment; this spending is consumption. Its continuation depends on the city having enough revenue from the return on its infrastructure investments.

When we treat infrastructure spending like it's consumption spending, something like having a police department or maintaining a park, we tend to make stupid infrastructure decisions. Again, we’ve documented this many, many, many, many times. And the further we go down this path, the less we can afford those consumption services — even the ones like public safety that we don’t generally think of as optional — and that impairs the tax base.

You can’t just triple your infrastructure spending, even if that was enough to meet your obligations (it almost certainly isn’t) without negatively impacting the feedback loop that includes property values, tax capacity and municipal services. This entire argument is based on a fundamental misunderstanding of the complex ways things actually work in a real city.

Urban Costs Are Typically Higher Than Suburban Costs

Arpitrage argues that “the real crux of the issue is that infrastructure costs are typically higher in urban areas than suburban ones.” That may or may not be true — the oddly chosen data provided in the article really doesn’t show that — but what is notable is the lack of curiosity as to why this might be the case.

If urban infrastructure is really more expensive than suburban infrastructure (and I’m inclined to think it is), there are a lot of reasonable explanations that are consistent with a Strong Towns thesis.

First, and most obvious, is that older cities have older infrastructure that requires immediate attention, while a brand-new suburb has most of its present infrastructure costs floated in residential mortgages and commercial real estate loans. Comparing the budget of a young suburb to a 200-year-old city is going to show the one with immediate needs spending more on infrastructure. This says nothing about their long-term solvency.

Second, it is true that it is typically far more expensive, on a per-foot basis, to lay pipes and build streets in established urban areas than it is in a greenfield, suburban location. That is true even in small towns that have fewer established constraints to work around. Just trying to achieve positive drainage from a stoop to the curb on a block of existing buildings is a massive challenge, let alone everything else needed to maintain critical infrastructure. The new suburb has few of these immediate challenges.

Yet, obviously, things need to be maintained, so it’s not clear what the assertion that maintenance of an existing city’s infrastructure costs more than a new suburb is getting at. That suburb will eventually need to be maintained, and then their costs will go up. That’s what we’ve always suggested will happen.

Plus — and this is even more core to our observations at Strong Towns — that foot of maintenance in the city’s core may cost five or 10 times as much as the same maintenance on a suburban cul-de-sac, but the tax base is going to be 50 to 100 times higher per foot in the city, too. Costs mean nothing without an understanding of the financial productivity derived from those investments.

Finally, and this is a bit gratuitous, core cities spend an obscene amount of infrastructure money subsidizing the commutes of suburban dwellers, including things like excessive transportation infrastructure or even the draining of water from parking lots. And those are examples of hard costs; the soft costs of reduced property values and lower quality of life are also very real. Cities are waking up to this reality. I suspect, as federal transportation subsidies wane further and old-school staff enter retirement, that cities are going to spend less on self-harm and more on becoming strong and resilient. That should narrow the gap somewhat.

Toward Better Urban Governance

This is one place where I broadly agree with the Arpitrage column, although I suspect we use different reasoning to get there. Yes, I agree that we desperately need better urban governance. Desperately.

And, for my friends who are reflexively pro-government, please understand that I’m not making an anti-government statement. Local government is really important, and elevating it is a core principle of Strong Towns. And because we believe it is important, we also believe that it needs to be better.

In a prior section, the author asserted that “cities are more costly to run than suburbs.” He suggests that this is true on a per capita basis, and I have no reason to suspect that it isn’t. I think what is going on, though, is a bit more complex than a simple apples-to-apples comparison.

A core insight of the Strong Towns approach is that cities function best when they are treated like complex, adaptive systems. In fact, our observation is that cities were complex and adaptive prior to the onset of the Suburban Experiment. Centralized economic policy after World War II sought to turn cities into engines of growth. The catastrophic capital — to borrow a phrase from Jane Jacobs — funneled into America’s Suburban Experiment precipitated a phase shift in our cities, turning them from complex to merely complicated.

Complex cities are adaptable and grow strong over time. Complicated cities (cities treated like machines) are nonadaptive and grow fragile over time. The catastrophic capital that created this shift — the effect of an abundance of resources that is familiar to those who have studied complexity — manifests in different ways in different places.

It is my observation that, in large cities, the catastrophic capital of the Suburban Experiment has made the government itself — the people and the systems they work in — less responsive and innovative than they otherwise would be. There are enough resources to prop up existing systems, so there is enormous resistance internally to even modest reform, despite the great pain that these systems create. In our cultural conversation, this historically has resulted in conservatives (often disingenuously) pointing out the obvious problems and progressives feeling obligated to defend these systems in public while expressing frustrations in private. This is a tragedy.

In America’s suburbs, the catastrophic capital of the Suburban Experiment manifests differently. It is my observation that the part of the suburban (and small town) system that becomes nonresponsive and lacking in innovation is the development pattern itself. Many suburbs have very efficient governments, and they are efficiently executing a horrible investment strategy that is leading them to insolvency. Once again, our culture reinforces the lack of innovation with conservatives blindly defending the subsidies, mandates and intrusive government necessary to prop up suburban development patterns and the suburban way of life, while progressives are satisfied with trading truck-loads of highway spending for a shovelful of money for sidewalks and transit. This is also a tragedy.

I can’t find a place where I’ve written this before, although I’ve said it publicly during Q&A sessions many times: If we wiped out all of the debt and pension obligations of our major cities, while simultaneously reforming their governments to be less siloed and hierarchical and more organized around geographically focused teams, the enormous productivity of these places would instantly give them a prosperity that they could sustain indefinitely.

If we wiped out all of the debt and pension liabilities of a second- or third-ring suburb and then prompted them to continue operating with their current business model, in a generation they would be taking on massive amounts of debt, raising taxes and cutting services. The underlying lack of financial productivity in this pattern of development can’t be sustained. 

The Arpitrage column reveals a glaring blindspot in the author’s worldview with the statement that “the competition between suburbs and cities is one of the main restraining forces keeping cities as functional as they are.” In other words, cities would be an even bigger mess if they weren’t pushed to compete with the suburbs.

The problem with this statement is that cities are not competing with suburbs. In fact, as I wrote in "Strong Towns: A Bottom-Up Revolution to Rebuild American Prosperity," cities are playing what game theory describes as an infinite game. In an infinite game, the primary objective isn’t to win; it’s to keep playing. Any serious discussion of reforming urban governance needs to start with understanding the implications of playing an infinite game.

An Invitation to Academics

The Arpitrage arguments in the "Contra Strong Towns" article reminded me a lot of the arguments I used to get from the Market Urbanists. In response, I wrote a seven part series called "My Journey from Free Market Ideologue to Strong Towns Advocate." I don’t hear from them anymore, so either I convinced them or they moved on. Or I just wore them down.

I am eager to engage academics because one of the things that the Strong Towns movement needs and is ready for is academic research and validation. I’m not an academic. I’m a practitioner. I love writing, but I loathe writing research papers. I am curious and can be an obsessive autodidactic when I need to understand something, but I lack the kind of patience that is needed to go super deep on one technically narrow topic.

The early days of Strong Towns involved me chasing ideas and insights wherever they took me. I did dozens of financial evaluations of different developments, anything I could get numbers for. I wasn’t out to prove a point; I was trying to figure it out. I haven’t done an evaluation like that in more than a decade, not because it isn’t important, but because it’s not interesting to me. I don’t need to see another insolvent cul-de-sac or suburban big box development to know the numbers don’t add up. It was never close anyway.

So, to finish this response to Arpitrage, I’d like to offer some missing insights of the Strong Towns conversation that I think academics should pursue. I’d love to work with someone, or some team, that is interested in these inquiries (or others, if they are interesting).

1. Financial Correlations. If we are correct, there should be strong correlations between the age of postwar suburbs and things like debt levels, tax rates, and even fines, fees, and other charges. As a place ages, all of those things increase. I pick these because they are knowable numbers — unlike infrastructure liabilities — and gathering them should become easier as AI technology allows us to pull numbers from old budgets and reports.

2. Productivity Analysis. We’ve done a lot of this, and Urban3 has done way, way more, yet this is an area where an academic approach would be useful and accessible to more people. What is the real financial return of different development patterns? How does this relate to easier metrics to compute, like value per acre? There are old textbooks and design manuals that show these calculations, so part of this would be resurrecting lost knowledge. We need that kind of renaissance.

3. Narrative Analysis. I’m really fascinated with the work that Ben Hunt at Epsilon Theory does tracking narratives (check out their Fiat Narrative Dashboard). These are the stories that we tell ourselves, and they can be as predictive as any other statistical data source. There are decades of articles and newspaper clips documenting growth and change throughout different communities at different periods of development. The Strong Towns thesis suggests that these narratives change over time as a city matures, going from blind and naive optimism about growth and development to adversarial as the interests of residents and their local government diverge.

4. Analysis of Governance. Earlier this year, I interviewed Eric Goldwyn (urban scholar and program director at the Marron Institute of Urban Management) about his fascinating research detailing why American transit is so much more expensive to build and operate than other transit systems around the world. This is a derivation of the catastrophic capital insight from earlier in this column. We need to expand this research to all parts of local government and give well-intentioned reformers both a roadmap and institutional cover for making necessary reforms.

If you are an academic and are working on this or related work, we’d love to hear from you. And if you have a critique of Strong Towns or a question you want answered, we have a place for that, too.

---

---

RELATED STORIES