The $375,000 Drinking Fountains

How a system no one designed and no one controls became what's breaking America

Renderings of what the $375,000 drinking fountains will look like once they're installed in mid-2027. (Images are from slides presented by the NYC Dept. of Parks & Recreation to Manhattan Community Board 7, October 2025.)

Last year I was appointed to Manhattan Community Board 7, one of New York City’s official neighborhood boards. At a Parks & Environment Committee meeting last October, a project manager from the parks department presented 23 slides about plans to replace two drinking fountains in Riverside Park, one broken and one outdated. The presentation covered flood zone analysis, tree root inventories, backflow prevention requirements, and the construction of an entirely new water line for the fountain at 96th Street (because the existing pipe was roughly 90 years old, the city had no records of where it ran, and parks directives required replacing it). The budget for replacing the two fountains: $375,000.

Rsd Park Drinking Fountains Reconstruction 10

5.86MB ∙ PDF file

Download

Download

I sat there wondering how replacing two drinking fountains could cost what a four-bedroom house costs in suburban Kansas City. But I’ve spent enough time in and around the public sector to understand how we got there. Before moving to New York, I was a lawyer for the City of Seattle for more than a decade. I advised mayors and councilmembers, worked on the city’s $15 minimum wage law, and served as in-house counsel for agencies handling everything from animal control to fair housing. Years of watching government work from the inside taught me that the story I’m about to tell is bigger than one big-city parks department or even New York’s many particular pathologies.

Each rule governing the Riverside Park drinking fountain project had its own logic. The backflow prevention systems protect the public water supply and are mandated by the Department of Environmental Protection. The dual-bowl fountains satisfy ADA requirements. Under current parks directives, water lines as old as the existing 90-year-old one have to be replaced when any new work is done. Because parks chose to rebuild the fountain at its existing location and there are no records showing where the old water line ran, a new trench has to be cut. A tree root survey is required because digging that trench risks damaging root systems that took decades to grow. The competitive bidding and procurement procedures were put in place to prevent corruption. The parks project manager presenting the slides was professional and thorough. Every box was checked. And the outcome was a third of a million dollars for two drinking fountains.

It’s actually worse than that. The community, through the city’s participatory budgeting process, had allocated the $375,000 for five modern hydration stations with bottle-filling capability spread throughout the park. The plan that emerged from the parks department had been whittled down to replacing two existing fountains with code-compliant versions of what had been there for decades. No new locations, no bottle fillers. Five hydration stations became two ordinary fountains.¹ And the replacement fountains won’t even be installed until 2027.²

Drinking fountains are not, on their own, a national crisis. But what happened in Riverside Park illustrates what’s wrong with almost everything else.

A four bedroom, two bathroom house in the Kansas City suburbs you can buy for less than it costs to replace two drinking fountains in Riverside Park. (Screenshot from Zillow listing, taken on April 30, 2026.)

A Problem Bigger Than Two Expensive Drinking Fountains

The dynamic that produces a $375,000 price tag for two drinking fountains scales. The same combination of individually reasonable requirements adding up to absurd outcomes is why it costs $2.5 billion per mile to build a subway in New York City, roughly 8 to 12 times what comparable projects cost in Italy, Sweden, Spain, France, Turkey, or Germany.³ In San Francisco, more than a decade after a developer first proposed replacing a laundromat with apartments, the lot sits empty after a fight over whether the laundromat was “historic” and three studies of the shadows the new building would cast.⁴ President Biden’s $7.5 billion federal program to build EV chargers, passed in November 2021, had produced only 57 operational stations across 15 states by the time he left office in January 2025.⁵ The TransWest Express transmission line spent 16 years getting permitted despite being put on the Obama administration’s “fast track” in 2011.

The system that makes everything cost too much and take too long doesn’t distinguish between the public and private sectors. It kneecaps both.

American life is becoming increasingly unaffordable. This is driving growing cynicism about our institutions. When a median-income household in 1990 could expect to afford the median-priced home in 75 of the 100 largest metro areas but today can do that in only three, people notice.⁶ When working-class Americans without college degrees see their life expectancy falling, people notice. When trust in institutions like government and media has been in structural decline for decades, it isn’t because people are irrational. It’s because these institutions haven’t been earning our trust.

Populism Is a Mirage

Our system feels broken because in many ways it is broken, and the establishment hasn’t come up with good answers for fixing it. Populists on the right and on the left have tapped into public frustration using narratives that blame some combination of coastal elites, billionaires, big businesses, or immigrants. The populist narratives aren’t identical; the right and the left disagree about who the villains are. But they share an approach: name villains, promise consequences for those villains, and insist that meting out those consequences will fix things. These narratives resonate because people’s grievances are real; our institutions have failed to deliver what they promised. But populism mistakes symptoms for causes.⁷ No combination of tariffs, wealth taxes, mass deportation, or antitrust enforcement will address what’s actually producing the symptoms. The populists’ villains may get their comeuppance, but housing remains unaffordable.

The villain — the thing breaking America — isn’t a person, a party, or a conspiracy. It’s a system, a system built over decades by people making a series of individually rational decisions that added up to something no one designed and no one would build on purpose.

I’m going to call that system the Machine. This name might bring to mind political machines like Tammany Hall or Mario Savio‘s 1964 “bodies upon the gears” speech at U.C. Berkeley. I’m describing something different. Tammany was run by identifiable bosses with specific interests and objectives. Savio envisioned the university administration as a “machine” run by people he could, in principle, confront. The Machine I’m describing has no one at the controls. People may run its individual components, but no one operates or even fully understands its scope. This Machine is a complex structure of misaligned incentives that perpetuates itself. It does so without malice and, for the most part, without awareness. And this Machine routinely operates against the interests of the very people who keep it running — which is why the villain I’m naming is a way of operating, not a category of people. The fix isn’t punishing a class; it’s changing incentives.⁸

The Missing Story, Popularism, and Abundance

In a February 2026 essay for the States Forum, Marshall Kosloff of the Niskanen Center describes a problem he calls “the missing liberal story.” Kosloff argues that the center-left⁹ intellectual community has, in the wake of the 2024 election, done extensive tactical and analytical work — polling, autopsies, swing-district analysis, moderation strategy, policy detail — but hasn’t done the storytelling work that would make those answers resonate the way populism resonates. Drawing on work by Danielle Lee Tomson, a scholar of political media and narrative¹⁰ whose ideas Kosloff has built on in his podcast The Realignment, he describes an “authenticity gap” between what Americans have been told to expect from their institutions and what they actually experience. The populist right and the populist left have spoken to this gap more effectively than centrists or liberals have. The center-left has plenty of policy and analysis; what it lacks is a way of explaining what went wrong and how to fix it that lands with people who have every reason to be skeptical of the establishment’s promises. Kosloff has continued to develop this framing on his podcast, including in an April 2026 season finale with Tomson, but the story itself remains unwritten. In his essay, Kosloff issues a challenge to come up with a story. This essay is my attempt at an answer.

The answer I’m proposing builds on two projects the center-left is largely getting right: Popularism and Abundance.

Popularism tells the center-left how to win. The case Popularism makes is that Democrats have been losing ground with voters they used to rely on, especially working-class voters without college degrees, by taking unpopular positions on immigration, crime, and cultural issues while underweighting the cost-of-living and broader economic concerns voters actually prioritize. Popularism’s prescription is grounded in taking public opinion seriously: run on what people actually want, moderate on issues where the party is out of step, and stop losing elections by foregrounding things voters oppose or don’t see as priorities. The framework was built on David Shor’s polling work, further developed by Welcome PAC and its Deciding to Win report,¹¹ and sharpened through Matthew Yglesias’s writing and Third Way’s broader effort. (Popularism, despite its similar name, is not populism; it is the discipline of choosing and emphasizing popular positions on high-salience issues while not taking unpopular positions or focusing on low-salience issues.)

Abundance tells the center-left how to govern. Its diagnosis is that post-1970s liberalism stopped building, started focusing instead on managing growth, and set up a thicket of procedures and rules that made building or doing anything a bureaucratic ordeal. The Abundance prescription is to stop fighting over how to divide a shrinking pie and instead tackle scarcity directly. This is the framework Derek Thompson named in a 2022 Atlantic essay and developed with Ezra Klein in their 2025 book of the same name. Steve Teles, a political scientist who has done more than anyone to map Abundance’s internal range, describes its core as “reducing zero-sum conflicts by creating more — more energy, more housing, more high-quality schools, more scientific discovery, more world-leading firms, and more, cheaper healthcare.” Klein articulated his version of the animating question in a June 2025 New York Times piece: “What do we need more of, and what is stopping us from getting it?”

Neither Popularism nor Abundance serves as the story that answers Kosloff’s question, but they are both key plotlines. Popularism’s electoral utility works more as the foundation for a campaign than as the campaign itself. While candidates need to run on something beyond taking popular positions, avoiding unpopular positions, and prioritizing what voters prioritize, building a campaign on a Popularist foundation puts a candidate in the best position to do everything that follows.¹² Popularism isn’t merely poll-testing or triangulation. It is as much about priorities as positions: lead with what voters care most about, and, on lower-salience issues, respect that what voters want is often closer to “I don’t want to spend time thinking about this” than to any specific substantive answer. Doing these things isn’t a complete campaign, but it’s the strongest place to start.

Abundance, as an approach to policymaking and governing, works better as a reelection strategy than an election strategy. Voters won’t feel Abundance until it’s been delivered. Candidates have to win their first elections by saying what they want to do, and then govern in a way that gives voters a reason to keep them in office. Governing well, in the Abundance frame, means delivering things voters can see and use, making life tangibly more affordable, and accomplishing this in a reasonable amount of time and for a reasonable amount of money.¹³

What’s missing, even with Popularism and Abundance deployed together, is a larger context that connects them — a story that explains what went wrong, what we’re fighting against, what we’re fighting for, why we’ve been losing, and why winning elections and governing well are part of the same project.

The Machine

My story pitch is the Machine. Its framework draws heavily on a body of work from what I’ll call the Abundance Expanded Universe: Klein and Thompson’s book, Marc Dunkelman’s Why Nothing Works, Jennifer Pahlka’s Recoding America, Francis Fukuyama on vetocracy, Nicholas Bagley on the procedure fetish.¹⁴ That literature has done much of the hard analytical labor documenting how we got here. Teles in particular has done critical work, notably his 2013 essay on kludgeocracy, 2017 book The Captured Economy (with Brink Lindsey), and his ongoing project of mapping different types of Abundance into a coherent intellectual framework. In his August 2025 “Varieties of Abundance” essay, Teles identifies two obstacles all versions of Abundance have to overcome: the asymmetric power of concentrated incumbents over diffuse challengers (yesterday’s winners maintaining a system that blocks new entrants), and government’s diminished capacity to manage complex tasks competently. What this literature hasn’t yet produced is a story that a voter who doesn’t read Niskanen Center white papers can absorb, recognize in their own life, and use to make sense of why replacing two drinking fountains costs $375,000.

Are Abundance and the Machine framework different names for the same thing? Yes and no. Abundance and the Machine framework approach the same problem from different angles. Abundance names the goal: more housing, more energy, more state capacity, more of the things that make life affordable and government functional. The Machine framework names the adversary: the self-perpetuating structure of misaligned incentives standing between us and the goal. A political story needs both — Abundance shows us where we want to go, and the Machine framework tells us what we’re fighting to get there.

America used to be able to do extraordinary things. We built infrastructure and housing, fostered a middle class that became the envy of the world, and put a man on the moon. Postwar America was far from perfect: many states still legally mandated segregation, society was more conformist and less tolerant, and people lived under the constant fear of nuclear war. But we had a sense of optimism, a feeling of forward motion, and a belief that the next generation could expect to do better than their parents did.

At some point, things began to change. Midcentury America leaned heavily on what Dunkelman calls a “Hamiltonian” ethic: confident, centralized, and capable of building things at scale. When the real abuses of that approach became clear in the 1960s and ‘70s — urban renewal bulldozing neighborhoods, Vietnam, Watergate, top-down planning with little regard for human cost, rivers catching fire and skies turning brown — progressives took what Dunkelman calls a “Jeffersonian” turn toward dispersing power, proceduralizing decisions, creating veto points, and giving a broad group of actors new ways to tie up projects in years of litigation. Jane Jacobs’s critique of top-down planning was part of what the country needed to hear — the Jeffersonian correction was necessary, but it overshot. The Jeffersonian mode became the default, the Hamiltonian capacity atrophied, and we stopped building.¹⁵

Homeowners understandably didn’t want their neighborhoods disrupted by new construction. Corporations discovered that lobbying for protection was cheaper and easier than continually improving their products and competing for market share. Environmental regulations initially put in place to solve real problems kept expanding in size and scope beyond their original purpose. Procurement rules designed to prevent corruption became barriers to entry only large, politically connected incumbents could navigate. Professional licensing regimes created to protect the public evolved into protective guilds. Special interest groups first set up to address specific injustices became victims of their own success, evolving from problem-solvers with defined goals into self-perpetuating ideological actors that, as part of “the Groups“ ecosystem, now wield more influence over policy and political discourse than the voters they nominally represent.

Many of the people who built the Machine weren’t necessarily seeking material advantage; they saw themselves as protecting the environment, preserving neighborhoods, and fighting corruption. No one woke up one morning and said: let’s make housing unaffordable, healthcare ruinously expensive, government contracts the exclusive province of insiders, and the regulatory environment so complex that new entrants can’t compete. That system is the aggregate output of thousands of decisions by people who mostly meant well.

The Machine framework addresses one of the center-left’s core problems by creating a path for us to be constructively angry about the status quo. We should, in fact, be fed up with how things work. Populists and extremists on both the left and the right are much better than center-left wonks at getting mad. But their solutions involve blaming and attacking only their chosen enemies. Our challenge is figuring out how to show authentic anger about the parts of the system that don’t work without forgetting that many things do work. We need to get mad and channel that anger into fixing things, not scapegoating or setting things on fire. Naming the Machine as our adversary lets us be anti-establishment in a way that is productive rather than destructive. Populists promise that defeating their named enemies will fix things, and extremists say burn it all down. The Machine critique names a system, explains what it’s doing, and demands structural change that can actually deliver results.

Kosloff argues that the center-left’s ideas are too measured and academic to compete. The analytical voice alone won’t do. Here, in a different register, is how we might articulate what it feels like to live under the Machine:

You can feel it even if you can’t name it. The sense that the game is rigged before you start, that working hard and playing by the rules used to mean something but no longer does. That the next generation should no longer expect to be better off than their parents. Nobody in charge seems particularly bothered by this or capable of doing much of anything about it. The institutions you were told to trust have stopped earning that trust in ways that are hard to explain but impossible to miss the minute you stop and look.

None of this is any one person’s fault; if anything, it’s everyone’s fault. That’s what’s so frustrating. There’s no one to be mad at, no villain to defeat, no quick fix. Just lots of people — ourselves included — living our lives, doing what seemed to make sense, and somehow we have a country where we don’t build anything and everything’s too expensive. We sleepwalked our way into this.

It costs us a third of a million dollars to replace two drinking fountains in a city park. It takes years to build housing and most of a decade to put in a semiconductor factory our leaders say we urgently need. And those are the things the system still produces. Most of the things people actually want — affordable housing, manageable healthcare costs, the ability to start a business without a lawyer and a compliance consultant on retainer — the system is quietly refusing to deliver at all.

What the Machine has taken from us, more than any specific outcome, is the expectation that the system will work. Our system doesn’t need to be perfect, seamless, or overly generous, but it needs to function at a basic level. The Machine has made that expectation feel naive. We want it to feel achievable again.

The broken 96th Street fountain in April 2026, looking uphill towards the parking lot.

Revisiting the Fountains

Let’s return to the Riverside Park drinking fountains. Their $375,000 price tag wasn’t corrupt or illegal. It didn’t violate any city policy. Indeed, the whole thing followed policy to a T. The flood zone analysis, the backflow prevention, the tree root survey, and the new pipe were all required by the rules or directives. The new pipe was required because parks chose to rebuild the fountain exactly where the old one had been. The parks department checked all the boxes, and the system, functioning as designed, produced a plan to spend $375,000 replacing two water fountains. What I saw was a process working exactly as it was designed to work and producing an outcome no reasonable person should defend.

The Machine makes it nearly impossible to say any of this out loud when you’re inside the system. Most of the fountain project’s individual requirements are reasonable enough. The solution isn’t casting aside any individual requirement; it’s asking the more basic structural questions the system currently avoids. The community voted to spend $375,000 for five new hydration stations throughout Riverside Park, and the Machine delivered a plan for two replacement drinking fountains at existing locations without any bottle fillers.¹⁶ Each requirement may have made sense on its own, but they added up to a result that delivers too little public benefit for too much taxpayer money.

Spend enough time inside the Machine and it will numb almost anyone into accepting the results it produces. I’ve experienced this myself. Anyone at that parks committee meeting could have asked the larger question: whether this project, at these two sites, and at this cost, was the right use of the $375,000 originally designated for something bigger. But the parks department wasn’t there to have that conversation. They were there to present what had already been decided, and they weren’t asking the community board for input, support, or approval.¹⁷ The Machine teaches complacency; this is a big part of what makes it so difficult to fight.

Diagrams of the 96th Street drinking fountain site showing existing conditions and proposed work. (Image from slides presented by the NYC Dept. of Parks & Recreation to Manhattan Community Board 7, October 2025.)

The Machine Is a National Problem

The Machine operates wherever the regulatory and institutional apparatus that enables it is available. The Biden-era CHIPS Act funding for semiconductor fabs — funding for a critical industry with bipartisan support and a strong presence in business-friendly states — came burdened with federal conditions stacked on top of the core mandate to actually build the factories. The Machine has a federal layer that operates regardless of who runs the statehouse.¹⁸

Texas issues housing permits at rates blue states like California can’t approach, but Texas has not escaped the Machine. Houston, famously without formal zoning, uses lot size rules, parking requirements, and deed restrictions to do much of what zoning does elsewhere. Texas’s occupational licensing regime is more burdensome than its small-government reputation might suggest, and Florida’s is among the most burdensome in the country.¹⁹ Red-state governments also pick culture-war fights that crowd out serious governance, taking positions on immigration, abortion, and LGBTQ rights that work against their own economic interests and ability to attract talent.²⁰ The Machine operates in red states, plaguing them with different pathologies than you see in blue states.

Still, the part of the Machine I know best is the part that operates in blue states, and the center-left has an obligation to own the dysfunction that blue state and blue city governance has produced. Blue jurisdictions are typically governed by coalitions built around institutions that receive public money: public sector unions, government contractors, university systems, hospital networks. More spending always seems to be the easiest answer because it is usually the path of least resistance for coalition management. One visible consequence of this approach is a communications habit that treats dollars appropriated as the achievement.

Consider New York City’s last two mayors: Eric Adams spearheaded “City of Yes,” the city’s most substantial pro-housing zoning reform in decades. Zohran Mamdani, only months into his administration, has expedited land-use review and is working to shorten pre-development timelines for a small set of affordable housing projects. This qualifies as pro-housing by NYC standards. But neither mayor could stop touting how much money they’re spending. Adams cited a “historic $26 billion commitment” to affordable housing, a “$5 billion commitment” to infrastructure and housing, and a $24.7 billion capital plan. Mamdani campaigned on spending $100 billion over ten years to build 200,000 “publicly subsidized, affordable, union-built, rent-stabilized homes” — about $500,000, or nearly three drinking fountains, per home.²¹

Blue state and especially blue city political cultures teach elected officials and their staffs to spend more so they can trumpet bigger numbers — exactly the wrong approach when what they actually need is to get the most out of every taxpayer dollar. Asking whether the spending works is threatening, because the answer often points toward restructuring the institutions that support the governing coalition. Rational people respond to the incentive structure they inhabit. The blue state part of the Machine doesn’t need bad people to perpetuate itself; all it takes is good people acclimating to their political climate.

What Fighting the Machine Can Accomplish

The Machine was not created anew during the Jeffersonian overcorrection of the 1970s. It has always existed in some form, becoming meaningfully more powerful and pervasive over the last half century. We should not expect some final battle where we defeat the Machine. Rather, our achievable goal is to temper the Machine by shrinking its footprint.

What would a country with the Machine tempered look like? There would still be politics, institutional friction, and legitimate disagreement, some of which are inherent elements of any free society. But we could have a more functional, more effective government that usually accomplishes what it sets out to do, in reasonable time, and at reasonable cost. That government would replace drinking fountains for what it should cost to replace drinking fountains. Transit agencies would build subways using something closer to European costs and European timetables. Housing would be built fast enough so a teacher could afford to live where she works. We would streamline regulations so an immigrant trying to start a restaurant wouldn’t need a lawyer or a consultant to obtain permits. Markets would be fairer and more open, allowing new players to compete by offering something better instead of protected incumbents continually repackaging the same products.

People could have more agency; they could choose their own lives and pursue their own meaning. Government cannot solve all of our problems or manufacture meaning. But it can help people obtain the tools they need to flourish. We should have a representative democracy that serves the people it’s supposed to serve instead of catering to the organized incumbents and interest groups who show up at every meeting and specialize in navigating the system.

The Machine stands in the way of all this. We know it’s possible because we’ve seen it happen. Pennsylvania Governor Josh Shapiro led his state’s effort to reopen a collapsed section of I-95 in 12 days. He used emergency powers not normally available but warranted in this case by the circumstances of the overpass’s collapse and I-95’s economic importance. The project’s speed demonstrated what our system can accomplish when it decides something is urgent.

The Transit Costs Project at NYU’s Marron Institute has documented what subway construction could look like: Milan’s M4 line, a fully underground 9.4-mile driverless metro through the city’s dense historic center, opened between 2022 and 2024 and cost roughly $200 to $235 million per mile. Phase 1 of New York City’s Second Avenue Subway cost $2.5 billion per mile. Italy isn’t a low-wage country; the difference is procurement, project management, and what contractors call “the agency factor“ — the premium they add to their bids just to cover the cost of dealing with a transit agency’s red tape. None of Milan’s achievements required suspending democracy or destroying the environment; they required institutions that knew what their job was and that were allowed to do it.²²

The Machine is powerful and entrenched, and most of us have played our own roles in keeping it running. Confronting this will sometimes be uncomfortable. But it’s a fight worth having, because we really can do better.

The Anti-Machine Coalition

The “People” in People vs. the Machine don’t yet exist as an organized coalition. But they could.

Part of why the Machine remains so entrenched is that people who should be fighting it are distracted, squabbling among themselves over things that don’t threaten the Machine. When narrow ideologies on the left and the right dominate the political discourse, people who are angry about the big things — housing, healthcare costs, the sense that nothing works anymore — get sidetracked by cable news spats, online pile-ons, and fights over the correct name for a healthcare plan. The real antagonist stays comfortably out of view.

A coalition organized around fighting the Machine would cut across our usual political lines. Affordable housing activists and free-market YIMBYs disagree about the role of public housing, but they share an antagonist in the zoning and permitting regime that prevents anything from getting built and bakes huge costs into what does get built. Climate activists and energy-abundance advocates disagree about the pace and process of our transition to clean, renewable energy, but they share a common enemy in NEPA delays that can stall transmission lines, solar farms, and nuclear plants alike and keep electricity dirtier and more expensive than it needs to be.

The axis becomes functional vs. dysfunctional, doer vs. blocker — not left vs. right, and not one class of people pitted against another. Populism organizes against a category of people. Machine politics organizes against a way of operating, which leaves room for a broad ideological coalition. And recent polling in swing congressional districts suggests anti-Machine framing can outperform populist framing. When Blue Rose Research asked swing-district voters which of several options was most responsible for rising costs, “politicians who’d rather fight than fix problems” came in first, beating “wealthy elites and billionaires” and “big corporations” by ten points.²³

The Machine has made our public sector too incapable and expensive, and our markets too sclerotic and captured. Reasonable people disagree about where the line between government and private functions should fall. The price of entry into the anti-Machine coalition isn’t a particular view on how to balance state capacity and markets; it’s an understanding that both have important roles, they should both work as well as possible, and the Machine is the central antagonist preventing this.²⁴

A country with strong state capacity and well-functioning, competitive markets could debate how best to balance their roles. My own view is that the best answers come from considering individual questions on their own merits without any particular preference for public or private sector solutions; some problems are better candidates for government action (e.g., building transit), some for markets (e.g., building housing), and the honest work is in figuring out which is which. Others in this coalition will start with stronger priors in one direction or the other. That’s fine. This is the argument the anti-Machine coalition ought to have, an argument our current politics has foreclosed.²⁵

Winning the next election is a first step, not the end goal. Popularism is the most direct path into office, but a coalition swept in on Popularism’s foundation will eventually face the same structural obstacles that have frustrated everyone else. The Machine doesn’t block winning so much as it prevents winning coalitions from accomplishing anything once they’re in power. The path to a durable majority runs through actually delivering — building things, reducing tangible costs voters can feel, and measuring success in concrete results rather than dollars spent. That’s the only version of winning that compounds.

Back to the beginning: Replacing two drinking fountains in Riverside Park costs $375,000 and takes three years from funding to installation. With the system working as designed. We can resign ourselves to this being how the world works. That’s what will happen if we keep oscillating between the siren song of populist promises on one side and the hollow comfort of status quo restoration on the other, each failing in turn. I think we can do better. We should give it a try.

“The $375,000 Drinking Fountains” is the first piece in an ongoing project. People vs. the Machine’s About page describes what comes next.

1. Parks retitled the project from “Hydration Stations in Riverside Park“ to “Drinking Fountains Reconstruction,” citing reliability concerns with bottle-filler mechanisms.

2. At the October 2025 community board meeting, the parks department presenters said design was scheduled to finish in spring 2026, followed by roughly a year of procurement, with construction beginning in spring or summer 2027.

3. The NYU Marron Institute’s Transit Costs Project maintains a database of 755 urban rail lines across 60 countries. The weighted global average cost is approximately $406 million per mile (PPP and inflation adjusted for 2025). Phase 1 of the Second Avenue Subway cost approximately $2.5 billion per mile, roughly 8 to 12 times more than comparable projects in Italy, Sweden, Spain, Turkey, France, and Germany; Phase 2, now underway, is projected to cost around $4 billion per mile. The TCP Final Report concludes that the cost differential is not explained by geology, density, or wage differentials.

4. Project opponents first argued that the laundromat was a “historic resource” and then that the shadows the new building might cast on an adjacent preschool’s playground (during hours when the playground wasn’t being used) required further study. After the Board of Supervisors voted in 2018 to require a third shadow study, the developer sued, quietly won approval at the Planning Commission that October, and sold the entitled site for $13.5 million in 2019 with approval to build 75 units. The buyer demolished the laundromat in 2022. The lot remains vacant; years of delay likely pushed construction and financing costs past the point where the project still penciled out.

5. The $7.5 billion was appropriated by the Infrastructure Investment and Jobs Act, signed November 15, 2021, and split between two programs: $5 billion for the National Electric Vehicle Infrastructure (NEVI) Formula Program, allocated to states by formula to build charging stations along the federal highway system, and $2.5 billion for the Charging and Fueling Infrastructure Discretionary Grant Program, awarded competitively for community charging. All 50 state NEVI deployment plans were approved by September 2022, but the first NEVI-funded station did not open until December 2023. By February 2025, only $527 million of NEVI’s $5 billion share had been obligated, with more than half of the 57 completed stations located in Ohio and New York.

6. Based on metro areas with price-to-income ratios under 3; data from The State of the Nation’s Housing 2025.

7. By “populist” I mean the political mode that frames politics as a battle between a virtuous “people” and a corrupt “elite,” names villains, and promises that defeating those villains will fix things. I’m not using “populist” to refer to politics that attends to ordinary people’s preferences, which is just democratic politics working as it should.

8. Whether “the Machine” is the right name is less important than whether the diagnosis is correct. Call it the system, the blob, or anything else that paints the most effective mental picture of the concept. The point is that the Machine is not a person or a specific group of people; it’s a self-perpetuating structure of bad incentives, built over decades, that is now the primary obstacle to getting things built and getting government to work.

9. In “The Missing Liberal Story” and throughout the most recent season of The Realignment, Kosloff uses the terms “center-left” and “liberal” essentially interchangeably without precisely defining either. I use “center-left” in this essay. My coalitional idea of “center-left” is broad, encompassing both the center and some parts of the left. It can include anyone who isn’t a committed populist insistent upon always having an “elite” enemy, isn’t an ideologically driven member of the hard right or the hard left, and isn’t an orthodox libertarian opposed to any serious role for government. Mainstream liberals, centrists, heterodox independents, the self-described “politically homeless,” “liberaltarians,” and even parts of the center-right (such as “state capacity libertarians“) can be part of this coalition. The line I’m drawing is around people who believe in democratic institutions and think government should be capable, markets should be open to fair competition, and both have important roles to play.

10. Tomson writes the Failure to Communicate Substack and is the author of the forthcoming book Under the Influence: What’s Real When America Feels Fake. Her “authenticity gap” concept is the diagnostic tool Kosloff uses to make his “missing liberal story” argument. Tomson has also articulated her more colorfully titled “Does It F*ck” Test for assessing whether a political idea has genuine vitality. She has appeared on The Realignment three times in the past eight months to develop these ideas with Kosloff: an initial September 2025 conversation on story and authenticity, a January 2026 follow-up, and the April 2026 season finale.

11. The most prominent left-populist argument against Popularism and Abundance (often conflated by the left, which views them both as enemies) was a May 2025 Demand Progress/YouGov poll showing populist messaging outperforming Abundance messaging by about 12 points. The poll’s question design contrasted a well-crafted populist message against a deliberately anemic articulation of Abundance. At WelcomeFest in June 2025 (a Popularist conference that included significant discussion of Abundance), Kosloff asked Derek Thompson and Rep. Jake Auchincloss about the poll during a panel discussion that became a Realignment episode. Auchincloss’s response: “It’s what happens when you test an economics textbook against a romance novel and tell people, what do you like to read better?” Thompson added that the Demand Progress poll had tried to “test almost literally a sentence from page 167” of his Abundance book against “something that sings beautifully in the poetry of campaign language.” The broader empirical case for Popularism runs through Welcome PAC’s Deciding to Win report, based on Blue Rose Research polling of hundreds of thousands of voters over six months.

12. Popularism’s most common framing focuses on winning national races and swing or purple areas, but its methods are useful anywhere. Every election has a median voter, and no campaign builds a durable coalition without a serious effort to reach that median voter with something they find appealing. The specifics vary — the median voter in a safely blue city is different from the median voter in a swing district, and both differ from the median voter in a deep-red area — but the underlying approach is the same: take those voters seriously as they are, not as someone might wish they were.

13. Teles originally drew this distinction in a June 2025 Realignment episode: "Abundance is fundamentally a theory of governance. It's not primarily a theory about what voters do or don't want or what they will or will not respond to in an election… Abundance is more of a re-election strategy than an election strategy. It’s more of a theory about if you actually do win, how do you have a governance agenda that people will actually like and that you won’t get elected very barely by one and a half points … and then end up doing a bunch of stuff that alienates people and they don’t like. Clinton did it and then got rejected in the midterms. Obama did it, got rejected in the midterms. Biden did it, got rejected more in the general election. And so part of Abundance is a theory about how can you make sure that what we’re actually doing in governance leads voters to say, ‘okay, that sounds like you’re doing the kind of thing that I care about and you’re actually delivering and you’re trying to get reelected on something that makes sense.’”

14. Several of these works predate the Abundance label and their authors wouldn’t necessarily describe themselves as part of Abundance.

15. Dunkelman’s Why Nothing Works is the most complete recent account of this overcorrection and, more broadly, what he describes as the oscillating Hamiltonian and Jeffersonian impulses within progressivism.

16. The 76th Street fountain is being installed next to the existing parks department building with a bathroom that already has water service. This minimizes the need for new pipe and allows the required reduced-pressure-zone (RPZ) backflow prevention device to be housed inside the building. The 96th Street fountain, which the parks department believes was damaged by a vehicle, has no nearby building or water supply other than the previous fountain’s old pipe. So its new water line has to run across a parking lot (requiring tearing up part of the lot) and through trees (requiring hand digging to safeguard the roots), and its RPZ device needs its own new outdoor enclosure with plant screening. These elements are individually mandated because the parks department decided to install the replacement fountain exactly where the old fountain had been right along the Hudson River. But that location was a choice, not a requirement — it may have been possible to put the new fountain closer to the existing water main. At the meeting, parks didn’t explain why replacing the damaged fountain at its existing riverfront site was the right use for the bulk of the $375,000 allocated for five hydration stations. Parks may have evaluated alternatives internally, but the presentation gave no indication that had happened.

17. At the meeting, I asked parks why they planned to dig a new trench across a parking lot and through tree root zones to deliver water to a site that already had water service. The project manager gave an honest answer to the question I asked: parks directives require replacing older water lines, and, with no records of where the 90-year-old pipe runs, in-place replacement wouldn’t be possible. That answer wasn’t wrong, but with the benefit of 20/20 hindsight I should have asked the more foundational question about why parks couldn’t have installed the replacement fountain closer to the water main at a lower cost.

18. In “The Missing Liberal Story,” Kosloff describes Abundance as primarily a blue state self-critique. That framing sells Abundance short. Abundance’s diagnosis applies more broadly: the Machine operates in red states, in purple states, and at the federal level, and Abundance’s remedies apply wherever scarcity is a problem.

19. Florida ranks fifth most burdensome in the Institute for Justice’s License to Work report.

20. A few recent examples: Idaho made it a crime for transgender people to use gender-aligned bathrooms in private businesses, a law Idaho’s own sheriffs and police opposed as unenforceable. Texas, already plagued with maternity care deserts in nearly half its counties, has seen its post-Dobbs abortion ban drive an OB-GYN exodus that threatens to exacerbate the problem. After Florida’s 2023 immigration law took effect, construction sites in south Florida reported losing a quarter to half of their workers. Keeping politicians and the public alike distracted by these types of unnecessarily divisive culture-war battles is a key Machine self-preservation tactic. In addition to tarnishing the “open for business” image red states like to project, these fights often play out as choices between two ideological poles, neither of which delivers the types of socially moderate outcomes the median voter actually wants.

21. Dividing the Mamdani campaign’s stated $100 billion commitment by its goal of 200,000 new units yields approximately $500,000 per unit. That figure is likely conservative, since the $100 billion is supposed to cover doubling NYCHA preservation spending in addition to funding new construction. Actual delivered cost for union-built affordable housing in New York is almost certainly higher than $500,000 per unit. But even at that figure, New York City cannot subsidize its way to affordability. Public subsidy alone will never produce the volume of housing the city needs. When the regulatory environment allows enough housing to be built to meet demand, prices moderate on their own. The political energy spent fighting over subsidy allocation distracts from the more fundamental question of why private construction is so constrained and so expensive in the first place — a discretionary approval system that only politically connected developers can navigate drives out competition and keeps prices high.

22. Marron’s Transit Costs Project Final Report identifies a key part of the problem: American transit agencies have outsourced their internal engineering and project management expertise to private consultants with little incentive to control costs or timelines. Design and project management costs ran to 21 percent of construction costs on Phase 1 of the Second Avenue Subway ($656 million in consultant spending against $3.16 billion in construction) compared with 5 to 10 percent in peer cities, most commonly 7 to 8 percent. Meanwhile, American agencies lack the in-house staff to manage consultants effectively, creating a cycle where less internal capacity leads to more consultant dependence leads to less internal capacity. Milan’s publicly owned engineering firm, Metropolitana Milanese, retains expertise across projects, manages contractors directly, and has been consulted by cities across Europe and the Middle East. Notably, Metropolitana Milanese builds both transit and other types of municipal infrastructure but does not operate the transit system, unlike American agencies that generally build and operate transit.

23. March 2026 polling of voters in the fifty most competitive congressional districts, conducted by Blue Rose Research for Inclusive Abundance, found that 7 of the 10 Abundance-frame messages tested scored in the 90th percentile of everything Blue Rose has ever tested, and that Abundance frames tested best when paired with populist “villain” framing. Inclusive Abundance reads this as evidence for synthesizing Abundance with populism. I read it as supporting the Machine critique as an alternative to populism. The top-tested villain isn’t a category of people but rather a description of how those people act — “politicians who’d rather fight than fix problems” identifies anyone who behaves that way, regardless of party or background.

24. Marshall Kosloff has been developing a center-left version of this argument under the label “fusionism,” drawing on Frank Meyer‘s 1950s-1960s National Review framework; in a Realignment appearance, Danielle Lee Tomson sharpened it as “ideological factions that are clear about what they want to see in the world, with enough overlap and a common enemy.” Steve Teles and Robert Saldin’s 2020 National Affairs essay “The Future Is Faction” makes the related case that moderates specifically should organize as durable intra-party factions rather than chase third-party fantasies.

25. For the center-left, this means treating Popularism and Abundance not as tactical concessions to be bargained away once in office but as the idea an honest diagnosis actually points to. Popularism says stop doubling down on things voters don’t believe. The Machine argument says stop governing in ways that confirm what voters already suspect about how Democrats govern. These are both good ideas on their merits, not just electoral strategies. Within the Democratic coalition, the populists on the activist left have the wrong answers to what ails the country, and the center-left should say so clearly.