Why does it take so long and so much for our government to build today?

On the California High-Speed Rail, proceduralism, and vetocracy

Apparently, we’re the problem

If you told me that infrastructure projects, like bike lanes, subway lines, train halls, and homes, take decades and hundreds of millions to build, I wouldn’t bat an eye. I don’t know, maybe I’m inured, but these figures seem normal to me. I feel like my world has always operated this way.

That’s why I was shocked to learn that up until ~60 years ago we were much more efficient. The Empire State Building was built in 410 days. That’s 1 year and 1.5 months! The Alaska Highway, 1,700 miles of roadway across remote tundra, was built in 234 days. New York’s first 28 subway stations were built in ~5 years.

Such efficiency seems implausible and otherworldly now. Here are some recently completed infrastructure projects and their figures:

• “San Francisco proposed a new bus lane on Van Ness in 2001. It opened in 2022, yielding a project duration of around 7,600 days…The project cost $346 million, i.e. $110,000 per meter.”

• It took 30 years and $930 million to renovate Penn Station post-1963 destruction (side note, Penn Station pre-destruction is awe-inspiring. See pictures here). Their starting quote was $315 million.

• The MTA decided to build the Second Avenue Subway in 2000. The first phase, with 3 stations, opened in 2017. The latest upgrade is the extended Q Train line. These 1.8 miles cost $4.6 billion and ~37 years to build, from planning to completion.

• The California High Speed Rail (CAHSR) project to connect SF to LA was authorized by voters in 2008 with a target completion date of 2020 and a budget of $33 billion. Now, the target completion date has shifted to the 2030s just for Merced to Bakersfield, there’s no date for the full line from SF to LA, and projected costs have ballooned to ~$100 billion.

Can we conclude, then, that our government doesn’t care to build anymore, that they’ve gotten lax about the dollars and delays these projects incur? Do they not care to advance transportation? What about housing and clean energy, to fix our housing and climate crises? Or forget societal advancement, what about simple things like meeting deadlines and budgets?

I think we’d get a chorus of yes’ — it doesn’t feel like our government wants to build anymore. And for the longest time, I’d have agreed. But then I read this:

I’ve covered Congress for almost 20 years. The place is littered with proposals to construct universal pre-K and reimagine the health system, to decarbonize the US economy and incentivize drug development through prizes and solve the housing crisis. They just don’t pass. It’s become a running joke in Washington that every week is “infrastructure week.” But we’re not rebuilding American infrastructure.

Which gave me pause. Is it true that we want to build? Then, what’s stopping us?

This is a hot topic, as it turns out. It’s what “Abundance” by Ezra Klein and Derek Thompson, the book du jour for politics enthusiasts, is all about. The “Abundance” answer is that we (as in liberal governance) do in fact want to build, but what’s stopping us…is us.

That’s a provocative claim. So much so that it’s held my attention for the past two months as I’ve searched for all the ways we’re getting in our own way. In this essay, I’ll share what I’ve learned (so far, and subject to change).

First, a brief history lesson

There’s two ways we’re hamstringing our own progress. Before I explain, though, we need to apply the principle of Chesterton’s Fence. As in, we should understand the historical decisions that created today’s dysfunction.

Like I said above, up until the 1960/70s we were pretty speedy about building infrastructure! The reason being: government, corporations, and key decision-makers could run their projects without too much oversight. That means a single actor like Robert Moses could actually see his highway project to completion, but also bulldoze neighborhoods and raze cities, displacing at least “[a] quarter of a million New Yorkers” for his own prejudiced pleasures. Or governments caused environmental disasters like the Santa Barbara oil spills and the Cuyahoga River fire. Or Francis Fukuyama writes that “regulatory agencies founded during the New Deal…work[ed] hand-in-glove with the auto industry, pharmaceutical companies, oil and gas interests, and other sectors contributing to pollution, unsafe vehicles, consumer fraud, and other wrongs.”

We were horrified to see the power we put in our leaders’ hands exploited at a national scale. Afraid of Robert Moses’ second coming, many left-wing judges introduced guardrails to more actively curtail government overreach, like suing the agencies that green-lighted environmentally damaging company projects instituting a sue-first mentality, or enacting the National Environmental Policy Act (NEPA), effectively making new infrastructure harder to build. There’s no doubt these were well-intentioned approaches. But an unforeseen and/or overlooked byproduct of making new infrastructure harder to build then, is making new infrastructure harder to build now like the housing and clean energy we need to mitigate both crises.

Our vision was to tame a government we viewed as violent (not without reason). We accomplished this with proceduralism and vetocracy, to diffuse its power throughout its layers.

But they’re also two ways we hamstring our progress. If everyone has power, then no one has power. We’re left with a government that has the will to do many noble things, but not the way.

A fetish for procedures

Let’s study the CAHSR project. A procedure fetish, or more formally proceduralism, for environmental reviews has thrown this railway off its tracks.

Quick definition: Proceduralism is the idea that modern politics loves to layer on procedures because people believe they rein in a “constitutionally suspect” state and prevent special interest group capture. But instead of progress, procedures paralyze.

Infrastructure projects have to start by submitting an Environmental Impact Report (EIR) and/or just an Environmental Impact Statement (EIS) — essentially, long documents that disclose potential environmental risks and mitigation steps — depending on if they have to comply with both federal and state policy, or just federal. This took over 10 years to complete (which actually feels too short) because

First, the procedural burden doubled: Initially, the project was under the assumption that they only needed to comply with the federal law (NEPA), and not also the state law (the California Environmental Quality Act, or CEQA). However, in 2017 the California Supreme Court overrode that assumption, meaning they had to comply with both.

Now, they had to complete a lot of the same procedures twice. You’d imagine that the government prunes duplicate procedures in this case, but no. Developers had to submit paperwork for approval by federal authorities — like the Environmental Protection Agency, the Bureau of Land Management, the U.S. Forest Service, the Fish and Wildlife Service, the Army Corps of Engineers — and then state and local authorities, all of whom have their own processes. The same goes for funding, permitting, land acquisition, and even water.

Second, environmental reviews are unwieldy: You might think that the project had to submit just a handful of environmental reviews, but no. From 2010-2020, more than 4000 EIRs under the CEQA were filed, 3 times the number of EISs filed for the rest of the country. And these reports are now thousands of pages to ward off NEPA lawsuits, when they were only dozens in the 1970s.

Third, the scope kept changing: The project’s scope changed 1000+ times, in part due to lawsuits that forced redesigns, new regulations that undid previous environmental clearances, utility relocations, and more. Each change can trigger a whole new round of environmental reviews.

And fourth, files are reviewed sequentially, not concurrently: Francis Fukuyama writes in a broader critique of the NEPA review process that “each must be reviewed by interested agencies and the reviews are often done sequentially, adding to delays. Simply choosing the “lead agency” can consume months of debate…Failure to take account of an agency position in the initial draft sometimes leads to the document being completely re-written or blocked, or else gives rise to litigation at a later stage.”

But this isn’t news to the government — there’ve been many attempts to relax procedural burdens. For instance, in 2015, the FAST Act was signed into law which forms a council to “coordinat[e] permits across different federal agencies, thereby streamlining and shortening the overall process for some large projects.” Obama issued an Executive Order to reduce the review time for permits and so did Trump.

Have they worked? Anecdotally, it doesn’t seem like it given the state of the CAHSR. But I won’t dive into that in this essay. The topic deserves its own space to ask: Why haven’t these laws worked? Does it just seem like they didn’t work, but there are indeed success stories? More broadly, how do you make laws effective? What are our success metrics?

Reinforces a fetish for vetoes

The CAHSR has also been affected by what I think is another fetish: a veto fetish. This is what political scientist Francis Fukuyama calls a vetocracy, or a dysfunctional form of governance where stakeholders exercise their veto rights, through suing, to block projects (often not in their interests). I’d claim that this reinforces proceduralism by layering on procedures to resolve lawsuits, which can force redesigns and trigger more environmental reviews, which can trigger new vetoes…and so on this Möbius strip goes.

In the CAHSR project, the government has exploited veto structures like federalism, and the public adversarial legalism. We actually covered federalism above. It aims to divide power between federal and state/local governments, but can end up duplicating regulation.

Adversarial legalism looks like policymaking influenced by bottoms-up lawsuits and not traditional top-down bureaucracy. The NEPA and CEQA are often leveraged in lawsuits. In fact, the CEQA bakes in veto points by “giv[ing] standing to all 40 million residents of [California] to sue any project, public or private.”

(That sounds egregious, but if you’re curious there’s a rational and good-faith reason for this — there always is, I’m learning. The CEQA is an example of the private right of action (PRA), a legal tool where one citizen can sue another if the first deems the second in violation of the law. The PRA exists because of “America’s racial history. Back in the 1950s and 60s, democratically-elected legislatures could not be relied upon to repeal laws legalizing segregation. The Civil Rights movement was kicked off by the 1954 Brown v. Board of Education Supreme Court ruling under which a private nonprofit group, the NAACP, sued the Topeka, Kansas Board of Education for segregating its public schools…setting the precedent for later civil rights cases regarding interracial marriage (Loving v. Virginia), abortion (Roe v. Wade), and gay marriage (Obergefell v. Hodges).” The PRA was cornerstone in affording us the human rights we have today, and was therefore extended to protecting our environment through the CEQA.)

These groups have sued/withheld support for the CAHSR under the CEQA:

• LA county withheld support due to concerns about community disruption, historic site preservation, and environmental justice. The project changed an early route to get LA county’s backing, which added about $8 billion.

• “Farmers were afraid that the railway would reduce the acreage of productive agricultural land, separate irrigation routes, and destroy permanent crops.” So the project considered alternatives, “creating a $5 million fund to buy agricultural ‘conservation easements,’ and coughing up $1 million in legal fees.”

• Auto and airplane companies saw the railway as a threat to their business and a chance to lose profits.

I’ll be honest, I can see real concerns with environmental damage, historic site preservation, and loss of productive land (or maybe that’s good marketing and I’m easily swayed). As an individual, your pains feel more acute whereas the broader project’s pains seem abstract. In fact, their pains are real. Re: farmers, “the state is legally required to compensate farmers for financial setbacks and the use of their land, [but] some payments have been significantly delayed, sometimes by up to a year.” I realize there’s the added complication of arguing against moral veto reasons. Another essay for another time is, how do you get individual groups to look past their grievances and see the bigger picture, while accounting for their concerns?

But other groups sued for less sound reasons:

• “Wealthy cities [like]…Atherton, Menlo Park and Palo Alto…banded together to challenge the authority under CEQA. Their motivation? Concern over noise generated by the train whipping through their cities at 110 mph…the lawsuit advanced to the courts, which upheld the authority’s environmental review. The authority spent at least $1.2 million to fight the case” just to land where they started.

• “The rail authority also paid out $1 million to Corcoran, a city of 24,813 in Kings County, to ‘make up for aesthetic effects from the rail route.’”

Now this I can’t empathize with. This is a classic example of Not in My Backyard-ism, or NIMBYism. It’s this mentality that protects self-interest, for what I think inane reasons. There’s so many examples of NIMBYism not just in this project, but also in housing and green energy: “people block housing construction because they fear living next to the people that our economic system provides so little income to. They block nearby transportation and energy infrastructure because they don’t want to tumble down the economic ladder by impairing the value of their personal real estate assets.” It’s why Red states are building more green energy and housing than Blue states, even though these are leftist ideals. It’s a form of economic inegalitarianism, when it’s coming from the folks that campaign egalitarianism.

The price of progress, of a good life, is inconvenience. It’s looking past your self-interest and out towards the greater good. It’s getting out of your own way to let progress in.

(Note: These were just delays at planning/approvals phase. Downstream factors, like incentive misalignments (e.g. the federal ARRA grant forcing construction to begin even though designs were only 15% complete) and operational challenges (e.g. funding, siloed risk management, third party coordination, etc.) have also delayed the CAHSR project.)

So, what do we do about this?

If we’re talking solutions, I know this essay sounds like I’m saying, “Let’s start slashing regulations and checks and balances guys!” The evidence for is certainly compelling. And sure, we could do some cleanup. But I’m not confident that we should go on a slashing spree, for a few reasons.

As we covered, there’s a rational historical precedent for these structures (an omitted perspective from “Abundance”). And seeing DOGE cut the USAID and other authoritarian moves made by Trump, I feel like we need to hold on tight to these structures. Pluralism is an important democratic ideal, one that’s fueled much progress like convincing Governor Cuomo to finally take congestion pricing seriously or funding more school breakfast programs.

And, let’s say we do remove extraneous procedures and veto points. Will that fix the problem? Probably not. Things like processing paperwork quicker or digitizing systems will help, but it’s likely not enough. We need a whole system rewrite. For example, to build out clean energy infrastructure,

[We’ll need to] dramatically streamline solar permitting in three thousand counties and almost twenty thousand towns and cities, [and] navigate stakeholder input processes in the countless jurisdictions new power lines must travel through…

So, what do we do about this? I’m not sure. But I do know that my next step is to keep learning.

I certainly need to study the operational bottlenecks. It’d also be interesting to learn from global successes like Spain or China that built transit systems fast and cheap. How did they do it? What makes sense for us to adopt? But more than anything I want to be on the ground, like a fractional government official. I want to study infrastructure projects up close, from start to finish. The end goal is to shadow a full project, but I’ll start small by attending community input meetings or learning from local groups.

While this ending is honest, I also know it’s inconclusive. After all this research, I was hoping I’d have an opinion to share and felt a touch disappointed that I don’t. What was all this work for? But in talking to friends and reflecting on my own, I realized that this nuanced problem requires more diligence than an armchair analysis — a realization that self-started a fire to participate not from the sidelines, but at the helm of the action by seeking ways to engage in local civics. (Younger me would have never seen this plot twist coming.)

I haven’t felt this way — like a beginner in a wondrous and expansive world — in a while, and that has to mean something. I’m excited to keep learning and sharing.

---

A note on my research: I’m sure my learnings are tenuous and porous. If you notice a gap or a wrong claim, let me know!

Here are some interesting examples at the local level:

• For four years, Zach Klein has been asking the city of San Francisco to install a speed bump on the notoriously dangerous street he lives on, a block the city knows is dangerous themselves. Temporarily, he took matters into his own hands — he installed an Amazon-bought speed bump which seemed to work. He “sat there for an hour recording car after car taking notice and slowing down.” But without proper signage it could lead to accidents, so he eventually had to remove it. It’s been 3 years since that experiment, and still no speed bump. Just a canned email response to every status update request.

  The city said it would take three years to install a speed bump. So I bought my own on the Internet.

• Setting up a micro-school in your neighbor’s basement unit is a “nightmarish”, months long process encumbered by regulations. This is also true if you were “converting your garage into a tiny walk in tea shop [which] is barred by municipal regulations in nearly every major US city; even running a home daycare is restricted by zoning laws in many of them. If you want to build [an] ADU in your backyard for your aging parents, you often need to spend years working through incomprehensible and expensive processes for permission. You need to be a determined, stubborn, bureaucratically fluent person with an absurd amount of time on your hands in order to do things in your own home, in your own community, to make it richer and healthier and better.”

And header image creds here.