Why the US Census is on the brink

I would like to tell you I am not the kind of person to talk about the census at a party, but if you’ve been reading Porch Party, you probably already know that’s not true. One of the conversations that made me launch this project happened at a large outdoor get-together last May, where I ran into my old college friend danah boyd. danah, who spells her name all lower-case, is an ethnographer who has turned her expertise on data, privacy, online socialization, and related topics into a career spanning academia, industry, the nonprofit world, and government. She was about to take a job as a professor at Cornell, she told me, but what we mainly ended up talking about was the Census Bureau, where she had spent years embedded as an ethnographer and which she was now writing about, as it suffered new blows in Trump’s second term.

“Data Are Made, Not Found: A Story of Politics, Power, and the Civil Servants Who Saved the US Census,” the book that resulted, was published last week. It is, in short, an emotional story about data and the people who quietly but sometimes heroically assemble the critical statistics on who and where Americans are, information that powers Congressional apportionment and a whole lot more. Right now, as this book makes clear, that effort is in serious peril.

danah largely sticks to writing about the 2020 decennial census rather than wade into the highly fraught, post-DOGE conditions of the bureau today.  But even before the current troubles, the 2020 census took place amidst a brand-new pandemic that scattered both the staff of this bureau within the Department of Commerce and the Americans they were trying to count. The Covid epidemic exacerbated the already huge challenges of hiring the country’s largest temporary work force (peaking around 288,000 people, in the tech-powered 2020 version) to do census enumeration while establishing effective systems to account for the last, hardest-to-reach individuals. The count also took place under a president who had no qualms about trying to influence the system.

Somehow, the bureau pulled it off. danah compares the continually beleaguered, high-pressure census to the game of Jenga, where people remove more and more game pieces from a tower and add them to the top until it falls. “As with many other types of infrastructure, people often think that they understand how the system works—until it breaks,” she writes. “In writing this book, I want to reveal some of what happens in the hope that an informed public can protect democracy’s data before it disintegrates.”

The book arrives with some urgency. By last summer, staffing at the census was down by around 25% compared to the previous decennial cycle. Three weeks ago, new rules were proposed that would curtail full data collection on US residents and likely skew the count to benefit Republicans, with a public comment period that is open until October 13. Meanwhile, elections loom for the next Congress, which will oversee plans for the 2030 census during the period where there’s still time to make changes.

Most of all, danah highlights how trust in institutions and our fellow Americans has cratered under Trump, but describes how the civil servants she came to know demonstrated the opposite spirit: “They were able to do the impossible because they functioned as a team—loyal to one another and to the mission—even as the internal social fabric that made this teamwork possible was under attack.”

danah spoke to me in the run-up to the Washington, DC, launch of the book, in a room filled with present and former Census Bureau staffers who seemed delighted to see each other and gratified by this account of their work. (Most people I spoke to had left their jobs in 2025.) This interview has been edited for length and clarity.

Porch Party: What were the biggest surprises you found at the Census Bureau?

danah boyd: First, as an American, I had grown up with the narrative that, you know, civil servants are dumb, civil servants are incompetent, civil servants could be easily replaced by companies, all of these things. I had heard this growing up in Pennsylvania, and I had absorbed a lot of it. And so I was first shocked by how dedicated these people were, how much they knew, how much they had built robust processes where they thought through every issue.

As a critical scholar, I had also grown up with the idea that government doesn’t understand that all numbers have politics; they don’t understand anything about the social aspect. Well, they understood it all.

Like, Ron Jarmin, the then acting director, he stands up in 2021 and announces, “The population is 331,449,281.” He’s an economist! He knows that none of this goes down to that level of precision. But when you ask him, he’s like, “Well, this is the ritual. And we’ve been doing this since 1791. So that is my responsibility.” He’s like, “Is there a better number? I don’t know a better number.”

The other thing that becomes clear about civil servants is that they’re not in it for the fame and attention. They’re really opposed to being recognized. You’re like, “Can you please actually take credit for something you did?” And they’re like, “No, no, we all did it.” Very unlike academia, unlike industry. You just had to appreciate these folks, because you realize how important they are to our democracy.

PP: You talk a little bit about some of the attempts to involve private companies in the census process. Did this project convince you of the reasons to keep it inside government?

db: Absolutely. I thought I understood management. I thought I understood how systems work. And, for example, there was this moment where I had gone on this meeting with the census official in charge of advertising and communication. He was meeting with a new vendor that they’d hired to do this mass advertising campaign, and he wanted me to observe this conversation.

And the vendors were proudly puffed up about how they could reach anybody. Coca-Cola had been one of their big clients. And these census folks were just sort of like, “Mm-hm! Okay. So here’s the thing, when you were trying to sell Coca-Cola, how did you reach people who didn't have enough money to buy a Coca-Cola?” And they’d be like, well, we just didn’t. “And how did you reach small children who you wouldn’t sell a Coke to?” Well, we didn’t. “And how would you reach…” and it kept going, right? So the entire responsibility at the census is to reach those folks, that last 1%. And industry, it doesn’t have any reason to. It doesn’t bother.

It’s not that hard to get the first 90 percent. It’s not even that hard to get the next eight or nine percent. There’s some issues and some imperfection, but the difficulty is all in that tail. Even something as simple as that advertising campaign—if they had their druthers, they would devote the entire advertising campaign and all of the dollars to that 1%. But Congress approves the advertising campaign. And Congress wants the money for that advertising campaign to go to their district.

PP: Because they’re trying to get robust response in their area?

db: Because they want to allocate money to various partners in their area. It’s like, oh, I don't care that you’ve already gotten everybody here, but you should make sure for fairness that you've advertised to all of our local news organizations that want that advertising money. That’s not the most efficient plan, but Congress holds the purse strings.

Jobs are another one where it’s like, you have to hire all the people. There was a big lawsuit after the 2010 census because with some people who applied, the background checks took too long and they didn’t get the jobs. And so a bunch of organizations sued, saying that this was unfair and that the people who didn’t get those jobs were more likely to be Black and Latino and other marginalized populations.

In that settlement, they went back and forth over which acts that got you to jail allowed you to be excluded from being hired by the census. And I remember at one point, the guy at the Census Bureau who was in charge of this was like, “Great, now I get to hire arsonists. Because it’s deemed at this point that it’s unfair if I exclude somebody who’s been convicted of arson.”

In his mind he’s like, “I am trying to run the largest temporary work program in the nation, by far. And I’ve got a management challenge that not a single industry can think through.” 

PP: And you’re sending these workers to people's homes, right?

db: I know! I loved talking to the FBI people because they would be like, “Oh yeah, the census is up again. You want to know the problems we have to deal with?” And then you’re like, ah right, because homicide and rape have occurred in previous censuses. It’s not been a problem this decade. But even just the number of people who have heart attacks on the job. People who die on the job.

PP: And this is a temporary workforce. Nobody knows each other well. And it’s a lot of elderly people and people who are not in the full-time workforce.

db: Yeah. The more you think about it, the more you’re like, how do they even do this?

PP: Most of your book is about the 2020 decennial census. What did you see as the most difficult issues the bureau was grappling with? My sense is that there were twin hurdles of the pandemic and then, as time went on, attempted interference under the first Trump administration.

db: It was the pandemic, 100%. For census officials, it was not as simple as the political interference of just the Trump administration. The choices of the Obama administration to not fund things became a problem up front. Then you have different challenges within the Trump administration that are sometimes political interference, sometimes reputational.

For the people inside the house, the partisan interference and the political interference span administrations. And they go back further, right? They point out all of the problems of the Clinton administration and the Bush administration and all of these choices. And that's one of the reasons that in the book I use the frame of Jenga politics. Because for me, it’s the realization that it’s not just one actor. They’ve all been weakening pieces and putting pressure on it in different ways, but they always blame the other.

PP: Heading into the 2030 census, what is the mood of staffers there that you know?

db: Abysmal.

PP: What are their biggest concerns?

db: The biggest concern, I would argue, is legitimacy. Legitimacy and data quality are not the same, which is to say that the 2020 data are actually better by almost every metric than in the 19th or 20th century. People don’t believe that they are good. And that means that they don’t want to use them, and they have all these political fights because they can delegitimize the data, delegitimize the workers.

We are in a massive delegitimization crash as far as trust in the institution. If we delegitimize this, nothing else matters, right? Because we can’t actually function on these data if we don’t treat them legitimately.

The other concern is that they’re going to attack procedure, and the actual functioning of the Census Bureau. Normally, years ending in zero are peak in terms of the number of people that work for the census. And then it declines, and then it starts to build back up mid-decade. We should be building up. We are still declining.

PP: Why is that?

db: There's a complete hiring freeze, and there has been for two years. Contracts need [Secretary of Commerce Howard] Lutnick’s approval. So all of these systems are not in motion, even though they should be. Keep in mind that people think of the decennial census as happening in 2030. But all of the major systems and everything critical will be decided and fully operationalized by the end of year eight.

This November is the election that matters, because whoever is sitting in Congress will have final say. And they can either abscond with their responsibility, which they've been doing a very good job of, or they can take responsibility. But either way, they are the ones responsible, and it’s this upcoming Congress who’s in charge.

The interventions that are happening right now are on schedule. One is the change of the residence rules, which happened on September 10th. And these are crazypants, if you look at these residence rules. It’s normal to be publishing them now, and then you get feedback. Congress holds hearings or reviews them. And then they come back with a final recommendation and that’s what they’re going to operationalize.

PP: When you say the residence rules, this is about, for example, where college students and people in prisons are counted?

db: Who counts where. The goal is to count everybody once, only once, and in the right location. The residence rules are about what is the right location, but also who counts. So it’s this weird esoteric document. And this is the big fight that happened on September 10th. This set of rules comes out and it’s like, rather than like micro changes, which is what’s been happening for decades, the new one says, you’re no longer allowed to collect race or ethnicity for the 2030 census.

We have countless laws that depend on those data, starting with the Voting Rights Act. We need those data to uphold the laws. So if you don’t want to uphold the law, you just get rid of the data. This is a form of policymaking by other means, and that is not a civil servant decision. And nobody tried to even justify it on research. That was a political decision.

PP: This list of rules came from where? From the administration?

db: Came from the Census Bureau. But it’s not written by the Census Bureau, right? It’s written by the political side. Another rule is about citizenship.

PP: What is the fate of that residence memo now?

db: It’s in the Federal Register Notice. We have until October 13 where people can write comments. And I strongly encourage everyone to write comments. [A coalition of groups called Hands Off Our Census—you can guess their position—has an online guide here on how and why to comment on the new rules.]

In theory, the Census Bureau is supposed to respond to these materials. In reality, the expectation is this is so partisan that they will not listen. So one of the main reasons why opponents to this change are asking the public to write materials is that litigation is following. This is going to be heavily litigated.

And it’s also a matter of going to Congress and being like, guys, Article 1, Section 2, of the Constitution, you’re in charge! Would you take responsibility? But the changes you should know about: First, race and ethnicity are off the table. Second, don’t count certain kinds of non-citizens at all—which is another way of getting at the agenda that the first Trump administration had, now by just not having the data.

And then the third one is creating all these micro changes to specific populations. For example, the norm for counting residence since 1790 is the usual place of abode over time. Where do you live across the whole year?

Now they’re saying, oh, well, we only care where you typically live during the census window of January to April 1. Why does that matter? People who have residences in both northern states and southern states, like snowbirds, are to be only counted in the south. People who are migratory due to crop harvesting are to be counted in the south.

PP: Oh my God. It’s like weather-based discrimination, where they’re like, how can we knock out cold states?

db: Which we’ve deemed to be Democrat and not Republican.

PP: Wow.

db: I mean, it’s obvious what they’re doing. But is Congress going to take responsibility? Open question. Because if you are actually a true believer in constitutionalism, it’s very clear this is not how this is supposed to be. The decisions have to be finalized next year.

The 2027 Congress makes critical decisions, and then there’s another wave of them for whoever’s seated in 2028. But anybody who’s seated in 2029, it’s a done deal already. 

PP: So those next two elections will be critical.

db: People think that they don't have to pay attention to the census until 2030. And part of why I wanted to publish this book now is I know that this is when the critical decisions start really happening. 

PP: A question I’m asking of people in government as I talk to them is, given what you know, if you had a chance to rebuild or change this bureau after a new Congress or a new presidential administration, what would you change? How much would you build back what was there, or what kind of changes demand to be made?

db: The most critical thing that I would do would be to make it an independent statistical agency. There’s no reason that this should be governed by the executive branch. It should be governed directly by the legislative branch, because they are ordered to do so in the Constitution. But I think there’s ways you can structure it to be managed as an independent agency, where you forcibly allow the best decision-making. 

I think the second clear one to me is that, especially post the Loper Bright decision [which held that courts would no longer defer to agency expertise to reasonably interpret laws passed by Congress], Congress needs to step in and explicitly be clear to the Census Bureau that the bureau needs to govern the best methodologies through a statistical method, rather than through the courts and through the fights over this. In a case like the Census, which is the foundation of political power, there is no way to do this as a constant fight in courts. That is actually devastating for it.

So Congress needs to explicitly tell the Census Bureau, you have the flexibility to make these decisions, and we will step in and say when you can’t do this or you must do that. For the core decennial census, I believe Congress has the right. If we don’t like that, we have to think about our Constitution; that’s a bigger fight. But this is Congress’s right in our democracy, and we need to empower them.

I recently showed my young kids the 2010 documentary “Babies” (we are expecting some new cousins and have babies on the mind) and they then surprised me by wanting to watch it three nights in a row, requesting a screening with their friends, etc. I don’t know quite why this practically wordless movie, which follows four very cute babies in Namibia, Mongolia, Japan, and the United States through their first year of life, is so transfixing, but it still is. Of course, kids always find fellow kids interesting, but for adults the beauty is partly in the reassuring message that a huge range of baby-raising tactics work out just fine, whether your kid is staring out the window of a high-rise or being licked by a goat. Just don’t forget to get them vaccinated against measles.

If you’ve read this far, thanks for your support. What topics or people should Porch Party cover this fall? Got pitches for a guest piece? Drop me a line. And if you haven’t done so already, please consider becoming a paid subscriber, which will allow you to comment and get invites to future in-person events.

Hope you’re keeping your footing in the whirlwind that is this fall. See you next time.