One Man’s Quest to Break Open the Secretive World of American Voting Machines

When President Donald Trump began spreading the conspiracy theories about why Joe Biden beat him in November’s election, he swung at all the glitzy targets: The media, Democrats, China. But he saved his most spectacular accusations for a more obscure and implausible target: The companies that make America’s voting machines. Over the winter, Trump publicly harangued Dominion Voting, a small company with a headquarters in Denver, and another company called Smartmatic, which barely has a footprint in the United States. Week after week, Trump’s lawyers cast them as the hub of a grandiose conspiracy to alter the vote, one machine at a time.

The wonky set of experts and academics who actually study U.S. voting infrastructure watched in shock. It was obvious the Big Lie about the election was patently absurd. And when it came to voting machines, it was also ironic: Precisely thanks to the serious efforts of Trump’s own Department of Homeland Security, the 2020 contest was, at least technologically speaking, the most secure election in modern history.

Yet these experts also understood that voting companies were an easy mark for a reason. By just about any measure, they are some of the murkiest and inscrutable firms in the civilian private sector.

Relative to their importance to society, few technology sectors are less well-understood than the voting systems market in the U.S. It is an industry funded entirely by taxpayers, and administered mostly by private equity, on whose work democracy completely depends—and which even committed analysts have struggled to understand. Last year on Capitol Hill, when CEOs from each of the three major voting companies testified before Congress, then-Rep. Susan Davis, a Democrat from California, asked each of them if they could share their annual profits. All three declined. As one CEO, Tom Burt of ES&S, politely explained, “Congresswoman, we’re a private company, so we’ll keep that information private.”

As 2020 fades, a growing vanguard of academics, good-government advocates and even Silicon Valley executives are turning their attention from the security of voting technology toward the companies who build and sell it.

Long before 2020, experts have been concerned, even flummoxed, by the voting-machine industry. One of them is Lee Bollinger, the president of Columbia University. In 2017, Bollinger co-chaired a commission by the National Academies of Sciences to study reforms to America’s election system. “The discussion about voting machines was fairly striking to me,” Bollinger said, recalling the early sessions in which he first grappled with the private industry. “I could tell from the beginning—and this may not be an entirely appropriate metaphor—but the kind of ‘black box’ quality to this whole arena was a mystery to me. A puzzle.”

What was going on in the black box? In the end, even Bollinger’s commission would not quite crack it open. “That struck me as strange,” Bollinger sighed. “I mean, it’s sort of demanding of an explanation.”

Someone did have an explanation. It wasn’t a White House lawyer, or a senator, or the chair of a Washington commission. Instead, it was a student at the University of Pennsylvania named Matthew Caulfield.

Caulfield grew up in Ocean City, N.J., where he attended public schools. In seventh grade, he calmly decided he would go to Wharton, the prestigious business school. Seven years later, that’s precisely what he did. He arrived on campus with an interest in investment banking. But by the end of sophomore year, Caulfield found himself drawn to policy. In 2014, he took an unusual internship at Third Way in Washington, where he co-authored one of the early policy papers on a quirky phenomenon called Bitcoin. In a school full of whiz kids, Caulfield was a Whiz Kid, graduating summa cum laude.

During the final semester of his senior year, he inherited a puzzle that would torment him for five years. In a public policy seminar, Caulfield’s professor assigned his students a research project: How big, precisely, was the U.S. elections industry? It was the kind of assignment that was routine at Wharton, where students frequently were told to evaluate, say, the international market for coffee beans. “It was supposed to be a one-semester thing,” Caulfield recalled. “Easy.”

Caulfield began to consume all the information he could. He learned the history of what happened after Bush v. Gore. Following that fiasco, Congress poured billions into state coffers to upgrade their voting machines. After the flood came the famine, and when the money dried up, the private market began a dramatic consolidation. Big firms gobbled up the smaller ones, until 20 years later, the number of major companies had whittled down to three: ES&S in Nebraska, Hart InterCivic in Texas, and Dominion, whose main headquarters is based in Canada. Today, these companies control just under 90 percent of the American market. All three were majority-owned by private equity companies—Staple Street Capital, Enlightenment Capital and McCarthy Capital. But their business model was unlike any Caulfield had seen before. They sold voting technology to their customers more or less individually, servicing the 3,000-odd small-budgeted counties and jurisdictions around the United States, one clerk at a time. Although their products were complex and vitally important, and used by governments nationwide, their model looked more like door-to-door globe salesmen than a sizable technology manufacturer.

After huddling together, the Wharton students agreed on a plan. Step One would be to size up the market, in order to get “the basic lay of the land,” as Caulfied put it. Then they would regroup.

Just as quickly as it began, however, Step One took a turn for the strange. That week, Caulfield and his classmates hit a brick wall: “We couldn’t find anything.”

The first sign something was amiss came when they started making phone calls. It was protocol for students to call up a company directly to discuss the state of their industry. In response, executives usually leapt at such a chance—a great way for a small coffee-bean producer to get noticed by the likes of Wharton. Caulfield had never encountered companies that appeared as if they didn’t want to be noticed. When he contacted election vendors, seeking an interview, he was essentially ignored, or in some cases given a pro forma hearing before being politely stonewalled. As a Plan B, the students queried PricewaterhouseCoopers, the big consulting firm. But they didn’t have any information on the companies, either. In a stroke of creativity, Caulfield made a public records request to the state of Maryland, seeking the financial history documents provided by the state’s voting machine vendor, ES&S. “That way, I could know their revenues, their costs, across a couple of years,” he explained. From this, he was able to glean a single breadcrumb: the number of full-time employees. The rest of the key data, contained in the all-important financial history portion, came back with every figure redacted.

It began to dawn on Caulfield, slowly at first, that the amount the public didn’t know about these companies was vast. Quarterly profits, regional market share, R&D budgets, even the number of employees—often, there was simply nothing. “Basic, basic data—the basic layout of the industry—was just not out there,” Caulfield recalls. “Eventually, we realized that it didn’t exist.”

There was nothing wrong about this, exactly. As Tom Burt told Congress, private companies are private, and they don’t have to disclose information like annual earnings or market share. Nor did Caulfield detect any malign ambitions by the companies or the people who ran them. As he would come to understand, most employees working in voting technology were drawn by a love of civics; a few had once served as local election officials. The oldest of the companies, Hart InterCivic, dated back to 1912, mainly printing ballots.

Today, these companies manufacture and sell all of the equipment that dot the walls of elementary school gymnasiums on Election Day: voting machines with names like ExpressVote or the ImageCast Evolution, but also ballot scanners and vote tabulators, electronic pollbooks and accessibility features for people with disabilities. Caulfield’s concern wasn’t the security of the machines or the software that ran them. He was a business student, and simply wanted to know how the companies ran their business, why they did it that way and how they made money. And it struck him as odd that, for an industry that was entirely supported by taxpayer dollars, “no one had been able to pierce the veil.”

Today, Caulfield is 27, and has ruddy, round cheeks and downy brown hair. “It was supposed to be a one semester, five-to-six month thing,” he recalled, laughing about that first assignment. “We were a little naive.”

The students worked until semester’s end. When it came time for graduation, the report wasn’t finished. So, while his fellow students took jobs in finance and consulting, Caulfield decided to stay on campus. He worked feverishly that summer with Andy Coopersmith, a director at the school’s Public Policy Initiative. The two worked like private investigators, calling dozens of employees, gearheads, clerks—anyone who touched even a small part of the elephant of the private voting industry. With the help of another student, they also made clever use of a public database, built by the respected watchdog group Verified Voting, that gave them a starting point of which companies’ products were dominant, and where. That summer, as Caulfield went deeper than anyone had before, he realized the project had taken on dimensions far more important than just figuring out a balance sheet. By now, he got excited when he told people the question he was trying to answer: “What happens when the very foundation, the literal critical infrastructure of democracy”—voting—“is subject to the forces of a private market?”

Caulfield worked on the report well past graduation—throughout the 2016 election, and into the first weeks of the Trump administration. In March 2017, Wharton published Caulfield’s and Coopersmith’s 58-page report. It was called “The Business of Voting.” Around the halls of Wharton, the document was received politely. Inside the insular field of election administration, however, Caulfield’s findings arrived like a lightning storm.

Caulfield had found a private market that seemed not only hard to analyze, but congenitally incapable of innovation. The problem started with money: There wasn’t any. Local governments, which were usually the ones that bought voting equipment, had budgets that were the equivalent of fumes. Most individual clerks found themselves negotiating with the companies individually. “What you have is highly fragmented buyers. And if you’re a buyer, fragmentation hurts you,” Caulfield explained. In Caulfield’s view, local clerks had a similar kind of problem to the one that often leads economists to argue the federal government should negotiate for lower drug prices by bargaining with the largest pharmaceutical companies directly; the bigger you were, the better your bargaining power. But the federal government didn’t buy voting equipment, local governments did. And the money and stature of, say, Clark County, Wisconsin, was not enough to exert the demand that generally provokes the healthy rhythm of innovation that characterizes other technology sectors.

Yet it wasn’t simply the case that sellers had little incentive to innovate; in some ways, the report found, they actually had incentives not to. Caulfield had a term for what he was seeing: “market complacency.” During the 2000’s, as smaller voting companies had been eaten up, the pressure to innovate had fallen; fewer firms meant fewer rivals to build a hot new product. And there was also another reason sellers might not want to innovate: A better product wouldn’t yield bigger profits. If election clerks were lucky, their county governments might allow them to buy a new fleet of technology every 15 years. So, even if a company poured resources into building a better machine, it would collect dust in a warehouse for years before clients could afford to buy it. “Typically, you can innovate and make your market bigger,” Caulfield summarized it. “And there’s no option for that here.”

In theory, another company could enter the market and build a better machine. In practice, Caulfield discovered, it was nearly impossible to break in. One reason was the byzantine regulations that were passed after 2000. Almost all 50 states now require a certification process for their election technology, which is usually unique to them—meaning that a machine sold to Florida might not be suitable for a county in Idaho. The cost for companies to certify a new voting machine can routinely surpass $1 million. In one remarkable example, a voting company sank $12 million into certifying its new machine before giving up and pulling the plug.

But by far, Caulfield’s most significant discovery was to put a figure on the total size of the industry. He estimated the entire revenue footprint of all the companies in the United States was $350 million. That meant the entire elections industry in the world’s richest democracy was about the peak size of the R&D department of the camera company GoPro. The private voting sector wasn’t like a secretive and well-heeled defense contractor. It was more like the manufacturers of arcade machines or jukeboxes, grasping for market share with a product they could sell at best once a decade.

This insight went a long way toward explaining what had previously been regarded as the companies’ unusual behavior. In recent years, some vendors have sued (or threatened to sue) counties and governments when they attempted to buy from a rival company, with instances in Texas, Louisiana and Illinois. It was as if Coke had become so desperate for business that it sued a middle school for switching its vending machines to Pepsi.

Though the report landed without a sound at Wharton, in the world of elections, Caulfield was catapulted into minor celebrity—the author of the first modern market analysis of elections that most experts had seen in their entire careers. The 23-year-old Caulfield was flown to conferences in San Francisco and Miami, where crowds of senescent math professors gave him standing ovations. At one event, a CEO of an international elections company pulled Caulfield aside, and explained that he’d made the Wharton paper required reading for his global staff. Compared to the international market, America seemed like a strange beast.

Suddenly, Caulfield was being asked to join lawsuits as an expert witness. More strangely, he found himself showered with job offers. Caulfield didn’t have the heart to tell them it was all just a class assignment. “My professors said, ‘Focus on your research!’” he chuckled—which involved a blend of business ethics and tort theory. “This was just my hobby.”

Several times, Caulfield attempted to put his hobby to the side. But his timing was unlucky. The 2016 election sent shock waves through the country. Ambitious efforts were now underway—in the public as well as the academy, to say nothing of the Department of Homeland Security—to find answers to the types of questions Caulfield had asked.

Caulfield stayed at Wharton and pursued a Ph.D in business ethics. Then, in 2017, he received an invitation in his email inbox from the National Academies of Sciences, which was convening a commission to investigate the American voting system. Its co-chair was Lee Bollinger, the president of Columbia University, and its panelists included people like General Michael Hayden and the cybersecurity chief of the Department of Homeland Security, along with approximately 50 others.

In June, in a ballroom on Columbia’s campus, Caulfield arrived to present his work to the committee. They surrounded the witness table in an imposing horseshoe arrangement while Caulfield clicked through slides of his report on a giant projector screen. When he finished, the panel was opened for questions. The first question was given to Bollinger.

“So, a typical voting machine costs—what exactly?” Bollinger asked the 23-year-old. Fifty heads in unison turned to look at Caulfield.

Caulfield considered the question. “That’s a good question, sir,” he stammered. He thought a little more, then said, “That’s something I don’t think anyone actually knows.”

Bollinger peered over his reading glasses. “You’re saying we don’t know?”

It was true. Even though Caulfield had managed to calculate the size of the industry, no one truly knew what any of this stuff—a voting machine, a ballot-counter, an electronic poll book—actually cost. One organization, the National Conference of State Legislatures, had released a report asking the same question. “Does the United States spend a billion dollars a year running elections [or] $10 million?” the report asked. “No one knows.” Even clerks, who made the purchases for their counties individually, were mostly in the dark about what other clerks paid. At gatherings, Caulfield was somewhat struck by how little the question of business dynamics—much less the price of the equipment—came up in conversation among election officials, at least in public. They didn’t seem particularly eager to learn, for example, whether they had gotten a bargain, or whether they’d been stiffed.

Everybody seemed to recognize the absurdity of this situation. Bollinger and his panelists might as well have been chairing a greenhouse gas commission, only to discover that they were forbidden from knowing the price of gasoline—as if the price per gallon were a closely held secret between drivers and cashiers.

For their part, voting technology companies, including the small ones, did not publicly list the prices they were charging. Being privately owned, they didn’t have to. Mark Lindeman, a longtime voting expert who leads Verified Voting, explained the difficulty this poses for election clerks whenever they attempt to buy a new fleet of voting machines. “Election officials have no way of knowing what a fair market price could possibly be,” Lindeman said. He compared it to a recent experience he had buying his car, a Chevy Volt. At the dealership, “My wife and I got run in circles, while we waited for the dealer to give a price that was sensible,” he said. “How did we know it was sensible? Well, we went to the Consumer Reports website and got an idea of what a fair market value would be.” He added, “Election officials have no way to do that.”

Neither did the National Academies of Sciences, it turned out. In the end, what the commission discovered—just like commissions before it, and the conferences before those—was another enigma, forcing even Bollinger to raise the white flag. In their final report, the best the commission could do was point to a study, back in 2001, which declared that “even the most basic facts” of the cost of voting technology were still unknown. In a delicate flourish of bureaucratese, the National Academies concluded that, 20 years later, “This assessment remains applicable.”

Nevertheless, Bollinger had posed the question. What was America paying for its own voting machines? And in asking it, he sent Caulfield back down the rabbit hole.

In the spring of 2019, Caulfield launched a second investigation. This time, he would attempt to determine how much clerks were paying for their own voting equipment. To keep the project simple, he focused only on the cost of machines, ignoring the battery of technology—electronic poll books, centralized ballot counters, computer software—that often come attached. It was on this second go-round, while he submerged himself in ever-more obscure studies and journals, that Caulfield began to suspect that the smartest people in elections had been living, essentially, in a hall of mirrors.

How much did a voting machine cost? Caulfield chased down numbers with the doggedness of a prosecutor. The most popular number came from a state lobbying group—but it listed no source. That figure had been popularized by another report, which estimated the cost of replacing every voting machine in America at a billion dollars. But deep in the footnotes, Caulfield saw that the number was plucked at random from a comically huge interval: 2.9 billion dollars, give-or-take. And its source, the state lobbying group, would later stop referencing its own figures. The only other sources were a few stray remarks from legislators in New Jersey and North Carolina, who once made some guesses about what their states had paid for machines. Importantly, there had been a scattering of studies that collected a handful of vendor contracts. But those were mostly decades old, and useless for making sense of a national market. Within a matter of weeks, Caulfield had discovered a statistical merry-go-round, in which everyone had cited everyone else, and hard data never really appeared.

The federal government was charged with keeping track of certain voting machine data. But it didn’t have anything about the price or value of the technology in use. In 2014, the White House had convened its own report on election administration. But on the question of prices and markets, the information it relied on was pure anecdote. Even Caulfield didn’t appreciate how little anyone knew until he received an email from a private research firm. Caulfield had noticed the firm was updating its financial figures for the voting machine companies, and he emailed the firm excitedly to ask what they’d learned. In response, the research firm sent Caulfield an article from USA Today—referencing Caulfield’s own Wharton study. They hadn’t realized, evidently, that some of the estimates Caulfield included in that study had come, in fact, from the research firm itself.

Around this time, bedraggled and perplexed, Caulfield felt somewhat like the outcast traders of Wall Street during 2008, who went searching for the adults who understood the financial system, and discovered that there were no adults who understood the financial system. “You’d go to presentations. And at these things, people would make claims based on, ‘Oh, we collected contracts from three counties in Virginia!’” Caulfield told me. “People spend years going to conferences making speeches confidently proclaiming the answers to election technology. And then they don’t have an answer to any of these questions.”

It took Caulfield a little time before he figured out what to do. There were estimated to be approximately 3,000 counties in the United States, each with an office responsible for handling or tracking the purchases of its own voting equipment. Why not just ask them?

Caulfield linked up again with his old research partner, Coopersmith, and together they recruited a small team of undergraduates. For a few weeks, the team collected thousands of emails and phone numbers. Then they went about contacting them, one by one. The students’ request was put in simple terms: Did they still have the contract they paid for their voting machine? And would they share it with a gaggle of teenagers from Penn?

The outreach took about eight weeks. That summer, Caulfield was sitting at his desk in Philadelphia when the first responses began to trickle in. Many of the officials wrote back, apologetically, to say that there weren’t any records. They had purchased their machines over a decade ago, some said, or under a different clerk. A few simply relied on word-of-mouth. “One county said something like, “Well, it was back in 1996, and we don’t have the records anymore, but I think we paid X,” Caulfield recalled. “There were a non-trivial number of jurisdictions that literally did not know how much they paid for their own voting system.”

Then there were a few clerks who got angry. One election official in Texas reported Caulfield to the attorney general’s office and alerted the election companies to their mischief. (Caulfield described the experience as “a little intimidating.”) One pair of Massachusetts clerks wrote primly that their wooden box had served them just fine for more than a century, “and we don’t have any plans to update our system.”

But there were enough election officials who played along. After several weeks, the data Caulfield’s team had hauled in was significant: 356 jurisdictions agreed to share their vendor contracts, by far the largest such dataset ever assembled. Peering into the computer, Caulfield and his team were the first people to see what America had been paying for its voting machines, all this time. They began analyzing the contracts closely, one at a time, crunching the numbers of thousands of pages of documents. “Once we started doing that, it was really stunning,” Caulfield said. “There was no pattern to it.”

Plotted onto a graph, the variation in what counties were paying for their technology appeared to be practically random. After some analysis, Caulfield began to figure out why. In the first place, unit prices for the machines weren’t uniform. Voting machines had clusters that indicated a mean price; the ES&S DS200, thought to be the most popular voting machine model in use, had a median price tag of $5,750 listed in the contracts. But the variation could be wide. There was a second reason for the apparent differences in what counties were paying. Voting companies had been dispensing what they called “discounts” to local governments. These discounts were sprinkled across the contracts and came in dozens of forms. It wasn’t the discounts themselves that startled Caulfield. It was that their widespread use didn’t appear to have a shred of coherence.

I contacted all three major voting-machine companies for this story, asking whether Caulfield’s findings reflected their practice nationally, and why the prices appeared to vary so much from one buyer to the next. Dominion and Hart declined to comment. A spokesperson for ES&S, the country’s largest vendor, said that “all contracts are competitively priced” and reflected a variety of factors, including “early adoption of technology or how the county will pay for the voting system.”

In the data, Caulfield discovered clerks who were buying the same voting machines from the same company, but seeing significantly different discounts than their peers—sometimes in the same state. In California, Mono County and Placer County both purchased orders of Dominion equipment, including the ImageCast Evolution, a voting machine priced at $7,200. But Mono received a 5 percent discount off of its bottom-line order, while Placer, 128 miles away, saw a mark-off of nearly 25 percent. Volume didn’t necessarily matter, either: Dodge County, Wis. purchased the same ES&S machines that Polk County, Fla. did. Even though Polk County bought substantially more machines and equipment, Dodge got a discount seven times larger than Polk’s—a mismatch Caulfield spotted in other states, including Texas and Virginia.

In some cases, the variations veered toward the absurd. The county of Gila, Ariz., received an “Arizona Customer Discount,” while Coconino, Ariz., did not. In Virginia, Arlington County and Spotsylvania County both received what their contracts called a “Good Faith” discount—but one was 20 percent better than the other’s. “Like a Wild West,” Caulfield described it. As far as he could tell, “there was no logic to it.”

Because of this variation, the data made it hard to draw big conclusions about national trends. But if there was a visible pattern, it was a disquieting one. Counties that tended to be smaller and poorer also tended to receive a smaller discount. Caulfield was hesitant to speculate why this might be. But in his initial Wharton study, a picture emerged that offered a plausible explanation: the “buyer power” of a severely fragmented market. In the world of elections, business was so scarce—and the sharks so frenzied—that losing a contract to a big jurisdiction, like a major U.S. city, could potentially pose a serious loss to a company. Almost any discount might be justified, if the jurisdiction were large enough. By the same measure, a sale to a small, rural county with a few thousand residents would not present nearly the same urgency. To Caulfield, and anyone reading the same data, the voting companies threw discounts at rich jurisdictions because they had to. They charged more to the small ones because they could.

But if the prices showed such wild variation, how did the election companies consistently make money? In his massive dataset, Caulfield discovered the answer. The companies didn’t necessarily need to profit from the machines at all. Instead, they appeared to rake in revenue from the significant fees they reaped in afterward. Maintenance, upkeep, annual software licenses—all came with fees that local election officials continued to pay the private companies, year after year. The fees were sometimes so large that they eclipsed the cost of the machines themselves. Caulfield discovered several counties in Colorado, including Boulder, where the cost of fees after 10 years was 140 percent more than the price that taxpayers had footed for the initial machines. In one county in Oregon, it was 200 percent.

As Caulfield and Coopersmith wrote, the key to stable revenue meant finding ways “to sustain long-term relationships with counties”—and the key to that was “ongoing annual payments.” Perhaps that was another reason innovation was scarce. A business model in which repair and maintenance costs are the most stable source of revenue might not, to put it delicately, create optimal incentives for designing hassle-free technology. In fact, as Caulfield and Coopersmith suggested, it might not be financially wise to regularly roll out new models at all.

Oftentimes, election clerks were the very last people who cottoned to this screwball model. “How much choice do you have?” Gretchen Reinemeyer, the director of elections for Arlington County, Va., told me. Caulfield’s work had revealed that, in 2015, Arlington and nearby Spotsylvania County both purchased the same voting machines, but that Arlington’s discount was 20 percent lower—even though its purchase volume was essentially double. “We’re such a small department in Arlington County. We only make up 0.1 percent of the total budget,” Reinemeyer explained. In many cases, she went on, “counties are left having to procure equipment without much time and without being able to go through a thorough procurement process.” Nor do they have the wisdom of past experience. “This is the type of purchase we do once a career,” Reinemeyer said. “The equipment lasts ten years, and most people don't make it 20 years in this business.”

Shortly after Bush v. Gore, the Inspector General in Florida’s Miami-Dade County issued a report on the voting system there. It concluded that the relationship between voting companies and the local government “can be summed up in one word: dependence.” Twenty years later, it could be said that, invoking the words of Bollinger’s commission, this assessment remains applicable. Reinemeyer told me, “It is frustrating to know that our taxpayers are paying a lot of money for these machines. But we have to run elections. And we don’t have a lot of choices.”

The one company that did reply to my questions, ES&S, said it “work[s] with jurisdictions which lack funding to offer discounted pricing” and other incentives, and pointed out that it’s common in business for larger-volume sales, such as to big metro areas, to receive a greater mark-off. (The company didn’t address Caulfield’s finding that volume did not appear to correlate with price.)

The spokesperson for the company also noted that ES&S “devotes significant resources to research and development to ensure the long-term security, accuracy and reliability of voting systems,” pointing out the extraordinary regulatory thicket it must navigate among local, state and federal officials—as well as “the EAC, DHS, law enforcement, voting system manufacturers, and the election community.”

When I asked about the charge that voting machine companies don’t innovate at the pace of other technology businesses—ones that, for example, develop smartphones—the ES&S spokesperson said that a slower pace of innovation “is not an indicator of a lack of innovation.” Plus, the spokesperson said, the two sectors aren’t really comparable: “Consumer devices are developed and sold based on a rapidly evolving market and a large demand. In the much smaller elections industry, jurisdictions generally purchase new voting technology every 10 or more years.”

To the charge of failing to innovate, ES&S also proposed a solution: actually funding elections more generously. “Providing necessary sustained funding to jurisdictions will allow them to invest in newer technology, additional innovation, [and] security and resources,” the spokesperson said.

In March, Caulfield’s study on prices was published by Verified Voting. In various corners of the election world, the analysis was instantly hailed as a breakthrough. “This information, on this scale, hasn’t been widely available to anyone,” says Lindeman, Verified Voting’s co-director. “Until this report was provided, no one really knew much at all.”

You might expect someone who was fresh off his Ph.D. and a groundbreaking study to spike the football, but then again, Caulfield is not a typical twentysomething. He is circumspect about what his study means for the big questions about democracy. (“I like to stay in my lane,” he says.) To others, however, Caulfield’s work points toward something more radical than perhaps even its author intended: a new reason to question the marriage of election administration and private industry. “What kinds of machines would we make if we declared this a public good, and had it produced in public laboratories?” Bollinger asked when I called him this spring, just before Caulfield’s report went public. He drew an analogy to the life sciences and human biology, sectors where the U.S. government has made hundreds of billions of dollars in public research investment. “We have public funding for all kinds of development of things. But we leave this—.” Bollinger cut himself off and laughed. “We leave this essential object and thing, which is so critical, to the free market?”

These technologies need not be state-owned and operated; like the laboratories at prestigious universities, where federal funding bankrolled early efforts to study coronaviruses, then handed off the technology to Big Pharma, a public effort to develop critical systems in voting infrastructure could be the foundation of a public and private collaboration. (A version of this model is already playing out in miniature, in the laboratories of Microsoft Research.) But the private-only approach, Bollinger said, “has, in my view, sort of failed as a system. I don’t want to overstate it, but repeatedly, that’s not an exaggeration.” He hastened to add, just as Caulfield had found, that private companies only seemed to be doing precisely what we encouraged them to do. “We’re relying on organizations that need to make a profit,” Bollinger said, in an industry that’s “not a place to go if you want to turn a significant profit.” His tone was a blend between amused and perplexed. “From the standpoint of democracy, and citizens,” he continued, “this does not seem to make sense.”

Channeling that spirit, today there is a small but growing group of experts, academics and executives who are thinking about big ideas that could change the American voting system. Their work broadly follows two ideas. The first is what is thought of as the open-source movement. This view holds that election systems are generally more secure when their software code is open to scrutiny. But it also holds that voting systems would be more affordable, and more dependable, if entities were permitted to build voting equipment using commercial-off-the-shelf-technology—COTS, in the jargon—fashioning secure machines from monitors and printers that would otherwise be available at, say, Staples.

In making this broader case, reformers see a rough analogy to the computer industry. “The [current] voting system vendors are actually a throwback to the computer business of the 1960s,” says Dan Wallach, a computer science professor who has done pioneering work designing alternative voting machines. In the '60s, he said, “You went to IBM for everything. They sold you the hardware, the software and the support, all in one big contract,” just as voting companies do today. But with the arrival of personal computers that ran on common operating systems, computing became instantly cheaper, and allowed competitors to flood the market with machines that were smaller, cheaper and faster.

A second idea is broader in its ambition: to take the profit out of for-profit voting. This view has a small but growing chorus of advocates. One of their leaders is Ben Adida, the executive director of VotingWorks, a voting technology manufacturer that sells its equipment not as a company, but as a nonprofit. Adida, who holds a Ph.D. from MIT (where his specialty was cryptography applied to elections), worked in the upper echelons of Silicon Valley, including as the director of engineering at Mozilla and Square. In 2018, Adida went searching for information about the voting machine industry, and was disappointed that the companies were as secretive as they had been during his Ph.D. work a decade earlier. In the ensuing months, someone sent him a copy of “The Business of Voting.” Adida pored over Caulfield’s conclusions. A few months later, he resigned his posts and founded his voting nonprofit, which has a staff of 15. “The market is broken,” Adida told me last year, when the 2020 election was underway. He likes to point out VotingWorks is the first new vendor selling voting equipment in 13 years; the next-newest is 40. “They’re some of the most secretive companies out there,” says Adida. “We’re the polar opposite of that.”

VotingWorks’ debut voting machine, a ballot-marking device that Adida dubbed the VxMark, is beating a path forward for the open-source movement. It uses Lenovo and Microsoft tablets and HP printers. Its voting software is available to the public for inspection. In a departure from a model heavily reliant on maintenance fees, Adida’s team has designed software with an emphasis on simplicity and self-service. The organization’s first client is Choctaw County, Mississippi—a state where the regulatory burden for new election equipment is lower than most. Last November, about 4,000 people voted on the VxMark machines across Chocktaw, whose residents are among the poorest in the United States, and where voters could be seen driving combine harvesters to the polls.

Adida is troubled, but not surprised, by Caulfield’s latest findings. “These are signs of an unhealthy market,” Adida says, summarizing the report. He was unsettled by the possibility, raised in Caulfield’s data, that smaller, poorer counties weren't getting the same discounts richer ones were. VotingWorks recently published its official price listings on a public website, where the VxMark sells for $1,550, about a third the average cost of the machines studied in Caulfield’s report.

Adida hopes that Caulfield’s work, along with his own example, might prod policymakers to rethink aspects of the private system. He described his goal at VotingWorks as trying “to build the publicly owned operating system of democracy”—and one that is publicly accountable. In assessing Adida’s model, Caulfield and his co-authors called its incorporation as a nonprofit “radical,” along with VotingWorks’ “concomitant belief that something as sensitive and critical as the machinery of our electoral process should not be in the hands of private enterprises.” If widely adopted, Caulfield thinks such a model would represent “a major paradigm shift.”

Yet Caulfield is quick to pour cold water on these prophecies—or even assume it’s necessarily all good news. The private industry isn’t going anywhere, or not soon, at least. “I’m always skeptical,” he said. “This innovation won't remove the current vendors.” Caulfied isn’t an idealist; he’s a business professor. Last spring, Caulfield defended his Ph.D. thesis, and will begin an assistant professorship in the Management Department at West Chester University outside Philadelphia this fall.

Last year, Wharton announced it was shutting down the Public Policy Initiative, which had sponsored Caulfield’s first report, and was laying off its two staffers. One of them was Coopersmith, Caulfield’s longtime collaborator on both reports. But the school hasn’t been shy about promoting the report itself, telling prospective applicants (and their parents) on their website about the Wharton student who “Became An Election Technology Expert—As an Undergrad.”

“It was sort of an adventure,” Caulfied says bashfully. Then he added a stern reminder: “And just my hobby.”