The United States averted a national freight rail strike a few weeks ago, when the Biden administration stepped in to broker a deal between rail companies and their union employees. Rail workers have not yet ratified that agreement, and a strike is still on the table. According to a railway trade group, a shutdown could cost the economy more than $2 billion a day. Among the disasters that could ensue, power plants would lack the coal needed to produce electricity and water treatment plants would lack the chemicals needed to provide clean water. If the companies don’t come to an agreement with rail workers, eventually, grocery shelves would go empty.

Rail workers’ demands are not outrageous — union members have little to no predictability in their schedules and are subjected to draconian attendance policies. But the current standoff is about more, and is the result of a deliberate, half-century-long conversion of the nation’s rail system from a network that could deliver many kinds of goods to market (while also hosting hundreds of passenger train lines), to a fleet of land barges that are good for coal and containers — not much else.

After World War II, railroad companies ceded freight including mail and perishables to semi trucks and instead favored goods like newsprint, chemicals and steel coils, which could sit in rail yards for days. Many freight companies made this change enthusiastically; trucking’s rise gave them a strong reason to stop servicing goods that were financial and logistical pains to ship. So long to the rail cars full of livestock.

Despite shifting to nonperishable goods, railroads did not stay profitable. By 1976, several railroad companies had gone bankrupt, threatening to collapse rail shipping along with them. Sensing disaster, Congress stepped in to create Conrail, a mostly government-owned company to keep service functional in the Northeast and Midwest. Its executives saw austerity as the way to profitability.

Conrail cut costs everywhere it could, abandoning redundant tracks. It took a while, but it worked. The company was making strong profits by the ’80s and was re-privatized in 1987. Conrail may have begun by trimming excesses, but soon railroads across the industry would take those lessons in cost-cutting to an extreme. In pursuit of efficiency, railroad companies across the entire industry adopted the theory of precision scheduled railroading (P.S.R.), developed by the railroad executive E. Hunter Harrison in the 1990s. Before P.S.R., freight shipping meant rail cars full of goods typically crossed the country on multiple trains, getting dropped off and picked up in yards and terminals in a time-consuming process. Harrison, seeing idle train cars and sitting inventory as waste, envisioned a system of shipping freight directly to its destination with consistent schedules for crew and customers. Goods, in theory, could arrive just in time for our lean supply chains.

If only it worked out. Alas, the actual implementation of P.S.R. led neither to precision nor scheduling nor railroading.

When companies implemented P.S.R., they also adopted new technology that allowed for locomotive engines to be placed along the length of a train. Now, instead of engines pulling the train from the front, additional engines in the middle and the back help move even more train cars. Average train length grew around 25 percent from 2008 to 2017, and companies now regularly run trains that are three miles long.

Our infrastructure isn’t built for these “monster trains,” which are now so long that many no longer fit the tracks designed to allow trains to pass one another. These trains are almost always overseen by a crew of just two people, who must walk for miles if a problem is found, in all kinds of weather. The trains are difficult to control, and if weight is unevenly distributed along them, they may break apart or even derail.

Precision schedules imply that trains run on some semblance of a schedule. But monster trains and longer distances often lead to a series of small delays that can easily cascade into much longer ones. This means that when a rail crew’s shift ends, its replacement is often called at odd hours to their station, usually with less than two hours’ notice. Ten years ago, railroading was a middle-class job where workers might not get typical weekends, but they could at least get some equivalent time off. With new attendance policies, conductors and engineers would be disciplined for activities such as visiting the doctor or attending a funeral. Effectively, for these workers, a weekend or an eight-hour workday does not exist.

These problems are rarely highlighted on companies’ accounting metrics. During union contract negotiations, rail companies asserted that their capital investment, not worker’s labor, led to their profits. Union Pacific and B.N.S.F., two of the largest rail companies in the country, posted record figures in 2021. However, simply looking at profits hides a more complicated story.

Nearly two decades ago, companies used to spend around 80 percent of their revenues on running trains and covering operating expenses like payroll, fuel and maintenance. The remaining 20 percent could then be used for stock dividends and buybacks for shareholders. Today that operating ratio is much closer to 60 percent. Since 2010, rail companies have spent $196 billion in stock dividends and buybacks for shareholders. Pursuing these financial goals has actively surrendered railroads’ market share to trucks, delayed trains and angered both unions and customers. It’s not sustainable.

Capping the length of our monster trains should be the first step that the United States takes toward reform, and it should take lessons from other countries as it pursues substantive change. Countries with huge levels of freight — including China, Russia and India — have nationalized railroads that reduce the financial incentives that clog up tracks with the longest trains possible. Switzerland has resisted trucking dominance through a herculean effort that combines upgrades to tracks and stations with centralized management. This benefits railway shippers and passengers, not just the private profit of shareholders.

Rail companies seem set on their self-destructive tendencies, often proposing one-man crews in labor negotiations that would further squeeze workers. Even with a successful negotiation, train yards and mainlines would still be full of long, slow, trudging trains on congested tracks with overworked crews. Through disinvestment, private rail management has shown for decades a disinterest in building a rail system that works for workers and shippers. If they can’t figure out how to run a functioning railroad, maybe it’s time to take it out of their hands.