Although I was as dumbfounded on Election Day as the next D.C. bubble-dweller, I did feel that I had one scrap of insight into the working-class anger that helped power Donald Trump’s improbable victory. Last year, I got a taste of what many Americans are coping with in a globalized economy—a globalized economy, more specifically, in which many workers feel voiceless and powerless. A taste was more than enough.

Early last year, my husband, Michael, took a part-time service job with an international airline. He immediately encountered the practice called just-in-time scheduling. The company would distribute shift schedules in advance, but then it would adjust them at will, often with only a couple of days’ notice and without bothering to consult the affected workers. Michael might be told on a Thursday that he would be working the Saturday-afternoon shift, and never mind our plans to be at a wedding that day.

He didn’t have kids to pick up, or classes to attend, or a second job to work, or any of the other commitments that make just-in-time scheduling an intolerable burden for many workers. Even so, the disruption to our lives soon became unbearable. The unpredictable hours were problematic, but even more demoralizing was the sense of being treated like a machine part. Most frustrating of all was that the company’s random, dysfunctional scheduling practices could easily have been improved by worker input, had there been channels for worker input; but Michael’s supervisor seemed just as helpless as he was, and unsurprisingly, ground-based workers like Michael had no union representation.

I say “unsurprisingly” because private-sector unions are close to extinct. In the 1950s, more than one in three private-sector workers belonged to a union; today, unionization is down to 6 percent of the private-sector workforce, which is lower than it was a century ago, before the modern labor movement took off. And the rate continues to drop.

Even before getting my small taste of what working-class Americans are experiencing in horse doctors’ doses, I had come to see the decline of unions as one of the country’s most pressing problems—and at least as much a social and political problem as an economic one. Old-style, mid-20th-century industrial unions had their flaws, unquestionably. But when unions work as they should, they serve important social functions. They can smooth the jagged edges of globalization by giving workers bargaining power. They are associated with lower income inequality, as the accompanying graph shows. Perhaps most important, they offer workers a way to be heard. “Unions provide a mediating function,” Matthew Dimick, a labor-law expert at suny Buffalo’s law school, told me. “Their social-capital function creates ties that reduce anomie and the sense of being abandoned and forgotten.” No other social institution, or at least none yet discovered, can serve that mediating function for workers.

All workers do not suffer equally from the decline of unions: In today’s fragmented, hypercompetitive, and globalized workplace, high-powered professionals enjoy more autonomy and respect than ever. Less educated workers, by contrast, have lost agency and, in many cases, dignity. Edward Luce of the Financial Times puts the problem well in his new book, The Retreat of Western Liberalism: “In survey after survey, the biggest employee complaint is being treated with a lack of respect. Whether they work in an Amazon warehouse, serve fast food, or sit in a … customer-service cubicle, they feel diminished by how they are treated.” That has implications not just for the well-being of workers, but for the health of capitalism and even of democracy.

In America, the modern conservative movement was founded on anticommunism and antiunionism. Senator Barry Goldwater (“Mr. Conservative”) built his career bashing unions. President Ronald Reagan, although a former union leader himself, made his bones by breaking the air-traffic controllers’ union. Just this past February, Republicans succeeded in their long push for a right-to-work law in Missouri. But the conservative war on unions is beginning to look like a Faustian bargain. If 2016 taught us anything, it was that miserable workers are angry voters, and angry voters are more than capable of lashing out against trade, immigration, free markets, and for that matter liberal democracy itself.

This is not to say the old style of American industrial unions will come back, or should. The mid-20th-century enterprise model, as it was called, relied on confrontational tactics to organize particular companies or factories. That may have succeeded in an era of oligopolistic, locally rooted corporations. However, in an era when even a slight increase in labor costs at a North Carolina factory sends jobs to China, organizing just a single company can boomerang against workers and management alike.

Fortunately, other models have emerged elsewhere in the world, models that can benefit both companies and labor. A well-known example, popular in Europe, is the so-called works council, which gives workers a voice in company affairs without triggering the fraught, complex process of creating a formal union. In Germany, unions can organize entire sectors, rather than particular companies, giving employers and workers incentives to cooperate in ways that improve industries’ competitive position.

Even more intriguing is the Ghent system, successful in Denmark and Sweden, under which unions administer government-funded unemployment benefits. Providing that safety net helps unions to shift their focus from protecting individual jobs to maintaining workers’ overall income security; this in turn allows employers more flexibility in hiring and firing.

Emin Dinlersoz and Jeremy Greenwood, “The Rise and Fall of Unions in the U.S.” (University of Pennsylvania ScholarlyCommons)

In principle, unions could offer skills training that qualifies workers for better jobs, a role that individual employers are not always eager to fill (they might be training employees to go work somewhere else). Unions could act as employment agencies, matching workers with jobs. They could offer and manage health-insurance plans and benefits programs. They could administer wage insurance, thereby helping workers through disruptive job transitions.

I could go on, but the point is not to endorse those or other specific ideas; it’s to show that there is no shortage of ways to modernize unions. Unfortunately, in America in 2017, we don’t know how a truly modern union would look, because it is mostly illegal to find out.

On a spring morning last year, two men from opposite ends of the ideological spectrum met at a Manhattan diner for brunch and, somewhat to their own surprise, discovered they agreed on a way to address that problem. One was Eli Lehrer, who co-founded and runs the R Street Institute, a free-market-oriented, Republican-leaning think tank in Washington. Lehrer believes the time has come for the American right to reconsider its decades-long war on unions. Their collapse, he says, has fueled the growth of government and of the welfare state, which has stepped in to regulate workplaces and provide job security as unions have died out.

His unlikely dining companion was Andy Stern. As the president of the Service Employees International Union from 1996 to 2010, Stern had become the labor-movement equivalent of a rock star by more than doubling the union’s membership. Unions, he thinks, cannot survive unless they innovate and change, but laws intended to protect and preserve them get in the way. “Anytime anybody gets creative, these laws stop us,” he said when I spoke with him and Lehrer recently.

The laws he refers to are hard-won federal enactments dating back to the early decades of the last century. The most foundational is the 1935 National Labor Relations Act, which has not had a major revision since Dwight Eisenhower was president. The law, along with a complex superstructure of regulations and court rulings interpreting it, sets out how and when private-sector workers can organize, mandates how companies recognize and deal with unions, dictates the bundle of services and benefits unions can provide, and precludes many other forms of employer-employee interaction.

Rigid and archaic as that structure may be, to many on the American left, it is sacred writ. They regard it as a form of life support for unionization. Stern and Lehrer, though, believe the status quo has become something more like a death grip. Imagine that American retailers were locked by law into doing things the same way they did them in the 1950s, and you can see the problem.

Stern and Lehrer joined forces to argue their case in the journal National Affairs. “The fundamental federal rules governing employer-worker relations were written for a different era,” they explain. For example: “Right now, union officials can face criminal charges if they sell anything to employers—even services like a health plan that employers might be willing to spend good money to buy in a free and open market.” In creating legal barriers to change, labor law’s ossification has stifled fresh thinking. David Rolf, the president of Service Employees International Union 775 (which represents home-health-care workers in Washington State and Montana), told me, “If you’re a union leader—if you’re one of hundreds of people like me around the country—the reality is that your range of motion is very constrained. We’re so conditioned to the parameters of our model that most of us can’t even think about doing different things.”

Efforts to revise labor law at the federal level (most recently, during President Barack Obama’s first term) have gone nowhere. That seems unlikely to change anytime soon. So Stern and Lehrer propose a workaround. Why not give states the authority to grant labor-law waivers that would allow experimentation? If an employer and a union came up with an interesting model that met certain guidelines, they could try it. In education and health care, state waivers have sparked all kinds of experimentation. They might do the same for unions.

The Stern-Lehrer waiver idea is a no-brainer if we want to address the deeper causes of the malaise and distemper afflicting America’s lower-middle class. Although income stagnation is certainly one culprit, another, perhaps still more important, is the decline of the civic organizations and social institutions that help people feel connected and efficacious. Service fraternities, volunteer clubs, youth groups, churches, political parties, widespread military service, unions, and the rest—in their prime, all fostered social interaction and face-to-face collaboration, cultivating a sense of social cohesion even when times were much tougher than they are today.

Among those institutions, none matter more than unions. My late uncle, a blue-collar garment worker in New York City, was a devoted union member, and I still recall how his union brotherhood gave him a sense of solidarity and dignity that no paycheck could have provided. Facebook and Instagram, it turns out, cannot come close to replicating that function.

Pondering the working-class grievances that helped elevate Donald Trump to the Oval Office in 2016, I often ask myself: How different might the political climate have been if 25 percent of the private sector were unionized, as was the case in the early 1970s? If more working-class Americans felt listened to and represented? If modernized unions could buffer economic shocks and improve productivity? The decline of the business model of old-style industrial unions may have been economically inevitable, but the lack of any new model to replace it has been socially calamitous. Unions will not be easy to fix, but allowing them to innovate would be a first step, and possibly also a last chance.