Like many Americans, I've been wary of labour unions.
Full-time union stagehands at Carnegie Hall earning more than US$400,000 (S$544,000) a year? A union hailing its defence of a New York teacher who smelled of alcohol and passed out in class, with even the principal unable to rouse her? A police union in New York City that has a tantrum and goes on virtual strike?
More broadly, I disdained unions as bringing corruption, nepotism and rigid work rules to the labour market, impeding the economic growth that ultimately makes a country strong.
I was wrong. The abuses are real. But, as unions wane in American life, it is also increasingly clear that they were doing a lot of good in sustaining middle-class life - especially the private sector unions that are now dwindling.
Most studies suggest that about one-fifth of the increase in economic inequality in America among men in recent decades is the result of the decline in unions. It may be more: A study in the American Sociological Review, using the broadest methodology, estimates the decline of unions may account for one-third of the rise of inequality among men.
"To understand the rising inequality, you have to understand the devastation in the labour movement," says labour expert Jake Rosenfeld at the University of Washington and the author of What Unions No Longer Do. Take construction workers. A full-time construction worker earns about US$10,000 less per year now than in 1973, in today's dollars, according to Mr Rosenfeld. One reason is probably that the proportion who are unionised has fallen in that period from more than 40 per cent to just 14 per cent.
"All the focus on labour's flaws can distract us from the bigger picture," Mr Rosenfeld writes. "For generations now the labour movement has stood as the most prominent and effective voice for economic justice." I'm as appalled as anyone by silly work rules and US$400,000 stagehands, or teachers' unions shielding the incompetent. But unions also lobby for programmes like universal pre-kindergarten that help create broad-based prosperity. They are pushing for a higher national minimum wage, even though that would directly benefit mostly non-unionised workers.
I've also changed my mind because, in recent years, the worst abuses by far haven't been in the union shop but in the corporate suite. One of the things you learn as a journalist is that when there's no accountability, we humans are capable of tremendous avarice and venality. That's true of union bosses - and of corporate tycoons. Unions, even flawed ones, can provide checks and balances for flawed corporations.
Many Americans think unions drag down the economy overall, but scholars disagree. American auto unions are often mentioned, but Germany's car workers have a strong union, and so do Toyota's in Japan and Kia's in South Korea.
In Germany, the average car worker earns about US$67 per hour in salary and benefits, compared with US$34 in the United States. Yet Germany's car companies in 2010 produced more than twice as many vehicles as American car companies did, and they were highly profitable. It's too glib to say the problem in the US sector was just unions.
Or look at American history. The peak years for unions were the 1940s and 1950s, which were also some of the fastest-growing years for the US - and with broadly shared prosperity. Historically, the periods when union membership was highest were those when inequality was least.
Professor Richard Freeman, a Harvard labour expert, notes that unions sometimes bring important benefits to industry: They can improve morale, reduce turnover and provide a channel to suggest productivity improvements.
Experts disagree about how this all balances out, but it's clear that it is not a major drag.
"If you're looking for big negatives, everybody knows they don't exist," Prof Freeman says.
Nobel Prize-winning economist Joseph Stiglitz notes in his book The Price Of Inequality that when unions were strong in America, productivity and real hourly compensation moved together in manufacturing. But after 1980 (and especially after 2000) the link seemed to break and real wages stagnated.
It may be that as unions weakened, executives sometimes grabbed the gains from productivity. Perhaps that helps explain why chief executives at big companies earned, on average, 20 times as much as the typical worker in 1965, and 296 times as much in 2013, according to the Economic Policy Institute.
Professor Lawrence Katz, a Harvard labour economist, raises concerns about some aspects of public sector unions, but says that in the private sector (where only 7 per cent of workers are unionised): "I think we've gone too far in de-unionisation." He's right. This isn't something you often hear a columnist say but I'll say it again: I was wrong. At least in the private sector, we should strengthen unions, not try to eviscerate them.
NEW YORK TIMES