One of the defining political and economic facts of the first decade of the 21st century in the United States is that the percentage of private sector employees who are in private sector unions is at a record low. The percentage hasn't been this low since the 1920s, before good government statistics on labor union membership were kept and before federal law protected labor unions.
There are two main schools of thought regarding the reason that we've seen this decline. I will offer a third contributing cause.
Probably the most popular theory for the decline of American unions has been to blame a legal environment put into place by Ronald Reagan, symbolized by the firing of America's air traffic controllers in his administration, and linked to a number of specific changes in labor law, most notably the establishment of an employer's right to permanently replace employees that strike. Obviously, this dates from the 1980s and beyond, and this time period has also seen a marked decline in major strikes. But, the union movement was already on the wane when this happened, and why should new laws, which still provide partial protection for unions leave labor worse off than it was when there were no effective laws protecting unions in most states, and actively anti-union laws in many states?
The second most popular theory for the decline of the American union has been to blame the deindustrialization of the American economy in the wake of international trade. Cheap foreign labor in combination with free trade in goods, but not services, and a lax foreign regulatory environment governing both the treatment of workers and pollution controls, it is argued, has made the outsourcing of American manufacturing almost inevitable. Automation of factories has gutted the manufacturing jobs that outsourcing didn't shift. Unions have always dominanted big business in America, because employees have with a couple of notable exceptions, organized individual employers, rather than industries. So, unions have fallen with the industry that they hung their star upon.
Less charitable versions of the second story go further. They blame the pace of outsourcing, in part, on artificially high union wages and inflexible attitudes that hurt the quality and productivity of the American work force. No one disputes that union wages are higher than non-union wages, but foreign wages are lower still. The union movement has persausive statistics to show that productivity is higher in union shops, and identifying who is responsible for poor quality is always difficult.
The second theory has the virtue of going back further, to the 1970s, and of coinciding closely with the three and a half decades in which the bottom 80%-90% of the income distribution has seen stagnant wage growth, while the top 10%-20% (and particularly the very top 1%) has seen stunning improvements in income.
But, this second theory has warts as well. While manufacturing has been in a nearly never ending free fall in the rust belt, manufacturers like Toyota, Honda and General Motor's Saturn division have built new factories in rural areas in the South and in places like Ohio away from urban centers. These new factories have generally speaking, had higher quality than existing U.S. automaker's factories, have not been locked into high rust belt union wage scales, and have even where unions have been established not been marked by the same highly antagonistic union-management relations. In much of the South, only about 5% of private sector workforce is in a union. Unions aren't a natural part of the socio-economic environment, even when they are created for show.
There is another possibility, and it can take us back all the way to the mid-1950s, when the private sector labor movement peaked and started to fall. This is that meritocratic policies, like the G.I. Bill, increased support for state colleges, anti-discrimination laws, and affirmative action, stripped unions of potential organizers or leaders. These policies are all designed to allow bright people who were minorities or who were working or middle class, obtain college educations and managerial or professional jobs. They succeeded. The percentage of people graduating from college has tripled in the last forty years. The percentage of women among those graduates has gone from 40% to 58%, and has increased even more dramatically in professions like law, where only about 5% or less of graduates were women as recently as 1970. The black middle class is much larger than it once was, and laws banning discrimination in real estate, have in due course separated the black middle class from the ghettos.
A 1950s study I read found that the average union leader at that time was on a par with college graduates in vocabulary, a proxy for general socio-economic success about as strong as IQ. I also recall stories of pre-Jim Crow America that describe how the thin ranks of black professionals prior to the Civil Rights movement had to return to their segregated neighborhoods each day, providing positive role models for the neighborhood, and how segregated neighborhoods helped foster some black owned businesses to serve these communities.
Anecdotes side, no one disputes that pre-G.I. bill America denied people who couldn't afford to go to college, or who weren't permitted to go to college, opportunities to advance up the economic ladder in a way unthinkable now. There were a few scholarships for the most brilliant young minds, but they were the exception. Social advancement though young women marrying up was probably more common than social advancement secured with a college education.
This development shunted many men and some women who would have found jobs in factories and become union leaders and organizers into the managerial and professional class, co-opting them and undermining potential spokespesons for class struggle. Running a union is very different stuff from working on the factory floor, and without leadership, a union will wither. The rise of service unions, likewise, can be seen, in part of the return of an American underclass with little chance of advancement in its lifetime, as immigrants, legal and illegal alike, fill low wage jobs to which poor English skills relegate them (as well as the fact that service jobs are harder to outsource).
There is no definitive answer to the question "Who Killed American Labor Unions?", and I certainly don't favor a return to a deliberate policy of keeping down talented working class and minority kids so that they can run unions. But, if you don't consider all possible causes of a problem, and your solution doesn't address the real cause, your solution won't work.