Today’s column, which used the Wisconsin recall as an opportunity to think through a post-labor political system, has attracted some strange readings. I’m not arguing that unions don’t provide valuable services outside Washington. Their key role, of course, is in individual workplaces. Nor is it that they have, or recently had, too much power. That depends on your policy preferences. Rather, the question is whether their decline is irreversible and, if so, what’s to stop corporations from taking the political system over wholesale?

But perhaps it’s worth saying a bit more about why I’m skeptical that labor will turn itself around.

I don’t think anyone disputes that it’s harder to unionize in the modern economy. This chart shows unionization rates in the United States, the United Kingdom, Germany, France, Japan and Canada since 1960. It also shows average unionization across the 34 countries in the Organization for Economic Cooperation and Development — that’s the highlighted red line with the gray dots* — most of which are much friendlier to labor unions than we are. The decline is sharpest in the United States, but evident in all the included countries, and in the aggregate of all OECD countries.

(Source: OECD)

In the United States, the conversation over how to reverse this decline has focused heavily on political change. In recent years, the main idea was the Employee Free Choice Act, which would’ve permitted unionization through “card check” elections. Most every House Democrat and every major Democratic presidential candidate signed onto that legislation in advance of the 2008 election. Then Democrats took power, and did so with the largest majorities since the 1970s. But EFCA never came anywhere near passing.

That follows the failure to pass labor-law reform during the Carter administration, and the failure to even get it on the agenda during the Clinton administration. It seemed, for a moment, that Scott Walker’s efforts to restrict collective bargaining in Wisconsin might create sufficient countermobilization among union members and their allies that it actually would lead to a resurgence in their political power. But Walker survived the recall attempt, and emerged, if anything, strengthened.

Given this history, and the continuing decline of union membership and thus power, I don’t see a plausible scenario in which unions muster the political strength to substantially change the legislative environment in which organizing happens. And that’s before you even ask whether card check — much less weaker reforms — would actually be enough to reverse their decline. Canada, which does have card check elections, has seen union density fall by almost 10 points since the mid-’80s.

The defeat in Wisconsin is, of course, a political defeat, and so that’s the dimension of labor power I was focusing on in this morning’s article. Unions have traditionally been a counterbalance to corporate influence in politics. They’ve expended enormous energy and capital pushing for policies such as universal health care and paid sick leave that arguably benefit non-unionized employees more than unionized employees. And now, simultaneous with their decline, corporations are, in part due to Citizens United, becoming even more powerful politically. This creates a dangerous imbalance in the political system as both parties grow more and more dependent on corporate funding.

Which brings me to my overall point: If you don’t want a political system dominated by corporate interests, and you don’t believe there’s a plausible path through with unions returning to their former density, the alternative is to try and reform the political system such that it’s more resistant to special interests of all kinds. If it was more difficult for corporations to spend unlimited amounts of anonymous money to win elections, if candidates were able to fund competitive campaigns using small donations and/or public money, if the rules restricting lobbying were considerably stricter, that would reduce the power corporations have over political outcomes going forward. And I think those kinds of reforms are a lot likelier than a resurgence in organized labor.

That doesn’t say anything about how to replace the role unions play in individual workplaces. But that’s in part because I’m just not sure what can replace unions in most workplaces, much less what should.

Update: An earlier version of this post had mislabeled the “OECD” line in the graph. It’s since been fixed.