I have no problem with articles that are coauthored by two contributors. Often, that means you have the methods jock and the ideas guy collaborating. And you can easily tell who's done what. You can also tell that the article would never have been done unless they had collaborated, since neither could do the project on his own. That means the collaboration has been fruitful for them and for the discipline.
What's harder to take are the articles with four or five coauthors. You see some scholars whose CVs include a dozen articles with a long chain of the same collaborators. These scholars are well networked and are contributing something. But it looks more like factory production and hoop jumping than scholarship.
This sounds really familiar. I'm in one of these. I don't know if this is how it works for other people, but in my experience, it all comes down to senior people exploiting juniors. There are senior people who attach themselves to (funded) projects and don't let go, or they give you some data you wouldn't otherwise have access to. Then they provide zero intellectual contribution and can't even be bothered to do trivial work, but they're guaranteed a line in the author list forever and ever.
I thought that too (as I was part of such an arrangement for one paper as an RA), which is why I was fine at first with interviewing grad students who were third authors on papers. But then while on those interviews, our hiring committees discovered strange things - like they really didn't have a solid grasp of the theory presented in the articles they coauthored, or their diss project (almost invariably not tied to the articles they coauthored) were uninspiring at best (and entirely derivative at worst). Further discussion sometimes revealed other big concerns, such as poor modeling choices or clear problems in understanding research design.
In short, quite the opposite was happening from your (and my) previous experience: the advisors and professors of these students were "carrying" them on papers only to check boxes for employability. What's more, watching their trajectories after we didn't hire them, they kept carrying them for years, even as far as midpoint and tenure review. These students weren't developing into individual thinkers; they just kept being third (or sometimes second or first) coauthors with their trainers, sometimes branching out into being third or fourth coauthors with other groups like the ones I've described.
To be sure, we typically don't hire from too 10 places. But this makes us pretty much like every other school in the world. The incentive structure had completely changed the training practices for these folks, in a way that only made them attractive candidates if the hiring departments wanted a perpetual RA for someone else, as Petronel describes above. And we don't want that, obviously.
I'll be honest: my department internally ranks students and uses these rankings for funding, TA choices, etc. You get a lot of points for publications. The students often with publications are those who were picked up as RAs for egap projects or with faculty involved in egap that are typically completely unrelated to their diss work. And there is a massive quality difference in their coauthored vs. solo work often. Is it fair that students in the egap clique get first pick at everything for being listed as a coauthor? Even first author is often unearned. The senior profs list them first for job market purposes.